CN118839138A - Intelligent assessment method and system for freezing effect of spiral instant freezer - Google Patents

Abstract

The present invention provides an intelligent evaluation method and system for the freezing effect of a spiral quick freezer, which relates to the technical field of data processing. The method and system determine an expected indicator matrix based on a pre-control program, monitor and determine monitoring sensor data, and obtain an actual control indicator matrix, perform matrix differentiation measurement, and determine a primary evaluation result; perform a mechanical transmission performance evaluation to determine a secondary evaluation result, and perform a cross-correlation analysis to determine the freezing evaluation effect. The method solves the technical problems that the prior art lacks a systematic and complete evaluation method, is not intelligent enough, and has insufficient analysis depth and completeness, resulting in limited evaluation accuracy of the freezing effect and causing subsequent equipment operation and management restrictions. The freezing dimension and the mechanical transmission dimension are used as the primary and secondary directions of the evaluation, and the freezing stage division and the hierarchical feature enhancement processing of the monitoring data are performed. The real-time freezing features are accurately located, and a comprehensive analysis is performed in combination with the transmission features. The intelligent and systematic evaluation and analysis of the freezing effect of the equipment is realized, and the objective consistency of the evaluation results is guaranteed.

Description

Intelligent evaluation method and system for freezing effect of spiral quick freezer

Technical Field

The present invention relates to the technical field of data processing, and in particular to a method and system for intelligently evaluating the freezing effect of a spiral quick freezer.

Background Art

With the popularity of quick-frozen food and the high-quality standard requirements, quick-freezing equipment, as a necessary mechanical freezing equipment, has higher and higher technical requirements. Spiral quick-freezing machines have become the current advantageous quick-freezing equipment with their high efficiency, high quality, and no loss. At present, the freezing effect is mainly determined based on the qualification of the quality inspection results through regular quality inspection of frozen materials and equipment operation. The existing technology lacks a systematic and complete evaluation method, is not intelligent enough, and lacks analysis depth and completeness, resulting in limited accuracy in the evaluation of the freezing effect, which in turn limits the subsequent equipment operation and management.

Summary of the invention

The present application provides an intelligent evaluation method and system for the freezing effect of a spiral quick freezer, which is used to solve the technical problems existing in the prior art, such as the lack of a systematic and complete evaluation method, the lack of intelligence, the insufficient analysis depth and completeness, which result in limited accuracy in the evaluation of the freezing effect and cause restrictions on subsequent equipment operation and management.

In view of the above problems, the present application provides a method and system for intelligently evaluating the freezing effect of a spiral quick freezer.

In a first aspect, the present application provides a method for intelligently evaluating the freezing effect of a spiral quick freezer, the method comprising:

Determine an expected index matrix for measuring the expected freezing effect of the target frozen item based on a pre-control program set by the assembled programmable controller, wherein the expected index matrix is marked with a tolerance interval based on technical loss;

As the target frozen items fall, the spiral quick-freezing machine is started and freezing parameter control monitoring is performed to determine monitoring sensor data, wherein a feeding conveyor belt and a discharging conveyor belt of the spiral quick-freezing machine are connected to the production line;

Based on the monitoring sensor data, in combination with the feature extraction module, feature extraction based on the expected indicator matrix is performed, selective feature enhancement processing is performed based on the configured hierarchical data processing algorithm, inter-layer features are cascaded and analyzed, and an indicator feature cascade network is determined, wherein the indicator feature cascade network is a pyramid structure;

Based on the indicator feature cascade network, accurately locate the indicator feature value and generate the actual control indicator matrix;

Based on the expected indicator matrix and the actual control indicator matrix, matrix differentiation measurement is performed to determine the main evaluation result;

Based on the monitoring sensor data, the mechanical transmission performance of the spiral quick-freezing machine is evaluated to determine the secondary evaluation result;

Based on the primary evaluation result and the secondary evaluation result, the freezing evaluation effect of the spiral quick freezer is determined in combination with the cross-correlation analysis.

In a second aspect, the present application provides a freezing effect intelligent evaluation system for a spiral quick freezer, the system comprising:

An expected indicator matrix determination module, the expected indicator matrix determination module is used to determine an expected indicator matrix for measuring an expected freezing effect of a target frozen item based on a pre-control program set by an assembled programmable controller, wherein the expected indicator matrix is marked with a tolerance interval based on technical loss;

A sensor monitoring module, which is used to start the spiral quick-freezing machine and perform freezing parameter control monitoring as the target frozen goods fall, and determine monitoring sensor data, wherein the feeding conveyor belt and the discharging conveyor belt of the spiral quick-freezing machine are connected to the production line;

