CN119042929A

CN119042929A - Intelligent refrigeration equipment control system and method for high-temperature environment

Info

Publication number: CN119042929A
Application number: CN202411428084.9A
Authority: CN
Inventors: 王纬; 张涛; 宗正论; 李森; 巫云; 苏比依·艾尔肯
Original assignee: Karamay Fucheng Natural Gas Co ltd
Current assignee: Karamay Fucheng Natural Gas Co ltd
Priority date: 2024-10-14
Filing date: 2024-10-14
Publication date: 2024-11-29

Abstract

The application discloses an intelligent refrigeration equipment control system and method for a high-temperature environment, which adopt a data analysis and processing technology based on deep learning to carry out data normalization and time sequence association on working state parameters, and carry out local time sequence feature extraction and full time domain significant aggregation on temperature data of high-temperature natural gas, so as to control the rotation speed value of a compressor of a monitored refrigeration equipment at the next time point in a self-adaptive manner according to main component significant matching fusion features between the working state time sequence features of the refrigeration equipment and the aggregation features of the high-temperature natural gas temperature data, and correspondingly adjust the rotation speed value. In this way, the system is able to collect multiple parameters, providing a more comprehensive view of the system's state. Meanwhile, the change of external conditions can be responded immediately, the rotating speed of the compressor can be dynamically adjusted, so that the equipment can maintain the optimal running state under different working conditions, the refrigerating efficiency is improved, and the control intellectualization of the refrigerating equipment is realized.

Description

Intelligent refrigeration equipment control system and method for high-temperature environment

Technical Field

The present application relates to the field of intelligent control, and more particularly, to an intelligent refrigeration appliance control system and method for a high temperature environment.

Background

The MRC (Mixed REFRIGERANT CYCLE) refrigeration technology is a technology for reducing the gas temperature by utilizing the molecular sieve adsorption principle, is commonly used in the natural gas liquefaction process, and can effectively reduce the temperature of associated gas, thereby separating light hydrocarbon components and improving the recovery rate of the light hydrocarbon. The device comprises a station separation sled, a compressor, a molecular sieve sled, a cold box, a low-temperature separator, a reabsorption tower, a deethanizer, a liquefied gas tower and a central control system, wherein the station separation sled is used for carrying out liquid removal treatment on associated gas to obtain liquid-removed associated gas, the compressor is used for adding the liquid-removed associated gas to obtain pressurized associated gas, the molecular sieve sled is used for carrying out dehydration treatment on the pressurized associated gas to obtain dehydrated associated gas, the dehydrated associated gas is precooled through the cold box and then subjected to gas-liquid separation through the low-temperature separator and the reabsorption tower to obtain natural gas and hydrocarbon liquid parts, the deethanizer is used for rectifying the hydrocarbon liquid parts, and liquid obtained from the bottom of the deethanizer enters the liquefied gas tower to stabilize light hydrocarbons to obtain light hydrocarbons and liquefied petroleum gas.

It should be understood that natural gas is generally in a higher temperature state in the processing and is at the heart of recovering high-value Liquefied Petroleum Gas (LPG) and light hydrocarbons therein to improve comprehensive utilization efficiency and economic value of resources and reduce environmental pollution. The main aim of the light hydrocarbon recovery device is to improve the recovery rate of the light hydrocarbon, reduce the oil gas loss and simultaneously ensure that the natural gas meets the transportation standard and the quality requirement of the liquefied petroleum gas. To achieve this objective, a mixed refrigerant refrigeration plant (i.e., the cold box is a mixed refrigerant refrigeration plant) is generally employed that is capable of optimizing operating parameters based on the specific conditions of the feed gas to thereby effectively enhance the recovery of C3 (propane) and C4 (butane). Therefore, the mixed refrigerant refrigeration equipment plays a critical role in improving the overall recovery rate as a key equipment.

However, the conventional technology generally adopts preset fixed parameters when controlling the refrigeration equipment, and the method is simple and easy to implement, but has obvious defects. Specifically, due to lack of real-time monitoring and dynamic adjustment mechanisms, when external conditions change, such as pressure or temperature fluctuation of raw gas, the system cannot respond in time, so that refrigeration efficiency is reduced, and even product quality may be affected. The fixed parameter control mode ignores dynamic changes under actual working conditions, so that the equipment operation often deviates from an optimal state, energy is wasted, and abrasion of the equipment is possibly increased. Furthermore, conventional control typically focuses on only a single or a few parameters, often neglecting interdependencies and overall dynamic behavior between parameters in the system, which limits the ability to fully understand the system, thereby affecting the optimization of the control strategy.

Accordingly, an optimized refrigeration appliance control system for use in high temperature environments is desired.

Disclosure of Invention

The present application has been made to solve the above-mentioned technical problems. The embodiment of the application provides an intelligent refrigeration equipment control system and method for a high-temperature environment, which are characterized in that a sensor group is used for collecting a time sequence of working state parameters (inlet temperature, inlet pressure, outlet temperature and outlet pressure) of a refrigeration equipment to be monitored, a time sequence of temperature data of high-temperature natural gas is obtained, data normalization and time sequence association are carried out on the working state parameters by adopting a data analysis and processing technology based on deep learning, local time sequence feature extraction and full time sequence feature significant aggregation are carried out on the temperature data of the high-temperature natural gas, so that the rotation speed value of a compressor of the refrigeration equipment to be monitored at the next time point is adaptively controlled according to a main component significant matching fusion feature between the working state time sequence feature of the refrigeration equipment and the aggregation feature of the temperature data of the high-temperature natural gas, and the rotation speed value of the compressor is correspondingly regulated. In this way, the system is able to collect multiple parameters, providing a more comprehensive view of the system's state. Meanwhile, the change of external conditions can be responded immediately, the rotating speed of the compressor can be dynamically adjusted, so that the equipment can maintain the optimal running state under different working conditions, the refrigerating efficiency is improved, and the control intellectualization of the refrigerating equipment is realized.

According to one aspect of the application, an intelligent refrigeration equipment control system for a high-temperature environment is provided, which comprises a state parameter acquisition module, a control module and a control module, wherein the state parameter acquisition module is used for acquiring a time sequence of operating state parameters of monitored refrigeration equipment acquired by a sensor group, and the operating state parameters comprise an inlet temperature, an inlet pressure, an outlet temperature and an outlet pressure; the system comprises a temperature data acquisition module for acquiring a time sequence of temperature data of high-temperature natural gas, a temperature local time sequence feature extraction module for enabling the time sequence of the temperature data of the high-temperature natural gas to pass through a 1D-CNN-based sequence encoder to obtain a time sequence of high-temperature natural gas temperature local time sequence feature vectors, a high-temperature natural gas temperature time sequence aggregation module for enabling the time sequence of the high-temperature natural gas temperature local time sequence feature vectors to pass through a node significance attenuation-based temperature local time sequence semantic aggregation network to obtain high-temperature natural gas temperature time sequence node significance aggregation expression vectors, a data regulation module for enabling the time sequence of the working state parameters to be data regulated according to time dimension and working state parameter sample dimension to obtain a time sequence of a working state parameter matrix, a state parameter time sequence correlation feature extraction module for enabling the time sequence of the working state parameter matrix to pass through a working state parameter time sequence correlation feature extractor comprising a cavity convolutional neural network and a recurrent neural network to obtain working state time sequence correlation feature vectors, a temperature time sequence-working state significance fusion module, the system comprises a high-temperature natural gas temperature time sequence node semantic significant aggregation representation vector, a control result generation module, a rotation speed adjustment module and a frequency converter, wherein the high-temperature natural gas temperature time sequence node semantic significant aggregation representation vector and the working state time sequence association feature vector are subjected to significant fusion network optimization matching based on feature principal components to obtain a temperature time sequence state-working state sparse significant matching fusion representation vector, the control result generation module is used for obtaining a control result based on the temperature time sequence state-working state sparse significant matching fusion representation vector, and the rotation speed adjustment module is used for adjusting the rotation speed of a compressor of the monitored refrigeration equipment based on the control result.

