Intelligent learning and adaptive control method, device and equipment for heating system and medium

Abstract

This application relates to a method, device, equipment, and medium for intelligent learning and adaptive control of heating systems, applied in the field of heating control technology. The method includes: collecting and preprocessing raw heating monitoring data to obtain heating monitoring data; constructing a heat load prediction feature vector and a user behavior feature vector; generating user behavior pattern labels based on a user behavior recognition model and user behavior feature vectors; determining a predicted indoor temperature data sequence based on heating monitoring data and a thermal inertia model; determining a heat load prediction data sequence based on the heat load prediction feature vector, predicted indoor temperature data sequence, meteorological prediction data sequence, behavior pattern labels, and heat load prediction model; and obtaining control reference values ​​for the heat source station, pipeline network, and terminal control based on the heat load prediction data sequence, user behavior pattern labels, heating monitoring data, and global optimization relationships. This application improves the reliability of heating control.

Description

Technical Field

[0001] This application relates to the technical field of heating control, and in particular to a method, apparatus, equipment and medium for intelligent learning and adaptive control of heating systems. Background Technology

[0002] With social development and the improvement of people's living standards, heating systems play a vital role in daily life. The stable operation of heating systems not only affects residents' comfort but also has a profound impact on the rational use of energy and environmental protection. A good heating system can provide people with a warm and comfortable indoor environment, reducing the impact of cold weather on people's lives and health. At the same time, reasonable heating control helps reduce energy consumption and environmental pollution, aligning with the concept of sustainable development. In recent years, heating technology has continuously developed, gradually shifting from traditional, simple heating methods towards intelligent and refined systems to meet heating needs in different scenarios.

[0003] Current heating system control typically employs fixed-parameter control strategies. A common approach is to set fixed heating parameters based on experience, such as fixed water supply temperature and flow rate. However, this method fails to consider real-time changes in the indoor and outdoor environment, as well as users' actual heating needs. Another approach involves monitoring indoor temperature using simple temperature sensors, increasing heating output when the temperature is below a set value and decreasing it when it's above. However, this method lacks comprehensive consideration of multiple factors. Some methods adjust heating based on a general range of outdoor temperatures, but this adjustment is rather coarse and cannot precisely adapt to the individualized needs of different users. While these conventional methods can achieve basic heating functions to some extent, they struggle to meet the complex and ever-changing real-world heating scenarios, resulting in low reliability of heating control. Summary of the Invention

[0004] To improve the reliability of heating control, this application provides a method, device, equipment, and medium for intelligent learning and adaptive control of heating systems.

[0005] Firstly, this application provides an intelligent learning and adaptive control method for a heating system, employing the following technical solution:

[0006] A method for intelligent learning and adaptive control of a heating system, comprising:

[0007] Collect raw heating monitoring data, which includes raw indoor temperature data sequence, raw terminal heating condition data sequence, raw pipeline operation data sequence, raw heat source station operation data sequence, raw outdoor temperature data sequence, and raw meteorological forecast data sequence.

[0008] All the original heating monitoring data are preprocessed to obtain heating monitoring data;

[0009] Based on the heating monitoring data, a heat load prediction feature vector and a user behavior feature vector are constructed.

[0010] The user behavior feature vector is identified by a user behavior recognition model to generate user behavior pattern labels, which include window ventilation mode, room vacancy mode, and energy-saving adjustment mode.

[0011] Input the current indoor temperature data, terminal heating condition data, and outdoor temperature data into the thermal inertia model to obtain the predicted indoor temperature data sequence.

[0012] The heat load prediction feature vector, the predicted indoor temperature data sequence, the meteorological prediction data sequence, and the user behavior pattern label are input into the heat load prediction model to obtain the heat load prediction data sequence within the preset prediction time domain.

[0013] Based on the heat load prediction data sequence and the heating monitoring data, the heat source station control layer, the pipeline network control layer, and the end user control layer respectively generate the corresponding heat source station control quantity, pipeline network control quantity, and end user control quantity;

[0014] Based on the heat load prediction data sequence, the user behavior pattern tags, the heating monitoring data, and the global optimization relationship, the control quantities of the heat source station, the pipeline network, and the terminal control quantities are optimized to obtain the control reference values ​​of the heat source station, the pipeline network, and the terminal control.

[0015] The control reference values ​​for the heat source station, the control reference values ​​for the pipeline network, and the control reference values ​​for the terminal are respectively sent to the corresponding control layers for execution.

[0016] By adopting the above technical solutions, collecting and preprocessing various raw heating monitoring data can improve the accuracy and usability of the data; constructing heat load prediction feature vectors and user behavior feature vectors, and generating user behavior pattern labels by combining user behavior recognition models, can accurately grasp user behavior patterns; using thermal inertia models to obtain predicted indoor temperature data sequences, and combining heat load prediction feature vectors, meteorological prediction data sequences, and user behavior pattern labels into the heat load prediction model to obtain heat load prediction data sequences, can more accurately predict heat load; generating and optimizing control quantities at different control layers can achieve refined control of the heating system, meet the complex and ever-changing actual heating scenarios, and improve the reliability of heating control.

[0017] Optionally, the preprocessing of all the original heating monitoring data to obtain heating monitoring data includes:

[0018] The original heating monitoring data is time-aligned according to a preset sampling period using linear interpolation to obtain unified time-series data.

[0019] Anomaly detection is performed on the unified time-series data based on a preset threshold.

[0020] If outliers exist in the unified time series data, alternative values ​​are generated based on the neighborhood mean, and the outliers are corrected using the alternative values.

[0021] The unified time-series data after outlier processing is normalized to obtain the heating monitoring data.

[0022] By employing the above technical solutions to preprocess raw heating monitoring data, the accuracy, consistency, and usability of the data can be improved. Time alignment using linear interpolation yields unified time-series data, synchronizing data from different sources. Identifying outliers based on preset thresholds and correcting with neighborhood mean removes errors or anomalies from the data. Normalizing the outlier-processed data scales it to a uniform range, facilitating subsequent analysis and processing. This provides a high-quality data foundation for subsequent heat load forecasting and heating control, improving the reliability and accuracy of intelligent control of the heating system.

[0023] Optionally, the heating monitoring data includes indoor temperature data sequences, terminal heat consumption data sequences, terminal valve opening sequences, outdoor temperature data sequences, pipeline operation data sequences, and heat source station operation data sequences. The step of constructing a heat load prediction feature vector and a user behavior feature vector based on the heating monitoring data includes:

[0024] The characteristics of indoor temperature changes are determined based on the indoor temperature data sequence.

