CN117121025A - Management method, system and storage medium of heating ventilation air conditioning system - Google Patents

Abstract

A management method, system and storage medium for heating ventilation air conditioning system. The method comprises the following steps: determining a current scene category according to the acquired current scene data related to the operation of the heating ventilation air conditioning system and a predetermined reference scene classification strategy; activating a reference BES model corresponding to the current scene category from the predetermined building energy consumption simulation reference BES models corresponding to the scene categories to obtain a current reference BES model; determining an optimal control strategy of the heating ventilation air conditioning system by utilizing the current scene data and the current reference BES model; and controlling corresponding equipment in the heating, ventilation and air conditioning system according to the optimal control strategy. The method can improve the accuracy of the BES model in energy performance prediction, and is beneficial to building energy conservation optimization.

Description

Management methods, systems and storage media for HVAC systems Technical field

The invention relates to the field of building energy consumption, in particular to a management system, method and computer-readable storage medium for an HVAC system.

Background technique

Reducing building energy consumption is one of the top priorities of official energy policies in many countries, and a large part of building energy consumption in industrialized countries is used in heating, ventilation and air conditioning (HVAC) systems, hereinafter referred to as HVAC system. Implementing optimized operating parameters in HVAC systems, such as chilled water temperature and supply air temperature, can reduce building energy consumption without sacrificing thermal comfort or major modifications to the building/system.

However, due to high variability in the design and control of the building, HVAC system, and external environment, conventional setups may not necessarily be energy efficient in day-to-day operation. The building energy consumption digital twin system based on Building Energy Simulation (BES) is an energy-saving optimization method that has been applied in recent years to improve the energy efficiency of existing buildings.

Contents of the invention

In view of this, embodiments of the present invention provide, on the one hand, a management method for an HVAC system, and on the other hand, a management system and a computer-readable storage medium for the HVAC system, to improve the performance of the BES model. Accuracy in energy performance prediction is beneficial to building energy optimization.

An HVAC management method proposed in the embodiment of the present invention includes: determining the current scene category based on the obtained current scene data related to the operation of the HVAC system and a predetermined benchmark scene classification strategy; Activating the benchmark BES model corresponding to the current scenario category in the building energy consumption simulation benchmark BES model corresponding to each scenario category to obtain the current benchmark BES model; using the current scenario data and the current benchmark BES model to determine the HVAC system The optimal control strategy; controlling the corresponding equipment in the HVAC system according to the optimal control strategy.

In one embodiment, the method further includes: collecting a set number of historically collected historical reference data related to the HVAC system; the set number being greater than a set number threshold; using the historical reference data to Calibrate the scene-independent parameters in an initial BES model; determine the current scene classification strategy based on the historical reference data; divide the historical reference data into different scene categories according to the current scene classification strategy, and use each The scene-related parameters in the initial BES model are calibrated with the reference data of each scene category to obtain a candidate BES model corresponding to the scene category; the scene classification strategy is determined as the benchmark scene classification strategy, and the scene classification strategy is The candidate BES model is determined as the baseline BES model.

In one embodiment, before determining the scene classification strategy as the baseline scene classification strategy and determining the candidate BES model as the baseline BES model, the method further includes: according to the number of scene categories, the difference between scene categories Any one or any combination of the discrimination degree and the simulation accuracy of the candidate BES model corresponding to each scene category, conduct a comprehensive evaluation of the scene classification strategy; when the comprehensive evaluation results meet the set requirements, execute the described The scene classification strategy is determined as the benchmark scene classification strategy, and the candidate BES model is determined as the operation of the benchmark BES model; otherwise, when the comprehensive evaluation result does not meet the set requirements, the set strategy optimization algorithm is used Optimize the scene classification strategy, use the optimized scene classification strategy as the current scene classification strategy, and return to execute the current scene classification strategy to divide the historical reference data into different scene categories. operate.

In one embodiment, the initial BES model is a BES model established using known fixed information related to the operation of the HVAC system and an initial estimate of uncertain information, or a currently available BES model.

In one embodiment, before determining the current scene classification strategy based on the historical reference data, the method further includes: determining whether there is currently a historical baseline scene classification strategy, and if there is a historical baseline scene classification strategy, determining whether there is a historical baseline scene classification strategy. Determine the current scene classification strategy with reference to the data and the historical reference scene classification strategy. If there is no historical reference scene classification strategy, perform the operation of determining the current scene classification strategy based on the historical reference data.

In one embodiment, it further includes: displaying information related to the HVAC system; the information related to the HVAC system comes from the current scene data and/or control information of the HVAC system and/ or status information.

In one embodiment, the scene data includes at least one or any combination of the following information: weather information, holiday information, local event information, sensor collections in the building related to HVAC systems and/or building thermal performance information, and information related to the HVAC system that is manually entered in the building; the historical reference data includes historical scene data.

The management system of the HVAC system proposed in the embodiment of the present invention includes: at least one memory and at least one processor, wherein: the at least one memory is used to store the computer program; the at least one processor is used to call the The computer program stored in at least one memory causes the device to perform corresponding operations. The operations include: determining the current scene category based on the acquired current scene data related to the operation of the HVAC system and the predetermined baseline scene classification strategy. ; Activating the benchmark BES model corresponding to the current scenario category from the predetermined building energy consumption simulation benchmark BES model corresponding to each scenario category to obtain the current benchmark BES model; using the current scenario data and the current benchmark BES model to determine An optimal control strategy of the HVAC system is obtained; corresponding equipment in the HVAC system is controlled according to the optimal control strategy.