According to another aspect of the application, an intelligent refrigeration equipment control method for a high-temperature environment is provided, which comprises the steps of obtaining a time sequence of operating state parameters of monitored refrigeration equipment collected by a sensor group, wherein the operating state parameters comprise inlet temperature, inlet pressure, outlet temperature and outlet pressure, obtaining a time sequence of temperature data of high-temperature natural gas, obtaining a time sequence of high-temperature natural gas temperature local time sequence feature vectors through a 1D-CNN-based sequence encoder, obtaining a high-temperature natural gas temperature local time sequence feature vector through a node significance attenuation-based temperature local time sequence semantic aggregation network, obtaining a high-temperature natural gas temperature time sequence node semantic significance aggregation expression vector, obtaining a time sequence of an operating state parameter matrix through data normalization according to a time dimension and an operating state parameter sample dimension, obtaining an operating state time sequence correlation feature vector through a cavity convolutional neural network and a neural network, obtaining an operating state time sequence correlation feature vector through the time sequence feature extractor, obtaining an operating state correlation feature vector through a node significance attenuation-based temperature local time sequence semantic aggregation network, obtaining a high-natural gas temperature time sequence node significance node semantic significance fusion expression vector based on a node significance-sparse state control time sequence feature vector, obtaining a significant state-phase-sparse state control feature vector based on a significant state-phase-feature fusion-significant control feature vector, and adjusting the rotating speed of the compressor of the monitored refrigeration equipment through a frequency converter.

Compared with the prior art, the intelligent refrigeration equipment control system and method for the high-temperature environment are characterized in that a sensor group is used for collecting a time sequence of working state parameters (inlet temperature, inlet pressure, outlet temperature and outlet pressure) of the refrigeration equipment to be monitored, a time sequence of temperature data of high-temperature natural gas is obtained, data normalization and time sequence association are carried out on the working state parameters by adopting a data analysis and processing technology based on deep learning, local time sequence feature extraction and full-time-domain significant aggregation are carried out on the temperature data of the high-temperature natural gas, so that the compressor rotation speed value of the refrigeration equipment to be monitored at the next time point is adaptively controlled according to a main component significant matching fusion feature between the working state time sequence feature of the refrigeration equipment and the aggregation feature of the temperature data of the high-temperature natural gas, and the compressor rotation speed value is correspondingly adjusted. In this way, the system is able to collect multiple parameters, providing a more comprehensive view of the system's state. Meanwhile, the change of external conditions can be responded immediately, the rotating speed of the compressor can be dynamically adjusted, so that the equipment can maintain the optimal running state under different working conditions, the refrigerating efficiency is improved, and the control intellectualization of the refrigerating equipment is realized.

Drawings

The above and other objects, features and advantages of the present application will become more apparent by describing embodiments of the present application in more detail with reference to the attached drawings. The accompanying drawings are included to provide a further understanding of embodiments of the application and are incorporated in and constitute a part of this specification, illustrate the application and together with the embodiments of the application, and not constitute a limitation to the application. In the drawings, like reference numerals generally refer to like parts or steps.

Fig. 1 is a block diagram of an intelligent refrigeration appliance control system for a high temperature environment in accordance with an embodiment of the present application.

Fig. 2 is a data flow schematic diagram of an intelligent refrigeration appliance control system for a high temperature environment in accordance with an embodiment of the present application.

Fig. 3 is a flowchart of a method of controlling an intelligent refrigeration appliance for a high temperature environment according to an embodiment of the present application.

Detailed Description

Hereinafter, exemplary embodiments according to the present application will be described in detail with reference to the accompanying drawings. It should be apparent that the described embodiments are only some embodiments of the present application and not all embodiments of the present application, and it should be understood that the present application is not limited by the example embodiments described herein.

As used in the specification and in the claims, the terms "a," "an," "the," and/or "the" are not specific to a singular, but may include a plurality, unless the context clearly dictates otherwise. In general, the terms "comprises" and "comprising" merely indicate that the steps and elements are explicitly identified, and they do not constitute an exclusive list, as other steps or elements may be included in a method or apparatus.

Although the present application makes various references to certain modules in a system according to embodiments of the present application, any number of different modules may be used and run on a user terminal and/or server. The modules are merely illustrative, and different aspects of the systems and methods may use different modules.

A flowchart is used in the present application to describe the operations performed by a system according to embodiments of the present application. It should be understood that the preceding or following operations are not necessarily performed in order precisely. Rather, the various steps may be processed in reverse order or simultaneously, as desired. Also, other operations may be added to or removed from these processes.

The traditional technology generally adopts preset fixed parameters when controlling the refrigeration equipment, and the method is simple and easy to operate, but has obvious defects. Specifically, due to lack of real-time monitoring and dynamic adjustment mechanisms, when external conditions change, such as pressure or temperature fluctuation of raw gas, the system cannot respond in time, so that refrigeration efficiency is reduced, and even product quality may be affected. The fixed parameter control mode ignores dynamic changes under actual working conditions, so that the equipment operation often deviates from an optimal state, energy is wasted, and abrasion of the equipment is possibly increased. Furthermore, conventional control typically focuses on only a single or a few parameters, often neglecting interdependencies and overall dynamic behavior between parameters in the system, which limits the ability to fully understand the system, thereby affecting the optimization of the control strategy. Accordingly, an optimized refrigeration appliance control system for use in high temperature environments is desired.

In the technical scheme of the application, an intelligent refrigeration equipment control system for a high-temperature environment is provided. Fig. 1 is a block diagram of an intelligent refrigeration appliance control system for a high temperature environment in accordance with an embodiment of the present application. Fig. 2 is a data flow schematic diagram of an intelligent refrigeration appliance control system for a high temperature environment in accordance with an embodiment of the present application. As shown in fig. 1 and 2, an intelligent refrigeration equipment control system 300 for a high-temperature environment according to an embodiment of the present application includes a state parameter acquisition module 310 for acquiring a time sequence of operating state parameters of a monitored refrigeration equipment acquired by a sensor group, the operating state parameters including an inlet temperature, an inlet pressure, an outlet temperature and an outlet pressure, a temperature data acquisition module 320 for acquiring a time sequence of temperature data of high-temperature natural gas, a temperature local time sequence feature extraction module 330 for performing data normalization on the time sequence of temperature data of the high-temperature natural gas by a 1D-CNN-based sequence encoder to obtain a time sequence of high-temperature natural gas temperature local time sequence feature vectors, a high-temperature natural gas temperature time sequence aggregation module 340 for performing data normalization on the time sequence of operating state parameters by a node significance decay-based temperature local time sequence semantic aggregation network to obtain a high-temperature natural gas temperature time sequence node semantic significance aggregation representation vector, a data normalization module 350 for performing data normalization on the time sequence of the operating state parameters by a time dimension and an operating state parameter sample dimension to obtain a time sequence of operating state parameter matrix, a time sequence feature extraction module 360 for performing a time sequence feature matrix correlation by a node significance decay time sequence neural state feature fusion, a time sequence feature extraction module for performing a time sequence feature correlation by a time sequence feature correlation with the time sequence feature of the operating state feature vector, the system comprises a high-temperature natural gas temperature time sequence node semantic significant aggregation representation vector, a working state time sequence association feature vector, a control result generation module 380 and a rotation speed adjustment module 390, wherein the high-temperature natural gas temperature time sequence node semantic significant aggregation representation vector and the working state time sequence association feature vector are subjected to a significant fusion network based on feature principal component fine granularity optimization matching to obtain a temperature time sequence state-working state sparse significant matching fusion representation vector, the control result generation module 380 is used for obtaining a control result based on the temperature time sequence state-working state sparse significant matching fusion representation vector, and the rotation speed adjustment module 390 is used for adjusting the rotation speed of a compressor of the monitored refrigeration equipment through a frequency converter based on the control result.

In particular, the state parameter acquiring module 310 and the temperature data acquiring module 320 are configured to acquire a time sequence of operating state parameters of the monitored refrigeration device acquired by the sensor group, where the operating state parameters include an inlet temperature, an inlet pressure, an outlet temperature, and an outlet pressure, and acquire a time sequence of temperature data of the high-temperature natural gas. It should be appreciated that by significantly matching the fusion feature of the principal components between the operational state timing feature of the refrigeration equipment and the aggregate feature of the high temperature natural gas temperature data, the compressor speed value of the monitored refrigeration equipment at the next point in time can be adaptively controlled and adjusted accordingly. Therefore, the rotating speed of the compressor can be dynamically adjusted in real time in response to the change of external conditions, so that the equipment can maintain the optimal running state under different working conditions, the refrigerating efficiency is improved, and the control intellectualization of the refrigerating equipment is realized. Thus, first, a time series of operating state parameters of the monitored refrigeration equipment, including inlet temperature, inlet pressure, outlet temperature, and outlet pressure, acquired by the sensor group is acquired, and a time series of temperature data of the high temperature natural gas is acquired.