[0025] The characteristics of end-user heat variation are determined based on the end-user heat data sequence;

[0026] The characteristics of the end valve opening change are determined based on the end valve opening sequence;

[0027] The user behavior feature vector is generated based on the indoor temperature change characteristics, the terminal heat consumption change characteristics, the terminal valve opening change characteristics, and the outdoor temperature data sequence.

[0028] The heat load prediction feature vector is generated based on the indoor temperature data sequence, the terminal heat consumption data sequence, the terminal valve opening sequence, the outdoor temperature data sequence, the pipeline operation data sequence, and the heat source station operation data sequence.

[0029] By adopting the above technical solution, various types of raw heating monitoring data are collected and preprocessed. Based on the heating monitoring data, a heat load prediction feature vector is generated. The heating monitoring data is further analyzed to determine the variation characteristics of indoor temperature, terminal heat consumption, and terminal valve opening. In turn, a user behavior feature vector is generated. The two feature vectors can comprehensively consider multiple factors and more accurately reflect user behavior patterns and heat load conditions. This provides more comprehensive and accurate data support for subsequent heat load prediction and heating control, improves the reliability and adaptability of heating system control, and meets the needs of complex and ever-changing actual heating scenarios.

[0030] Optionally, the thermal inertia model is constructed based on an equivalent first-order heat capacity model;

[0031] The temperature prediction expression of the thermal inertia model is as follows:

[0032] ,

[0033] in, For the nth sampling time, For the (n+1)th sampling time, Let the indoor temperature be the temperature at the (n+1)th sampling time. Let the indoor temperature be the value at the nth sampling time. For the preset sampling period, For the effective heat capacity of the room, The equivalent heat transfer coefficient between the heat dissipation equipment and the indoor air. For the terminal heat data at the nth sampling time, The equivalent heat transfer coefficient between the building envelope and the external environment. This represents the outdoor temperature data at the nth sampling time.

[0034] By adopting the above technical solution, a thermal inertia model is constructed based on the equivalent first-order heat capacity model. Based on the current indoor temperature, terminal heat consumption and outdoor temperature, the indoor temperature at the next sampling time can be accurately predicted, providing more accurate indoor temperature data for heat load prediction. This improves the reliability and accuracy of heating system control and better adapts to complex and ever-changing actual heating scenarios.

[0035] Optionally, the step of generating corresponding heat source station control quantities, pipeline network control quantities, and terminal user control quantities at the heat source station control layer, pipeline network control layer, and terminal user control layer based on the heat load prediction data sequence and the heating monitoring data, respectively, includes:

[0036] Obtain the target indoor temperature;

[0037] The temperature deviation is determined based on the target indoor temperature and the indoor temperature data sequence.

[0038] The end control quantity is calculated based on the temperature deviation and the PI control algorithm.

[0039] The branch target heat load is calculated based on the preset load weighting coefficient and the heat load prediction data sequence;

[0040] The pipeline control quantity is calculated based on the target heat load of the branch and the preset pipeline calculation formula.

[0041] Calculate the total heat load forecast data based on the heat load forecast data sequence of all end users;

[0042] The control quantity of the heat source station is calculated based on the total heat load prediction data and the preset heat balance formula.

[0043] By adopting the above technical solution, the temperature deviation is determined based on the target indoor temperature and indoor temperature data, and the terminal control quantity is calculated by combining the PI control algorithm. The branch target heat load is calculated by using the preset load weight coefficient and heat load prediction data sequence, and the pipeline control quantity is obtained by combining the preset pipeline calculation formula. The total heat load prediction data is calculated based on the heat load prediction data of all end users, and the heat source station control quantity is obtained by combining the preset heat balance relationship. This can achieve accurate calculation of the control quantity of each control layer of the heating system, comprehensively consider multiple factors, make heating control more precise, meet the complex and ever-changing actual heating scenarios, improve the reliability of heating control, and at the same time help reduce energy consumption, which is in line with the concept of sustainable development.

[0044] Optionally, the global optimization relationship includes temperature deviation, energy input, and control variation.

[0045] The objective function corresponding to the global optimization relationship includes: , Where J is the objective function value, , , , as well as All are control strategy weighting coefficients, where i is the end-user ID and N is the number of end-users. For the indoor temperature data of the i-th end user, Let i be the target indoor temperature for the i-th end user. The actual water supply temperature of the heat source station. This is a reference value for the water supply temperature of the heat source station. This represents the actual circulating flow rate of the heat source station. This is a reference value for the circulating flow rate of the heat source station. For changes in the control parameters of the heat source station, k is the branch number, and K is the number of branches. This refers to changes in branch control variables.

[0046] By adopting the above technical solution, temperature deviation, energy input, and control changes are incorporated into the global optimization relationship. The objective function is used to optimize the control quantities of the heat source station, pipeline network, and terminal control quantities. This allows for a comprehensive consideration of the indoor temperature deviation of terminal users, the differences between the heat source station's water supply temperature and circulation flow rate and reference values, as well as the changes in control quantities. This enables more precise adjustment of the heating system's operating parameters, making the heating system more intelligent and efficient in meeting the personalized needs of different users, improving the reliability of heating control, achieving refined control of the heating system, reducing energy consumption, and minimizing environmental pollution.

[0047] Optionally, the method further includes:

[0048] Based on the deviation between actual operating data and model output, the heat load prediction model, the user behavior recognition model, and the thermal inertia model are adaptively updated.

[0049] By adopting the above technical solutions, the heat load prediction model, user behavior recognition model, and thermal inertia model can be adaptively updated based on the deviation between actual operating data and model output. This allows the models to better adapt to the actual operating conditions of the heating system, improve the reliability and accuracy of heating control, and meet the complex and ever-changing actual heating scenarios and personalized user needs.

[0050] Secondly, this application provides an intelligent learning and adaptive control device for a heating system, which adopts the following technical solution:

[0051] A heating system intelligent learning and adaptive control device, comprising:

[0052] The data acquisition module is used to collect raw heating monitoring data, which includes raw indoor temperature data sequences, raw terminal heating condition data sequences, raw pipeline operation data sequences, raw heat source station operation data sequences, raw outdoor temperature data sequences, and raw meteorological forecast data sequences.

