A multi-source heat supply system regulation method, device and medium

Abstract

This invention provides a method, device, and medium for regulating a multi-source heating system, belonging to the field of energy management technology. The method includes: acquiring operational data of the heating system; wherein the heating system includes at least two heat sources; inputting the operational data into a pre-trained hybrid model to obtain the actual heating capacity of each heat source; acquiring the heat load demand; and determining the optimal load allocation scheme for each heat source based on the heat load demand and the actual heating capacity; by constructing a hybrid model that integrates a mechanistic model and a residual network, based on the calculation of theoretical heat supply based on energy conservation and mass conservation, further compensating for the deviation between the theoretical heat supply output by the mechanistic model and the actual heat supply of each heat source measured on-site, it can achieve accurate output of the actual usable heating capacity of each heat source, thereby solving the technical problem that the calculation results of the pure mechanistic model are distorted and do not match the actual operating conditions due to equipment aging, etc.

Description

Technical Field

[0001] This invention belongs to the field of energy management technology, and specifically relates to a method, equipment and medium for regulating a multi-source heating system. Background Technology

[0002] With the rapid growth of new energy installed capacity and the continuous transformation of the power grid's power structure, thermal power units are gradually shifting from traditional "main power sources" to "regulatory power sources," with units frequently participating in deep peak shaving and operating routinely in low-load ranges. At the same time, in response to clean heating policies and to improve heating security, power plants are gradually introducing new heat source equipment such as electrode boilers and thermal storage tanks, which work in conjunction with the existing thermal power extraction steam system to form a "electricity-heat-steam" coupled heating system with multiple heat sources and complex and variable operating conditions.

[0003] Currently, most output calculations and scheduling analyses for multi-source heating systems still rely on pure mechanistic models, directly calculating the heat output of thermal power extraction steam, electrode boilers, and thermal storage tanks using theoretical formulas. However, in practical engineering applications, heating systems operate continuously for extended periods, and equipment commonly suffers from aging and wear, pipe scaling, and decreased heat exchange efficiency. Furthermore, under low-load conditions, unmodeled dynamic characteristics such as dead zones in the flow field, two-phase steam-water flow, and increased local thermal resistance can easily occur. Therefore, directly relying on pure mechanistic models for load allocation and scheduling decisions can easily lead to situations where theoretical output is sufficient but actual heat output is insufficient, failing to reliably meet heat load demands and severely impacting the operational stability and scheduling accuracy of multi-source coupled heating systems.

[0004] Furthermore, even ignoring the aforementioned biases, while the pure mechanistic model can output accurate theoretical heat supply from each heat source, its calculation results only reflect the maximum energy supply potential of each heat source and cannot provide scientific guidance for the coordinated scheduling of multiple heat sources. Specifically, the pure mechanistic model does not consider the multi-objective requirements of system operation, such as economy and environmental protection. The output result is essentially a rough allocation mode where each heat source operates at maximum output, rather than an optimal solution that takes into account multiple objectives. In actual operation, the scheduling of the heating system needs to simultaneously minimize the total system operating cost, the total carbon emissions, and the risk of hydraulic imbalance in the pipeline network. Relying solely on the theoretical heat supply calculation of the pure mechanistic model cannot balance the differences in operating costs and carbon emissions among heat sources, easily leading to problems such as energy waste, high operating costs, and failure to meet environmental standards. Summary of the Invention

[0005] To address the problems in the background art, this invention proposes a method, equipment, and medium for regulating a multi-source heating system.

[0006] To achieve the above objectives, the present invention adopts the following technical solution: In a first aspect, the present invention proposes a method for regulating a multi-source heating system, comprising: Acquire operational data of the heating system; wherein the heating system includes at least two heat sources; The operational data is input into a pre-trained hybrid model to obtain the actual heating capacity of each heat source; Obtain the heat load demand, and based on the heat load demand and the actual heating capacity, determine the optimal load allocation scheme for each heat source.

