Coordinated Control Method for Heating Systems Based on Air Source Heat Pumps and Building Heat Exchanger Units

Abstract

This invention discloses a coordinated control method for a heating system based on air source heat pumps and building heat exchange units, comprising: setting up a composite heating system consisting of air source heat pump units, building heat exchange units, heating pipe networks, and buildings, and establishing a digital twin model of the composite heating system; based on the time-period heat load forecast values ​​of each building, peak-valley electricity pricing strategies, and the time-period heating capacity range of the air source heat pump units, with the objective function of minimizing the comprehensive heating cost of each building, solving for the optimal heating mode of the composite heating system for each building in each time period and the optimized load ratio of the air source heat pump units and building heat exchange units under the composite heating mode; establishing a supply and return water temperature control model for the air source heat pump units and a pump and valve control model for the building heat exchange units to obtain the control strategies of the air source heat pump units and building heat exchange units under the corresponding heating modes.

Description

Technical Field

[0001] This invention belongs to the field of smart heating technology, specifically relating to a coordinated control method for a heating system based on an air source heat pump and a building heat exchange unit. Background Technology

[0002] Traditional heating methods in most parts of my country mainly include district boiler rooms, combined heat and power (CHP), and electric heating. These methods all have varying degrees of problems, such as low energy efficiency, serious pollution, unreasonable use of high-grade energy, and low utilization rate of heating equipment. There is an urgent need to find a new, environmentally friendly, efficient, and rational heating model for buildings. Air source heat pumps are a highly efficient, environmentally friendly, and energy-saving heating device. Based on the reverse Carnot cycle principle, this device can extract heat from low-temperature outdoor air and use it for building heating. It can significantly reduce pollution from traditional energy heating methods and conserve non-renewable energy. In recent years, it has received high attention from national energy conservation and environmental protection departments.

[0003] However, the heating capacity and efficiency of air source heat pumps decrease as the weather temperature drops, while the building load increases, leading to a situation where the heat supply cannot meet the building's heat demand. Furthermore, air source heat pumps are prone to frosting in low temperatures, further reducing their coefficient of performance (COP) and increasing energy consumption. To address the shortcomings of air source heat pumps, a common approach is to integrate them with heating networks deployed in conjunction with other heat sources. This allows the heat from the heating network to compensate for the heat loss from the air source heat pump. However, how to precisely coordinate the integration of air source heat pumps and heating networks to meet the building's heating needs remains a pressing issue.

[0004] Based on the above technical problems, a new coordinated control method for heating systems based on air source heat pumps and building heat exchange units needs to be designed. Summary of the Invention

[0005] The technical problem this invention aims to solve is to overcome the shortcomings of existing technologies and provide a coordinated control method for a heating system based on air source heat pumps and building heat exchange units. This method allows for the selection of heating modes based on peak and off-peak electricity prices, the heating capacity range of the air source heat pump units, and load value intervals. It can utilize both individual air source heat pump units and building heat exchange units for heating, effectively applying their heating capacities. A combined heating approach, where the building heat exchange units draw heat from the existing heating network by activating their pump valves, not only meets the building's heating needs but also offers advantages such as cleaner heating and higher economic efficiency. Furthermore, by establishing a control model for the corresponding heating mode, the accuracy of heating is improved, allowing for pre-emptive, time-segmented control of the heating system.

[0006] To solve the above-mentioned technical problems, the technical solution of the present invention is as follows:

[0007] This invention provides a coordinated control method for a heating system based on an air-source heat pump and a building heat exchanger unit, comprising:

[0008] Step S1: Install building heat exchange units at the heat inlet of each building and install air source heat pump units on the top floor of the building to form a composite heating system of air source heat pump units, building heat exchange units, heating network and building, and establish a digital twin model of the composite heating system.

[0009] Step S2: Based on weather data and historical load data of each building, predict the heat load demand of each building in different time periods;

[0010] Step S3: Based on the time-sharing heat load forecast of each building, peak-valley electricity pricing strategy, and time-sharing heating capacity range of air source heat pump units, with the objective function of minimizing the comprehensive heating cost of each building, solve to obtain the optimal heating mode of the composite heating system of each building for each time period and the optimized load ratio of air source heat pump units and building heat exchange units under the composite heating mode.

[0011] Step S4: Based on the heating mode of the composite heating system and the optimized load ratio of air source heat pump units and building heat exchange units under the composite heating mode, as well as the time-period heat load prediction values ​​of each building, establish the air source heat pump unit supply and return water temperature control model and the building heat exchange unit pump valve control model to obtain the control strategies of air source heat pump units and building heat exchange units under the corresponding heating modes.

[0012] Furthermore, step S1 includes:

[0013] S101. A building heat exchange unit is installed at the heat inlet of each building, and multiple air source heat pump units are installed on the top floor of the building, forming a load heating system consisting of air source heat pump units, building heat exchange units, the existing heating network, and the building; the air source heat pump units are connected in parallel and include evaporators, compressors, condensers, and expansion valves; the building heat exchange units are connected to the existing heating network and obtain heat supply by adjusting valve opening and pump frequency;

[0014] S102. Construct a virtual entity of a composite heating system, including air source heat pump units, building heat exchange units, heating pipe networks, and building structures, and establish a digital twin model of the composite heating system after connecting virtual and physical data, including:

[0015] A structural model, physical equipment entity model, behavioral model, and rule model of the composite heating system are constructed. The physical equipment entity model is obtained by adding physical attributes of the equipment. Based on the basic functional theory, a behavioral model is constructed to establish a virtual simulation system of the composite heating system with interactive functions and a simulated real operating environment. Finally, a rule model of the virtual entities is established to formulate the control strategy of the virtual entities.

[0016] By collecting actual operating data of physical equipment in the composite heating system to drive corresponding virtual equipment, a mapping relationship between virtual and real data is established to form a composite heating system operation strategy; by continuously iterating and optimizing the data acquisition and control process, the connection and dynamic interaction of real-time data between physical entities and virtual space are realized to establish a digital twin model of the composite heating system.

[0017] S103. Identify the digital twin model of the composite heating system, including:

[0018] The real-time operating data of the composite heating system under multiple operating conditions are integrated into the established digital twin model. The simulation results of the digital twin model are adaptively identified and corrected using the reverse identification method to obtain the identified and corrected digital twin model of the composite heating system.

[0019] Furthermore, step S2 includes:

[0020] Historical load data and local weather data for each building at different times during the early and late stages of heating are obtained. After data cleaning, missing value completion, and normalization preprocessing, the processed data are divided into training set, validation set, and test set. The weather data includes outdoor temperature, wind direction, humidity, precipitation, total horizontal radiation, and surface albedo.