[0053] The preprocessing module is used to preprocess all the original heating monitoring data to obtain heating monitoring data;

[0054] The vector construction module is used to construct heat load prediction feature vectors and user behavior feature vectors based on the heating monitoring data;

[0055] The behavior recognition module is used to identify the user behavior feature vector through the user behavior recognition model and generate user behavior pattern labels, including window ventilation mode, room vacancy mode and energy-saving adjustment mode.

[0056] The temperature prediction module is used to input the current indoor temperature data, terminal heating condition data and outdoor temperature data into the thermal inertia model to obtain the predicted indoor temperature data sequence.

[0057] The load prediction module is used to input the heat load prediction feature vector, the predicted indoor temperature data sequence, the meteorological prediction data sequence, and the user behavior pattern label into the heat load prediction model to obtain the heat load prediction data sequence within a preset prediction time domain.

[0058] The control quantity determination module is used to generate corresponding heat source station control quantities, pipeline network control quantities, and terminal user control quantities at the heat source station control layer, pipeline network control layer, and terminal user control layer, respectively, based on the heat load prediction data sequence and the heating monitoring data.

[0059] The global optimization module is used to optimize the control quantities of the heat source station, the pipeline network, and the terminal control quantities based on the heat load prediction data sequence, the user behavior pattern tags, the heating monitoring data, and the global optimization relationship, so as to obtain the control reference values ​​of the heat source station, the pipeline network, and the terminal control.

[0060] The control distribution module is used to distribute the control reference values ​​of the heat source station, the control reference values ​​of the pipeline network, and the control reference values ​​of the terminal to the corresponding control layers for execution.

[0061] By adopting the above technical solutions, collecting and preprocessing various raw heating monitoring data can improve the accuracy and usability of the data; constructing heat load prediction feature vectors and user behavior feature vectors, and generating user behavior pattern labels by combining user behavior recognition models, can accurately grasp user behavior patterns; using thermal inertia models to obtain predicted indoor temperature data sequences, and combining heat load prediction feature vectors, meteorological prediction data sequences, and user behavior pattern labels into the heat load prediction model to obtain heat load prediction data sequences, can more accurately predict heat load; generating and optimizing control quantities at different control layers can achieve refined control of the heating system, meet the complex and ever-changing actual heating scenarios, and improve the reliability of heating control.

[0062] Thirdly, this application provides an electronic device that adopts the following technical solution:

[0063] An electronic device includes a processor coupled to a memory;

[0064] The memory stores a computer program that can be loaded by a processor and executed by the intelligent learning and adaptive control method for the heating system described in any of the first aspects.

[0065] Fourthly, this application provides a computer-readable storage medium, which adopts the following technical solution:

[0066] A computer-readable storage medium storing a computer program capable of being loaded by a processor and executing the intelligent learning and adaptive control method for a heating system as described in any of the first aspects. Attached Figure Description

[0067] Figure 1 This is a schematic diagram of a heating system provided in an embodiment of this application.

[0068] Figure 2 This is a flowchart illustrating an intelligent learning and adaptive control method for a heating system provided in an embodiment of this application.

[0069] Figure 3 This is a structural block diagram of an intelligent learning and adaptive control device for a heating system provided in an embodiment of this application.

[0070] Figure 4 This is a structural block diagram of the electronic device provided in the embodiments of this application. Detailed Implementation

[0071] To make the objectives, technical solutions, and advantages of the embodiments of this application clearer, the technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this application, not all embodiments. Based on the embodiments of this application, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this application.

[0072] Furthermore, the term "and / or" in this article is merely a description of the relationship between related objects, indicating that three relationships can exist. For example, A and / or B can represent: A existing alone, A and B existing simultaneously, or B existing alone. Additionally, the character " / " in this article, unless otherwise specified, generally indicates that the preceding and following related objects have an "or" relationship.

[0073] like Figure 1 As shown, the heating system includes a heat source station, a pipe network, end-user units, and electronic equipment. The heat source station is the energy supply end of the entire heating system and is connected to the end-user units through the pipe network to realize the transmission and distribution of heat energy. The electronic equipment serves as a central control unit and establishes signal / control connections with the heat source station, the pipe network, and the end-user units respectively to monitor and regulate heating operation parameters.

[0074] A heat source station includes a heat source unit, heat exchange equipment, and a hydraulic circulation system. The heat source unit can be a boiler, district energy supply equipment, or other heat source equipment with heating capabilities. The heat exchange equipment is used to transfer heat from the heat source side to the pipe network and end-user units. The hydraulic circulation system includes components such as circulating pumps and flow regulating valves, used to regulate the water supply pressure and flow rate of the heat source station.

[0075] The pipeline network includes supply and return water networks, used to transport the heating medium from the heat source station. Multiple monitoring nodes are installed within the network, each equipped with temperature sensors, pressure sensors, flow sensors, etc. The network may also include valve actuators for regulating flow or differential pressure, such as electrically operated regulating valves or differential pressure control devices.

[0076] The end-user unit includes indoor heating equipment, terminal valves, and an indoor temperature acquisition device. The heating equipment can be radiators or underfloor heating systems. The terminal valves are automatically regulated by actuators. The indoor temperature acquisition device is used to collect the user's room temperature.

[0077] This application provides an intelligent learning and adaptive control method for a heating system. This method is applied to a heating system and can be executed by electronic devices within the system. These electronic devices can be servers or terminal devices. The server can be a standalone physical server, a server cluster or distributed system composed of multiple physical servers, or a cloud server providing cloud computing services. The terminal device can be a smartphone, tablet, desktop computer, etc., but is not limited to these.

[0078] like Figure 2 As shown, a method for intelligent learning and adaptive control of a heating system is described in the following steps (S101-S109):

[0079] Step S101: Collect raw heating monitoring data.

[0080] The original heating monitoring data includes original indoor temperature data sequences, original terminal heating condition data sequences, original pipeline operation data sequences, original heat source station operation data sequences, original outdoor temperature data sequences, and original meteorological forecast data sequences.