[0007] Preferably, the heating system includes: thermal power steam extraction heating, electrode boiler heating, and thermal storage tank heating; The operating data for thermal power extraction steam heating includes: thermal power extraction steam flow rate, extraction steam enthalpy, condensate enthalpy, and thermal power heat exchange efficiency; the operating data for electrode boiler heating includes: electrode boiler rated power and electrode boiler conversion efficiency; the operating data for thermal storage tank heating includes: thermal storage tank volume, current thermal storage tank charging temperature, thermal storage tank final heat release temperature, and thermal storage tank heat release duration.

[0008] Preferably, training a hybrid model includes: Based on the laws of conservation of energy and mass, a mechanism model is constructed; wherein, the mechanism model includes: a thermal power extraction steam supply model, an electrode boiler supply model, and a thermal storage tank supply model; Construct a residual network and obtain historical operating data; wherein, the historical operating data includes: the theoretical heat supply output by the mechanistic model and the actual heat supply of each heat source; Based on the theoretical heat supply output by the aforementioned mechanism model and the actual heat supply of each heat source, the residual corresponding to each set of historical data is determined. The historical running data is used as the input to the residual network, the residual corresponding to each set of historical data is used as the label, and the residual network is trained based on the loss function to obtain the trained residual network. Based on the aforementioned mechanism model and the trained residual network, a trained hybrid model is obtained.

[0009] Preferably, the expression for the thermal power plant steam extraction heating model is: In the formula, For the heating capacity of thermal power plants, This refers to the mass flow rate of steam extracted from thermal power plants. This is the enthalpy of extraction. For hydrophobic enthalpy, For the heat exchange efficiency of thermal power plants; The expression for the electrode boiler heating model is as follows: In the formula, This represents the actual heating capacity of the electrode boiler. The rated electrical power of the electrode boiler, For electrode boiler conversion efficiency; The expression for the heat supply model of the heat storage tank is: In the formula, To provide heat supply power to the thermal storage tank, The density of water, For the volume of the heat storage tank, The specific heat capacity of water, This refers to the heat release time of the heat storage tank.

[0010] Preferably, the expression for the loss function is: In the formula, For data-driven error terms, For energy conservation constraints, For parameter smoothing constraint terms, For mechanistic model parameters, These are the prior values ​​of the parameters of the mechanistic model. , Weights for constraint terms; The expression for the hybrid model is: In the formula, To determine the actual heating capacity of each heat source output by the hybrid model. The output value of the mechanistic model. In order to obtain the runtime data, For the parameters of the mechanistic model, The output value of the residual network. For the parameters of the residual network, This represents the historical output state of the residual network.

[0011] Preferably, determining the optimal load allocation scheme for each heat source specifically includes: Determine the constraints; wherein the constraints include: heat load balance constraints, unit operation constraints, and ramp rate constraints; Construct a multi-objective collaborative optimization function; Based on the heat load demand and the actual heating capacity, the optimal load allocation scheme for each heat source is obtained by using the multi-objective collaborative optimization function and constraints.

[0012] Preferably, the expression for the heat load balance constraint is: In the formula, Extracting steam from thermal power plants to provide heat. To supply heat to the electrode boiler, For the heat storage tank to release heat, To meet the system's real-time heat load requirements; The expression for the unit operation constraints is: , In the formula, These represent the minimum and maximum available output of steam extracted from thermal power plants, respectively. The actual output of steam extraction for thermal power plants; These are the minimum and maximum heat outputs of the electrode boiler, respectively. This refers to the actual heat supply of the electrode boiler; The expression for the climbing rate constraint is: In the formula, This represents the maximum ramp rate of a thermal power extraction steam turbine unit. This represents the maximum ramp-down rate for thermal power extraction steam turbine units. For the rate of change of output of thermal power extraction steam turbine units; The expression for the multi-objective collaborative optimization function is: In the formula, , , Pre-set weighting coefficients by humans The total operating cost of the heating system, Total carbon emissions Risk of hydraulic imbalance in the pipeline network.