In particular, the temperature local time sequence feature extraction module 330 is configured to pass the time sequence of the temperature data of the high-temperature natural gas through a 1D-CNN-based sequence encoder to obtain a time sequence of high-temperature natural gas temperature local time sequence feature vectors. In the technical scheme of the application, the time sequence of the temperature data of the high-temperature natural gas is processed by a sequence encoder based on the 1D-CNN to obtain the time sequence of the local time sequence feature vector of the temperature of the high-temperature natural gas, so that the specific mode of temperature change such as fluctuation, peak value or abnormal behavior can be identified. It is worth mentioning that the one-dimensional convolutional neural network (1D-CNN) is a deep learning model for processing one-dimensional data with time sequence correlation. It extracts features by applying convolution operations on one-dimensional space of the input data.

In particular, the high-temperature natural gas temperature time sequence aggregation module 340 is configured to obtain a high-temperature natural gas temperature time sequence node semantic significant aggregation representation vector by passing the time sequence of the high-temperature natural gas temperature local time sequence feature vector through a temperature local time sequence semantic aggregation network based on node significant attenuation. In consideration of the fact that different time points in the time sequence of the high-temperature natural gas temperature local time sequence feature vector have different influences and correlations between the current time points, in order to identify and strengthen key nodes in temperature data, namely, the most important parts for understanding the whole time sequence change, so as to highlight the whole trend and mode of the natural gas temperature change, in the technical scheme of the application, the time sequence of the high-temperature natural gas temperature local time sequence feature vector is subjected to a temperature local time sequence semantic aggregation network based on node significance attenuation to obtain a high-temperature natural gas temperature time sequence node semantic significance aggregation expression vector. That is, the temperature local time sequence semantic aggregation network based on node significance attenuation is an innovative extraction and aggregation feature model based on significant feature analysis.

In the embodiment of the application, the time sequence of the high-temperature natural gas temperature local time sequence feature vector is obtained by a temperature local time sequence semantic aggregation network based on node significance attenuation to obtain a high-temperature natural gas temperature time sequence node semantic significance aggregation expression vector, which comprises the steps of firstly, calculating feature significance description factors of each high-temperature natural gas temperature local time sequence feature vector in the time sequence of the high-temperature natural gas temperature local time sequence feature vector, wherein the feature significance description factors relate to the mean value and the variance of each high-temperature natural gas temperature local time sequence feature vector, particularly, the feature significance factors are calculated based on the variance and the variance of each high-temperature natural gas temperature local time sequence feature vector, the mean value provides the average state of the natural gas temperature in a period of time, the variance represents the degree of dispersion among the temperatures, and more key features can be highlighted by calculating the feature significance factors. And then, constructing a characteristic significance attenuation factor of each high-temperature natural gas temperature local time sequence characteristic vector based on the distance span between each high-temperature natural gas temperature local time sequence characteristic vector and the current high-temperature natural gas temperature local time sequence characteristic vector in the time queue of the high-temperature natural gas temperature local time sequence characteristic vector, so that the model dynamically adjusts the weight of each characteristic according to the distance between the characteristic vectors, and endows higher weight to the characteristic with a closer distance from the current node. And then modulating the characteristic saliency description factors of the high-temperature natural gas temperature local time sequence feature vectors based on the characteristic saliency attenuation factors of the high-temperature natural gas temperature local time sequence feature vectors to obtain a sequence of the high-temperature natural gas temperature local time sequence feature saliency attenuation description factors, reflecting and highlighting the features consistent with the current features in such a way, and reducing the features with longer distance or larger difference to obtain the sequence of the high-temperature natural gas temperature local time sequence feature saliency attenuation description factors. And inputting the sequence of the high-temperature natural gas temperature local time sequence characteristic significance attenuation describing factors into a gating mask module to obtain a sequence of high-temperature natural gas temperature local time sequence characteristic significance attenuation weighting factors, namely, ensuring that each characteristic has equal importance in the model through normalization and gating masking processing, and dynamically selecting or inhibiting certain characteristics through a gating mechanism so that the model can concentrate on the most important information to obtain the sequence of the high-temperature natural gas temperature local time sequence characteristic significance attenuation weighting factors. And finally, taking the sequence of the high-temperature natural gas temperature local time sequence characteristic significance attenuation weight factors as a weight sequence, and calculating the weighted sum of the time sequence of the high-temperature natural gas temperature local time sequence characteristic vectors to obtain the high-temperature natural gas temperature time sequence node semantic significance aggregation expression vector.

The method comprises the steps of calculating characteristic significance description factors of each high-temperature natural gas temperature local time sequence characteristic vector in a time sequence of the high-temperature natural gas temperature local time sequence characteristic vector, wherein the characteristic significance description factors are related to the mean value and variance of each high-temperature natural gas temperature local time sequence characteristic vector, the method comprises the steps of calculating the mean value and variance of the high-temperature natural gas temperature local time sequence characteristic vector to obtain a high-temperature natural gas temperature local time sequence characteristic mean value and a high-temperature natural gas temperature local time sequence characteristic variance respectively, calculating the difference value according to the position of the high-temperature natural gas temperature local time sequence characteristic vector and the high-temperature natural gas temperature local time sequence characteristic mean value to obtain a high-temperature natural gas temperature local time sequence difference vector, calculating the fourth power of each characteristic value in the high-temperature natural gas temperature local time sequence difference vector to obtain a high-temperature natural gas temperature local time sequence expected value, calculating the expected value of the high-temperature natural gas temperature local time sequence expected value and the high-temperature natural gas temperature local time sequence expected value to obtain the high-temperature natural gas temperature local time sequence expected value, and the high-temperature natural gas temperature local time sequence characteristic difference vector to obtain the high-temperature natural gas temperature local time sequence characteristic difference vector corresponding to the high-temperature natural gas time sequence characteristic difference factor square value.

More specifically, the process of constructing the feature significance attenuation factor of each high-temperature natural gas temperature local time sequence feature vector based on the distance span between each high-temperature natural gas temperature local time sequence feature vector and the current high-temperature natural gas temperature local time sequence feature vector in the time queue of the high-temperature natural gas temperature local time sequence feature vector comprises the steps of extracting the maximum value of each high-temperature natural gas temperature local time sequence feature vector to obtain the sequence of the maximum value of the high-temperature natural gas temperature local time sequence feature; the method comprises the steps of obtaining a current high-temperature natural gas temperature local time sequence characteristic vector, extracting the maximum value of the current high-temperature natural gas temperature local time sequence characteristic vector to obtain a current high-temperature natural gas temperature local characteristic maximum value, calculating the position-wise subtraction between the current high-temperature natural gas temperature local characteristic maximum value and the high-temperature natural gas temperature local time sequence characteristic maximum value to obtain a high-temperature natural gas temperature time sequence offset value sequence, counting the distance span values between each high-temperature natural gas temperature local time sequence characteristic vector and the current high-temperature natural gas temperature local time sequence characteristic vector to obtain a high-temperature natural gas temperature distance span value sequence, calculating the square of each value in the high-temperature natural gas temperature distance span value sequence to obtain a high-temperature natural gas temperature distance span modulation value sequence, and dividing the high-temperature natural gas temperature time sequence offset value sequence and the high-temperature natural gas temperature distance span modulation value sequence by position to obtain the characteristic significant attenuation factors of each high-temperature natural gas temperature local time sequence characteristic vector.

The process of inputting the sequence of the high-temperature natural gas temperature local time sequence characteristic significant attenuation describing factors into a gating mask module to obtain the sequence of the high-temperature natural gas temperature local time sequence characteristic significant attenuation weight factors comprises the steps of carrying out normalization processing on each high-temperature natural gas temperature local time sequence characteristic significant attenuation describing factor in the sequence of the high-temperature natural gas temperature local time sequence characteristic significant attenuation describing factors based on a Sigmoid function to obtain a set of normalized high-temperature natural gas temperature local time sequence characteristic significant attenuation describing factors, and inputting each normalized high-temperature natural gas temperature local time sequence characteristic significant attenuation describing factor in the set of normalized high-temperature natural gas temperature local time sequence characteristic significant attenuation describing factors into a gating function to carry out masking processing to obtain the sequence of the high-temperature natural gas temperature local time sequence characteristic significant attenuation weight factors.