[0021] Select load and weather data series from the same historical time period and calculate the MIC value between the two series;

[0022] The calculated MIC values ​​are approximated by symbolic aggregation and the correlation mean is solved. After sorting the mean values ​​of each weather variable, the input variables of the load forecasting model are selected.

[0023] The training and validation sets are input into the improved LSTNet model for training and learning, and a load prediction model for each building is established to obtain the heat load prediction values ​​of each building at different times.

[0024] Based on the predicted heat load of each building at different times, the target values ​​of the supply and return water temperatures that meet the comfort conditions of the building are calculated by simulating the digital twin model of the composite heating system under different predicted heat load conditions.

[0025] Furthermore, the calculation of the MIC value between the two sequences includes:

[0026] The mutual information MI between the load data series L and the weather data series C is calculated and expressed as:

[0027]

[0028] p(l,c) is the joint probability density of the load data sequence L and the weather data sequence C; p(l) and p(c) are the marginal probability densities of the load data sequence L and the weather data sequence C, respectively.

[0029] Calculate the mutual information between the load data sequence L and the weather data sequence C in grid G, and iterate through all grids to obtain the largest MI value for a given integer (x, y), expressed as:

[0030] f′ MI (D,G(x,y))=max f MI (D,G(x,y));

[0031] f MI (D,G(x,y)) represents the MI value of dataset D=(L,C) on G(x,y);

[0032] The maximum MI value in different grids is normalized and expressed as:

[0033]

[0034] Take all M G(x,y) The maximum value of is taken as the MIC value of the load data series L and the weather data series C, and expressed as:

[0035]

[0036] B(n) is a function of n.

[0037] Furthermore, the step of performing symbolic aggregation approximation and correlation mean calculation on the calculated MIC values ​​includes:

[0038] The MIC is quantized using a piecewise aggregation approximation method, expressed as:

[0039]

[0040] y i k is the average value of the i-th segment; i Let x be the i-th time breakpoint; j The value of the j-th MIC;

[0041] The sequence after segmented aggregation approximation is symbolized and converted into discrete strings;

[0042] Analyze the occurrence frequency of each character, calculate the average of all characters to obtain the correlation, expressed as:

[0043]

[0044] β i For correlation; f k S is the number of times the k-th character appears; k Let N be the mathematically normalized value of the k-th character; P is the total number of occurrences of all characters; N is the total number of occurrences of all characters. i The number of characters represented by the MIC symbol.

[0045] Furthermore, the LSTNet model includes convolutional layers, recurrent layers and recurrent skip layers, fully connected layers, and an autoregressive model;

[0046] The improved LSTNet model includes:

[0047] The channel attention mechanism is introduced into the convolutional layer. Different channels are assigned corresponding weights, and then the channel information is weighted and summed to obtain the output of the convolutional layer.

[0048] Bidirectional recurrent neural network (BiLSTM) is used to replace the basic units of recurrent layers and recurrent skip layers, while simultaneously acquiring bidirectional data information;

[0049] The temporal attention mechanism is introduced into the recurrent layer and the recurrent skip layer. After the initial state vector is obtained by processing each feature vector output by BiLSTM through the attention mechanism, the final output vector is obtained by weighted summation through weight coefficients.

[0050] Furthermore, step S3 includes:

[0051] Based on weather data and air source heat pump performance, the maximum and minimum heating values ​​of air source heat pump units equipped in each building are predicted in different time periods to obtain the time period heating capacity range of air source heat pump units.

[0052] Based on the time-period heat load forecasts for each building, peak-valley electricity pricing strategies, and the time-period heating capacity range of air source heat pump units, the objective function is to minimize the overall heating cost for each building.

[0053] Set constraints for the composite heating system, including: heat power balance constraints, air source heat pump unit heating output constraints, and building heat exchanger unit heating output constraints.

[0054] The optimal heating mode for each building's composite heating system in different time periods and the optimal load ratio of air source heat pump units and building heat exchange units under the composite heating mode are obtained by solving the intelligent optimization algorithm.

[0055] The heating modes include: separate heating mode for air source heat pump units, separate heating mode for building heat exchange units, and a combined heating mode in which air source heat pump units provide heating and building heat exchange units obtain heat from the heating network.

[0056] Furthermore, step S4 includes:

[0057] When the heating mode of the composite heating system is that the air source heat pump unit provides heating alone, the operating data of the air source heat pump unit and outdoor meteorological parameters are obtained.

[0058] Based on the operating data of the air source heat pump unit and outdoor meteorological parameters, combined with the predicted heat load of the building, a first control model for the supply and return water temperature of the air source heat pump unit is established to control the operating status and parameters of the air source heat pump unit.

[0059] When the heating mode of the composite heating system is that the building heat exchange unit provides heating alone, the operating data of the building heat exchange unit and outdoor meteorological parameters are obtained.

[0060] Based on the operating data of the building heat exchange unit and outdoor meteorological parameters, combined with the predicted heat load of the building, a first control model for the pump valves of the building heat exchange unit is established to control the pump valves of the building heat exchange unit.

[0061] as well as,

[0062] When the heating mode of the composite heating system is a composite heating mode in which the air source heat pump unit provides heating and the building heat exchange unit obtains heat from the heating network, the operating data of the air source heat pump unit, outdoor meteorological parameters, and operating data of the building heat exchange unit are obtained.

[0063] Based on the operating data of air source heat pump units, outdoor meteorological parameters, and operating data of building heat exchange units, combined with the optimized load ratio of air source heat pump units and building heat exchange units under the combined heating mode and the predicted heat load of buildings, a second control model for the supply and return water temperature of air source heat pump units and a second pump and valve control model for building heat exchange units are established to regulate the operating status and parameters of air source heat pump units and control the pumps and valves of building heat exchange units, respectively.

[0064] The operating data and outdoor meteorological parameters of the air source heat pump unit include the target setpoints for the supply and return water temperatures of the air source heat pump unit, the measured values ​​of the supply and return water temperatures, the defrost status information, the compressor start / stop information of the air source heat pump unit, the expansion valve operating parameters, and the ambient temperature where the air source heat pump unit is located; the target setpoints for the supply and return water temperatures of the air source heat pump unit are set based on the predicted heat load of the building.

[0065] The operating data of the building heat exchange unit includes the target setpoint for the secondary supply and return water temperature, the measured value of the secondary supply and return water temperature, the secondary circulating water flow rate, the pressure difference between the secondary supply and return water in the secondary pipe network, the measured value of the primary supply and return water temperature in the primary pipe network, the opening degree of the electric regulating valve on the primary supply pipe, and the operating frequency of the circulating pump on the secondary return water network; the target setpoint for the secondary supply and return water temperature of the building heat exchange unit is set based on the predicted heat load of the building.