[0013] Preferably, after obtaining the heat load demand and determining the optimal load allocation scheme for each heat source based on the heat load demand and the actual heating capacity, the method further includes: Based on the pressure and temperature values ​​at each node of the pipeline network, the gradient change of the pipeline color is calculated, and the steam flow rate is simulated using a particle system.

[0014] In a second aspect, the present invention proposes a device including a memory and a processor, wherein the memory stores computer instructions that can be executed on the processor, and the processor executes a multi-source heating system control method of the first aspect when executing the computer instructions.

[0015] Thirdly, the present invention proposes a computer-readable storage medium storing computer instructions thereon, which, when executed, can realize the multi-source heating system control method described in the first aspect.

[0016] The beneficial effects of this invention are: 1. The method of the present invention constructs a hybrid model that integrates the mechanistic model and the residual network. Based on the calculation of theoretical heat supply based on energy conservation and mass conservation, it further compensates for the deviation between the theoretical heat supply output by the mechanistic model and the actual heat supply of each heat source measured on site. This enables the accurate output of the actual usable heat supply capacity of each heat source, thereby solving the technical problem that the calculation results of the pure mechanistic model are distorted and do not match the actual working conditions due to equipment aging, flow field distortion, and lack of dynamic modeling.

[0017] 2. The method of the present invention constructs a multi-objective collaborative optimization model with constraints based on the actual heating capacity and real-time heat load demand, and solves the optimal load allocation scheme for each heat source. This can achieve the comprehensive minimization of system operating costs, carbon emissions and hydraulic imbalance risks, thereby solving the technical problems of high energy consumption, poor economy and insufficient environmental protection caused by the traditional extensive allocation based solely on the maximum output of the heat source.

[0018] Other features and advantages of the invention will be set forth in the following description, 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 may be realized and obtained by means of the structures pointed out in the description and the drawings. Attached Figure Description

[0019] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the description of the 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 based on these drawings without creative effort.

[0020] Figure 1 A flowchart of a multi-source heating system control method according to the present invention is shown. Detailed Implementation

[0021] To make the objectives, technical solutions, and advantages of the embodiments of the present invention clearer, the technical solutions of the embodiments 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.

[0022] Reference Figure 1 As shown, a method for regulating a multi-source heating system specifically includes the following steps: S10. Obtain the operating data of the heating system; wherein the heating system includes at least two heat sources; S20. Input the operating data into the pre-built hybrid model to obtain the actual heating capacity of each heat source; S30. Obtain the heat load demand and determine the optimal load allocation scheme for each heat source based on the heat load demand and the actual heating capacity.

[0023] In step S10 above, the heating system includes: thermal power extraction steam heating, electrode boiler heating, and thermal storage tank heating. The operating data for thermal power extraction steam heating includes: thermal power extraction steam flow rate, extraction steam enthalpy, condensate enthalpy, and thermal power heat exchange efficiency; the operating data for electrode boiler heating includes: electrode boiler rated power and electrode boiler conversion efficiency; and the operating data for thermal storage tank heating includes: thermal storage tank volume, current charging temperature of the thermal storage tank, final heat release temperature of the thermal storage tank, and heat release duration of the thermal storage tank.

[0024] In step S20 above, the hybrid model is constructed, which specifically includes the following steps: S201. Based on the laws of conservation of energy and mass, construct a mechanism model; among which, the mechanism model includes: a thermal power extraction steam supply model, an electrode boiler supply model, and a thermal storage tank supply model; Construct a model for the heating supply of steam extracted from thermal power plants, specifically: Within the heating system, the rate of change of fluid mass = total inflow mass The total mass flowing out is considered, therefore, the mass conservation equation is: In the formula, The rate of change of fluid mass within the system. The total mass flowing into the system, This represents the total mass flowing out of the system.

[0025] During steady-state operation of a heating system, its mass does not accumulate; therefore, we can conclude that: ,Right now: As can be seen, the water supply flow rate is equal to the return flow rate.