In one embodiment, the operations further include: collecting a set number of historically collected historical reference data related to the HVAC system; the set number being greater than a set number threshold; utilizing the historical reference data Calibrate the scene-independent parameters in an initial BES model; determine the current scene classification strategy based on the historical reference data; divide the historical reference data into different scene categories according to the current scene classification strategy, using The reference data of each scene category is used to calibrate the scene-related parameters in the initial BES model to obtain a candidate BES model corresponding to the scene category; the scene classification strategy is determined as the benchmark scene classification strategy, and the The candidate BES model is determined as the base BES model.

In one embodiment, before determining the scene classification strategy as the baseline scene classification strategy and determining the candidate BES model as the baseline BES model, the method further includes: according to the number of scene categories, the distance between scene categories Any one or any combination of the discrimination and the simulation accuracy of the candidate BES model corresponding to each scene category, conduct a comprehensive evaluation of the scene classification strategy; when the comprehensive evaluation results meet the set requirements, execute the scene classification strategy The classification strategy is determined as the benchmark scene classification strategy, and the candidate BES model is determined as the operation of the benchmark BES model; otherwise, when the comprehensive evaluation result does not meet the set requirements, the set strategy optimization algorithm is used to The scene classification strategy is optimized, the optimized scene classification strategy is used as the current scene classification strategy, and the operation of dividing the historical reference data into different scene categories according to the current scene classification strategy is performed.

In one embodiment, before determining the current scene classification strategy based on the historical reference data, the operation further includes: determining whether there is currently a historical baseline scene classification strategy, and if there is a historical baseline scene classification strategy, then determining according to The historical reference data and the historical reference scene classification strategy determine the current scene classification strategy. If there is no historical reference scene classification strategy, the operation of determining the current scene classification strategy based on the historical reference data is performed.

In one embodiment, the operations further include: displaying information related to the HVAC system; the information related to the HVAC system comes from the current scene data and/or the control of the HVAC system information and/or status information.

Another management system for the HVAC system proposed in the embodiment of the present invention includes: a data collection module for collecting current scene data related to the operation of the HVAC system; a scene decision-making module for based on the current scene The data and the predetermined benchmark scenario classification strategy are used to determine the current scenario category of the building; the model determination module is used to activate the benchmark corresponding to the current scenario category from the predetermined benchmark building energy consumption simulation BES model corresponding to each scenario category. BES model to obtain the current baseline BES model; the control strategy determination module is used to call the current baseline BES model based on the set optimization control algorithm and the current scene data, and determine HVAC based on the output of the current baseline BES model. The optimal control strategy of the system; the system control module is used to control the corresponding equipment in the HVAC system according to the optimal control strategy.

In one embodiment, the method further includes: a data recording module, configured to record a set number of historical reference data related to the HVAC system collected by the data collection module and/or other external modules; the setting The number is greater than a set quantity threshold; the first calibration module is used to use the historical reference data to calibrate the scene-independent parameters of an initial BES model; the classification strategy determination module is used to determine the current scene classification strategy; a data division module, used to divide the historical reference data into different scene categories according to the current scene classification strategy; a second calibration module, used to use the reference data of each scene category to Calibrate the scene-related parameters in the initial BES model to obtain a candidate BES model corresponding to the scene category; an evaluation module is used to calculate the number of scene categories, the distinction between scene categories, and the candidates corresponding to each scene category. Any one or any combination of the simulation accuracy of the BES model is used to comprehensively evaluate the scene classification strategy; the result determination module is used to determine the scene classification strategy as a benchmark when the comprehensive evaluation results meet the set requirements. The scene classification strategy is provided to the scene decision-making module, the candidate BES model is determined as the benchmark BES model, and provided to the BES model determination module; when the comprehensive evaluation result does not meet the set requirements, the strategy optimization module is instructed to perform Strategy optimization; the strategy optimization module is used to optimize the scene classification strategy using the set strategy optimization algorithm, use the optimized scene classification strategy as the current scene classification strategy, and instruct the data partition module to execute the scene classification strategy according to the The current scene classification strategy is an operation of dividing the historical reference data into different scene categories.

A computer-readable storage medium proposed in the embodiment of the present invention has a computer program stored thereon; the computer program can be executed by a processor and implement the management method of the HVAC system as described in any of the above embodiments.

It can be seen from the above solution that in the embodiment of the present invention, multiple scene categories of the building are determined based on a large amount of historical reference data collected historically, and corresponding BES models are set corresponding to each scene category. After that, the current scene category can be determined based on the currently collected scene data, and the BES model corresponding to the current scene category can be called to determine the optimal control strategy of the HVAC system, and then the HVAC system can be controlled based on the optimal control strategy. The corresponding equipment in the system, such as heating equipment, ventilation equipment and air conditioning equipment, etc., thereby improving the accuracy of the BES model in energy performance prediction and conducive to building energy saving optimization.

Description of drawings

Preferred embodiments of the present invention will be described in detail below to make the above and other features and advantages of the present invention more apparent to those skilled in the art with reference to the accompanying drawings, in which:

Figure 1 is an exemplary flow chart of a management method for an HVAC system in an embodiment of the present invention.