A feature extraction module, wherein the feature extraction module is used to extract features based on the expected indicator matrix based on the monitoring sensor data in combination with the feature extraction module, perform selective feature enhancement processing based on the configured hierarchical data processing algorithm, cascade analyze inter-layer features, and determine an indicator feature cascade network, wherein the indicator feature cascade network is a pyramid structure;

A real control indicator matrix generation module, the real control indicator matrix generation module is used to accurately locate the indicator characteristic value and generate the real control indicator matrix based on the indicator characteristic cascade network;

A main evaluation result determination module, the main evaluation result determination module is used to perform matrix differentiation measurement based on the expected indicator matrix and the actual control indicator matrix to determine the main evaluation result;

A secondary evaluation result determination module, the secondary evaluation result determination module is used to evaluate the equipment mechanical transmission performance of the spiral quick-freezing machine based on the monitoring sensor data and determine a secondary evaluation result;

A freezing evaluation effect determination module is used to determine the freezing evaluation effect of the spiral quick freezer based on the main evaluation result and the secondary evaluation result in combination with a cross-correlation analysis.

One or more technical solutions provided in this application have at least the following technical effects or advantages:

The method for intelligently evaluating the freezing effect of a spiral quick-freezer provided in an embodiment of the present application determines an expected index matrix for measuring the expected freezing effect of a target frozen item based on a pre-control program set in an assembled programmable controller, starts the spiral quick-freezer and performs freezing parameter control monitoring as the target frozen item is dropped, determines monitoring sensor data, and combines a feature extraction module to perform hierarchical feature extraction mapping and cascade analysis based on the expected index matrix, determines an index feature cascade network, accurately locates the index feature value and generates an actual control index matrix, performs differentiated measurement between the expected index matrix and the actual control index matrix, and determines a main evaluation result; based on the monitoring sensor data, a hierarchical feature extraction mapping and cascade analysis based on the expected index matrix are performed, and an index feature cascade network is determined, so as to accurately locate the index feature value and generate an actual control index matrix, and performs differentiated measurement between the expected index matrix and the actual control index matrix, and determines a main evaluation result; based on the monitoring sensor data, a hierarchical feature extraction mapping and cascade analysis based on the expected index matrix are performed, and ... The mechanical transmission performance evaluation is performed based on the sensor data to determine the secondary evaluation result, and the cross-correlation analysis is performed to determine the freezing evaluation effect of the spiral quick freezer, which solves the problems existing in the prior art, such as the lack of a systematic and complete evaluation method, lack of intelligence, insufficient analysis depth and completeness, which leads to limited accuracy in the evaluation of the freezing effect and technical problems that restrict the subsequent equipment operation and management. The freezing dimension and the mechanical transmission dimension are taken as the primary and secondary directions of the evaluation, the freezing stage is divided and the hierarchical feature enhancement processing of the monitoring data is carried out, the real-time freezing features are accurately located, and a comprehensive analysis is carried out in combination with the transmission characteristics, so as to realize the intelligent and systematic evaluation and analysis of the freezing effect of the equipment and ensure the objectivity and consistency of the evaluation results.

BRIEF DESCRIPTION OF THE DRAWINGS

FIG1 is a schematic diagram of a flow chart of a method for intelligently evaluating the freezing effect of a spiral quick freezer provided in the present application;

FIG2 is a schematic diagram of the structural connection flow in the method for intelligently evaluating the freezing effect of a spiral quick freezer provided in the present application;

FIG3 is a schematic diagram of the structure of the intelligent evaluation system for the freezing effect of the spiral quick freezer provided in the present application.

Explanation of the accompanying drawings: expected indicator matrix determination module 11, sensor monitoring module 12, feature extraction module 13, actual control indicator matrix generation module 14, primary evaluation result determination module 15, secondary evaluation result determination module 16, freezing evaluation effect determination module 17.

DETAILED DESCRIPTION

The present application provides an intelligent evaluation method and system for the freezing effect of a spiral quick freezer, determines an expected indicator matrix based on a pre-control program, performs freezing parameter control monitoring of the spiral quick freezer to determine monitoring sensor data, determines an actual control indicator matrix based on the freezing effect, performs matrix differentiation measurement, and determines a primary evaluation result; performs a mechanical transmission performance evaluation to determine a secondary evaluation result, and performs a cross-correlation analysis to determine the freezing evaluation effect, so as to solve the technical problems existing in the prior art, such as the lack of a systematic and complete evaluation method, the lack of intelligence, the lack of analysis depth and completeness, which results in limited accuracy in the evaluation of the freezing effect, and causes restrictions on subsequent equipment operation and management.