[0026] The rate of change of internal energy in a heating system = inflow energy Energy outflow + heating amount Since work is done on the surroundings, the energy conservation equation can be obtained as follows: In the formula, The rate of change of internal energy in the heating system. The enthalpy of the inflowing fluid. The enthalpy of the outflowing fluid is denoted as . The amount of heat absorbed by the system The work done by the system for external communication.

[0027] Similarly, during steady-state operation of the heating system, and when the heat exchanger does not perform external work, that is... Therefore, we can conclude that: .

[0028] Furthermore, there is no water accumulation or leakage inside the heat exchanger; therefore, the inflow mass equals the outflow mass.

[0029] Therefore, by considering the addition time and the heat exchange efficiency of thermal power plants, we can obtain the expression for the thermal power plant steam extraction heating model: In the formula, For the heating capacity of thermal power plants, This refers to the mass flow rate of steam extracted from thermal power plants. This is the enthalpy of extraction. For hydrophobic enthalpy, The heat exchange efficiency of thermal power is given by 3600, which is the unit conversion factor (converting kJ / h to MW).

[0030] A heating model for an electrode boiler is constructed as follows: Within the electrode boiler system, electrical energy is primarily converted into thermal energy to heat the fluid, and its internal energy change rate equals the electrical energy input. Therefore, considering the heat loss, the energy conservation equation for the electrode boiler can be obtained as follows: In the formula, The rate of change of internal energy of the fluid inside the electrode boiler. The electrical energy input to the electrode boiler, This refers to the heat energy loss during the operation of the electrode boiler. During steady-state operation of the heating system, the internal energy of the fluid inside the electrode boiler does not accumulate. Therefore, we can conclude that: ,Right now: In the formula, This refers to the heat output from the electrode boiler. Furthermore, the conversion efficiency of electrode boilers is introduced. Efficiency is used to characterize the effectiveness of converting electrical energy into heat energy. Efficiency is defined as the ratio of the output heat supply to the input electrical energy of the electrode boiler. Therefore, we can obtain .

[0031] Since the electrical power input to the electrode boiler is equal to its rated electrical power (the power is stable during steady-state operation), that is... Furthermore, the heat supply must ultimately be converted into power units (MW). Electrode boilers have no additional time loss during the conversion between electrical and thermal energy (electrical energy is converted into thermal energy in real time). Therefore, the formula for the heat supply of an electrode boiler can be obtained as follows: In the formula, This represents the actual heating capacity of the electrode boiler. The rated electrical power of the electrode boiler, This refers to the conversion efficiency of the electrode boiler.

[0032] A mechanism model of the thermal storage tank is constructed as follows: Within the thermal storage tank system, the change in internal energy stored in the fluid equals the heat released (or absorbed) by the thermal storage tank. Therefore, the energy conservation equation for the thermal storage tank can be obtained as follows: In the formula, This represents the change in internal energy of the fluid inside the heat storage tank. This represents the total heat released by the heat storage tank. According to the fundamental thermodynamic formulas, the change in internal energy of a fluid In the formula, The total mass of the fluid inside the heat storage tank. This is the specific heat capacity of the fluid (which in this system is the specific heat capacity of water, and is a constant term). This represents the temperature difference of the fluid in the heat storage tank before and after heat release. Total mass of fluid in the heat storage tank It can be determined by the density of water and the volume of the heat storage tank The calculation shows that, In the formula, The inherent density of water, This refers to the inherent volume of the heat storage tank. Will Substitution The total heat storage capacity of the heat storage tank (i.e., the total heat released) can be obtained as follows: In the formula, This represents the current charging temperature of the heat storage tank. The final temperature at which the heat storage tank releases heat. This represents the temperature difference before and after heat release. Since subsequent load distribution calculations require the use of heating power, the total heat output needs to be converted into heat output per unit time, i.e., divided by the heat release duration. Simultaneously, unit conversion is performed (converting kJ / h to MW). Therefore, the expression for the heat supply model of the thermal storage tank can be obtained as follows: In the formula, To provide heat supply power to the thermal storage tank, The density of water, For the volume of the heat storage tank, The specific heat capacity of water, The heat release time of the thermal storage tank is 3600, which is the unit conversion factor (converting kJ / h to MW).