[0081] Each of the above data sequences includes multiple data points at various data acquisition times within a time window (e.g., within 2 hours, without specific limitation here). The raw indoor temperature data sequence can be obtained from the indoor temperature acquisition devices of end users. The raw terminal heating condition data sequence includes raw terminal heat consumption data and raw terminal valve opening sequence. The raw terminal heat consumption data is collected by heat meters, and the raw terminal valve opening sequence is collected by valve opening detection equipment. The raw pipeline operation data sequence includes the raw pipeline supply water temperature, raw pipeline return water temperature, raw pipeline flow rate, and raw pipeline pressure at each data acquisition time, which can be collected by temperature sensors, flow meters, and pressure sensors. The raw heat source station operation data sequence includes the raw heat source station outlet temperature, raw heat source station return water temperature, raw heat source station circulation flow rate, and raw heat source station outlet pressure at each data acquisition time, which can be collected by temperature sensors, flow meters, and pressure sensors. The raw outdoor temperature data sequence and the raw weather forecast data sequence can be obtained from the meteorological system.

[0082] Step S102: Preprocess all raw heating monitoring data to obtain heating monitoring data.

[0083] The heating monitoring data includes indoor temperature data sequences, terminal heating condition data sequences, pipeline operation data sequences, heat source station operation data sequences, outdoor temperature data sequences, and meteorological forecast data sequences. The terminal heating condition data sequences include terminal heat consumption data sequences and terminal valve opening sequences. The pipeline operation data sequences include pipeline supply water temperature sequences, pipeline return water temperature sequences, pipeline flow sequences, and pipeline pressure sequences. The heat source station operation data sequences include heat source station outlet temperature sequences, heat source station return water temperature sequences, heat source station circulating flow sequences, and heat source station outlet pressure sequences.

[0084] Specifically, all raw heating monitoring data are preprocessed to obtain heating monitoring data, including: aligning the raw heating monitoring data according to a preset sampling period using linear interpolation to obtain unified time-series data; judging outliers in the unified time-series data based on a preset threshold; if outliers exist in the unified time-series data, generating alternative values ​​based on the neighborhood mean and correcting the outliers using the alternative values; and normalizing the unified time-series data after outlier processing to obtain heating monitoring data.

[0085] In this embodiment, the sampling periods of data from different sources may be different. To facilitate subsequent data analysis, it is necessary to unify the timestamps of all collected data sequences. The original data sequences in all the original heating monitoring data are time-aligned according to the preset sampling period using linear interpolation to obtain unified time-series data. Outlier judgment is performed on the unified time-series data according to the preset thresholds (including the upper and lower thresholds) corresponding to various data. If the data is higher than the upper threshold or lower than the lower threshold, the data is an outlier. If there are outliers in the unified time-series data, a replacement value is generated based on the neighborhood mean (i.e., the average of a preset number of adjacent data before and after the outlier). The outlier is replaced by the replacement value. The replacement value can also be generated using linear interpolation. Then, the maximum and minimum values ​​in the unified time-series data after outlier processing are used to normalize each data in the sequence to obtain the heating monitoring data.

[0086] Step S103: Construct a heat load prediction feature vector and a user behavior feature vector based on heating monitoring data.

[0087] Specifically, a heat load prediction feature vector and a user behavior feature vector are constructed based on heating monitoring data, including: determining indoor temperature change characteristics based on indoor temperature data sequences; determining terminal heat consumption change characteristics based on terminal heat consumption data sequences; determining terminal valve opening change characteristics based on terminal valve opening sequences; generating a user behavior feature vector based on indoor temperature change characteristics, terminal heat consumption change characteristics, terminal valve opening change characteristics, and outdoor temperature data sequences; and generating a heat load prediction feature vector based on indoor temperature data sequences, terminal heat consumption data sequences, terminal valve opening sequences, outdoor temperature data sequences, pipeline operation data sequences, and heat source station operation data sequences.

[0088] In this embodiment, the difference between two adjacent indoor temperature data points in the indoor temperature data sequence is calculated to obtain the indoor temperature change characteristics; the difference between two adjacent terminal heat consumption data points in the terminal heat consumption data sequence is calculated to obtain the terminal heat consumption change characteristics; the difference between two adjacent terminal valve openings in the terminal valve opening sequence is calculated to obtain the terminal valve opening change characteristics; the user behavior feature vector at a given moment is a vector composed of multiple features including the indoor temperature change characteristics, terminal heat consumption change characteristics, terminal valve opening change characteristics, and outdoor temperature data at that moment. It is worth noting that the above-mentioned change characteristics can also be represented by time window features. For example, the indoor temperature change characteristics can be the average rate of change of data or the maximum change in data within a time window; the terminal valve opening change characteristics can be the frequency of opening changes or the maximum change in opening within a time window; and the terminal heat consumption change characteristics can be the average heat consumption or the slope of heat consumption change within a time window. No specific limitations are made here.

[0089] Each sampling time (determined according to the preset sampling period) corresponds to a heat load prediction feature vector. The heat load prediction feature vector sequence is a vector sequence composed of multiple heat load prediction feature vectors within a time window. The heat load prediction feature vector is a vector composed of multiple features, including indoor temperature data, terminal heat consumption data, terminal valve opening, outdoor temperature data, pipeline operation data (including pipeline supply water temperature, pipeline return water temperature, pipeline flow rate, and pipeline pressure), and heat source station operation data (including original heat source station outlet temperature, original heat source station return water temperature, original heat source station circulation flow rate, and original heat source station outlet pressure).

[0090] Step S104: Identify user behavior feature vectors using a user behavior recognition model to generate user behavior pattern labels.

[0091] The user behavior recognition model can be built using a supervised classification model or an LSTM / GRU model. Examples of supervised classification models include Random Forest, XGBoost, Support Vector Machine, and Multilayer Perceptron. After selecting a model, it is trained using historical user behavior feature vectors labeled with user behavior patterns to obtain the user behavior recognition model. User behavior pattern labels include window ventilation mode, room vacancy mode, and energy-saving mode.

[0092] Input the user behavior feature vector corresponding to each time moment into the user behavior recognition model to obtain the user behavior pattern label at that time moment.

[0093] Step S105: Input the current indoor temperature data, terminal heating condition data and outdoor temperature data into the thermal inertia model to obtain the predicted indoor temperature data sequence.