Figure 2 is an exemplary flow chart of a method of establishing and updating a benchmark scene classification strategy and a benchmark BES model corresponding to each scene category in an example of the present invention.

Figure 3 is a schematic diagram of the process of classifying historical reference data, calibrating the initial BES model and obtaining candidate BES models corresponding to each scene category in an example of the present invention.

Figure 4 is a schematic structural diagram of a management system of an HVAC system in an embodiment of the present application.

Figure 5 is a schematic structural diagram of a system for establishing and updating a benchmark scene classification strategy and a benchmark BES model corresponding to each scene category in an embodiment of the present application.

FIG. 6 is a schematic structural diagram of another HVAC system management system in an embodiment of the present application.

Among them, the reference signs are as follows:

label meaning 101～104; 201～208 step 31 Reference data 32 Scene classification strategy 331, 332,…, 33n Data subsets under different scene categories 34 Model calibration 351, 352,…, 35n Candidate BES models under different scene categories 41 HVAC system management system 411 data collection module 412 Scenario decision module 413 model determination module 4131, 4132,…, 413n Baseline BES models under different scene categories 414 Control strategy determination module 415 System control module

42 architecture 421 sensor 422 control Panel 423 HVAC system 43 external data source 501 data logging module 502 first calibration module 503 Classification strategy determination module 504 Data partition module 505 Second calibration module 506 Assessment module 507 Result decision module 508 Strategy optimization module 61 memory 62 processor 63 bus

Detailed ways

In the embodiment of the present invention, considering that in order to optimize the energy consumption of the HVAC system while maintaining the prescribed building performance standards, building energy simulation (Building Energy Simulation, BES) is usually applied during the building operation stage to influence the external environment. Predict the impact of HVAC operating parameters.

BES models can generally be divided into rule-driven models and data-driven models. Data-driven BES models utilize monitoring data from a building to generate a model capable of predicting system behavior. The accuracy of the model largely depends on the quality and quantity of available training data. Rule-driven models based on heat balance equations and data-driven models can provide the most detailed predictions of building performance and accurate assessment of control strategies. Since hundreds or thousands of input parameters are used in rule-driven BES models, some of which are often unmeasurable, significant differences can be found between building energy consumption predicted by BES models and actual measured building energy consumption. Therefore, it is necessary to use monitoring data during building operation to calibrate and test the BES model. Considering the impact of the external environment on HVAC system performance, operators usually update the BES model at a fixed frequency based on rules of thumb, such as every three months.

However, in the current BES model calibration process, changes in model parameters cannot be fully reflected due to the influence of the external environment. For example, the ventilation rate of a building has a great relationship with the air pollution index and outdoor temperature. Therefore, in this embodiment, it is considered to determine the location based on a large amount of historically collected external information (such as weather information, holiday information, local events, etc.) and building status information (such as information collected by sensors, manual input information in the building, etc.) Describe multiple scene categories of the building, and set the corresponding BES model corresponding to each scene category. After that, the current scene category can be determined based on the currently collected external information and building status information, and the BES model corresponding to the current scene category can be called to determine the optimal control strategy of the HVAC system, and then control based on the optimal control strategy Corresponding equipment in HVAC systems, such as heating equipment, ventilation equipment, and air conditioning equipment.

In order to make the purpose, technical solutions and advantages of the present invention clearer, the following examples are given to further describe the present invention in detail.

Figure 1 is an exemplary flow chart of a management method for an HVAC system in an embodiment of the present invention. As shown in Figure 1, the method in this embodiment may include the following steps:

Step 101: Determine the current scene category based on the acquired current scene data related to the operation of the HVAC system and the predetermined baseline scene classification strategy.

In this embodiment, the scene data related to the operation of the HVAC system may include: building status data from within the building and/or external data from outside the building, etc. Among them, the building status data from the building may include: information collected by sensors in the building related to the HVAC system and/or the building's thermal performance; and manually input information related to the HVAC system in the building. Among them, the information collected by the sensor may include: chilled water/hot water temperature, refrigerator/heater power consumption, fan coil unit (Fan Control Unit, FCU) terminal flow rate, indoor temperature and humidity, etc. Manually entered information can include: room comfort requirements, etc. External data from outside the building can include: weather information, holiday information, local event information, etc. In actual applications, specific scene data can be determined according to actual needs and are not limited here.

In this embodiment, during specific implementation, the artificial intelligence network can be trained based on a predetermined benchmark scene classification strategy, using a large number of groups of historical scene data as input samples, and using the scene categories corresponding to each group of historical scene data as output samples. , and obtain a scene classification model, and then the current scene data can be used as the input of the scene classification model, and the scene classification model can directly output the corresponding scene category.

In addition, a scene database can also be set up based on a predetermined baseline scene classification strategy. The scene database can store scene data for each scene category. After receiving the current scene data, various artificial intelligence algorithms can be used, such as Algorithms such as K-means Clustering (KMeans) or Density-Based Spatial Clustering Algorithm with Noise (DBSCAN) determine the scene category corresponding to the current scene data from the scene database.

Step 102: Activate the benchmark BES model corresponding to the current scene category from the predetermined benchmark BES models corresponding to each scene category to obtain the current benchmark BES model.