[0066] The target set values ​​for the supply and return water temperatures of the air source heat pump unit are different for establishing the first and second control models of the supply and return water temperatures; the target set values ​​for the supply and return water temperatures of the building heat exchanger unit are different for establishing the first and second pump valve control models of the building heat exchanger unit.

[0067] Furthermore, the first control model for the supply and return water temperatures of the air source heat pump unit, based on the input-output relationship of the model, is expressed as follows:

[0068] V1=f(T as,g,s1 ,T as,h,s1 ,T as,g,m ,T as,h,m ,T as,d ,T as,q ,T as,p ,T as,h Q as,1 )

[0069] V1 is the output variable of the first control model, including the operating status and parameters of the air source heat pump unit; T as,g,s1 The first target setpoint for the water supply temperature of the air source heat pump unit; T as,h,s1 The first target setpoint for the return water temperature of the air source heat pump unit; T as,g,m The measured water supply temperature of the air source heat pump unit; T as,h,m This refers to the measured return water temperature of the air source heat pump unit; T as,d This is for defrosting status information; T as,q This refers to compressor start / stop information in air source heat pump units; T as,p For the operating parameters of the expansion valve; T as,h Q represents the ambient temperature at which the air source heat pump unit operates; as,1 This refers to the heat load value when the air source heat pump unit is in stand-alone heating mode.

[0070] The second control model for the supply and return water temperatures of the air source heat pump unit, based on the input-output relationship of the model, is expressed as follows:

[0071] V2=f(T as,g,s2 ,T as,h,s2 ,T as,g,m ,T as,h,m ,Tas,d ,T as,q ,T as,p ,T as,h Q as,2 )

[0072] V2 is the output variable of the second control model, including the operating status and parameters of the air source heat pump unit; T as,g,s2 The second target setpoint for the water supply temperature of the air source heat pump unit; T as,h,s2 The second target setpoint for the return water temperature of the air source heat pump unit; Q as,2 The heat load value of the air source heat pump unit when the combined heating mode of the air source heat pump unit and the heating network is used;

[0073] The first control model for the pump valves of the building heat exchanger unit, based on the input-output relationship of the model, is expressed as follows:

[0074] F1=f(T he,g,s1 ,T he,h,s1 ,T he,g,m2 ,T he,h,m2 ,T he,cf ,T he,p ,T he,g,m1 ,T he,h,m1 ,T he,e ,T he,pf Q he,1 );

[0075] F1 is the output variable of the first control model of the pump and valve, including the operating parameters of the pump and valve; T he,g,s1 The first target setpoint for the secondary water supply temperature of the building heat exchanger unit; T he,h,s1 The first target setpoint for the secondary return water temperature of the building heat exchanger unit; T he,g,m2 This refers to the measured temperature of the secondary water supply for the building's heat exchanger unit; T he,h,m2 This refers to the measured value of the secondary return water temperature of the building heat exchanger unit; T he,cf T represents the secondary circulating water flow rate; he,p For the supply and return water pressure difference of the secondary side pipeline network; T he,g,m1 This refers to the measured temperature of the primary side water supply network; T he,h,m1 This refers to the measured return water temperature of the primary side pipe network; T he,e The opening degree of the electric regulating valve on the primary water supply pipe; T he,pf Q is the operating frequency of the circulating pump on the secondary return water network; he,1 This refers to the heat load value when the building's heat exchange unit is in standby heating mode.

[0076] The second control model for the pump valves of the building heat exchanger unit, based on the input-output relationship of the model, is expressed as follows:

[0077] F2=f(T he,g,s2 ,The,h,s2 ,T he,g,m2 ,T he,h,m2 ,T he,cf ,T he,p ,T he,g,m1 ,T he,h,m1 ,T he,e ,T he,pf Q he,2 );

[0078] F2 is the output variable of the second control model for the pump and valve, including the operating parameters of the pump and valve; T he,g,s2 The second target setpoint for the secondary water supply temperature of the building heat exchanger unit; T he,h,s2 The second target setpoint for the secondary return water temperature of the building heat exchanger unit; Q he,2 The heat load value of the building heat exchange unit when the heating mode is a combination of air source heat pump unit and heating network.

[0079] Furthermore, the first control model for the supply and return water temperature of the air source heat pump unit, the second control model for the supply and return water temperature of the air source heat pump unit, the first control model for the pump valve of the building heat exchanger unit, and the second control model for the pump valve of the building heat exchanger unit are trained using the already trained TabNet-LSTM model.

[0080] The beneficial effects of this invention are:

[0081] This invention establishes a composite heating system comprising air-source heat pump units, building heat exchange units, heating pipe networks, and buildings by installing building heat exchange units at the heat inlets of each building and air-source heat pump units on the top floor of the buildings. A digital twin model of this composite heating system is also established. Based on weather data and historical load data of each building, the heat load demand of each building is predicted in different time periods. Based on the predicted heat load values ​​of each building in different time periods, peak-valley electricity pricing strategies, and the time-period heating capacity range of the air-source heat pump units, the optimal heating mode and the optimized load ratio of air-source heat pump units and building heat exchange units under the composite heating mode are obtained by solving the problem with the objective function of minimizing the overall heating cost of each building. Based on the heating mode of the composite heating system and the optimized load ratio of air-source heat pump units and building heat exchange units under the composite heating mode, the optimal heating mode and the optimized load ratio of air-source heat pump units and building heat exchange units under the composite heating mode are determined. This study uses time-of-day heat load forecasts for buildings to establish temperature control models for the supply and return water of air-source heat pump units and pump valve control models for building heat exchanger units. This allows for the acquisition of control strategies for air-source heat pump units and building heat exchanger units under corresponding heating modes. The study enables the selection of heating modes based on peak-valley electricity prices, the heating capacity range of air-source heat pump units, and load value intervals. It allows for the effective utilization of the heating capacity of both air-source heat pump units and building heat exchanger units through separate heating and combined heating. By activating the pump valves of the building heat exchanger units to obtain heat from the existing heating network, combined with air-source heat pump heating, the study not only meets the building's heating needs but also offers advantages such as cleaner heating and higher economic efficiency. Furthermore, by establishing control models for corresponding heating modes, the study improves the accuracy of heating and allows for advance time-of-day control of the heating system.

[0082] Other features and advantages will be set forth in the description which follows, and will be apparent in part from the description, or may be learned by practicing the invention. The objects and other advantages of the invention are realized and obtained through the structures particularly pointed out in the description and the drawings.

[0083] To make the above-mentioned objects, features and advantages of the present invention more apparent and understandable, preferred embodiments are described below in detail with reference to the accompanying drawings. Attached Figure Description

[0084] To more clearly illustrate the specific embodiments of the present invention or the technical solutions in the prior art, the drawings used in the description of the specific embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are some embodiments of the present invention. For those skilled in the art, other drawings can be obtained from these drawings without creative effort.