In summary, in the above embodiment, the method for obtaining the high-temperature natural gas temperature time sequence node semantic significant aggregation expression vector by passing the time sequence of the high-temperature natural gas temperature local time sequence feature vector through a node significant attenuation-based temperature local time sequence semantic aggregation network includes that the time sequence of the high-temperature natural gas temperature local time sequence feature vector is processed through the node significant attenuation-based temperature local time sequence semantic aggregation network by using the following semantic aggregation formula to obtain the high-temperature natural gas temperature time sequence node semantic significant aggregation expression vector, wherein the semantic aggregation formula is as follows: wherein, the method comprises the steps of, Representing the time sequence of the local time sequence characteristic vector of the high-temperature natural gas temperature,、AndRespectively the first time sequence of the local time sequence characteristic vector of the high-temperature natural gas temperatureFirst, secondAnd (b)A local time sequence characteristic vector of the high-temperature natural gas temperature,In order to extract the maximum value of the vector,Representation ofAndA value of the span of the distance between them,Is the firstThe characteristic significance attenuation factors of the local time sequence characteristic vectors of the high-temperature natural gas temperature,Is the firstThe first high-temperature natural gas temperature local time sequence characteristic vectorThe characteristic value of the individual position is used,AndRespectively the firstThe mean and square of variance of the local time sequence feature vectors of the high temperature natural gas temperature,Is the firstThe high-temperature natural gas temperature local time sequence characteristic corresponding to the high-temperature natural gas temperature local time sequence characteristic vector obviously attenuates the descriptive factors,Is thatThe function of the function is that,Is the first in the set of the normalized high-temperature natural gas temperature local time sequence characteristic significant decay descriptive factorsThe normalized high-temperature natural gas temperature local time sequence characteristic significance decay descriptive factor,In order to implement the masking process,Is the first in the sequence of the high-temperature natural gas temperature local time sequence characteristic significant decay weight factorThe local time sequence characteristic of the high-temperature natural gas temperature obviously attenuates the weight factor,For a predetermined threshold value,Is the semantic significant aggregate expression vector of the high-temperature natural gas temperature time sequence node.

In particular, the data normalization module 350 is configured to normalize the time sequence of the working state parameters according to a time dimension and a working state parameter sample dimension to obtain a time sequence of a working state parameter matrix. Considering that the time sequence of the working state parameters has time sequence characteristics in a time dimension and that each working parameter has mutual influence and correlation in time, in the technical scheme of the application, the time sequence of the working state parameters is subjected to data normalization according to the time dimension and the working state parameter sample dimension to obtain the time sequence of the working state parameter matrix.

In particular, the state parameter time sequence correlation feature extraction module 360 is configured to pass the time sequence of the operating state parameter matrix through an operating state parameter time sequence correlation feature extractor including a hole convolutional neural network and a recurrent neural network to obtain an operating state time sequence correlation feature vector. In the technical scheme of the application, the time sequence of the working state parameter matrix is processed by a working state parameter time sequence association feature extractor comprising a cavity convolutional neural network and a recurrent neural network to obtain a working state time sequence association feature vector, so that the features of the working state parameters in the time dimension and the space dimension are captured and mined more accurately. It is worth mentioning that the hole convolutional neural network is a Convolutional Neural Network (CNN) that uses hole convolutional operations to expand the receptive field while maintaining spatial resolution. A Recurrent Neural Network (RNN) is a neural network that processes sequential data, such as text, time series, and audio. RNNs have feedback connections that allow information to propagate across time steps in the network.

In particular, the temperature time sequence state-working state significant fusion module 370 is configured to obtain a temperature time sequence state-working state sparse significant matching fusion expression vector by using the high-temperature natural gas temperature time sequence node semantic significant aggregation expression vector and the working state time sequence related feature vector to optimize and match a significant fusion network based on feature principal components in a fine granularity. Considering that the semantic significant aggregation expression vector of the high-temperature natural gas temperature time sequence node expresses the most critical and prominent temperature change characteristics of the high-temperature natural gas in the whole time domain. The operating state time sequence associated feature vector reveals the dynamic change of the operating state parameters of the equipment along with time and the relation between the operating state parameters and the time sequence associated feature vector. And the temperature change of the high-temperature natural gas can influence the operation parameters of the refrigeration equipment, such as the load and the efficiency of the compressor, and the changes can be reflected in the time sequence correlation characteristic vector of the working state, and the performance of the refrigeration equipment can be better understood and predicted by analyzing the main component correlation between the two vectors, so that the rotating speed value of the compressor can be controlled more accurately. Therefore, in the technical scheme of the application, the high-temperature natural gas temperature time sequence node semantic significant aggregate expression vector and the working state time sequence associated feature vector are subjected to a significant fusion network based on feature principal component fine granularity optimization matching to obtain a temperature time sequence state-working state sparse significant matching fusion expression vector. In particular, the feature principal component fine granularity optimization matching-based salient fusion network is a technology integrating feature sparsification, principal Component Analysis (PCA) and salient feature fusion and is used for constructing key and salient feature correlation representations among feature vectors.

In the embodiment of the application, the high-temperature natural gas temperature time sequence node semantic significance aggregate representation vector and the working state time sequence association feature vector are subjected to standardization processing through a feature principal component-based feature fusion network which is optimized and matched in a detail mode to obtain a temperature time sequence state-working state sparse significance match fusion representation vector, so as to obtain a standardized high-temperature natural gas temperature time sequence node semantic significance aggregate representation vector and a standardized working state time sequence association feature vector, then, a sample covariance matrix of the standardized high-temperature natural gas temperature time sequence node semantic significance aggregate representation vector and the standardized working state time sequence association feature vector is calculated to obtain a high-temperature natural gas temperature node semantic significance aggregate sample covariance matrix and a working state time sequence association covariance matrix, feature vector extraction based on matrix decomposition is performed on the high-temperature natural gas temperature node semantic significance aggregate sample covariance matrix and the working state time sequence association covariance matrix to obtain a standardized high-temperature natural gas temperature node significant aggregate principal component feature vector and a standardized working state time sequence association feature vector, namely important feature vector can be better removed based on the feature principal component, namely important feature information in a set can be better extracted, and important feature information can be better captured, and important feature information can be better retained in a set. And inputting the set of the high-temperature natural gas temperature node semantic significant time sequence aggregation principal component feature vectors and the set of the working state time sequence association principal component feature vectors into a maximum approximate query matching network to obtain a set of optimal matching pairs of the high-temperature natural gas temperature node semantic significant time sequence aggregation principal component feature vectors and the working state time sequence association principal component feature vectors, further inputting the optimal matching pairs of the high-temperature natural gas temperature node semantic significant time sequence aggregation principal component feature vectors and the working state time sequence association principal component feature vectors into a semantic fine granularity gating joint module to obtain a set of high-temperature natural gas temperature-working state component fusion feature vectors, and cascading the set of the high-temperature natural gas temperature-working state component fusion feature vectors to obtain the temperature time sequence state-working state sparse significant matching fusion representation vector.

The process of carrying out standardized treatment on the high-temperature natural gas temperature time sequence node semantic significant aggregation expression vector and the working state time sequence associated feature vector to obtain a standardized high-temperature natural gas temperature time sequence node semantic significant aggregation expression vector and a standardized working state time sequence associated feature vector comprises the steps of respectively calculating the mean value and standard deviation of the high-temperature natural gas temperature time sequence node semantic significant aggregation expression vector to obtain a high-temperature natural gas temperature time sequence node semantic significant aggregation feature mean value and a high-temperature natural gas temperature time sequence node semantic significant aggregation feature standard deviation; the method comprises the steps of carrying out position-by-position subtraction on a high-temperature natural gas temperature time sequence node semantic significant aggregation representation vector and a high-temperature natural gas temperature time sequence node semantic significant aggregation feature mean value, carrying out position-by-position division on the calculated high-temperature natural gas temperature time sequence offset vector and the high-temperature natural gas temperature time sequence node semantic significant aggregation feature standard deviation to obtain a standardized high-temperature natural gas temperature time sequence node semantic significant aggregation representation vector, respectively calculating the mean value and standard deviation of the working state time sequence association feature vector to obtain a working state time sequence association feature mean value and a working state time sequence association feature standard deviation, carrying out position-by-position subtraction on the working state time sequence association feature vector and the working state time sequence association feature mean value, and carrying out position-by-position division on the calculated working state time sequence offset vector and the working state time sequence association feature standard deviation to obtain the standard working state time sequence association feature vector.