[0094] Specifically, the thermal inertia model is constructed based on an equivalent first-order heat capacity model. That is, the thermal inertia model uses an equivalent first-order heat capacity model to describe the relationship between indoor temperature changes and heat input and outdoor meteorological conditions. After discretizing the equivalent first-order heat capacity model (i.e., the thermal inertia model) according to a preset sampling period, the resulting temperature prediction expression is: ,in, For the nth sampling time, For the (n+1)th sampling time, Let the indoor temperature be the temperature at the (n+1)th sampling time. Let the indoor temperature be the value at the nth sampling time. For the preset sampling period, For the effective heat capacity of the room, The equivalent heat transfer coefficient between the heat dissipation equipment and the indoor air. For the terminal heat data at the nth sampling time, The equivalent heat transfer coefficient between the building envelope and the external environment. The outdoor temperature data at the nth sampling time, the indoor effective heat capacity, and various equivalent heat transfer coefficients can all be obtained from the database or from staff, or they can be obtained by data fitting calculations of historical data using methods such as least squares or recursive least squares.

[0095] Input the current indoor temperature data, terminal heating condition data, and outdoor temperature data into the thermal inertia model. The thermal inertia model can obtain a sequence of predicted indoor temperature data through the above temperature prediction expression.

[0096] Step S106: Input the heat load prediction feature vector, the predicted indoor temperature data sequence, the meteorological prediction data sequence, and the user behavior pattern label into the heat load prediction model to obtain the heat load prediction data sequence within the preset prediction time domain.

[0097] The heat load prediction model can be constructed using time series models, machine learning models, etc. It is also trained based on historical data. The heat load prediction feature vector (or a vector sequence within a window, depending on the model requirements), the predicted indoor temperature data sequence, the meteorological prediction data sequence, and the user behavior pattern label are input into the heat load prediction model to obtain the heat load prediction data sequence within the preset prediction time domain, that is, the predicted end-use heat values ​​at multiple times in the future.

[0098] Step S107: Generate corresponding heat source station control quantities, pipeline network control quantities, and terminal user control quantities in the heat source station control layer, pipeline network control layer, and terminal user control layer respectively based on the heat load prediction data sequence and heating monitoring data.

[0099] Specifically, the heat source station control layer, pipeline network control layer, and end-user control layer generate corresponding heat source station control quantities, pipeline network control quantities, and end-user control quantities based on the heat load prediction data sequence and heating monitoring data, respectively. This includes: obtaining the target indoor temperature; determining the temperature deviation based on the target indoor temperature and the indoor temperature data sequence; calculating the end-user control quantity based on the temperature deviation and the PI control algorithm; calculating the branch target heat load based on the preset load weight coefficient and the heat load prediction data sequence; calculating the pipeline network control quantity based on the branch target heat load and the preset pipeline network calculation formula; calculating the total heat load prediction data based on the heat load prediction data sequence of all end users; and calculating the heat source station control quantity based on the total heat load prediction data and the preset heat balance relationship.

[0100] In this embodiment, the end control quantity includes the end valve control quantity, which is obtained from the staff to obtain the target indoor temperature; temperature deviation = target indoor temperature - indoor temperature data; the end control quantity at each time is calculated based on the temperature deviation at each time and the PI control algorithm (a conventional method in the field, which will not be described in detail here).

[0101] The predicted heat load data of all end users within a pipeline branch are summed according to time intervals to obtain the branch heat load prediction data of each pipeline branch at each time interval, i.e., the branch heat load prediction data sequence. Each branch heat load prediction data at each time interval in the branch heat load prediction data sequence corresponds to a preset load weighting coefficient (used to adjust the degree of influence of different prediction steps, which is not specifically limited here). The branch target heat load of all end users corresponding to a pipeline branch is calculated by weighted summation. The pipeline control variables include the pipeline target flow rate and / or the pipeline target pressure difference. The preset pipeline calculation formulas include: pipeline target flow rate of each branch = branch target heat load / (specific heat capacity of heat medium × pipeline supply and return water temperature difference), and pipeline target pressure difference = heat medium resistance coefficient × the square of pipeline target flow rate.

[0102] The control parameters of the heat source station include the reference value of the heat source station supply water temperature, and the summation of the heat load prediction data of all end users to obtain the total heat load prediction data. The preset heat balance relationship includes: reference value of heat source station supply water temperature = heat source station return water temperature + total heat load prediction data / (specific heat capacity of heat medium × circulation flow rate of heat source station), where the specific heat capacity of heat medium can be obtained from the database or from the staff.

[0103] Step S108: Optimize the control quantities of heat source stations, pipeline networks, and terminal units based on the heat load prediction data sequence, user behavior pattern tags, heating monitoring data, and global optimization relationships to obtain the control reference values ​​of heat source stations, pipeline networks, and terminal units.

[0104] The data involved in the global optimization relationship is divided into three levels: heat source station, pipeline network, and terminal. The control reference values ​​of the heat source station include the heat source station circulation flow reference value and the heat source station water supply temperature reference value. The control reference values ​​of the pipeline network include the pipeline network target flow reference value and the pipeline network target differential pressure reference value. The terminal control reference value includes the target indoor temperature.

[0105] Specifically, the global optimization relationship includes temperature deviation, energy input, and control variations, and the objective function corresponding to the global optimization relationship includes: , Where J is the objective function value, For temperature deviation, For energy input, To control the amount of change. , , , as well as All are control strategy weight coefficients. , It can be adjusted according to different user behavior pattern tags. , Relationship with user behavior pattern tags , as well as All settings are pre-defined and are not specifically limited here. i represents the end-user ID, and N represents the number of end-users. For the indoor temperature data of the i-th end user, Let i be the target indoor temperature for the i-th end user. The actual water supply temperature of the heat source station. This is a reference value for the water supply temperature of the heat source station. This represents the actual circulating flow rate of the heat source station. This is a reference value for the circulating flow rate of the heat source station. To describe the changes in the control parameters of the heat source station, various data types can be used, such as the change in the water supply temperature of the heat source station. No specific limitation is made here; k is the branch (pipeline) number, and K is the number of branches. For changes in branch control quantities, various types of data can be used to describe the branch control quantities, such as changes in valve opening, without specific limitations here.

[0106] The objective function is solved based on the preset constraints (e.g., the target temperature does not exceed the preset target temperature threshold, the valve opening does not exceed the preset opening threshold, etc.). During the solution process, the control quantity obtained in step S107 (e.g., the reference value of the heat source station water supply temperature) can be used as the initial value. The parameter combination with the minimum objective function value is determined as the objective solution. The objective solution includes the target indoor temperature, the reference value of the heat source station circulation flow, the reference value of the heat source station water supply temperature, the branch valve opening, etc. Based on the objective solution, the target flow reference value and the target pressure difference reference value of the pipeline network are indirectly calculated according to the preset pipeline network calculation formula in step S107.