[0085] Figure 1 This is a flowchart of a coordinated control method for a heating system based on an air source heat pump and a building heat exchanger unit, according to the present invention.

[0086] Figure 2 This is a schematic diagram of the combined heating structure of the air source heat pump and building heat exchange unit of the present invention;

[0087] Figure 3 This is a block diagram of the improved LSTNet model structure of the present invention;

[0088] Figure 4 This is a block diagram of the TabNet model structure of the present invention. Detailed Implementation

[0089] To make the objectives, technical solutions, and advantages of the embodiments of the present invention clearer, the technical solutions of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0090] Example 1

[0091] Figure 1 This is a flowchart of a coordinated control method for a heating system based on an air source heat pump and a building heat exchange unit, which is involved in this invention.

[0092] Figure 2 This is a schematic diagram of the combined heating structure of the air source heat pump and building heat exchange unit involved in this invention.

[0093] like Figure 1-2 As shown, this embodiment 1 provides a coordinated control method for a heating system based on an air source heat pump and a building heat exchanger unit, which includes:

[0094] Step S1: Install building heat exchange units at the heat inlet of each building and install air source heat pump units on the top floor of the building to form a composite heating system of air source heat pump units, building heat exchange units, heating network and building, and establish a digital twin model of the composite heating system.

[0095] Step S2: Based on weather data and historical load data of each building, predict the heat load demand of each building in different time periods;

[0096] Step S3: Based on the time-sharing heat load forecast of each building, peak-valley electricity pricing strategy, and time-sharing heating capacity range of air source heat pump units, with the objective function of minimizing the comprehensive heating cost of each building, solve to obtain the optimal heating mode of the composite heating system of each building for each time period and the optimized load ratio of air source heat pump units and building heat exchange units under the composite heating mode.

[0097] Step S4: Based on the heating mode of the composite heating system and the optimized load ratio of air source heat pump units and building heat exchange units under the composite heating mode, as well as the time-period heat load prediction values ​​of each building, establish the air source heat pump unit supply and return water temperature control model and the building heat exchange unit pump valve control model to obtain the control strategies of air source heat pump units and building heat exchange units under the corresponding heating modes.

[0098] In this embodiment, step S1 includes:

[0099] S101. A building heat exchange unit is installed at the heat inlet of each building, and multiple air source heat pump units are installed on the top floor of the building, forming a load heating system consisting of air source heat pump units, building heat exchange units, the existing heating network, and the building; the air source heat pump units are connected in parallel and include evaporators, compressors, condensers, and expansion valves; the building heat exchange units are connected to the existing heating network and obtain heat supply by adjusting valve opening and pump frequency;

[0100] S102. Construct a virtual entity of a composite heating system, including air source heat pump units, building heat exchange units, heating pipe networks, and building structures, and establish a digital twin model of the composite heating system after connecting virtual and physical data, including:

[0101] A structural model, physical equipment entity model, behavioral model, and rule model of the composite heating system are constructed. The physical equipment entity model is obtained by adding physical attributes of the equipment. Based on the basic functional theory, a behavioral model is constructed to establish a virtual simulation system of the composite heating system with interactive functions and a simulated real operating environment. Finally, a rule model of the virtual entities is established to formulate the control strategy of the virtual entities.

[0102] By collecting actual operating data of physical equipment in the composite heating system to drive corresponding virtual equipment, a mapping relationship between virtual and real data is established to form a composite heating system operation strategy; by continuously iterating and optimizing the data acquisition and control process, the connection and dynamic interaction of real-time data between physical entities and virtual space are realized to establish a digital twin model of the composite heating system.

[0103] S103. Identify the digital twin model of the composite heating system, including:

[0104] The real-time operating data of the composite heating system under multiple operating conditions are integrated into the established digital twin model. The simulation results of the digital twin model are adaptively identified and corrected using the reverse identification method to obtain the identified and corrected digital twin model of the composite heating system.

[0105] It should be noted that the heat source for building heat exchange units is municipal hot water, while the heat source for air source heat pump units is ambient air.

[0106] In this embodiment, step S2 includes:

[0107] Historical load data and local weather data for each building at different times during the early and late stages of heating are obtained. After data cleaning, missing value completion, and normalization preprocessing, the processed data are divided into training set, validation set, and test set. The weather data includes outdoor temperature, wind direction, humidity, precipitation, total horizontal radiation, and surface albedo.

[0108] Select load and weather data series from the same historical time period and calculate the MIC value between the two series;

[0109] The calculated MIC values ​​are approximated by symbolic aggregation and the correlation mean is solved. After sorting the mean values ​​of each weather variable, the input variables of the load forecasting model are selected.

[0110] The training and validation sets are input into the improved LSTNet model for training and learning, and a load prediction model for each building is established to obtain the heat load prediction values ​​of each building at different times.

[0111] Based on the predicted heat load of each building at different times, the target values ​​of the supply and return water temperatures that meet the comfort conditions of the building are calculated by simulating the digital twin model of the composite heating system under different predicted heat load conditions.

[0112] In this embodiment, calculating the MIC value between two sequences includes:

[0113] The mutual information MI between the load data series L and the weather data series C is calculated and expressed as:

[0114]

[0115] p(l,c) is the joint probability density of the load data sequence L and the weather data sequence C; p(l) and p(c) are the marginal probability densities of the load data sequence L and the weather data sequence C, respectively.

[0116] Calculate the mutual information between the load data sequence L and the weather data sequence C in grid G, and iterate through all grids to obtain the largest MI value for a given integer (x, y), expressed as:

[0117] f′ MI (D,G(x,y))=max f MI (D,G(x,y));

[0118] f MI (D,G(x,y)) represents the MI value of dataset D=(L,C) on G(x,y);

[0119] The maximum MI value in different grids is normalized and expressed as:

[0120]

[0121] Take all M G(x,y) The maximum value of is taken as the MIC value of the load data series L and the weather data series C, and expressed as:

[0122]

[0123] B(n) is a function of n.

[0124] In this embodiment, the step of performing symbolic aggregation approximation and correlation mean calculation on the calculated MIC values ​​includes:

[0125] The MIC is quantized using a piecewise aggregation approximation method, expressed as:

[0126]

[0127] y i k is the average value of the i-th segment; i Let x be the i-th time breakpoint; j The value of the j-th MIC;

[0128] The sequence after segmented aggregation approximation is symbolized and converted into discrete strings;

[0129] Analyze the occurrence frequency of each character, calculate the average of all characters to obtain the correlation, expressed as:

[0130]

[0131] β i For correlation; f k S is the number of times the k-th character appears; k Let N be the mathematically normalized value of the k-th character; P is the total number of occurrences of all characters; N is the total number of occurrences of all characters. i The number of characters represented by the MIC symbol.