In this embodiment, instead of just setting a single BES model for a building, a baseline BES model is set corresponding to different scene categories. In this way, after the current scene category is determined in step 101, the benchmark BES model corresponding to the scene category can be activated as a simulation model of building performance and energy consumption, thereby making the simulation results of the benchmark BES model more accurate.

Step 103: Use the current scenario data and the current baseline BES model to determine the optimal control strategy of the HVAC system.

In this step, the optimal control strategy of the HVAC system can be determined based on an optimization control algorithm, such as genetic algorithm (GA) or particle swarm algorithm (PSO) or a combination algorithm of GA and PSO, to ensure that the building under the current scenario performance indicators and has optimal energy-saving performance. Specifically, the current BES model can be called based on the set optimization control algorithm and the current scene data. For example, the current scene data and other different input parameters obtained based on the optimization control algorithm can be used as the current benchmark. BES model input, and obtain different outputs of the current baseline BES model, and determine the optimal control strategy of the HVAC system based on the output of the current baseline BES model.

In one example, the operating parameters that need to be optimized in the optimal control strategy may be the temperature of chilled/hot water, the air flow rate of each FCU, etc.

Step 104: Control corresponding equipment in the HVAC system according to the optimal control strategy.

In this step, corresponding control signals can be sent to the heating, ventilation and air conditioning system equipment in the building according to the optimal control strategy.

In addition, in this embodiment, information related to the HVAC system may be further displayed. The information related to the HVAC system comes from the current scene data and/or the control information and/or status information of the HVAC system. For example, information related to the building's thermal and HVAC systems such as indoor temperature, energy consumption, HVAC equipment status, etc. can be displayed.

The above method can be executed periodically according to a preset time interval. Alternatively, in some applications, it can be performed in real time.

There are many specific implementation methods for the establishment and update process of the benchmark scene classification strategy and the benchmark BES model corresponding to each scene category in the above steps 101 and 102. One of them is listed below.

FIG. 2 is an exemplary flow chart of a method for establishing and updating a benchmark scene classification strategy and a benchmark BES model corresponding to each scene category in an example of the present invention. In this example, the update process can be performed periodically according to a preset time interval. As shown in Figure 2, the method may include the following steps:

Step 201: Collect a set number of historically collected historical reference data related to the HVAC system; the set number is greater than a set number threshold. Among them, the set quantity threshold refers to a large enough value to meet the requirements.

In this step, historical reference data may include historical scene data and other related data, such as control strategies, energy consumption, etc. The specifics can be determined according to actual needs and are not limited here.

In practical applications, with the accumulation of historical reference data, when the benchmark scene classification strategy and the benchmark BES model corresponding to each scene category are subsequently updated, the amount of historical reference data will also increase. Correspondingly, the updated baseline scene classification strategy and the baseline BES model corresponding to each scene category will be increasingly optimized.

Step 202: Calibrate the scene-independent parameters of an initial BES model 20 using the historical reference data.

In this step, the initial BES model can be based on the use of known fixed information related to the operation of the HVAC system, such as geometric information, structural information, HVAC system information, etc., and uncertain information such as HVAC equipment performance, occupancy rate, etc. The BES model established based on the initial valuation. Alternatively, if there are no major renovations to the building and HVAC system, the initial BES model could be a currently available BES model that already exists.

In this step, parameters unrelated to the scene can be some structural feature data, etc.

Step 203: When initially establishing a benchmark scene classification strategy and a benchmark BES model corresponding to each scene category, in this step, the current scene classification strategy can be determined based on the historical reference data. When the baseline scene classification strategy and the baseline BES model corresponding to each scene category are subsequently updated, the current scene classification strategy can be determined directly based on the historical reference data, or the previous baseline scene classification strategy can be referred to, that is, when there is history When the baseline scene classification strategy is determined, the current scene classification strategy may be determined based on the historical reference data and the historical baseline scene classification strategy.

In this step, the scene classification strategy can be determined based on the actual situation and is not limited here. For example, the classification criteria may be maximum temperature, minimum temperature, average temperature, precipitation, air pollution index, wind speed, weekdays/weekends, other parameters related to weather and/or building operation and/or a combination of these parameters.

Step 204: Divide the historical reference data into different scene categories according to the current scene classification strategy, use the reference data of each scene category to calibrate the scene-related parameters in the initial BES model, and obtain the corresponding Candidate BES models for the scene category.

The process in this step can be seen in Figure 3. Figure 3 shows a schematic diagram of the process of classifying historical reference data, calibrating the initial BES model, and obtaining candidate BES models corresponding to each scene category. As shown in Figure 3, this process may include:

The reference data 31 is divided into different scene categories based on the scene classification strategy 32, and a data subset 331 under scene category 1, a data subset 332 under scene category 2, ..., and a data subset 33n under scene category n are obtained. The initial BES model is calibrated 34 using the data subset under each scene category, and a candidate BES model 351 under scene category 1, a candidate BES model 352 under scene category 2, ..., and a candidate BES model under scene category n are obtained. Model 35n.

If it can be determined that the current scene classification strategy and candidate BES model can meet the requirements, for example, the current scene classification strategy and candidate BES model are the scene classification strategy and candidate BES model obtained after multiple updates, then the scene can be classified directly The strategy is determined as the benchmark scene classification strategy, and the candidate BES model is determined as the benchmark BES model.