Embodiment 1

As shown in FIG. 1 and FIG. 2 , the present application provides an intelligent evaluation method for the freezing effect of a spiral quick freezer, the method comprising:

S1: determining an expected index matrix for measuring an expected freezing effect of a target frozen item based on a pre-control program set by an assembled programmable controller, wherein the expected index matrix is marked with a tolerance interval based on technical loss;

Wherein, the determination of the expected index matrix for measuring the expected freezing effect of the target frozen item, the present application S1 also includes:

S11: setting a plurality of freezing stages based on the pre-control program and the target frozen product, wherein the plurality of freezing stages at least include a liquid quick-freezing stage and a crystallization stage;

S12: configuring evaluation indicators for measuring freezing effects, traversing the multiple freezing stages to determine expected characteristic values of stage indicators, and obtaining multiple groups of indicator characteristic values;

S13: constructing an expected indicator matrix using the evaluation indicators as matrix rows, the multiple freezing stages as matrix columns, and the multiple groups of indicator feature values as distribution matrix items.

With the popularity of quick-frozen food and the high-quality standard requirements, quick-freezing equipment, as a necessary mechanical freezing equipment, has higher and higher technical requirements. Spiral quick-freezing machines have become the current advantageous quick-freezing equipment with their high efficiency, high quality, and no loss. The intelligent evaluation method for the freezing effect of a spiral quick-freezing machine provided in this application takes the freezing dimension and the mechanical transmission dimension as the primary and secondary directions of the evaluation, divides the freezing stage and strengthens the hierarchical characteristics of the monitoring data, accurately locates the real-time freezing characteristics, and conducts a comprehensive analysis in combination with the transmission characteristics, so as to realize the intelligent and systematic evaluation and analysis of the freezing effect of the equipment and ensure the objective consistency of the evaluation results.

The programmable controller is the control center during the operation of the spiral quick-freezing machine, and is used to conduct the command operation of the equipment. The pre-control program set in the programmable controller is determined, that is, the braking control program for the spiral quick-freezing operation. The pre-control program is adapted to the target frozen items and is configured by a technician in this field based on the quick-freezing requirements. Furthermore, based on the pre-control program, the freezing effect measurement index of the target frozen items under the desired state is determined.

Specifically, the freezing stages of the target frozen items are divided to determine the multiple freezing stages, and the multiple freezing stages correspond to multiple stage-by-stage pre-control programs. The multiple freezing stages include at least the liquid quick-freezing stage and the crystallization stage, and the specific stage-by-stage standards are not specifically limited. For each freezing stage, based on the mapped stage-by-stage pre-control program, the stage-by-stage freezing effect measurement indicators are determined. For example, for the crystallization stage, the crystal size, crystallization uniformity, crystallization rate, etc. are included. The best freezing control needs to be achieved on the basis of ensuring the cellular moisture, original quality, etc. of the target frozen items. Determine the indicator characteristic values that are suitable for the corresponding stage-by-stage pre-control program, and integrate to determine the multiple groups of indicator characteristic values, wherein the multiple groups of indicator characteristic values correspond one-to-one to the multiple freezing stages.

Furthermore, completeness evaluation indicators covering multiple freezing stages are extracted and used as matrix rows, and the multiple freezing stages are used as matrix columns. The matrix items of the multiple groups of indicator characteristic values are distributed, wherein the matrix directions corresponding to the evaluation indicators that are not available in each freezing stage are blanked or marked as 0, and the completed expected indicator matrix is obtained. Among them, the tolerance interval based on technical losses is the inevitable loss in the freezing process of the equipment, which cannot be avoided. When performing differential analysis of indicators later, the tolerance interval needs to be excluded from the characteristic deviation. The expected indicator matrix is a freezing effect measurement indicator under ideal conditions that is consistent with the pre-control program, and it is used as a benchmark for evaluating the freezing effect of the spiral freezer.

S2: As the target frozen item falls, the spiral quick-freezing machine is started and freezing parameter control monitoring is performed to determine monitoring sensor data, wherein the feeding conveyor belt and the discharging conveyor belt of the spiral quick-freezing machine are connected to the production line;

S3: Based on the monitoring sensor data, in combination with the feature extraction module, feature extraction based on the expected indicator matrix is performed, selective feature enhancement processing is performed based on the configured hierarchical data processing algorithm, inter-layer features are cascaded and analyzed, and an indicator feature cascade network is determined, wherein the indicator feature cascade network is a pyramid structure;

Among them, the feed conveyor belt and the discharge conveyor belt of the spiral freezer are connected to the production line. As a part of the production process, they cooperate with the production line to freeze and transport the transported items. Generally speaking, the spiral freezer has a high-efficiency refrigeration effect, and the freezing time is generally more than ten minutes. The target frozen items are produced and transported in the production line, and are transferred to the spiral quick-freezer based on the feed conveyor belt. As the target frozen items are dropped, the spiral quick-freezing is started synchronously, and braking control is performed based on the pre-control program, and the freezing parameter control process is monitored. The monitoring data of the freezing process is time-series integrated and integrated as the monitoring sensor data. The monitoring sensor data is the collected data source for evaluating the freezing effect. Furthermore, based on the monitoring sensor data and combined with the expected indicator matrix, the data indicator values for measuring the freezing effect are extracted, and selective hierarchical enhancement processing is performed based on the extracted feature state to ensure the accuracy of the features, and the indicator feature cascade network characterized as a pyramid structure is built.