The process of calculating the sample covariance matrix of the normalized high-temperature natural gas temperature time sequence node semantic significant aggregate representative vector and the normalized working state time sequence associated feature vector comprises the steps of multiplying the transpose vector of the normalized high-temperature natural gas temperature time sequence node semantic significant aggregate representative vector by the normalized high-temperature natural gas temperature time sequence node semantic significant aggregate representative vector, then dividing the obtained value of the length of the normalized high-temperature natural gas temperature time sequence node semantic significant aggregate representative vector by the position to obtain the high-temperature natural gas temperature node semantic significant aggregate sample covariance matrix, multiplying the transpose vector of the normalized working state time sequence associated feature vector by the normalized working state time sequence associated feature vector, and dividing the obtained value of the length of the normalized working state time sequence associated feature vector by the position to obtain the working state time sequence associated covariance matrix.

More specifically, the process of inputting the set of high-temperature natural gas temperature node semantic significant time sequence aggregation principal component feature vectors and the set of working state time sequence association principal component feature vectors into a maximum approximate query matching network to obtain a set of optimal matching pairs of the high-temperature natural gas temperature node semantic significant time sequence aggregation principal component feature vectors and the working state time sequence association principal component feature vectors comprises the steps of extracting a preset high-temperature natural gas temperature node semantic significant time sequence aggregation principal component feature vector in the set of high-temperature natural gas temperature node semantic significant time sequence aggregation principal component feature vectors, calculating cosine similarity between each working state time sequence association principal component feature vector in the set of preset high-temperature natural gas temperature node semantic significant time sequence aggregation principal component feature vectors and the working state time sequence association principal component feature vector to obtain a set of matching query similarity, and taking the working state time sequence association principal component feature vector corresponding to the maximum matching query similarity in the set of matching query similarity and the preset high-temperature natural gas temperature node semantic significant time sequence aggregation principal component feature vector as the preset high-temperature natural gas temperature node semantic significant time sequence aggregation principal component feature vector and the optimal matching time sequence feature vector.

More specifically, the process of inputting the optimal matched pairs of the semantic significant time sequence aggregation principal component feature vector of the high-temperature natural gas temperature node and the working state time sequence correlation principal component feature vector in the set of the optimal matched pairs of the semantic significant time sequence aggregation principal component feature vector of the high-temperature natural gas temperature node and the working state time sequence correlation principal component feature vector into a semantic fine granularity gating and combining module to obtain the set of the high-temperature natural gas temperature-working state component fusion feature vector comprises the steps of respectively calculating the difference according to positions, multiplying according to positions and adding according to positions between the optimal matched pairs of the semantic significant time sequence aggregation principal component feature vector of the high-temperature natural gas temperature node and the working state time sequence correlation principal component feature vector to obtain a high-temperature natural gas temperature-working state time sequence principal component difference vector, a high-temperature natural gas temperature-working state time sequence principal component multiplication vector and a high-temperature natural gas temperature-working state time sequence principal component addition vector; and carrying out one-dimensional convolution coding after cascading the high-temperature natural gas temperature-working state time sequence main component differential vector, the high-temperature natural gas temperature-working state time sequence main component dot product vector and the high-temperature natural gas temperature-working state time sequence main component addition vector to obtain a high-temperature natural gas temperature-working state time sequence main component multi-dimensional fusion vector, and carrying out local window-based maximum pooling treatment on the high-temperature natural gas temperature-working state time sequence main component multi-dimensional fusion vector to obtain the high-temperature natural gas temperature-working state component fusion feature vector.

In summary, in the above embodiment, the processing of the high-temperature natural gas temperature time sequence node semantic significant aggregate expression vector and the working state time sequence associated feature vector through a significant fusion network based on feature principal component fine granularity optimization matching to obtain a temperature time sequence state-working state sparse significant match fusion expression vector includes processing the high-temperature natural gas temperature time sequence node semantic significant aggregate expression vector and the working state time sequence associated feature vector through a significant fusion network based on feature principal component fine granularity optimization matching with the following principal component feature fusion formula to obtain the temperature time sequence state-working state sparse significant match fusion expression vector, where the principal component feature fusion formula is: wherein, the method comprises the steps of, AndRespectively representing the semantic significant aggregate representation vector of the high-temperature natural gas temperature time sequence node and the working state time sequence associated feature vector,AndThe average value of the semantic significant aggregate expression vector of the high-temperature natural gas temperature time sequence node and the working state time sequence associated characteristic vector is respectively obtained,AndRespectively and obviously aggregating the standard deviation of the representing vector and the working state time sequence associated characteristic vector for the high-temperature natural gas temperature time sequence node semanteme,AndA representation vector and the normalized operation state time sequence association characteristic vector are semantically and obviously aggregated for the normalized high-temperature natural gas temperature time sequence node respectively,AndRespectively isAndIs used to determine the transposed vector of (c),AndThe normalized high-temperature natural gas temperature time sequence node semantically significant aggregate expression vector and the length of the normalized operation state time sequence associated feature vector are respectively,AndThe high-temperature natural gas temperature node semantic significant time sequence aggregate sample covariance matrix and the working state time sequence associated covariance matrix are respectively,AndThe main component orthogonal matrix is polymerized by the high-temperature natural gas temperature time sequence and the main component orthogonal matrix is related by the working state time sequence,AndRespectively a high-temperature natural gas temperature time sequence aggregation diagonal matrix and a working state time sequence association diagonal matrix,For elements on diagonal in the matrixThe diagonal matrix is polymerized according to the high-temperature natural gas temperature time sequence,The weight values of the principal component feature vectors are aggregated for each high-temperature natural gas temperature node semantically significant time sequence respectively,For elements on diagonal in the matrixThe operating state timing of (a) is associated with the diagonal matrix,The weight values of the principal component feature vectors are associated for each operating state timing sequence respectively,AndRespectively isAndIs used to determine the transposed matrix of (a),The main component feature vectors are aggregated for each high-temperature natural gas temperature node semantic significant time sequence in the set of the high-temperature natural gas temperature node semantic significant time sequence,For each operating state time sequence associated principal component feature vector in the set of operating state time sequence associated principal component feature vectors,To calculate theAndThe inner product of the vectors between them,In order to calculate a norm of the vector,For returning to maximum valueThe value of the sum of the values,Is the maximum approximate match value and,、AndRespectively the difference value according to the position, the multiplication according to the position point and the addition according to the position,In the case of a cascade of processes,Is a one-dimensional convolutional encoding operation,Is the operation of the maximum pooling,Is the first in the set of the high-temperature natural gas temperature-working state component fusion characteristic vectorsThe high-temperature natural gas temperature-working state components are fused with feature vectors,Is the number of the characteristic vectors in the high-temperature natural gas temperature-working state component fusion characteristic vector set,The temperature time sequence state-working state sparse significant matching fusion expression vector is obtained.

In particular, the control result generating module 380 and the rotation speed adjusting module 390 are configured to obtain a control result based on the sparse and significant matching fusion expression vector of the temperature time sequence state and the working state, and adjust the rotation speed of the compressor of the monitored refrigeration equipment through the frequency converter based on the control result. In a specific example of the application, the temperature time sequence state-working state sparse significant matching fusion expression vector is passed through a classifier-based compressor speed controller to obtain the control result, wherein the control result is used for indicating that the compressor speed value of the monitored refrigeration equipment at the next time point should be increased, decreased or unchanged. The temperature time sequence state-working state sparse significant matching fusion expression vector obtained by sparse significant fusion is classified by utilizing the high-temperature natural gas temperature time sequence node semantic significant aggregation expression vector and the working state time sequence association feature vector, so that the compressor rotating speed value of the monitored refrigeration equipment at the next time point is adaptively controlled, and the compressor rotating speed value is correspondingly adjusted. In this way, the system is able to collect multiple parameters, providing a more comprehensive view of the system's state. Meanwhile, the change of external conditions can be responded immediately, the rotating speed of the compressor can be dynamically adjusted, so that the equipment can maintain the optimal running state under different working conditions, the refrigerating efficiency is improved, and the control intellectualization of the refrigerating equipment is realized.