[0132] Figure 3 This is a block diagram of the improved LSTNet model structure involved in this invention.

[0133] like Figure 3 As shown, in this embodiment, the LSTNet model includes convolutional layers, recurrent layers and recurrent skip layers, fully connected layers and an autoregressive model;

[0134] The improved LSTNet model includes:

[0135] The channel attention mechanism is introduced into the convolutional layer. Different channels are assigned corresponding weights, and then the channel information is weighted and summed to obtain the output of the convolutional layer.

[0136] Bidirectional recurrent neural network (BiLSTM) is used to replace the basic units of recurrent layers and recurrent skip layers, while simultaneously acquiring bidirectional data information;

[0137] The temporal attention mechanism is introduced into the recurrent layer and the recurrent skip layer. After the initial state vector is obtained by processing each feature vector output by BiLSTM through the attention mechanism, the final output vector is obtained by weighted summation through weight coefficients.

[0138] It's important to note that the LSTNet model consists of both linear and nonlinear parts. The nonlinear part includes convolutional layers, recurrent layers, and recurrent skip layers, while the linear part comprises an autoregressive model. This combination captures both long-term and short-term patterns in the data, improving model accuracy. Replacing the basic units of the recurrent and recurrent skip layers with a bidirectional recurrent neural network (BiLSTM) allows for simultaneous acquisition of bidirectional data information, aiming for better prediction results. BiLSTM consists of a forward LSTM and a backward LSTM, both fed forward to the same output layer. Introducing channel attention into the convolutional layers, assigning corresponding weights to different channels, allows for the assessment of the importance of information contained in different channels, enhancing the model's learning of features from important channels. Introducing temporal attention into the recurrent and recurrent skip layers can uncover dependencies in time series data, highlighting information from important time steps.

[0139] In this embodiment, step S3 includes:

[0140] Based on weather data and air source heat pump performance, the maximum and minimum heating values ​​of air source heat pump units equipped in each building are predicted in different time periods to obtain the time period heating capacity range of air source heat pump units.

[0141] Based on the time-period heat load forecasts for each building, peak-valley electricity pricing strategies, and the time-period heating capacity range of air source heat pump units, the objective function is to minimize the overall heating cost for each building.

[0142] Set constraints for the composite heating system, including: heat power balance constraints, air source heat pump unit heating output constraints, and building heat exchanger unit heating output constraints.

[0143] The optimal heating mode for each building's composite heating system in different time periods and the optimal load ratio of air source heat pump units and building heat exchange units under the composite heating mode are obtained by solving the intelligent optimization algorithm.

[0144] The heating modes include: separate heating mode for air source heat pump units, separate heating mode for building heat exchange units, and a combined heating mode in which air source heat pump units provide heating and building heat exchange units obtain heat from the heating network.

[0145] It should be noted that air source heat pumps extract low-grade heat energy from outdoor air, raise its temperature to usable high-grade heat energy, and deliver it indoors to achieve heating purposes. They offer numerous advantages such as high operating efficiency, convenient installation, small footprint, wide applicability, and no environmental pollution. However, their performance varies with atmospheric temperature. At the same ambient temperature, the higher the outlet water temperature, the lower the COP. Conversely, at the same outlet water temperature, the lower the ambient temperature, the lower the COP. Since the COP of an air source heat pump changes with outdoor temperature, the unit's heating capacity also decreases as the outdoor temperature drops. Therefore, when using an air source heat pump alone for heating during the heating season, on the one hand, as the ambient temperature decreases, the heating capacity of the air source heat pump decreases, while the building's heat load increases, potentially leading to insufficient heat supply to meet the building's heating needs. On the other hand, at low ambient temperatures, the air source heat pump may experience frosting, further reducing its coefficient of performance (COP) and increasing system energy consumption. Therefore, when the outdoor ambient temperature is low (the lower the ambient temperature, the higher the building's heat load), a combined heating mode of air source heat pump and building heat exchange unit can be adopted.

[0146] In this embodiment, step S4 includes:

[0147] When the heating mode of the composite heating system is that the air source heat pump unit provides heating alone, the operating data of the air source heat pump unit and outdoor meteorological parameters are obtained.

[0148] Based on the operating data of the air source heat pump unit and outdoor meteorological parameters, combined with the predicted heat load of the building, a first control model for the supply and return water temperature of the air source heat pump unit is established to control the operating status and parameters of the air source heat pump unit.

[0149] When the heating mode of the composite heating system is that the building heat exchange unit provides heating alone, the operating data of the building heat exchange unit and outdoor meteorological parameters are obtained.

[0150] Based on the operating data of the building heat exchange unit and outdoor meteorological parameters, combined with the predicted heat load of the building, a first control model for the pump valves of the building heat exchange unit is established to control the pump valves of the building heat exchange unit.

[0151] as well as,

[0152] When the heating mode of the composite heating system is a composite heating mode in which the air source heat pump unit provides heating and the building heat exchange unit obtains heat from the heating network, the operating data of the air source heat pump unit, outdoor meteorological parameters, and operating data of the building heat exchange unit are obtained.

[0153] Based on the operating data of air source heat pump units, outdoor meteorological parameters, and operating data of building heat exchange units, combined with the optimized load ratio of air source heat pump units and building heat exchange units under the combined heating mode and the predicted heat load of buildings, a second control model for the supply and return water temperature of air source heat pump units and a second pump and valve control model for building heat exchange units are established to regulate the operating status and parameters of air source heat pump units and control the pumps and valves of building heat exchange units, respectively.

[0154] The operating data and outdoor meteorological parameters of the air source heat pump unit include the target setpoints for the supply and return water temperatures of the air source heat pump unit, the measured values ​​of the supply and return water temperatures, the defrost status information, the compressor start / stop information of the air source heat pump unit, the expansion valve operating parameters, and the ambient temperature where the air source heat pump unit is located; the target setpoints for the supply and return water temperatures of the air source heat pump unit are set based on the predicted heat load of the building.

[0155] The operating data of the building heat exchange unit includes the target setpoint for the secondary supply and return water temperature, the measured value of the secondary supply and return water temperature, the secondary circulating water flow rate, the pressure difference between the secondary supply and return water in the secondary pipe network, the measured value of the primary supply and return water temperature in the primary pipe network, the opening degree of the electric regulating valve on the primary supply pipe, and the operating frequency of the circulating pump on the secondary return water network; the target setpoint for the secondary supply and return water temperature of the building heat exchange unit is set based on the predicted heat load of the building.