Wherein, the combined feature extraction module performs hierarchical feature extraction mapping and cascade analysis based on the expected indicator matrix, and the present application S3 also includes:

S31: The feature extraction module includes multiple fully connected network layers, wherein each network layer is configured with a data processing algorithm;

S32: Taking the evaluation index as a benchmark and combining it with the data screening layer in the feature extraction module, extract multiple source data mapped to each evaluation index;

S33: Transferring the multiple source data flows to a post-positioned multi-level functional layer, performing feature extraction and hierarchical selective enhancement processing, cascading indicator features based on hierarchical mapping, and obtaining the indicator feature cascade network.

Among them, the feature extraction and hierarchical selective enhancement processing, the present application S33 also includes:

S331: configuring a feature intensity threshold for measuring feature clarity;

S332: Taking the feature strength threshold as a constraint, hierarchical processing is performed on the features at each functional layer, and the features are selectively transferred to the post-functional layer to select the features to be enhanced.

Among them, this application also has S333, including:

S3331: Determine the adaptive feature type based on the data processing algorithm of the network configuration;

S3332: Mapping the adaptability feature type to the multi-level functional layer as a hierarchical data processing constraint;

S3333: Based on the hierarchical data processing constraints, selective processing is performed on the features to be enhanced to determine hierarchical enhanced features, wherein the selective processing refers to enhancement processing or empty flow processing.

Wherein, the feature extraction module is a functional model constructed for extracting index features from the monitoring sensor data, and the feature extraction module includes the data screening layer and the multi-level functional layer, wherein the multi-level functional layer is respectively configured with different data processing algorithms, such as feature scaling, feature enhancement, contrast enhancement, etc., and a hierarchical connection is established between the data screening layer and the multi-level functional layer to generate the feature extraction module. Further, the monitoring sensor data is input into the feature extraction module, and based on the data screening layer and the evaluation index as a benchmark, the mapping monitoring sensor data of each evaluation index is identified and extracted to determine the multiple source data, wherein the multiple source data corresponds one-to-one with the evaluation index.

Then, the multiple source data streams are transferred to the functional layer after the monitoring sensor data, and feature extraction corresponding to each evaluation index is performed, wherein each evaluation index corresponds to at least one extracted feature, and a primary extracted feature is determined. Furthermore, based on the post-functional layer, the feature strength of the primary extracted feature is sequentially enhanced.

Specifically, the feature strength threshold for measuring feature clarity is configured, and the feature strength threshold can be customized by a person skilled in the art based on the information completeness requirement of the feature. After the first-time extracted feature is transferred to the post-functional layer, the feature strength threshold is used as a constraint to calibrate the first-time extracted feature, determine the indicator feature set whose feature strength is less than the feature strength threshold, i.e., the feature to be enhanced, and mark it.

Further, based on the data processing algorithm configured by the transferred functional layer, the data feature type adapted by the data processing algorithm is determined, for example, the image feature enhancement algorithm is adapted to the image convolution feature, and the feature type applicable to the algorithm configured by the functional layer is used as the adaptive feature type. The data processing algorithm configured by each functional layer is determined, the corresponding adaptive feature type is determined, and a mapping between the adaptive feature type and the multi-level functional layer is established, and it is used as a feature constraint for functional layer data processing, and determined as the hierarchical data processing constraint.

Based on the adaptive feature type of the transferred functional layer, the features to be enhanced in the once extracted features are screened, the identification target features that meet the adaptive feature type are determined, and the identification target features are enhanced based on the configured data processing algorithm. The identification target features after the enhancement process are further integrated with the remaining index features and transferred to the post-functional layer, and the feature strength threshold is again determined and screened, and the adaptive feature type of the functional layer is screened and the index feature enhancement process is performed. Repeat the above steps to complete the feature enhancement process of the multi-level functional layers in turn, and complete the comprehensive enhancement process of the extracted index features. By performing hierarchical selective enhancement processing, the targeted processing and completeness of the differentiated features are guaranteed, and the processed features are guaranteed to meet the feature strength requirements, so as to improve the feature clarity, facilitate the measurement of the index feature values, and improve the accuracy of determining the index feature values.