In a preferred example, the step of passing the temperature time sequence state-working state sparse significant matching fusion representation vector through a classifier-based compressor speed controller to obtain a control result includes calculating a sum of absolute values of respective eigenvalues of the temperature time sequence state-working state sparse significant matching fusion representation vector to obtain a first temperature time sequence state-working state sparse significant matching fusion spatial structure value, and calculating a square root of a square sum of respective eigenvalues of the temperature time sequence state-working state sparse significant matching fusion representation vector to obtain a second temperature time sequence state-working state sparse significant matching fusion spatial structure value; multiplying each characteristic value of the temperature time sequence state-working state sparse significant matching fusion representation vector by the first temperature time sequence state-working state sparse significant matching fusion space structure value and the second temperature time sequence state-working state sparse significant matching fusion space structure value respectively to obtain a first temperature time sequence state-working state sparse significant matching fusion structure reference value and a second temperature time sequence state-working state sparse significant matching fusion structure reference value corresponding to each characteristic value The method comprises the steps of obtaining a first temperature time sequence state-working state sparse significant matching fusion conversion adjustment value, obtaining a second temperature time sequence state-working state sparse significant matching fusion conversion adjustment value by dividing a first temperature time sequence state-working state sparse significant matching fusion structure reference value by a difference value between a first temperature time sequence state sparse significant matching fusion space structure value and a first temperature time sequence state sparse significant matching fusion scale conversion value, obtaining a first temperature time sequence state-working state sparse significant matching fusion conversion adjustment value by dividing a second temperature time sequence state-working state sparse significant matching fusion structure reference value by a difference value between a second temperature time sequence state sparse significant matching fusion space structure value and a second temperature time sequence state sparse significant matching fusion scale conversion value, obtaining each characteristic value of an optimized temperature time sequence state-working state sparse significant matching fusion representation vector, and obtaining a classification result of the optimized temperature time sequence state-working state sparse significant matching fusion representation vector through a compressor based on a controller.

Here, the temperature time sequence state-working state sparse significant matching fusion expression vector is recorded asThe optimization of (c) is expressed as: wherein, the method comprises the steps of, Fusing representative vectors for the temperature time sequence state-working state sparse significant matching,Represents a set of real numbers,Representing the temperature time sequence state-working state sparse significant matching fusion representation vectorThe characteristic value of the individual position is used,Representing the length of the temperature time sequence state-working state sparse significant matching fusion representation vector,The first temperature time sequence state-working state sparse significant matching fusion space structure value is represented,Representing a second temperature time sequence state-working state sparse significant matching fusion space structure value,Representing the multiplication by the position point,Indicating that the addition is by location,Represents a second temperature time sequence state-working state sparse significant matching fusion transformation adjustment vector,The second temperature time sequence state-working state sparse significant matching fusion transformation regulating value is represented,Represents a first temperature time sequence state-working state sparse significant matching fusion transformation adjustment vector,The first temperature time sequence state-working state sparse significant matching fusion transformation regulating value is represented,The weighted super-parameter is represented by a weighted super-parameter,And (5) representing the optimized temperature time sequence state-working state sparse significant matching fusion representation vector.

Here, in the preferred example, considering that the high-temperature natural gas temperature timing node semantic saliency aggregation representation vector and the operating state timing association feature vector respectively represent the timing node saliency attenuation aggregation feature of the one-dimensional local timing association feature of the high-temperature natural gas temperature data and the sample-timing cross dimension local association feature of the operating state parameter, when the significant fusion based on feature principal component query matching is performed on the high-temperature natural gas temperature timing node semantic saliency aggregation representation vector, principal component feature differences in the timing dimension and the timing-sample cross dimension have different saliency fusion weights based on the query matching, so that the temperature timing state-operating state sparse saliency matching fusion representation vector also has diversified aggregate expression distribution of fusion features, it is desirable to promote balance between regression mapping accuracy and integrity when the temperature timing state-operating state sparse saliency matching fusion representation vector is subjected to quasi regression through a classifier-based compressor rotation speed controller, and thus promote accuracy of a control result obtained.

In the preferred example, for the spatial structure information of the feature set of the temperature time sequence state-working state sparse significant matching fusion representation vector in a high-dimensional space, the spatial transformation (translation, scaling and rotation) invariance of the temperature time sequence state-working state sparse significant matching fusion representation vector under the interaction of the feature space is ensured by taking the quasi-norm spatial structured representation of the temperature time sequence state-working state sparse significant matching fusion representation vector as a reference window to perform scale-based frame transformation of each feature value of the temperature time sequence state-working state sparse significant matching fusion representation vector, and realizing frame attention weight adjustment of each feature value of the temperature time sequence state-working state sparse significant matching fusion representation vector based on a spatial structure, so that the control result (translation, scaling and rotation) invariance of the temperature time sequence state-working state sparse significant matching fusion representation vector under the interaction of the feature space is realized, and the equilibrium performability between mapping accuracy and mapping completeness is improved in the quasi-regression process based on the discretization feature distribution of the temperature time sequence state-working state sparse significant matching fusion representation vector. In this way, the system is able to collect multiple parameters, providing a more comprehensive view of the system's state. Meanwhile, the change of external conditions can be responded immediately, the rotating speed of the compressor can be dynamically adjusted, so that the equipment can maintain the optimal running state under different working conditions, the refrigerating efficiency is improved, and the control intellectualization of the refrigerating equipment is realized.

As described above, the intelligent-refrigerator control system 300 for a high-temperature environment according to an embodiment of the present application may be implemented in various wireless terminals, for example, a server or the like having an intelligent-refrigerator control algorithm for a high-temperature environment. In one possible implementation, the intelligent refrigeration appliance control system 300 for a high temperature environment according to an embodiment of the present application may be integrated into a wireless terminal as one software module and/or hardware module. For example, the intelligent refrigerant device control system 300 for a high temperature environment may be a software module in the operating system of the wireless terminal or may be an application developed for the wireless terminal, although the intelligent refrigerant device control system 300 for a high temperature environment may be one of a plurality of hardware modules of the wireless terminal.

Alternatively, in another example, the intelligent cooling apparatus control system 300 for a high temperature environment and the wireless terminal may be separate apparatuses, and the intelligent cooling apparatus control system 300 for a high temperature environment may be connected to the wireless terminal through a wired and/or wireless network and transmit interactive information in a contracted data format.

Further, an intelligent refrigeration equipment control method for the high-temperature environment is also provided.

Fig. 3 is a flowchart of a method of controlling an intelligent refrigeration appliance for a high temperature environment according to an embodiment of the present application. S1, acquiring a time sequence of operating state parameters of monitored refrigeration equipment acquired by a sensor group, wherein the operating state parameters comprise inlet temperature, inlet pressure, outlet temperature and outlet pressure; S2, acquiring a time sequence of temperature data of high-temperature natural gas, S3, passing the time sequence of the temperature data of the high-temperature natural gas through a 1D-CNN-based sequence encoder to obtain a time sequence of high-temperature natural gas temperature local time sequence feature vectors, S4, passing the time sequence of the high-temperature natural gas temperature local time sequence feature vectors through a node significance attenuation-based temperature local time sequence semantic aggregation network to obtain high-temperature natural gas temperature time sequence node significance aggregation expression vectors, S5, carrying out data normalization on the time sequence of the working state parameters according to time dimension and working state parameter sample dimension to obtain a time sequence of a working state parameter matrix, S6, passing the time sequence of the working state parameter matrix through a working state parameter time sequence correlation feature extractor comprising a cavity convolutional neural network and a recursive neural network to obtain a working state time sequence correlation feature vector, S7, passing the high-temperature natural gas temperature time sequence node significance aggregation expression vectors and the working state time sequence correlation feature vectors through a feature vector fusion network based on feature principal component fine granularity optimization matching to obtain a temperature state sparse time sequence-working state sparse significant fusion expression vector, S8, obtaining a working state sparse matching result based on the control result, and adjusting the rotating speed of the compressor of the monitored refrigeration equipment through a frequency converter.