Of course, if you are not sure whether the current scene classification strategy and candidate BES model meet the requirements, you can continue to perform the following steps as shown in Figure 2:

Step 205: Conduct a comprehensive evaluation of the scene classification strategy based on any one or any combination of the number of scene categories, the distinction between scene categories, and the simulation accuracy of the candidate BES models corresponding to each scene category.

In this step, when calculating the simulation accuracy of the candidate BES models corresponding to each scene category, the error between the simulation results of the candidate BES models and the results of historical samples can be calculated for each candidate BES model, such as the average Error or root mean square error, etc., to obtain an evaluation score. Then, the evaluation scores of the candidate BES models corresponding to each scene category are comprehensively processed, such as weighted summation, to obtain a comprehensive score, and then the comprehensive score can be Compare with a threshold to determine whether the simulation accuracy of the candidate BES model corresponding to each scene category meets the requirements.

Step 206: Determine whether the evaluation result meets the requirements. When the comprehensive evaluation result meets the set requirements, execute step 207; otherwise, execute step 208.

Step 207: Determine the scene classification strategy as the base scene classification strategy, and determine the candidate BES model as the base BES model.

Step 208: Use the set strategy optimization algorithm to optimize the scene classification strategy, use the optimized scene classification strategy as the current scene classification strategy, and return to step 204.

In this step, artificial intelligence algorithms such as genetic algorithm (GA) or particle swarm algorithm (PSO) or a combination algorithm of GA and PSO can be used to optimize the scene classification strategy. In addition, simulation acceleration algorithms such as surrogate models can also be used to generate optimized scene classification strategies.

The management method of the HVAC system in the embodiment of the present invention has been described in detail above. The management system of the HVAC system in the embodiment of the present invention will be described in detail below. The management system of the HVAC system in the embodiment of the present invention can be used to implement the management method of the HVAC system in the embodiment of the present invention. For details not disclosed in the system embodiment of the present invention, please refer to the method embodiment of the present invention. The corresponding descriptions will not be repeated here.

Figure 4 is an exemplary structural diagram of a management system of an HVAC system in an embodiment of the present invention. As shown in FIG. 4 , the management system 41 of the HVAC system may include: a data collection module 411 , a scenario decision module 412 , a model determination module 413 , a control strategy determination module 414 and a system control module 415 .

Among them, the data collection module 411 is used to collect current scene data related to the operation of the HVAC system, process the collected data, and send the processed data to the digital interface of the scene decision-making module 412. As shown in Figure 4, the data collection module 411 can collect information collected by sensors 421 in the building 42 related to the HVAC system and/or the thermal performance of the building; it can also collect information manually input through the control panel 422 in the building 42. ; Information from external data sources 43 may also be collected, such as weather information, holiday information, local event information, etc.

The scene decision module 412 is used to determine the current scene category of the building based on the current scene data and the predetermined benchmark scene classification strategy.

The model determination module 413 is used to activate the benchmark BES model corresponding to the current scene category from the predetermined benchmark BES models 4131, 4132, ..., 413n corresponding to each scene category, to obtain the current benchmark BES model.

Among them, the scene decision module 412, the model determination module 413, and the BES models corresponding to each scene category can be components of the digital twin system DT of the HVAC system.

The control strategy determination module 414 is configured to call the current BES model based on the set optimal control algorithm and the current scene data, and determine the optimal control strategy of the HVAC system based on the output of the current baseline BES model.

The system control module 415 is used to control corresponding equipment in the HVAC system 423 in the building 42 according to the optimal control strategy, such as heating equipment, ventilation equipment, air conditioning equipment, etc.

In addition, the management system of the HVAC system in the embodiment of the present invention may further include a display module 416 as shown in the dotted line in Figure 4. The display module 416 is used to read the current baseline BES from the data collection module 411. Information related to the building thermal and HVAC system of at least one of the model, and system control module 415, and display the information.

Figure 5 is a schematic structural diagram of a system for establishing and updating a benchmark scene classification strategy and a benchmark BES model corresponding to each scene category in an embodiment of the present application. As shown in Figure 5, the system may include: data recording module 501, first calibration module 502, classification strategy determination module 503, data partition module 504, second calibration module 505, evaluation module 506, result determination module 507 and strategy optimization Module 508.

The data recording module 501 is configured to record a set number of historical reference data related to the HVAC system collected by the data collection module and/or other external modules; the set number is greater than a set number threshold.

The first calibration module 502 is used to calibrate scene-independent parameters in an initial BES model using the historical reference data.

The classification strategy determination module 503 is used to determine the current scene classification strategy based on the historical reference data. When a historical reference scene classification strategy exists, the classification strategy determination module 503 may further refer to the historical reference scene classification strategy. For example, the current scene classification strategy may be determined based on the historical reference data and the historical reference scene classification strategy.

The data dividing module 504 is configured to divide the historical reference data into different scene categories according to the current scene classification strategy.

The second calibration module 505 is configured to use the reference data of each scene category to calibrate scene-related parameters in the initial BES model to obtain a candidate BES model corresponding to the scene category.

The evaluation module 506 is used to conduct a comprehensive evaluation of the scene classification strategy based on any one or any combination of the number of scene categories, the distinction between scene categories, and the simulation accuracy of the candidate BES models corresponding to each scene category. .