[0156] The target set values ​​for the supply and return water temperatures of the air source heat pump unit are different for establishing the first and second control models of the supply and return water temperatures; the target set values ​​for the supply and return water temperatures of the building heat exchanger unit are different for establishing the first and second pump valve control models of the building heat exchanger unit.

[0157] In this embodiment, the first control model for the supply and return water temperatures of the air source heat pump unit, based on the input-output relationship of the model, is expressed as follows:

[0158] V1=f(T as,g,s1 ,T as,h,s1 ,T as,g,m ,T as,h,m ,T as,d ,T as,q ,T as,p ,T as,h Q as,1 )

[0159] V1 is the output variable of the first control model, including the operating status and parameters of the air source heat pump unit; T as,g,s1 The first target setpoint for the water supply temperature of the air source heat pump unit; T as,h,s1 The first target setpoint for the return water temperature of the air source heat pump unit; T as,g,m The measured water supply temperature of the air source heat pump unit; T as,h,m This refers to the measured return water temperature of the air source heat pump unit; T as,d This is for defrosting status information; T as,q This refers to compressor start / stop information in air source heat pump units; T as,p For the operating parameters of the expansion valve; T as,h Q represents the ambient temperature at which the air source heat pump unit operates; as,1 This refers to the heat load value when the air source heat pump unit is in stand-alone heating mode.

[0160] The second control model for the supply and return water temperatures of the air source heat pump unit, based on the input-output relationship of the model, is expressed as follows:

[0161] V2=f(T as,g,s2 ,T as,h,s2 ,T as,g,m ,T as,h,m ,T as,d ,T as,q ,T as,p ,T as,h Q as,2 )

[0162] V2 is the output variable of the second control model, including the operating status and parameters of the air source heat pump unit; T as,g,s2 The second target setpoint for the water supply temperature of the air source heat pump unit; T as,h,s2 The second target setpoint for the return water temperature of the air source heat pump unit; Q as,2 The heat load value of the air source heat pump unit when the combined heating mode of the air source heat pump unit and the heating network is used;

[0163] The first control model for the pump valves of the building heat exchanger unit, based on the input-output relationship of the model, is expressed as follows:

[0164] F1=f(T he,g,s1 ,T he,h,s1 ,T he,g,m2 ,T he,h,m2 ,T he,cf ,T he,p ,T he,g,m1 ,T he,h,m1 ,T he,e ,T he,pf Q he,1 );

[0165] F1 is the output variable of the first control model of the pump and valve, including the operating parameters of the pump and valve; T he,g,s1 The first target setpoint for the secondary water supply temperature of the building heat exchanger unit; T he,h,s1 The first target setpoint for the secondary return water temperature of the building heat exchanger unit; T he,g,m2 This refers to the measured temperature of the secondary water supply for the building's heat exchanger unit; T he,h,m2 This refers to the measured value of the secondary return water temperature of the building heat exchanger unit; T he,cf T is the secondary circulating water flow rate; he,p For the supply and return water pressure difference of the secondary side pipeline network; T he,g,m1 This refers to the measured temperature of the primary side water supply network; T he,h,m1 This refers to the measured return water temperature of the primary side pipe network; T he,e The opening degree of the electric regulating valve on the primary water supply pipe; T he,pf Q is the operating frequency of the circulating pump on the secondary return water network; he,1 This refers to the heat load value when the building's heat exchange unit is in standby heating mode.

[0166] The second control model for the pump valves of the building heat exchanger unit, based on the input-output relationship of the model, is expressed as follows:

[0167] F2=f(T he,g,s2 ,T he,h,s2 ,T he,g,m2 ,T he,h,m2 ,T he,cf ,T he,p ,T he,g,m1 ,T he,h,m1 ,T he,e ,T he,pf Q he,2 );

[0168] F2 is the output variable of the second control model for the pump and valve, including the operating parameters of the pump and valve; T he,g,s2 The second target setpoint for the secondary water supply temperature of the building heat exchanger unit; T he,h,s2 The second target setpoint for the secondary return water temperature of the building heat exchanger unit; Q he,2 The heat load value of the building heat exchange unit when the heating mode is a combination of air source heat pump unit and heating network.

[0169] In this embodiment, the first control model for the supply and return water temperature of the air source heat pump unit, the second control model for the supply and return water temperature of the air source heat pump unit, the first control model for the pump valve of the building heat exchanger unit, and the second control model for the pump valve of the building heat exchanger unit are trained using the pre-trained TabNet-LSTM model.

[0170] It should be noted that the process of training the model using the TabNet-LSTM model includes: after batch standardization of the model's input data, it is fed into the additive model consisting of multiple steps in the TabNet algorithm model. The outputs of each additive model are then added together to obtain the feature vector, which is used as the input vector of the fully connected layer to obtain the final feature vector. Finally, the LSTM model is used for training to learn and establish the control model.

[0171] Figure 4 This is a block diagram of the TabNet model structure involved in this invention.

[0172] like Figure 4 As shown, the input to the TabNet model is features of dimension B×D, where B is the batch size and D is the dimension of the feature. The additive model includes Attention Transformer, Mask, Feature Transformer, Split, and ReLU, and the processing includes:

[0173] The features after BN batch normalization are input into the Feature transformer module for feature processing, and the output is obtained as a[i-1] through the split module.

[0174] a[i-1] is processed by the Attention transformer module to obtain M[i], which is then multiplied with the features processed by BN to obtain the Mask matrix to achieve the purpose of feature selection;

[0175] After inputting M[i]·f into the Feature transformer module, we obtain [d[i],a[i]]=f i After (M[i]·f) is split, d[i] is activated by the ReLU function and used as the output of this step, and a[i] is fed into the Attentiontransformer module in the next step for a new round of loop;

[0176] Sum the outputs of d[i] after ReLU transformation in all steps to obtain

[0177] The final output is obtained by performing a linear mapping through a fully connected layer: output = FC(d) out ).

[0178] In the several embodiments provided in this application, it should be understood that the disclosed systems and methods can also be implemented in other ways. The system embodiments described above are merely illustrative; for example, the flowcharts and block diagrams in the accompanying drawings illustrate the architecture, functionality, and operation of possible implementations of systems, methods, and computer program products according to various embodiments of the present invention. In this regard, each block in a flowchart or block diagram may represent a module, segment, or portion of code, which contains one or more executable instructions for implementing a specified logical function. It should also be noted that in some alternative implementations, the functions marked in the blocks may occur in a different order than those marked in the drawings. For example, two consecutive blocks may actually be executed substantially in parallel, and they may sometimes be executed in reverse order, depending on the functions involved. It should also be noted that each block in a block diagram and / or flowchart, and combinations of blocks in block diagrams and / or flowcharts, can be implemented using a dedicated hardware-based system that performs the specified function or action, or using a combination of dedicated hardware and computer instructions.