Among them, the feature intensity threshold is used as the functional layer selective flow checkpoint to screen the features to be strengthened, and only the features that need to be strengthened are processed, thereby reducing the redundant feature data in the data processing process to improve the processing efficiency. The adaptive feature type is used as the hierarchical data processing constraint to selectively process the features to be strengthened based on the algorithm fitness, that is, the indicator features that meet the adaptive feature type are strengthened, and the indicator features that do not meet the adaptive feature type are subjected to empty flow processing, that is, the hierarchical flow of data is directly performed without processing. By sequentially screening the indicator features at the primary and secondary checkpoints, the indicator features that meet the complex processing requirements are accurately defined, and the algorithm is processed in a targeted manner.

Furthermore, hierarchical attribution and inter-layer correlation mapping are performed on the indicator features after hierarchical processing, and the constructed indicator feature cascade network is obtained by cascading the hierarchical indicator features. The indicator feature cascade network is the relevant extracted features corresponding to the evaluation indicators accurately located based on the monitoring sensor data.

S4: Based on the indicator feature cascade network, accurately locate the indicator feature value and generate the actual control indicator matrix;

S5: Based on the expected indicator matrix and the actual control indicator matrix, perform matrix differentiation measurement to determine the main evaluation result;

Traverse the indicator feature cascade network, perform feature matching of the evaluation indicators, determine at least one indicator feature corresponding to each evaluation indicator, and determine the indicator feature value of the evaluation indicator based on the indicator feature state. That is, if the evaluation indicator corresponds to one indicator feature, it is directly determined based on the indicator feature; if the evaluation indicator corresponds to multiple indicator features, weight configuration is performed based on the feature importance, and the weighted sum of multiple indicator features is performed as the indicator feature value of the evaluation indicator. The indicator feature values are arranged in the same mode as the expected indicator matrix to generate the actual control indicator matrix.

Furthermore, the actual control indicator matrix and the expected indicator matrix are mapped and measured in terms of matrix items, and the main evaluation result is determined based on the difference indicator value. Exemplarily, based on the deviation indicator characteristic value of each evaluation indicator, multiple single evaluation results are determined, wherein the single evaluation result is positively correlated with the deviation degree. The multiple single evaluation results are weighted averaged to obtain the main evaluation result, wherein the specific weight configuration is consistent with the importance of the indicator.

S6: Based on the monitoring sensor data, perform an equipment mechanical transmission performance evaluation of the spiral quick-freezing machine to determine a secondary evaluation result;

S7: Based on the primary evaluation result and the secondary evaluation result, the freezing evaluation effect of the spiral quick freezer is determined in combination with a cross-correlation analysis.

Among them, the mechanical transmission performance evaluation of the spiral quick-freezing machine is performed, and S6 of the present application also includes:

S61: Based on the monitoring sensor data, identifying and screening braking source data;

S62: Reading the production specification information of the spiral quick-freezing machine to determine the effective braking characteristics;

S63: extracting actual braking characteristics based on the braking source data, and checking the effective braking characteristics to determine the mechanical transmission state of the equipment;

S64: Configure a standard performance level and determine the secondary evaluation result based on the mechanical transmission state.

Then, a performance evaluation is performed on the mechanical transmission dimension. Specifically, the transmission monitoring data is screened and extracted based on the monitoring sensor data as the braking source data. The production specification information of the spiral quick freezer is further read, and it can be directly identified and determined based on the production work order. Based on the production specification information, the braking characteristics of the spiral quick freezer under the standard operating state are determined, for example, the cumulative transmission deviation within the allowable range, mechanical stability, wind circulation smoothness, etc., as the effective braking characteristics. Based on the effective braking characteristics, corresponding identification and extraction are performed in the braking source data as the actual braking characteristics.

The effective braking feature and the actual braking feature are further mapped and proofread, and the mechanical transmission state of the equipment is determined based on the feature deviation value, wherein, since the mechanical performance of the equipment is an additional evaluation dimension, in order to improve the efficiency of the evaluation process, feature extraction and differentiation analysis are directly performed on it. The standard performance level is a plurality of defined levels for measuring the mechanical transmission states of different equipment, which are set by technicians in this field based on the operating regulations of the spiral quick freezer. The standard performance level is traversed, the mechanical transmission state is matched, the matching level is determined, and the characteristic deviation degree is combined as the secondary evaluation result. The main evaluation result and the secondary evaluation result are integrated and integrated, and the primary and secondary correlation analysis is performed. For example, the freezing effect abnormality caused by the equipment transmission deviation is correlated and mapped to obtain the freezing evaluation effect of the spiral quick freezer. The freezing evaluation result is highly consistent with the actual operation of the spiral quick freezer.

Among them, this application also has S8, including:

S81: configure alarm constraint threshold;

S82: If the freezing evaluation effect does not meet the alarm constraint threshold, an abnormal alarm is issued and a human-computer interaction instruction is generated synchronously, wherein the differentiated traceability information is additional output information;

S83: Transmitting the human-computer interaction instruction to the personnel mobile terminal, and performing operation control of the spiral quick freezer based on the differentiated traceability information.