In summary, the intelligent refrigeration equipment control method for a high-temperature environment according to the embodiment of the application is clarified, which acquires a time sequence of working state parameters (inlet temperature, inlet pressure, outlet temperature and outlet pressure) of the refrigeration equipment to be monitored through a sensor group, acquires a time sequence of temperature data of high-temperature natural gas, adopts a data analysis and processing technology based on deep learning to carry out data normalization and time sequence association on the working state parameters, carries out local time sequence feature extraction and full-time-domain significant aggregation on the temperature data of the high-temperature natural gas, and accordingly adaptively controls the rotation speed value of a compressor of the refrigeration equipment to be monitored at the next time point according to a main component significant matching fusion feature between the working state time sequence feature of the refrigeration equipment and the aggregation feature of the temperature data of the high-temperature natural gas, and correspondingly adjusts. In this way, the system is able to collect multiple parameters, providing a more comprehensive view of the system's state. Meanwhile, the change of external conditions can be responded immediately, the rotating speed of the compressor can be dynamically adjusted, so that the equipment can maintain the optimal running state under different working conditions, the refrigerating efficiency is improved, and the control intellectualization of the refrigerating equipment is realized.

The foregoing description of the embodiments of the present disclosure has been presented for purposes of illustration and description, and is not intended to be exhaustive or limited to the embodiments disclosed. Many modifications and variations will be apparent to those of ordinary skill in the art without departing from the scope and spirit of the various embodiments described. The terminology used herein was chosen in order to best explain the principles of the embodiments, the practical application, or the improvement of technology in the marketplace, or to enable others of ordinary skill in the art to understand the embodiments disclosed herein.

Claims

1. An intelligent refrigeration equipment control system for a high temperature environment, characterized in that it comprises: a state parameter acquisition module, used to acquire the time series of the working state parameters of the monitored refrigeration equipment collected by the sensor group, the working state parameters including inlet temperature, inlet pressure, outlet temperature and outlet pressure; a temperature data acquisition module, used to acquire the time series of temperature data of high-temperature natural gas; a temperature local time series feature extraction module, used to pass the time series of the temperature data of the high-temperature natural gas through a sequence encoder based on 1D-CNN to obtain the time series of the local time series feature vector of the temperature of the high-temperature natural gas; a high-temperature natural gas temperature time series aggregation module, used to pass the time series of the local time series feature vector of the temperature of the high-temperature natural gas through a temperature local time series semantic aggregation network based on node significance attenuation to obtain a high-temperature natural gas temperature time series node semantic significant aggregation representation vector; a data regularization module, used to convert the time series of the working state parameters into a local time series feature vector of the temperature of the high-temperature natural gas according to the time series of the working state parameters ... The time dimension and the working state parameter sample dimension are data regularized to obtain the time series of the working state parameter matrix; the state parameter time series association feature extraction module is used to obtain the working state time series association feature vector by a working state parameter time series association feature extractor including a void convolutional neural network and a recursive neural network; the temperature time series state-working state significant fusion module is used to obtain the temperature time series state-working state sparse significant matching fusion representation vector by a significant fusion network based on the fine-grained optimization matching of the feature principal component through the semantic significant aggregation representation vector of the high-temperature natural gas temperature time series node and the working state time series association feature vector; the control result generation module is used to obtain the control result based on the temperature time series state-working state sparse significant matching fusion representation vector; the speed adjustment module is used to adjust the compressor speed of the monitored refrigeration equipment through the inverter based on the control result.

2. The intelligent refrigeration equipment control system for high temperature environment according to claim 1, characterized in that the high temperature natural gas temperature time series aggregation module comprises: a high temperature natural gas temperature local time series feature significance description factor calculation unit, used to calculate the feature significance description factor of each high temperature natural gas temperature local time series feature vector in the time series of the high temperature natural gas temperature local time series feature vector, wherein the feature significance description factor is related to the mean and variance of each high temperature natural gas temperature local time series feature vector; a high temperature natural gas temperature local time series feature significance attenuation factor construction unit, used to construct the feature significance attenuation factor of each high temperature natural gas temperature local time series feature vector based on the distance span between each high temperature natural gas temperature local time series feature vector in the time queue of the high temperature natural gas temperature local time series feature vector and the current high temperature natural gas temperature local time series feature vector; a high temperature natural gas temperature local time series feature significance description factor modulation unit, used to modulate the feature significance description factor of each high temperature natural gas temperature local time series feature vector based on the feature significance attenuation factor of each high temperature natural gas temperature local time series feature vector to obtain a sequence of high temperature natural gas temperature local time series feature significance attenuation description factors; high temperature natural gas temperature local time series feature significance attenuation factor modulation unit The feature significance attenuation weight factor calculation unit is used to input the sequence of the high-temperature natural gas temperature local temporal feature significance attenuation description factors into the gated mask module to obtain the sequence of the high-temperature natural gas temperature local temporal feature significance attenuation weight factors; the high-temperature natural gas temperature temporal node semantic significance aggregation unit is used to use the sequence of the high-temperature natural gas temperature local temporal feature significance attenuation weight factors as the sequence of weights, calculate the weighted sum of the time series of the high-temperature natural gas temperature local temporal feature vectors to obtain the high-temperature natural gas temperature temporal node semantic significance aggregation representation vector; wherein, the high-temperature natural gas temperature local temporal feature significance attenuation weight factor sequence is used to ... The weight factor calculation unit is used to: perform normalization processing based on the Sigmoid function on each significant attenuation description factor of the local time series characteristic of the high-temperature natural gas temperature in the sequence of the significant attenuation description factors of the local time series characteristic of the high-temperature natural gas temperature to obtain a set of normalized significant attenuation description factors of the local time series characteristic of the high-temperature natural gas temperature; input each normalized significant attenuation description factor of the local time series characteristic of the high-temperature natural gas temperature in the set of normalized significant attenuation description factors of the local time series characteristic of the high-temperature natural gas temperature into the gating function for masking processing to obtain a sequence of significant attenuation weight factors of the local time series characteristic of the high-temperature natural gas temperature.

3. The intelligent refrigeration equipment control system for high temperature environment according to claim 2 is characterized in that the high temperature natural gas temperature local time series feature significance description factor calculation unit is used to: respectively calculate the mean and variance of the high temperature natural gas temperature local time series feature vector to obtain the high temperature natural gas temperature local time series feature mean and the high temperature natural gas temperature local time series feature variance; calculate the position difference between the high temperature natural gas temperature local time series feature vector and the high temperature natural gas temperature local time series feature mean to obtain the high temperature natural gas temperature local time series difference vector; calculate the fourth power of each eigenvalue in the high temperature natural gas temperature local time series difference vector to obtain the high temperature natural gas temperature local time series modulation difference vector; calculate the expected value of the high temperature natural gas temperature local time series modulation difference vector to obtain the high temperature natural gas temperature local time series expected value; divide the high temperature natural gas temperature local time series expected value by the square of the high temperature natural gas temperature local time series feature variance to obtain the feature significance description factor corresponding to the high temperature natural gas temperature local time series feature vector.

4. The intelligent refrigeration equipment control system for high temperature environment according to claim 3, characterized in that the high temperature natural gas temperature local time series feature significance attenuation factor construction unit is used to: extract the maximum value of each high temperature natural gas temperature local time series feature vector to obtain a sequence of high temperature natural gas temperature local time series feature maximum values; extract the maximum value of the current high temperature natural gas temperature local time series feature vector to obtain the current high temperature natural gas temperature local feature maximum value; calculate the positional subtraction between the current high temperature natural gas temperature local feature maximum value and the sequence of the high temperature natural gas temperature local time series feature maximum values to obtain a sequence of high temperature natural gas temperature time series offset values; count the distance span values between each high temperature natural gas temperature local time series feature vector and the current high temperature natural gas temperature local time series feature vector to obtain a sequence of high temperature natural gas temperature distance span values, and calculate the square of each value in the sequence of the high temperature natural gas temperature distance span values to obtain a sequence of high temperature natural gas temperature distance span modulation values; divide the sequence of the high temperature natural gas temperature time series offset values by the sequence of the high temperature natural gas temperature distance span modulation values by position to obtain the feature significance attenuation factor of each high temperature natural gas temperature local time series feature vector.