[0107] Step S109: Send the control reference values ​​of the heat source station, the control reference values ​​of the pipeline network, and the control reference values ​​of the terminal to the corresponding control layers for execution.

[0108] After obtaining the control reference values ​​for the heat source station, the pipeline network, and the terminal control, the reference values ​​are sent to the corresponding control layers for execution, thereby completing the intelligent control of the heating system.

[0109] Furthermore, the method also includes: adaptively updating the heat load prediction model, user behavior recognition model, and thermal inertia model based on the deviation between actual operating data and model output.

[0110] In this embodiment, the parameters in the heat load prediction model, user behavior recognition model, and thermal inertia model are all adaptively updated in real time. The adaptive update is based on the deviation between the actual operating data and the model output, and the update method is recursive. Each parameter in the model is represented in vector form.

[0111] For example: The parameter vector of the heat load prediction model at the current moment = the parameter vector at the previous moment + the first learning rate parameter × the deviation between the actual running data and the model output × the heat load prediction feature vector at the previous moment;

[0112] The parameter vector of the user behavior recognition model at the current moment = the parameter vector at the previous moment + the second learning rate parameter × the deviation between the actual running data and the model output × the user behavior feature vector at the previous moment.

[0113] The parameter vector of the thermal inertia model at the current moment = the parameter vector at the previous moment + the third learning rate parameter × the deviation between the actual running data and the model output × the update vector;

[0114] The first learning rate parameter, the second learning rate parameter, the third learning rate parameter, and the update vector are all preset, and no specific limitations are made here.

[0115] Figure 3 This is a structural block diagram of a heating system intelligent learning and adaptive control device 200 provided in an embodiment of this application.

[0116] like Figure 3 As shown, the intelligent learning and adaptive control device 200 for the heating system mainly includes:

[0117] The data acquisition module 201 is used to collect raw heating monitoring data, which includes raw indoor temperature data sequence, raw terminal heating condition data sequence, raw pipeline operation data sequence, raw heat source station operation data sequence, raw outdoor temperature data sequence, and raw meteorological forecast data sequence.

[0118] Preprocessing module 202 is used to preprocess all raw heating monitoring data to obtain heating monitoring data;

[0119] Vector construction module 203 is used to construct heat load prediction feature vectors and user behavior feature vectors based on heating monitoring data;

[0120] The behavior recognition module 204 is used to identify user behavior feature vectors through a user behavior recognition model and generate user behavior pattern labels, which include window ventilation mode, room unoccupied mode and energy-saving adjustment mode.

[0121] The temperature prediction module 205 is used to input the current indoor temperature data, terminal heating condition data and outdoor temperature data into the thermal inertia model to obtain the predicted indoor temperature data sequence.

[0122] The load prediction module 206 is used to input the heat load prediction feature vector, the predicted indoor temperature data sequence, the meteorological prediction data sequence, and the user behavior pattern label into the heat load prediction model to obtain the heat load prediction data sequence within the preset prediction time domain.

[0123] The control quantity determination module 207 is used to generate corresponding heat source station control quantities, pipeline network control quantities, and terminal control quantities at the heat source station control layer, pipeline network control layer, and terminal user control layer, respectively, based on the heat load prediction data sequence and heating monitoring data.

[0124] The global optimization module 208 is used to optimize the control quantities of heat source stations, pipeline networks, and terminal control quantities based on the heat load prediction data sequence, user behavior pattern tags, heating monitoring data, and global optimization relationships, so as to obtain the control reference values ​​of heat source stations, pipeline networks, and terminal control.

[0125] The control distribution module 209 is used to distribute the control reference values ​​of the heat source station, the pipeline network, and the terminal control reference values ​​to the corresponding control layers for execution.

[0126] As an optional implementation of this embodiment, the preprocessing module 202 is specifically used to preprocess all the original heating monitoring data to obtain heating monitoring data, including: aligning the original heating monitoring data according to a preset sampling period based on linear interpolation to obtain unified time-series data; judging outliers in the unified time-series data based on a preset threshold; if there are outliers in the unified time-series data, generating a substitute value based on the neighborhood mean, and correcting the outliers through the substitute value; and normalizing the unified time-series data after outlier processing to obtain heating monitoring data.

[0127] As an optional implementation of this embodiment, the heating monitoring data includes indoor temperature data sequences, terminal heat consumption data sequences, terminal valve opening sequences, outdoor temperature data sequences, pipeline operation data sequences, and heat source station operation data sequences. The vector construction module 203 is specifically used to construct heat load prediction feature vectors and user behavior feature vectors based on the heating monitoring data, including: determining indoor temperature change characteristics based on the indoor temperature data sequences; determining terminal heat consumption change characteristics based on the terminal heat consumption data sequences; determining terminal valve opening change characteristics based on the terminal valve opening sequence; generating user behavior feature vectors based on indoor temperature change characteristics, terminal heat consumption change characteristics, terminal valve opening change characteristics, and outdoor temperature data sequences; and generating heat load prediction feature vectors based on indoor temperature data sequences, terminal heat consumption data sequences, terminal valve opening sequences, outdoor temperature data sequences, pipeline operation data sequences, and heat source station operation data sequences.

[0128] As an optional implementation of this embodiment, the thermal inertia model is constructed based on an equivalent first-order heat capacity model; the temperature prediction expression of the thermal inertia model is: ,in, For the nth sampling time, For the (n+1)th sampling time, Let the indoor temperature be the temperature at the (n+1)th sampling time. Let the indoor temperature be the value at the nth sampling time. For the preset sampling period, For the effective heat capacity of the room, The equivalent heat transfer coefficient between the heat dissipation equipment and the indoor air. For the terminal heat data at the nth sampling time, The equivalent heat transfer coefficient between the building envelope and the external environment. This represents the outdoor temperature data at the nth sampling time.

[0129] As an optional implementation of this embodiment, the control quantity determination module 207 is specifically used to generate corresponding heat source station control quantities, pipeline control quantities, and terminal control quantities in the heat source station control layer, pipeline control layer, and terminal user control layer respectively based on the heat load prediction data sequence and heating monitoring data. This includes: acquiring the target indoor temperature; determining the temperature deviation based on the target indoor temperature and the indoor temperature data sequence; calculating the terminal control quantity based on the temperature deviation and the PI control algorithm; calculating the branch target heat load based on the preset load weighting coefficient and the heat load prediction data sequence; calculating the pipeline control quantity based on the branch target heat load and the preset pipeline calculation formula; calculating the total heat load prediction data based on the heat load prediction data sequence of all terminal users; and calculating the heat source station control quantity based on the total heat load prediction data and the preset heat balance relationship.