When the freezing evaluation effect does not meet the standard, the operation and maintenance management of the spiral quick freezer must be carried out in a timely manner. Specifically, based on the operation and maintenance standards of the spiral quick freezer, the alarm constraint threshold is configured, that is, the defined range for measuring the normal operating state. When it is within the alarm constraint threshold, it indicates that the equipment is in a normal operating state. If the freezing evaluation effect does not meet the alarm constraint threshold, it indicates that the freezing effect of the equipment does not meet the standard, an abnormal alarm is issued, and the human-computer interaction instruction is generated synchronously. Based on the indicator feature cascade network of the main evaluation result and the feature deviation degree of the secondary evaluation result, differential traceability is performed, and the collection source of the deviated feature is determined, which is used as the differentiated traceability information, and the human-computer interaction instruction is synchronized for output. The staff who performs the operation and maintenance management of the spiral quick freezer is determined, and the human-computer interaction instruction is further transmitted to the personnel mobile terminal. The operation of the spiral quick freezer is regulated based on the differentiated traceability information to ensure the freezing effect of the equipment.

The intelligent evaluation method for the freezing effect of a spiral quick freezer provided in this application has the following technical effects:

1. The existing technology lacks a systematic and complete evaluation method, is not intelligent enough, and lacks analysis depth and completeness, which limits the accuracy of the evaluation of the freezing effect and restricts the subsequent equipment operation and management. Taking the freezing dimension and mechanical transmission dimension as the primary and secondary directions of the evaluation, the freezing stage division and hierarchical feature enhancement processing of the monitoring data are carried out, the real-time freezing characteristics are accurately located, and a comprehensive analysis is performed in combination with the transmission characteristics, so as to realize the intelligent and systematic evaluation and analysis of the equipment freezing effect and ensure the objective consistency of the evaluation results.

2. Combined with the feature extraction module with hierarchical functionality, with the set feature intensity threshold and the hierarchical configured data processing algorithm as constraints, feature extraction and hierarchical adaptive selection and enhancement processing are performed on the monitoring data to ensure that the processing process is consistent with the requirements of feature extraction, so as to perform feature enhancement processing to ensure the clarity and accuracy of the features and improve the accuracy of freezing effect evaluation.

Embodiment 2

Based on the same inventive concept as the method for intelligently evaluating the freezing effect of a spiral quick freezer in the aforementioned embodiment, as shown in FIG3 , the present application provides an intelligent evaluation system for the freezing effect of a spiral quick freezer, the system comprising:

An expected indicator matrix determination module 11, the expected indicator matrix determination module 11 is used to determine an expected indicator matrix for measuring the expected freezing effect of the target frozen item based on a pre-control program set by the assembled programmable controller, wherein the expected indicator matrix is marked with a tolerance interval based on technical loss;

A sensor monitoring module 12, which is used to start the spiral quick-freezing machine and perform freezing parameter control monitoring as the target frozen goods fall, and determine monitoring sensor data, wherein the feeding conveyor belt and the discharging conveyor belt of the spiral quick-freezing machine are connected to the production line;

A feature extraction module 13, wherein the feature extraction module 13 is used to extract features based on the expected indicator matrix based on the monitoring sensor data in combination with the feature extraction module, perform selective feature enhancement processing based on the configured hierarchical data processing algorithm, cascade analyze inter-layer features, and determine an indicator feature cascade network, wherein the indicator feature cascade network is a pyramid structure;

A real control indicator matrix generation module 14, the real control indicator matrix generation module 14 is used to accurately locate the indicator characteristic value and generate the real control indicator matrix based on the indicator characteristic cascade network;

A main evaluation result determination module 15, wherein the main evaluation result determination module 15 is used to perform matrix differentiation measurement based on the expected indicator matrix and the actual control indicator matrix to determine the main evaluation result;

A secondary evaluation result determination module 16, the secondary evaluation result determination module 16 is used to evaluate the mechanical transmission performance of the spiral quick-freezing machine based on the monitoring sensor data and determine a secondary evaluation result;

The freezing evaluation effect determination module 17 is used to determine the freezing evaluation effect of the spiral quick freezer based on the primary evaluation result and the secondary evaluation result in combination with a cross-correlation analysis.

Wherein, the expected indicator matrix determination module 11 also includes:

A freezing stage setting module, the freezing stage setting module is used to set a plurality of freezing stages based on the pre-control program and the target frozen item, wherein the plurality of freezing stages at least include a liquid quick freezing stage and a crystallization stage;

A characteristic value acquisition module, the characteristic value acquisition module is used to configure an evaluation index for measuring the freezing effect, traverse the multiple freezing stages to determine the expected characteristic values of the stage indicators, and obtain multiple groups of indicator characteristic values;

A matrix building module is used to build an expected indicator matrix using the evaluation indicators as matrix rows, the multiple freezing stages as matrix columns, and the multiple groups of indicator feature values as distribution matrix items.