5. According to the intelligent refrigeration equipment control system for high temperature environment of claim 4, it is characterized in that the temperature time series state-working state significant fusion module comprises: a high temperature natural gas temperature working state feature standardization unit, which is used to standardize the semantic significant aggregation representation vector of the high temperature natural gas temperature time series node and the working state time series associated feature vector to obtain a standardized high temperature natural gas temperature time series node semantic significant aggregation representation vector and a standardized working state time series associated feature vector; a high temperature natural gas temperature working state sample covariance calculation unit, which is used to calculate the sample covariance matrix of the standardized high temperature natural gas temperature time series node semantic significant aggregation representation vector and the standardized working state time series associated feature vector to obtain a high temperature natural gas temperature node semantic significant aggregation sample covariance matrix and a working state A time series association covariance matrix; a high-temperature natural gas temperature working state principal component feature extraction unit, used to perform matrix decomposition-based feature vector extraction on the semantically significant time series aggregation sample covariance matrix of the high-temperature natural gas temperature node and the working state time series association covariance matrix to obtain a set of semantically significant time series aggregation principal component feature vectors of the high-temperature natural gas temperature node and a set of working state time series association principal component feature vectors; a high-temperature natural gas temperature working state feature query matching unit, used to input the set of semantically significant time series aggregation principal component feature vectors of the high-temperature natural gas temperature node and the set of working state time series association principal component feature vectors into a maximum approximate query matching network to obtain a set of optimal matching pairs of semantically significant time series aggregation principal component feature vectors of the high-temperature natural gas temperature node and working state time series association principal component feature vectors; The high-temperature natural gas temperature working state feature semantic association unit is used to input the best matching pairs of the high-temperature natural gas temperature node semantic significant temporal aggregation principal component feature vector and the working state temporal association principal component feature vector in the set of the best matching pairs of the high-temperature natural gas temperature node semantic significant temporal aggregation principal component feature vector and the working state temporal association principal component feature vector into the semantic fine-grained gated joint module to obtain a set of high-temperature natural gas temperature-working state component fusion feature vectors; the temperature temporal state-working state sparse significant matching fusion unit is used to cascade the set of the high-temperature natural gas temperature-working state component fusion feature vectors to obtain the temperature temporal state-working state sparse significant matching fusion representation vector; wherein, the high-temperature natural gas temperature working state feature query matching unit is used to: extract the A predetermined high-temperature natural gas temperature node semantically significant temporal aggregation principal component feature vector in a set of high-temperature natural gas temperature node semantically significant temporal aggregation principal component feature vectors; calculating the cosine similarity between the predetermined high-temperature natural gas temperature node semantically significant temporal aggregation principal component feature vector and each working state temporal association principal component feature vector in the set of working state temporal association principal component feature vectors to obtain a set of matching query similarities; and taking the working state temporal association principal component feature vector corresponding to the largest matching query similarity in the set of matching query similarities and the predetermined high-temperature natural gas temperature node semantically significant temporal aggregation principal component feature vector as the best matching pair of the predetermined high-temperature natural gas temperature node semantically significant temporal aggregation principal component feature vector and the working state temporal association principal component feature vector.

6. According to claim 5, the intelligent refrigeration equipment control system for high temperature environment is characterized in that the high temperature natural gas temperature working state feature standardization unit is used to: respectively calculate the mean and standard deviation of the semantically significant aggregation representation vector of the high temperature natural gas temperature time series node to obtain the mean of the semantically significant aggregation feature of the high temperature natural gas temperature time series node and the standard deviation of the semantically significant aggregation feature of the high temperature natural gas temperature time series node; after subtracting the semantically significant aggregation representation vector of the high temperature natural gas temperature time series node from the mean of the semantically significant aggregation feature of the high temperature natural gas temperature time series node by position, the high temperature natural gas temperature time series offset calculated The vector is divided by position with the standard deviation of the semantically significant aggregation feature of the high-temperature natural gas temperature time series node to obtain the standardized high-temperature natural gas temperature time series node semantically significant aggregation representation vector; the mean and standard deviation of the working state time series association feature vector are calculated respectively to obtain the working state time series association feature mean and the working state time series association feature standard deviation; after subtracting the working state time series association feature vector from the working state time series association feature mean by position, the calculated working state time series offset vector is divided by the standard deviation of the working state time series association feature by position to obtain the standardized working state time series association feature vector.

7. The intelligent refrigeration equipment control system for a high temperature environment according to claim 6 is characterized in that the high temperature natural gas temperature working state sample covariance calculation unit is used to: after multiplying the transposed vector of the standardized high temperature natural gas temperature time series node semantically significant aggregation representation vector with the standardized high temperature natural gas temperature time series node semantically significant aggregation representation vector, the obtained standardized high temperature natural gas temperature time series association matrix is divided by the value obtained by subtracting one from the length of the standardized high temperature natural gas temperature time series node semantically significant aggregation representation vector to obtain the high temperature natural gas temperature node semantically significant time series aggregation sample covariance matrix; after multiplying the transposed vector of the standardized working state time series association feature vector with the standardized working state time series association feature vector, the obtained standardized working state time series association matrix is divided by the value obtained by subtracting one from the length of the standardized working state time series association feature vector to obtain the working state time series association covariance matrix.

8. The intelligent refrigeration equipment control system for high temperature environment according to claim 7, characterized in that the high temperature natural gas temperature working state feature semantic association unit is used to: respectively calculate the position difference, position point multiplication and position addition between the best matching pair of the high temperature natural gas temperature node semantic significant time series aggregation principal component feature vector and the working state time series association principal component feature vector to obtain the high temperature natural gas temperature-working state time series principal component difference vector, the high temperature natural gas temperature-working state time series principal component point product vector and the high temperature natural gas temperature-working state time series principal component sum vector; cascade the high temperature natural gas temperature-working state time series principal component difference vector, the high temperature natural gas temperature-working state time series principal component point product vector and the high temperature natural gas temperature-working state time series principal component sum vector, and perform one-dimensional convolution coding to obtain the high temperature natural gas temperature-working state time series principal component multi-dimensional fusion vector; perform local window-based maximum pooling processing on the high temperature natural gas temperature-working state time series principal component multi-dimensional fusion vector to obtain the high temperature natural gas temperature-working state component fusion feature vector.

9. The intelligent refrigeration equipment control system for high temperature environment according to claim 8 is characterized in that the control result generation module is used to: pass the temperature time series state-working state sparse significant matching fusion representation vector through a classifier-based compressor speed controller to obtain the control result, and the control result is used to indicate whether the compressor speed value of the monitored refrigeration equipment at the next time point should increase, decrease or remain unchanged.

10. A method for controlling intelligent refrigeration equipment for high temperature environment, characterized in that it comprises: obtaining a time series of working state parameters of the monitored refrigeration equipment collected by a sensor group, wherein the working state parameters include inlet temperature, inlet pressure, outlet temperature and outlet pressure; obtaining a time series of temperature data of high temperature natural gas; passing the time series of temperature data of high temperature natural gas through a sequence encoder based on 1D-CNN to obtain a time series of local time series feature vectors of temperature of high temperature natural gas; passing the time series of local time series feature vectors of temperature of high temperature natural gas through a temperature local time series semantic aggregation network based on node significance attenuation to obtain a high temperature natural gas temperature time series node semantic significant aggregation representation vector; and dividing the time series of working state parameters according to time. The dimensions of the working state parameter samples are regularized to obtain the time series of the working state parameter matrix; the time series of the working state parameter matrix is passed through a working state parameter time series association feature extractor including a void convolutional neural network and a recurrent neural network to obtain a working state time series association feature vector; the high-temperature natural gas temperature time series node semantic significant aggregation representation vector and the working state time series association feature vector are passed through a significant fusion network based on feature principal component fine-grained optimization matching to obtain a temperature time series state-working state sparse significant matching fusion representation vector; based on the temperature time series state-working state sparse significant matching fusion representation vector, a control result is obtained; based on the control result, the compressor speed of the monitored refrigeration equipment is adjusted through a frequency converter.