[0179] Furthermore, the functional modules in the various embodiments of this invention can be integrated together to form an independent part, or each module can exist independently, or two or more modules can be integrated to form an independent part. If the function is implemented as a software functional module and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of this invention, or the part that contributes to the prior art, or a part of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods in the various embodiments of this invention. The aforementioned storage medium includes: USB flash drives, portable hard drives, read-only memory (ROM), random access memory (RAM), magnetic disks, optical disks, and other media capable of storing program code.

[0180] Based on the above-described preferred embodiments of the present invention, and through the foregoing description, those skilled in the art can make various changes and modifications without departing from the inventive concept. The technical scope of this invention is not limited to the contents of the specification, but must be determined according to the scope of the claims.

Claims

1. A coordinated control method for a heating system based on an air source heat pump and a building heat exchanger unit, characterized in that, It includes: Step S1: Install building heat exchange units at the heat inlet of each building and install air source heat pump units on the top floor of the building to form a composite heating system of air source heat pump units, building heat exchange units, heating network and building, and establish a digital twin model of the composite heating system. Step S2: Based on weather data and historical load data of each building, predict the heat load demand of each building in different time periods; Based on the predicted heat load of each building at different times, the target values ​​of the building supply and return water temperatures that meet the comfort conditions of the building are calculated by simulating the digital twin model of the composite heating system under different predicted heat load conditions. Step S3: Based on the time-sharing heat load forecast of each building, peak-valley electricity pricing strategy, and time-sharing heating capacity range of air source heat pump units, with the objective function of minimizing the comprehensive heating cost of each building, solve to obtain the optimal heating mode of the composite heating system of each building for each time period and the optimized load ratio of air source heat pump units and building heat exchange units under the composite heating mode. Step S4: Based on the heating mode of the composite heating system and the optimized load ratio of air source heat pump units and building heat exchange units under the composite heating mode, as well as the time-period heat load prediction values ​​of each building, establish a supply and return water temperature control model for the air source heat pump units and a pump and valve control model for the building heat exchange units. Obtain the control strategies for the air source heat pump units and building heat exchange units under the corresponding heating modes, including: When the heating mode of the composite heating system is that the air source heat pump unit provides heating alone, the operating data of the air source heat pump unit and outdoor meteorological parameters are obtained. Based on the operating data of the air source heat pump unit and outdoor meteorological parameters, combined with the predicted heat load of the building, a first control model for the supply and return water temperature of the air source heat pump unit is established to control the operating status and parameters of the air source heat pump unit. When the heating mode of the composite heating system is that the building heat exchange unit provides heating alone, the operating data of the building heat exchange unit and outdoor meteorological parameters are obtained. Based on the operating data of the building heat exchange unit and outdoor meteorological parameters, combined with the predicted heat load of the building, a first control model for the pump valves of the building heat exchange unit is established to control the pump valves of the building heat exchange unit. as well as, When the heating mode of the composite heating system is a composite heating mode in which the air source heat pump unit provides heating and the building heat exchange unit obtains heat from the heating network, the operating data of the air source heat pump unit, outdoor meteorological parameters, and operating data of the building heat exchange unit are obtained. Based on the operating data of air source heat pump units, outdoor meteorological parameters, and operating data of building heat exchange units, combined with the optimized load ratio of air source heat pump units and building heat exchange units under the combined heating mode and the predicted heat load of buildings, a second control model for the supply and return water temperature of air source heat pump units and a second pump and valve control model for building heat exchange units are established to regulate the operating status and parameters of air source heat pump units and control the pumps and valves of building heat exchange units, respectively. The operating data and outdoor meteorological parameters of the air source heat pump unit include the target setpoints for the supply and return water temperatures of the air source heat pump unit, the measured values ​​of the supply and return water temperatures, the defrost status information, the compressor start / stop information of the air source heat pump unit, the expansion valve operating parameters, and the ambient temperature where the air source heat pump unit is located; the target setpoints for the supply and return water temperatures of the air source heat pump unit are set based on the predicted heat load of the building. The operating data of the building heat exchange unit includes the target setpoint for the secondary supply and return water temperature, the measured value of the secondary supply and return water temperature, the secondary circulating water flow rate, the pressure difference between the secondary supply and return water in the secondary pipe network, the measured value of the primary supply and return water temperature in the primary pipe network, the opening degree of the electric regulating valve on the primary supply pipe, and the operating frequency of the circulating pump on the secondary return water network; the target setpoint for the secondary supply and return water temperature of the building heat exchange unit is set based on the predicted heat load of the building. The target set values ​​for the supply and return water temperatures of the air source heat pump unit are different for establishing the first and second control models of the supply and return water temperatures; the target set values ​​for the supply and return water temperatures of the building heat exchanger unit are different for establishing the first and second pump valve control models of the building heat exchanger unit.

2. The coordinated control method for a heating system according to claim 1, characterized in that, Step S1 includes: S101. A building heat exchange unit is installed at the heat inlet of each building, and multiple air source heat pump units are installed on the top floor of the building, forming a load heating system consisting of air source heat pump units, building heat exchange units, the existing heating network, and the building; the air source heat pump units are connected in parallel and include evaporators, compressors, condensers, and expansion valves; the building heat exchange units are connected to the existing heating network and obtain heat supply by adjusting valve opening and pump frequency; S102. Construct a virtual entity of a composite heating system, including air source heat pump units, building heat exchange units, heating pipe networks, and building structures, and establish a digital twin model of the composite heating system after connecting virtual and physical data, including: A structural model, physical equipment entity model, behavioral model, and rule model of the composite heating system are constructed. The physical equipment entity model is obtained by adding physical attributes of the equipment. Based on the basic functional theory, a behavioral model is constructed to establish a virtual simulation system of the composite heating system with interactive functions and a simulated real operating environment. Finally, a rule model of the virtual entities is established to formulate the control strategy of the virtual entities. By collecting actual operating data of physical equipment in the composite heating system to drive corresponding virtual equipment, a mapping relationship between virtual and real data is established to form a composite heating system operation strategy; by continuously iterating and optimizing the data acquisition and control process, the connection and dynamic interaction of real-time data between physical entities and virtual space are realized to establish a digital twin model of the composite heating system. S103. Identify the digital twin model of the composite heating system, including: The real-time operating data of the composite heating system under multiple operating conditions are integrated into the established digital twin model. The simulation results of the digital twin model are adaptively identified and corrected using the reverse identification method to obtain the identified and corrected digital twin model of the composite heating system.