[0130] As an optional implementation of this embodiment, the global optimization relationship includes temperature deviation, energy input, and control variation, and the objective function corresponding to the global optimization relationship includes: , Where J is the objective function value, , , , as well as All are control strategy weighting coefficients, where i is the end-user ID and N is the number of end-users. For the indoor temperature data of the i-th end user, Let i be the target indoor temperature for the i-th end user. The actual water supply temperature of the heat source station. This is a reference value for the water supply temperature of the heat source station. This represents the actual circulating flow rate of the heat source station. This is a reference value for the circulating flow rate of the heat source station. For changes in the control parameters of the heat source station, k is the branch number, and K is the number of branches. This refers to changes in branch control variables.

[0131] As an optional implementation of this embodiment, the intelligent learning and adaptive control device 200 for heating systems is also specifically used to: adaptively update the heat load prediction model, user behavior recognition model, and thermal inertia model based on the deviation between actual operating data and model output.

[0132] In one example, the module in any of the above devices may be one or more integrated circuits configured to implement the above methods, such as one or more application-specific integrated circuits (ASICs), or one or more digital signal processors (DSPs), or one or more field-programmable gate arrays (FPGAs), or a combination of at least two of these integrated circuit forms.

[0133] For example, when modules in a device can be implemented via a processing element scheduler, the processing element can be a general-purpose processor, such as a central processing unit (CPU) or other processor capable of calling programs. Alternatively, these modules can be integrated together as a system-on-a-chip (SOC).

[0134] Those skilled in the art will clearly understand that, for the sake of convenience and brevity, the specific working process of the above-described device and module can be referred to the corresponding process in the foregoing method embodiments, and will not be repeated here.

[0135] Figure 4 This is a structural block diagram of an electronic device 300 provided in an embodiment of this application.

[0136] like Figure 4 As shown, the electronic device 300 includes a processor 301 and a memory 302, and may further include one or more of an information input / output (I / O) interface 303, a communication component 304, and a communication bus 305.

[0137] The processor 301 controls the overall operation of the electronic device 300 to complete all or part of the steps of the aforementioned intelligent learning and adaptive control method for the heating system. The memory 302 stores various types of data to support the operation of the electronic device 300. This data may include, for example, instructions for any application or method operating on the electronic device 300, as well as application-related data. The memory 302 can be implemented by any type of volatile or non-volatile storage device or a combination thereof, such as one or more of the following: Static Random Access Memory (SRAM), Electrically Erasable Programmable Read-Only Memory (EEPROM), Erasable Programmable Read-Only Memory (EPROM), Programmable Read-Only Memory (PROM), Read-Only Memory (ROM), magnetic storage, flash memory, magnetic disk, or optical disk.

[0138] I / O interface 303 provides an interface between processor 301 and other interface modules, such as keyboards, mice, and buttons. These buttons can be virtual or physical. Communication component 304 is used for wired or wireless communication between electronic device 300 and other devices. Wireless communication includes Wi-Fi, Bluetooth, Near Field Communication (NFC), 2G, 3G, or 4G, or a combination thereof. Therefore, the corresponding communication component 304 may include a Wi-Fi component, a Bluetooth component, and an NFC component.

[0139] The electronic device 300 may be implemented by one or more application-specific integrated circuits (ASICs), digital signal processors (DSPs), digital signal processing devices (DSPDs), programmable logic devices (PLDs), field-programmable gate arrays (FPGAs), controllers, microcontrollers, microprocessors, or other electronic components to execute the intelligent learning and adaptive control method for the heating system given in the above embodiments.

[0140] The communication bus 305 may include a path for transmitting information between the aforementioned components. The communication bus 305 may be a PCI (Peripheral Component Interconnect) bus or an EISA (Extended Industry Standard Architecture) bus, etc. The communication bus 305 may be divided into an address bus, a data bus, a control bus, etc.

[0141] Electronic device 300 may include, but is not limited to, mobile terminals such as mobile phones, laptops, digital radio receivers, PDAs (personal digital assistants), PADs (tablet computers), PMPs (portable multimedia players), and in-vehicle terminals (such as in-vehicle navigation terminals), as well as fixed terminals such as digital TVs and desktop computers, and may also be servers.

[0142] This application also provides a computer-readable storage medium storing a computer program, which, when executed by a processor, implements the steps of the above-described intelligent learning and adaptive control method for a heating system.

[0143] The computer-readable storage medium may include various media capable of storing program code, such as USB flash drives, portable hard drives, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks.

[0144] The terms “comprising,” “including,” or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such process, method, article, or apparatus.

[0145] The above description is merely a preferred embodiment of this application and an explanation of the technical principles employed. Those skilled in the art should understand that the scope of this application is not limited to technical solutions formed by specific combinations of the above-described technical features, but should also cover other technical solutions formed by arbitrary combinations of the above-described technical features or their equivalents without departing from the foregoing application concept. For example, technical solutions formed by substituting the above features with (but not limited to) technical features with similar functions claimed in this application.

Claims

1. A method for intelligent learning and adaptive control of a heating system, characterized in that, include: Collect raw heating monitoring data, which includes raw indoor temperature data sequence, raw terminal heating condition data sequence, raw pipeline operation data sequence, raw heat source station operation data sequence, raw outdoor temperature data sequence, and raw meteorological forecast data sequence. All the original heating monitoring data are preprocessed to obtain heating monitoring data; Based on the heating monitoring data, a heat load prediction feature vector and a user behavior feature vector are constructed. The user behavior feature vector is identified by a user behavior recognition model to generate user behavior pattern labels, which include window ventilation mode, room vacancy mode, and energy-saving adjustment mode. Input the current indoor temperature data, terminal heating condition data, and outdoor temperature data into the thermal inertia model to obtain the predicted indoor temperature data sequence. The heat load prediction feature vector, the predicted indoor temperature data sequence, the meteorological prediction data sequence, and the user behavior pattern label are input into the heat load prediction model to obtain the heat load prediction data sequence within the preset prediction time domain. Based on the heat load prediction data sequence and the heating monitoring data, the heat source station control layer, the pipeline network control layer, and the end user control layer respectively generate the corresponding heat source station control quantity, pipeline network control quantity, and end user control quantity; Based on the heat load prediction data sequence, the user behavior pattern tags, the heating monitoring data, and the global optimization relationship, the control quantities of the heat source station, the pipeline network, and the terminal control quantities are optimized to obtain the control reference values ​​of the heat source station, the pipeline network, and the terminal control. The control reference values ​​for the heat source station, the control reference values ​​for the pipeline network, and the control reference values ​​for the terminal are respectively sent to the corresponding control layers for execution.