[0033] As can be seen, the mechanistic model is only a theoretical calculation model under ideal operating conditions and cannot be directly applied to the load distribution of actual heating systems. This is because the mechanistic model makes idealized assumptions during its derivation and does not consider various unmodeled dynamic factors in actual field operation, such as pipe bubble thermal resistance, flow field dead zones, equipment aging losses, and network heat dissipation losses. These factors will cause deviations between the calculation results of the mechanistic model and the actual available heating capacity on site. For example, when the real-time heat load demand of the system is 40MW, the mechanistic model may calculate that a certain heat source can supply 50MW of heat, which seems to meet the load demand. However, due to various losses on site, the actual available heating capacity of this heat source is only 35MW, which cannot reach the calculated value of the mechanistic model. Consequently, the total heat supply is less than 40MW, which cannot meet the user's heat load demand. Therefore, a hybrid model needs to be built on the basis of the mechanistic model before it can be used in practical applications.

[0034] S202. Construct a hybrid model based on residual networks and mechanistic models; Specifically, the hybrid model adopts a structure of "mechanistic baseline and AI residual compensation", and its overall expression is as follows: In the formula, To determine the actual usable heating capacity of each heat source in the final output of the hybrid model, This is the output value of the mechanistic model (theoretical heat supply power). This includes various operational data collected in real time on-site (including operational data from thermal power steam extraction, electrode boilers, and thermal storage tanks, as well as operating condition data such as outdoor ambient temperature, pipeline supply and return water pressure, and supply and return water temperature). These are parameters for the mechanistic model (such as efficiency, density, specific heat capacity, etc.). This is the output value (bias compensation amount) of the residual network. For the parameters of the residual network, This represents the historical output state of the residual network.

[0035] First, determine the input and output parameters of the hybrid model: the input parameters are all the operational data collected in real time on site, which are consistent with the data collected when calculating the mechanism model. At the same time, supplementary pipeline operating data (outdoor ambient temperature, pipeline supply and return water pressure, supply and return water temperature) are used to learn the influence of unmodeled dynamic factors in the residual network; the output parameters are the actual available heating capacity of each heat source (including the upper and lower limits of output), which correspond to the core parameters required for subsequent load allocation optimization.

[0036] Secondly, a residual network is constructed and trained offline: historical data from long-term field operation is collected, including various types of collected data, theoretical values ​​output by the mechanistic model, and actual heat supply data from each heat source measured on-site. The residual corresponding to each set of data is calculated, i.e. In the formula, The residual is the deviation between the mechanistic model and the measured value. This represents the actual heat output power of the heat source as measured on-site. The theoretical heat supply power calculated for the mechanistic model.

[0037] The collected data is used as input to the residual network, and the calculated residuals are used as the network output labels. A loss function from a hybrid model is also introduced to constrain the model training process, ensuring that the model output closely matches the actual measured data while adhering to physical laws. The expression for this loss function is: In the formula, This is a data-driven error term used to measure the output value of the mixture model. Compared with the actual measured value To minimize deviations and ensure that the output of the hybrid model closely matches reality, This is an energy conservation constraint, a mathematical expression of the law of conservation of energy. It is used to force the output of the hybrid model to not violate energy conservation, thus avoiding physically unreasonable results. For parameter smoothing constraint terms, For mechanistic model parameters, These are the prior values ​​of the mechanistic model parameters, used to constrain the fluctuation range of the mechanistic model parameters and improve the stability of model training; , The weights of the constraint terms are used to adjust the degree of influence of each constraint term and can be determined offline through calibration based on field operation data. This loss function is used to synchronously optimize the residual network parameters during multiple iterations of the residual network training process. and mechanism model parameters This enables the residual network to accurately learn the mapping relationship between "collected data and residuals". That is, after inputting real-time collected data, the network can output residuals that are consistent with the actual deviation on site, thereby achieving accurate compensation for the deviation of the mechanism model.