3. The coordinated control method for a heating system according to claim 1, characterized in that, Step S2 includes: Historical load data and local weather data for each building at different times during the early and late stages of heating are obtained. After data cleaning, missing value completion, and normalization preprocessing, the processed data are divided into training set, validation set, and test set. The weather data includes outdoor temperature, wind direction, humidity, precipitation, total horizontal radiation, and surface albedo. Select load and weather data series from the same historical time period and calculate the MIC value between the two series; The calculated MIC values ​​are approximated by symbolic aggregation and the correlation mean is solved. After sorting the mean values ​​of each weather variable, the input variables of the load forecasting model are selected. The training and validation sets are input into the improved LSTNet model for training and learning, and a load prediction model for each building is established to obtain the heat load prediction values ​​for each building at different times.

4. The coordinated control method for a heating system according to claim 3, characterized in that, The calculation of the MIC value between two sequences includes: The mutual information MI between the load data series L and the weather data series C is calculated and expressed as: ； Let L be the joint probability density of the load data sequence L and the weather data sequence C; , These are the marginal probability densities of the load data sequence L and the weather data sequence C, respectively. Calculate the mutual information between the load data sequence L and the weather data sequence C in grid G, and iterate through all grids to obtain the largest MI value for a given integer (x, y), expressed as: ； For dataset exist MI value; The maximum MI value in different grids is normalized and expressed as: ； Take all The maximum value of is taken as the MIC value of the load data series L and the weather data series C, and expressed as: ； It is a function of n.

5. The coordinated control method for a heating system according to claim 3, characterized in that, The process of performing symbolic aggregation approximation and correlation mean calculation on the calculated MIC values ​​includes: The MIC is quantized using a piecewise aggregation approximation method, expressed as: ； For the first The average value of the segment; For the first A time breakpoint; For the first The value of each MIC; The sequence after segmented aggregation approximation is symbolized and converted into discrete strings; Analyze the occurrence frequency of each character, calculate the average of all characters to obtain the correlation, expressed as: ； Correlation; For the first The number of times each character appears; For the first The mathematical normalized value of each character; This represents the number of all occurrences of a character. The number of characters represented by the MIC symbol.

6. The coordinated control method for a heating system according to claim 3, characterized in that, The LSTNet model includes convolutional layers, recurrent layers and recurrent skip layers, fully connected layers, and an autoregressive model. The improved LSTNet model includes: The channel attention mechanism is introduced into the convolutional layer. Different channels are assigned corresponding weights, and then the channel information is weighted and summed to obtain the output of the convolutional layer. Bidirectional recurrent neural network (BiLSTM) is used to replace the basic units of recurrent layers and recurrent skip layers, while simultaneously acquiring bidirectional data information; The temporal attention mechanism is introduced into the recurrent layer and the recurrent skip layer. After the initial state vector is obtained by processing each feature vector output by BiLSTM through the attention mechanism, the final output vector is obtained by weighted summation through weight coefficients.

7. The coordinated control method for a heating system according to claim 1, characterized in that, Step S3 includes: Based on weather data and air source heat pump performance, the maximum and minimum heating values ​​of air source heat pump units equipped in each building are predicted in different time periods to obtain the time period heating capacity range of air source heat pump units. Based on the time-period heat load forecasts for each building, peak-valley electricity pricing strategies, and the time-period heating capacity range of air source heat pump units, the objective function is to minimize the overall heating cost for each building. Set constraints for the composite heating system, including: heat power balance constraints, air source heat pump unit heating output constraints, and building heat exchanger unit heating output constraints. The optimal heating mode for each building's composite heating system in different time periods and the optimal load ratio of air source heat pump units and building heat exchange units under the composite heating mode are obtained by solving the intelligent optimization algorithm. The heating modes include: separate heating mode for air source heat pump units, separate heating mode for building heat exchange units, and a combined heating mode in which air source heat pump units provide heating and building heat exchange units obtain heat from the heating network.

8. The coordinated control method for a heating system according to claim 1, characterized in that, The first control model for the supply and return water temperatures of the air source heat pump unit, based on the input-output relationship of the model, is expressed as follows: ； The output variables of the first control model include the operating status and parameters of the air source heat pump unit; The first target setpoint for the water supply temperature of the air source heat pump unit; The first target setpoint for the return water temperature of the air source heat pump unit; The measured water supply temperature of the air source heat pump unit; This is the measured return water temperature of the air source heat pump unit. This is information indicating the defrosting status; This provides compressor start / stop information for air source heat pump units. These are the operating parameters for the expansion valve; The ambient temperature at which the air source heat pump unit is located; This refers to the heat load value when the air source heat pump unit is in stand-alone heating mode. The second control model for the supply and return water temperatures of the air source heat pump unit, based on the input-output relationship of the model, is expressed as follows: ； The output variables of the second control model include the operating status and parameters of the air source heat pump unit; The second target setpoint for the water supply temperature of the air source heat pump unit; The second target setpoint for the return water temperature of the air source heat pump unit; The heat load value of the air source heat pump unit when the combined heating mode of the air source heat pump unit and the heating network is used; The first control model for the pump valves of the building heat exchanger unit, based on the input-output relationship of the model, is expressed as follows: ； The output variables of the first control model for pumps and valves include the operating parameters of the pumps and valves; The first target setpoint for the secondary water supply temperature of the building heat exchange unit; The first target setpoint for the secondary return water temperature of the building heat exchanger unit; The measured value of the secondary water supply temperature for the building heat exchange unit; This refers to the measured value of the secondary return water temperature of the building heat exchanger unit. This refers to the secondary circulating water flow rate; The pressure difference between the supply and return water in the secondary side pipeline network; This refers to the measured temperature of the primary side water supply network. This refers to the measured return water temperature of the primary side pipe network. The opening degree of the electric regulating valve on the primary water supply pipe; This refers to the operating frequency of the circulating pump on the secondary return water pipeline network; This refers to the heat load value when the building's heat exchange unit is in standby heating mode. The second control model for the pump valves of the building heat exchanger unit, based on the input-output relationship of the model, is expressed as follows: ； The output variables of the second control model for pumps and valves include the operating parameters of the pumps and valves; The second target setpoint for the secondary water supply temperature of the building heat exchange unit; The second target setpoint for the secondary return water temperature of the building heat exchanger unit; The heat load value of the building heat exchange unit when the heating mode is a combination of air source heat pump unit and heating network.

9. The coordinated control method for a heating system according to claim 8, characterized in that, The first control model for supply and return water temperature of the air source heat pump unit, the second control model for supply and return water temperature of the air source heat pump unit, the first control model for pump valves of the building heat exchanger unit, and the second control model for pump valves of the building heat exchanger unit are trained using the pre-trained TabNet-LSTM model.

Images