Wherein, the feature extraction module 13 also includes:

A structure analysis module, wherein the structure analysis module is used for the feature extraction module and includes multiple fully connected network layers, wherein each network layer is configured with a data processing algorithm;

A data screening module, the data screening module is used to extract multiple source data mapped to each evaluation index based on the evaluation index and in combination with the data screening layer in the feature extraction module;

An indicator feature cascade network acquisition module is used to transfer the multiple source data flows to a post-positioned multi-level functional layer, perform feature extraction and hierarchical selective enhancement processing, cascade indicator features based on hierarchical mapping, and obtain the indicator feature cascade network.

Wherein, the indicator feature cascade network acquisition module also includes:

A feature intensity threshold configuration module, wherein the feature intensity threshold configuration module is used to configure a feature intensity threshold for measuring feature clarity;

The feature screening module to be strengthened is used to use the feature strength threshold as a constraint, hierarchically process the features at each functional layer, and selectively transfer the features to the post-functional layer to screen the features to be strengthened.

Wherein, the system further comprises:

A feature type determination module, the feature type determination module is used to determine the adaptive feature type based on the data processing algorithm configured by the network;

a hierarchical data processing constraint determination module, the hierarchical data processing constraint determination module being used to map the adaptability feature type with the multi-level functional layer as a hierarchical data processing constraint;

A hierarchical enhancement feature determination module is used to perform selective processing of the features to be enhanced based on the hierarchical data processing constraints to determine hierarchical enhancement features, wherein selective processing refers to enhancement processing or empty flow processing.

Wherein, the secondary evaluation result determination module 16 further includes:

A braking source data screening module, the braking source data screening module is used to identify and screen the braking source data based on the monitoring sensor data;

An effective braking feature determination module, the effective braking feature determination module is used to read the production specification information of the spiral quick-freezing machine and determine the effective braking feature;

A device mechanical transmission state determination module, the device mechanical transmission state determination module is used to extract actual braking characteristics based on the braking source data, and calibrate the effective braking characteristics to determine the device mechanical transmission state;

A result determination module is used to configure a standard performance level and determine the secondary evaluation result based on the mechanical transmission state.

Wherein, the system further comprises:

An alarm constraint threshold configuration module, wherein the alarm constraint threshold configuration module is used to configure the alarm constraint threshold;

An instruction generation module, wherein if the freezing evaluation effect does not meet the alarm constraint threshold, the instruction generation module is used to issue an abnormal alarm and simultaneously generate a human-computer interaction instruction, wherein the differentiated traceability information is additional output information;

An operation control module, wherein the operation control module is used to transmit the human-computer interaction instruction to the personnel mobile terminal, and perform operation control of the spiral quick freezer based on the differentiated traceability information.

Through the above-mentioned detailed description of the intelligent evaluation method for the freezing effect of a spiral quick freezer in this specification, those skilled in the art can clearly understand the intelligent evaluation method and system for the freezing effect of a spiral quick freezer in this embodiment. For the device disclosed in the embodiment, since it corresponds to the method disclosed in the embodiment, the description is relatively simple, and the relevant parts can be referred to the description of the method part.

The above description of the disclosed embodiments enables those skilled in the art to implement or use the present application. Various modifications to these embodiments will be apparent to those skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of the present application. Therefore, the present application will not be limited to the embodiments shown herein, but will conform to the widest scope consistent with the principles and novel features disclosed herein.

Claims (8)

1. The intelligent assessment method for the freezing effect of the spiral instant freezer is characterized by comprising the following steps of:

determining an expected index matrix for measuring expected freezing effects of target frozen articles based on a pre-control program set by an assembled programmable controller, wherein the expected index matrix identifies a tolerance interval based on technical loss;

starting the spiral instant freezer and performing freezing parameter control monitoring along with blanking of the target frozen articles to determine monitoring sensing data, wherein a feeding conveying belt and a discharging conveying belt of the spiral instant freezer are connected with a production line;

based on the monitoring sensing data, combining a feature extraction module to perform feature extraction based on the expected index matrix, performing selective feature strengthening processing based on a configured hierarchical data processing algorithm, and performing cascade analysis on interlayer features to determine an index feature cascade network, wherein the index feature cascade network is in a pyramid structure;

based on the index feature cascade network, accurately positioning index feature values and generating a real control index matrix;

based on the expected index matrix and the actual control index matrix, performing matrix differentiation measurement to determine a main evaluation result;

Based on the monitoring sensing data, evaluating the mechanical transmission performance of the equipment of the spiral instant freezer, and determining a secondary evaluation result;

And determining the freezing evaluation effect of the spiral instant freezer based on the main evaluation result and the secondary evaluation result by combining cross correlation analysis.