The result determination module 507 is configured to determine the scene classification strategy as the benchmark scene classification strategy and provide it to the scene decision module 412 when the comprehensive evaluation result meets the setting requirements, and determine the candidate BES model as the benchmark BES model, And provided to the BES model determination module 413; when the comprehensive evaluation result does not meet the set requirements, the strategy optimization module 508 is instructed to perform strategy optimization.

The strategy optimization module 508 is used to optimize the scene classification strategy using the set strategy optimization algorithm, use the optimized scene classification strategy as the current scene classification strategy, and instruct the data partitioning module 504 to execute the scene classification strategy according to the current scene classification strategy. The scene classification strategy is an operation of dividing the historical reference data into different scene categories.

Corresponding to the method shown in Figure 2, in some applications, the evaluation module 506 and the strategy optimization module 508 may not be included, and the result judgment module 507 directly determines the scene classification strategy as a benchmark scene classification strategy and provides To the scene decision-making module 412, the candidate BES model is determined as the baseline BES model and provided to the BES model determination module 413.

During specific implementation, the system shown in Figure 5 can also be merged into the system shown in Figure 4.

Figure 6 is a schematic structural diagram of another HVAC system management system in an embodiment of the present application. This system can be used to implement the method shown in Figures 1-3, or to implement the system shown in Figures 4-5 . As shown in FIG. 6 , the system may include: at least one memory 61 and at least one processor 62 . In addition, some other components may also be included, such as communication ports, etc. These components communicate via bus 63.

Among them, at least one memory 61 is used to store computer programs. In one embodiment, the computer program can be understood as including various modules of the management system of the HVAC system shown in Figures 4-5. In addition, at least one memory 61 may also store an operating system and the like. Operating systems include but are not limited to: Android operating system, Symbian operating system, Windows operating system, Linux operating system, etc.

At least one processor 62 is configured to call a computer program stored in at least one memory 61 to execute the management method of the HVAC system described in the embodiment of the present application. The processor 62 may be a CPU, a processing unit/module, an ASIC, a logic module or a programmable gate array, etc. It can receive and send data through the communication port.

Specifically, at least one processor 62 is used to call a computer program stored in at least one memory 61 to cause the system to perform corresponding operations. The operation may include: determining the current scene category based on the acquired current scene data related to the operation of the HVAC system and the predetermined benchmark scene classification strategy; activating the corresponding scene category from the predetermined benchmark BES model corresponding to each scene category. The baseline BES model of the current scene category is used to obtain the current baseline BES model; the optimal control strategy of the HVAC system is determined using the current scene data and the current baseline BES model; and the optimal control strategy is controlled according to the optimal control strategy. Describe the corresponding equipment in the HVAC system.

In one embodiment, the operations may further include: collecting a set number of historically collected historical reference data related to the HVAC system; the set number being greater than a set number threshold; utilizing the historical reference The data calibrates scene-independent parameters in an initial BES model; determines a current scene classification strategy based on the historical reference data; divides the historical reference data into different scene categories according to the current scene classification strategy, Use the reference data of each scene category to calibrate the scene-related parameters in the initial BES model to obtain a candidate BES model corresponding to the scene category; determine the scene classification strategy as the baseline scene classification strategy, and The candidate BES model is determined as the baseline BES model.

In one embodiment, before determining the scene classification strategy as the baseline scene classification strategy and determining the candidate BES model as the baseline BES model, the method further includes: according to the number of scene categories, the distance between scene categories Any one or any combination of the discrimination and the simulation accuracy of the candidate BES model corresponding to each scene category, conduct a comprehensive evaluation of the scene classification strategy; when the comprehensive evaluation results meet the set requirements, execute the scene classification strategy The classification strategy is determined as the benchmark scene classification strategy, and the candidate BES model is determined as the operation of the benchmark BES model; otherwise, when the comprehensive evaluation result does not meet the set requirements, the set strategy optimization algorithm is used to The scene classification strategy is optimized, the optimized scene classification strategy is used as the current scene classification strategy, and the operation of dividing the historical reference data into different scene categories according to the current scene classification strategy is performed.

In one embodiment, before determining the current scene classification strategy based on the historical reference data, the operation further includes: determining whether there is currently a historical baseline scene classification strategy, and if there is a historical baseline scene classification strategy, then determining according to The historical reference data and the historical reference scene classification strategy determine the current scene classification strategy. If there is no historical reference scene classification strategy, the operation of determining the current scene classification strategy based on the historical reference data is performed.

In one embodiment, the operations further include: displaying information related to the HVAC system; the information related to the HVAC system comes from the current scene data and/or the control of the HVAC system information and/or status information.

In one embodiment, the scene data includes at least one or any combination of the following information: weather information, holiday information, local event information, sensor collections in the building related to HVAC systems and/or building thermal performance information, manually entered information related to HVAC systems within the building. The historical reference data includes historical scene data.

It should be noted that not all steps and modules in the above-mentioned processes and structure diagrams are necessary, and some steps or modules can be ignored according to actual needs. The execution order of each step is not fixed and can be adjusted as needed. The division of each module is only for the convenience of describing the functional division. In actual implementation, one module can be implemented by multiple modules, and the functions of multiple modules can also be implemented by the same module. These modules can be located on the same device. , or it can be on a different device.