2. The method of claim 1, wherein, The preprocessing of all the original heating monitoring data to obtain heating monitoring data includes: The original heating monitoring data is time-aligned according to a preset sampling period using linear interpolation to obtain unified time-series data. Anomaly detection is performed on the unified time-series data based on a preset threshold. If outliers exist in the unified time series data, alternative values ​​are generated based on the neighborhood mean, and the outliers are corrected using the alternative values. The unified time-series data after outlier processing is normalized to obtain the heating monitoring data.

3. The method of claim 1, wherein, The heating monitoring data includes indoor temperature data sequences, terminal heat consumption data sequences, terminal valve opening sequences, outdoor temperature data sequences, pipeline operation data sequences, and heat source station operation data sequences. The construction of heat load prediction feature vectors and user behavior feature vectors based on the heating monitoring data includes: The characteristics of indoor temperature changes are determined based on the indoor temperature data sequence. The characteristics of end-user heat variation are determined based on the end-user heat data sequence; The characteristics of the end valve opening change are determined based on the end valve opening sequence; The user behavior feature vector is generated based on the indoor temperature change characteristics, the terminal heat consumption change characteristics, the terminal valve opening change characteristics, and the outdoor temperature data sequence. The heat load prediction feature vector is generated based on the indoor temperature data sequence, the terminal heat consumption data sequence, the terminal valve opening sequence, the outdoor temperature data sequence, the pipeline operation data sequence, and the heat source station operation data sequence.

4. The method of claim 1, wherein, The thermal inertia model is constructed based on an equivalent first-order heat capacity model; The temperature prediction expression of the thermal inertia model is as follows: ， wherein, is the nth sampling time, is the (n+1)th sampling time, is the indoor temperature at the (n+1)th sampling time, is the indoor temperature at the nth sampling time, is the preset sampling period, is the effective indoor thermal capacity, is the equivalent heat transfer coefficient between the heat dissipation device and the indoor air, is the end-use heat data at the nth sampling time, is the equivalent heat transfer coefficient between the building envelope and the external environment, is the outdoor temperature data at the nth sampling time.

5. The method of claim 1, wherein, The generation of corresponding heat source station control quantities, pipeline network control quantities, and terminal user control quantities at the heat source station control layer, pipeline network control layer, and terminal user control layer based on the heat load prediction data sequence and the heating monitoring data, respectively, includes: Obtain the target indoor temperature; The temperature deviation is determined based on the target indoor temperature and the indoor temperature data sequence. The terminal control quantity is calculated based on the temperature deviation and the PI control algorithm. The branch target heat load is calculated based on the preset load weighting coefficient and the heat load prediction data sequence; The pipeline control quantity is calculated based on the target heat load of the branch and the preset pipeline calculation formula. Calculate the total heat load forecast data based on the heat load forecast data sequence of all end users; The control quantity of the heat source station is calculated based on the total heat load prediction data and the preset heat balance formula.

6. The method of claim 1, wherein, The global optimization relationship includes temperature deviation, energy input, and control variations; The objective function corresponding to the global optimization relationship includes: ， Where J is the objective function value, , , , as well as All are control strategy weighting coefficients, where i is the end-user ID and N is the number of end-users. For the indoor temperature data of the i-th end user, Let i be the target indoor temperature for the i-th end user. The actual water supply temperature of the heat source station. This is a reference value for the water supply temperature of the heat source station. This represents the actual circulating flow rate of the heat source station. This is a reference value for the circulating flow rate of the heat source station. For changes in the control parameters of the heat source station, k is the branch number, and K is the number of branches. This refers to changes in branch control variables.

7. The method of claim 1, wherein, The method further includes: Based on the deviation between actual operating data and model output, the heat load prediction model, the user behavior recognition model, and the thermal inertia model are adaptively updated.

8. An intelligent learning and adaptive control device for a heating system, characterized in that, include: The data acquisition module is used to collect raw heating monitoring data, which includes raw indoor temperature data sequences, raw terminal heating condition data sequences, raw pipeline operation data sequences, raw heat source station operation data sequences, raw outdoor temperature data sequences, and raw meteorological forecast data sequences. The preprocessing module is used to preprocess all the original heating monitoring data to obtain heating monitoring data; The vector construction module is used to construct heat load prediction feature vectors and user behavior feature vectors based on the heating monitoring data; The behavior recognition module is used to identify the user behavior feature vector through the user behavior recognition model and generate user behavior pattern labels, including window ventilation mode, room vacancy mode and energy-saving adjustment mode. The temperature prediction module is used to input the current indoor temperature data, terminal heating condition data and outdoor temperature data into the thermal inertia model to obtain the predicted indoor temperature data sequence. The load prediction module is used to input the heat load prediction feature vector, the predicted indoor temperature data sequence, the meteorological prediction data sequence, and the user behavior pattern label into the heat load prediction model to obtain the heat load prediction data sequence within a preset prediction time domain. The control quantity determination module is used to generate corresponding heat source station control quantities, pipeline network control quantities, and terminal user control quantities at the heat source station control layer, pipeline network control layer, and terminal user control layer, respectively, based on the heat load prediction data sequence and the heating monitoring data. The global optimization module is used to optimize the control quantities of the heat source station, the pipeline network, and the terminal control quantities based on the heat load prediction data sequence, the user behavior pattern tags, the heating monitoring data, and the global optimization relationship, so as to obtain the control reference values ​​of the heat source station, the pipeline network, and the terminal control. The control distribution module is used to distribute the control reference values ​​of the heat source station, the control reference values ​​of the pipeline network, and the control reference values ​​of the terminal to the corresponding control layers for execution.

9. An electronic device, comprising: Includes a processor, which is coupled to a memory; The processor is configured to execute a computer program stored in the memory to cause the electronic device to perform the method as described in any one of claims 1 to 7.

10. A computer-readable storage medium, characterized in that, It includes a computer program or instructions that, when run on a computer, cause the computer to perform the method as described in any one of claims 1 to 7.

Images