2. The method of claim 1, wherein the determining a desired index matrix that measures a desired freezing effect of the frozen article comprises:

Setting a plurality of freezing stages based on the pre-control program and the target frozen article, wherein the plurality of freezing stages at least comprise a liquid quick freezing stage and a crystallization stage;

Configuring evaluation indexes for measuring freezing effects, traversing the plurality of freezing stages to determine expected characteristic values of the stage indexes, and obtaining a plurality of groups of index characteristic values;

And constructing a desired index matrix by taking the evaluation index as a matrix row, taking the plurality of freezing stages as a matrix array and taking the plurality of groups of index characteristic values as distribution matrix items.

3. The method of claim 2, wherein the combined feature extraction module performs a hierarchical feature extraction mapping and cascade analysis based on the desired index matrix, the method comprising:

The feature extraction module comprises a plurality of fully-connected network layers, wherein each network layer is configured with a data processing algorithm;

extracting a plurality of source data mapped to each evaluation index by taking the evaluation index as a reference and combining a data screening layer in the characteristic extraction module;

And transferring the multiple source data streams to a rear-mounted multi-stage functional layer, performing feature extraction and hierarchical selectivity enhancement processing, and cascading index features based on hierarchical mapping to obtain the index feature cascading network.

4. A method according to claim 3, wherein the feature extraction and hierarchical selective enhancement processes are performed, the method comprising:

configuring a characteristic intensity threshold for measuring characteristic definition;

And taking the characteristic intensity threshold value as constraint, carrying out hierarchical processing on the functional layers, wherein the hierarchical processing is characterized in that the post functional layer selectively flows to the card, and the characteristics to be reinforced are screened.

5. The method of claim 4, wherein the method comprises:

determining an adaptive feature type based on a data processing algorithm of network configuration;

Mapping the adaptive feature type and the multi-level functional layer as a hierarchical data processing constraint;

and based on the hierarchical data processing constraint, carrying out selective processing of the feature to be enhanced, and determining the hierarchical enhancement feature, wherein the selective processing refers to enhancement processing or air flow processing.

6. The method of claim 1, wherein the assessment of the mechanical transmission performance of the equipment of the spiral freezer is performed, the method comprising:

identifying and screening brake source data based on the monitoring sensing data;

Reading production specification information of the spiral instant freezer, and determining effective braking characteristics;

Extracting actual braking characteristics based on the braking source data, checking the effective braking characteristics and determining the mechanical transmission state of the equipment;

and configuring a standard performance grade, and determining the secondary evaluation result based on the mechanical transmission state.

7. The method of claim 1, characterized in that the method comprises:

Configuring an alarm constraint threshold;

if the freezing evaluation effect does not meet the alarm constraint threshold, carrying out abnormal warning and synchronously generating a man-machine interaction instruction, wherein the differential tracing information is additional output information;

and transmitting the man-machine interaction instruction to a personnel mobile terminal, and performing operation regulation and control of the spiral instant freezer based on the differential traceability information.

8. Intelligent assessment system for freezing effect of spiral instant freezer, which is characterized in that the system comprises:

The system comprises an expected index matrix determining module, a target freezing object determining module and a target freezing object determining module, wherein the expected index matrix determining module is used for determining an expected index matrix for measuring expected freezing effects of target frozen objects based on a pre-control program set by an assembled programmable controller, and the expected index matrix identifies a tolerance interval based on technical loss;

the sensing monitoring module is used for starting the spiral instant freezer along with blanking of the target frozen article and performing freezing parameter control monitoring to determine monitoring sensing data, wherein a feeding conveying belt and a discharging conveying belt of the spiral instant freezer are connected with a production line;

The feature extraction module is used for carrying out feature extraction based on the expected index matrix by combining with the feature extraction module based on the monitoring sensing data, carrying out selective feature strengthening processing based on a configured hierarchical data processing algorithm, carrying out cascade analysis on interlayer features, and determining an index feature cascade network, wherein the index feature cascade network is of a pyramid structure;

the real control index matrix generation module is used for precisely positioning index characteristic values and generating a real control index matrix based on the index characteristic cascade network;

The main evaluation result determining module is used for carrying out matrix differentiation measurement based on the expected index matrix and the actual control index matrix to determine a main evaluation result;

The secondary evaluation result determining module is used for evaluating the mechanical transmission performance of the equipment of the spiral instant freezer based on the monitoring sensing data and determining a secondary evaluation result;

And the freezing evaluation effect determining module is used for determining the freezing evaluation effect of the spiral instant freezer based on the main evaluation result and the secondary evaluation result and combining cross correlation analysis.