It can be understood that the hardware modules in the above embodiments can be implemented mechanically or electronically. For example, a hardware module may include specially designed permanent circuits or logic devices (such as a dedicated processor such as an FPGA or ASIC) to perform specific operations. Hardware modules may also include programmable logic devices or circuits (eg, including general-purpose processors or other programmable processors) temporarily configured by software to perform specific operations. As for the specific use of mechanical means, or the use of dedicated permanent circuits, or the use of temporarily configured circuits (such as configured by software) to implement the hardware modules, it can be decided based on cost and time considerations.

In addition, embodiments of the present application also provide a computer-readable storage medium on which a computer program is stored. The computer program can be executed by a processor and implement the management of the HVAC system described in the embodiments of the present application. method. Specifically, a system or device equipped with a storage medium may be provided, on which the software program code that implements the functions of any of the above embodiments is stored, and the computer (or CPU or MPU) of the system or device ) reads and executes the program code stored in the storage medium. In addition, the operating system operating on the computer can also complete some or all of the actual operations through instructions based on the program code. The program code read from the storage medium can also be written into a memory provided in an expansion board inserted into the computer or into a memory provided in an expansion unit connected to the computer, and then based on the instructions of the program code, the program code is installed in the computer. The CPU on the expansion board or expansion unit performs some and all actual operations, thereby realizing the functions of any of the above embodiments. Storage media implementations for providing program codes include floppy disks, hard disks, magneto-optical disks, optical disks (such as CD-ROM, CD-R, CD-RW, DVD-ROM, DVD-RAM, DVD-RW, DVD+RW), Tapes, non-volatile memory cards and ROM. Alternatively, the program code can be downloaded from the server computer via the communications network.

It can be seen from the above solution that in the embodiment of the present invention, multiple scene categories of the building are determined based on a large amount of historical reference data collected historically, and corresponding BES models are set corresponding to each scene category. After that, the current scene category can be determined based on the currently collected scene data, and the BES model corresponding to the current scene category can be called to determine the optimal control strategy of the HVAC system, and then the HVAC system can be controlled based on the optimal control strategy. The corresponding equipment in the system, such as heating equipment, ventilation equipment and air conditioning equipment, etc., thereby improving the accuracy of the BES model in energy performance prediction and conducive to building energy saving optimization.

The above descriptions are only preferred embodiments of the present invention and are not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of the present invention shall be included in the present invention. within the scope of protection.

Claims (15)

The management method of HVAC system is characterized by including: Determine the current scene category according to the acquired current scene data related to the operation of the HVAC system and the predetermined baseline scene classification strategy (101); Activate the benchmark BES model corresponding to the current scenario category from the predetermined building energy consumption simulation benchmark BES model corresponding to each scenario category to obtain the current benchmark BES model (102); Determine the optimal control strategy of the HVAC system using the current scene data and the current baseline BES model (103); Control corresponding equipment in the HVAC system according to the optimal control strategy (104). The management method of the HVAC system according to claim 1, characterized in that the method further includes: Collect a set number of historically collected historical reference data related to the HVAC system; the set number is greater than a set number threshold (201); Using the historical reference data to calibrate scene-independent parameters in an initial BES model (202); Determine the current scene classification strategy based on the historical reference data (203); According to the current scene classification strategy, the historical reference data is divided into different scene categories, and the reference data of each scene category is used to calibrate the scene-related parameters in the initial BES model to obtain the corresponding scene Candidate BES models for categories (204); The scene classification strategy is determined as the base scene classification strategy, and the candidate BES model is determined as the base BES model. The management method of an HVAC system according to claim 2, characterized in that, before determining the scene classification strategy as the base scene classification strategy and determining the candidate BES model as the base BES model, Further includes: Comprehensively evaluate the scene classification strategy according to any one or any combination of the number of scene categories, the distinction between scene categories, and the simulation accuracy of the candidate BES models corresponding to each scene category (205); When the comprehensive evaluation results meet the set requirements, perform the operation of determining the scene classification strategy as the benchmark scene classification strategy and determining the candidate BES model as the benchmark BES model (207); otherwise, When the comprehensive evaluation result does not meet the set requirements, the set strategy optimization algorithm is used to optimize the scene classification strategy, and the optimized scene classification strategy is used as the current scene classification strategy (208), and returns to execute the based on The current scene classification strategy is an operation of dividing the historical reference data into different scene categories (204). The management method of an HVAC system according to claim 2, wherein the initial BES model is a BES established using known fixed information related to the operation of the HVAC system and an initial estimate of uncertain information. model, or a currently available BES model. The management method of an HVAC system according to claim 2, wherein before determining the current scene classification strategy based on the historical reference data, the method further includes: Determine whether there is currently a historical benchmark scene classification strategy. If there is a historical benchmark scene classification strategy, determine the current scene classification strategy based on the historical reference data and the historical benchmark scene classification strategy. If there is no historical benchmark scene, classification strategy, then perform the operation of determining the current scene classification strategy based on the historical reference data. The management method of the HVAC system according to any one of claims 1 to 5, further comprising: displaying information related to the HVAC system; the information related to the HVAC system comes from the The current scene data, and/or the control information and/or status information of the HVAC system. The management method of an HVAC system according to any one of claims 1 to 5, wherein the scene data includes at least one or any combination of the following information: weather information, holiday information, and local event information. , information collected by sensors in the building related to the HVAC system and/or the thermal performance of the building, information manually entered in the building related to the HVAC system; The historical reference data includes historical scene data. A management system for an HVAC system, characterized in that it includes: at least one memory (61) and at least one processor (62), wherein: The at least one memory (61) is used to store computer programs; The at least one processor (62) is used to call the computer program stored in the at least one memory (61) to cause the device to perform corresponding operations. The operations include: Determine the current scene category based on the acquired current scene data related to the operation of the HVAC system and the predetermined baseline scene classification strategy; Activate the benchmark BES model corresponding to the current scenario category from the predetermined building energy consumption simulation benchmark BES model corresponding to each scenario category to obtain the current benchmark BES model; Determine the optimal control strategy of the HVAC system using the current scene data and the current baseline BES model; Control corresponding equipment in the HVAC system according to the optimal control strategy. The management system of the HVAC system according to claim 8, wherein the operations further include: Collect a set number of historically collected historical reference data related to the HVAC system; the set number is greater than a set number threshold; Using the historical reference data to calibrate scene-independent parameters in an initial BES model; Determine the current scene classification strategy based on the historical reference data; According to the current scene classification strategy, the historical reference data is divided into different scene categories, and the reference data of each scene category is used to calibrate the scene-related parameters in the initial BES model to obtain the corresponding scene Candidate BES models for categories; The scene classification strategy is determined as the base scene classification strategy, and the candidate BES model is determined as the base BES model. The management system of an HVAC system according to claim 9, wherein before determining the scene classification strategy as the base scene classification strategy and determining the candidate BES model as the base BES model, further include: Conduct a comprehensive evaluation of the scene classification strategy based on any one or any combination of the number of scene categories, the distinction between scene categories, and the simulation accuracy of the candidate BES models corresponding to each scene category; When the comprehensive evaluation results meet the set requirements, perform the operation of determining the scene classification strategy as the base scene classification strategy and the candidate BES model as the base BES model; otherwise, When the comprehensive evaluation result does not meet the set requirements, use the set strategy optimization algorithm to optimize the scene classification strategy, use the optimized scene classification strategy as the current scene classification strategy, and return to execute the scene classification strategy based on the current scene classification strategy. The scene classification strategy is an operation of dividing the historical reference data into different scene categories. The management system of an HVAC system according to claim 8, wherein before determining the current scene classification strategy based on the historical reference data, the operation further includes: Determine whether there is currently a historical benchmark scene classification strategy. If there is a historical benchmark scene classification strategy, determine the current scene classification strategy based on the historical reference data and the historical benchmark scene classification strategy. If there is no historical benchmark scene, classification strategy, then perform the operation of determining the current scene classification strategy based on the historical reference data. The management system of an HVAC system according to any one of claims 8 to 11, wherein the operation further includes: displaying information related to the HVAC system; The information comes from the current scene data, and/or control information and/or status information of the HVAC system. The management system of the HVAC system is characterized by including: a data collection module (411) for collecting current scenario data related to the operation of the HVAC system; The scene decision module (412) is used to determine the current scene category of the building based on the current scene data and the predetermined benchmark scene classification strategy; The model determination module (413) is used to activate the benchmark BES model corresponding to the current scenario category from the predetermined benchmark building energy consumption simulation BES model corresponding to each scenario category to obtain the current benchmark BES model; The control strategy determination module (414) is used to call the current baseline BES model based on the set optimal control algorithm and the current scene data, and determine the optimal control strategy of the HVAC system based on the output of the current baseline BES model. ; System control module (415), used to control corresponding equipment in the HVAC system according to the optimal control strategy. The management system of the HVAC system according to claim 13, further comprising: Data recording module (501), used to record a set number of historical reference data related to the HVAC system collected by the data collection module and/or other external modules; the set number is greater than a set number threshold; The first calibration module (502) is used to calibrate scene-independent parameters in an initial BES model using the historical reference data; Classification strategy determination module (503), used to determine the current scene classification strategy based on the historical reference data; A data dividing module (504), configured to divide the historical reference data into different scene categories according to the current scene classification strategy; The second calibration module (505) is used to calibrate the scene-related parameters in the initial BES model using the reference data of each scene category to obtain a candidate BES model corresponding to the scene category; The evaluation module (506) is used to evaluate the scene classification strategy according to any one or any combination of the number of scene categories, the distinction between scene categories, and the simulation accuracy of the BES model corresponding to each scene category. Comprehensive Evaluation; The result determination module (507) is used to determine the scene classification strategy as the benchmark scene classification strategy when the comprehensive evaluation result meets the setting requirements and provide it to the scene decision module (412), and determine the candidate BES model is the benchmark BES model and is provided to the BES model determination module (413); when the comprehensive evaluation result does not meet the set requirements, instruct the strategy optimization module (508) to perform strategy optimization; The strategy optimization module (508) is used to optimize the scene classification strategy using the set strategy optimization algorithm, use the optimized scene classification strategy as the current scene classification strategy, and instruct the data partitioning module (504) to execute the The operation of dividing the historical reference data into different scene categories according to the current scene classification strategy. A computer-readable storage medium with a computer program stored thereon; characterized in that the computer program can be executed by a processor and implement the management method of the HVAC system according to any one of claims 1 to 7.