[0038] Finally, the mechanistic model and the trained residual network are integrated to form a hybrid model. Real-time data collected on-site is synchronously input into both the mechanistic model and the residual network. The mechanistic model outputs the theoretical heating capacity, and the residual network outputs the corresponding deviation compensation. The sum of these two values ​​represents the actual usable heating capacity of each heat source. In the formula, The final output of the hybrid model, representing the actual heating capacity of each heat source, can be directly used for subsequent optimal load allocation calculations, ensuring that the calculation results closely match the actual operating conditions on site.

[0039] By constructing a hybrid model that integrates the mechanistic model and the residual network, based on the calculation of theoretical heat supply according to the laws of energy conservation and mass conservation, the deviation between the theoretical heat supply output by the mechanistic model and the actual heat supply of each heat source measured on site can be further compensated. This enables the accurate output of the actual usable heat supply capacity of each heat source, thereby solving the technical problem that the calculation results of the pure mechanistic model are distorted and do not match the actual working conditions due to equipment aging, flow field distortion, and lack of dynamic modeling.

[0040] In step S30 above, the actual heating capacity of each heat source output by the hybrid model and the real-time heat load demand collected on site are combined. Operating cost parameters of each heat source, carbon emission parameters, parameters related to the risk of hydraulic imbalance in the pipeline network, and preset weighting coefficients of the objective function. , , The data is compiled and summarized to serve as the basic input parameters for load allocation optimization.

[0041] Next, it is necessary to clarify the various constraints of load allocation to ensure that the optimization results meet the requirements of equipment operation safety and heating demand; among them, the various constraints include: heat load balance constraints, unit operation constraints and ramp rate constraints; Specifically, the heat load balance constraint refers to the requirement that the sum of the heat output from all heat sources in the heating system must meet the real-time heat load demand to ensure sufficient heating. Its expression is: In the formula, Extracting steam from thermal power plants to provide heat. To supply heat to the electrode boiler, For the heat storage tank to release heat, This is to meet the real-time heat load requirements of the system.

[0042] Unit operating constraints refer to the requirement that the output of each heat source must be within its actual usable upper and lower limits to ensure stable and safe operation of the equipment. The expression for this constraint is: , In the formula, These represent the minimum and maximum available output of steam extracted from thermal power plants, respectively. The actual output of steam extraction for thermal power plants; These are the minimum and maximum heat outputs of the electrode boiler, respectively. The actual heat supplied by the electrode boiler; the heat release tank. It also follows its own upper and lower limits of output constraints. This is consistent with the constraint logic mentioned above.

[0043] The ramp rate constraint refers to the requirement that the rate of change of output of a thermal power extraction steam turbine unit must be controlled within an allowable range to avoid sudden changes in output that could lead to equipment damage or system instability. Its expression is: In the formula, This refers to the maximum ramp rate of the thermal power extraction steam turbine unit (in MW / h). This represents the maximum rate of rate of decrease for thermal power extraction steam turbine units (in MW / h). This represents the rate of change in output of thermal power extraction steam turbine units.

[0044] Furthermore, it is necessary to achieve synergistic optimization that minimizes the total system operating cost, total carbon emissions, and the risk of hydraulic imbalance in the heating network. A multi-objective synergistic optimization function needs to be constructed, the expression of which is: In the formula, , , These are pre-set weighting coefficients used to adjust the degree of influence of the three optimization objectives. The total operating cost of the heating system, Total carbon emissions Risk of hydraulic imbalance in the pipeline network.

[0045] Specifically, total system operating cost It encompasses the fuel costs of thermal power units and the electricity purchase costs of electrode boilers, while deducting the revenue from deep peak-shaving ancillary services. Its expression is: In the formula, To optimize the total number of time periods, Fuel cost function of thermal power units (and actual output of thermal power extraction steam) Heating supply (related) Electricity purchase cost for electrode boilers (compared to the electricity purchased by electrode boilers) (related) For revenue from deep peak shaving ancillary services (and output from deep peak shaving of steam extracted from thermal power plants) (Related).

[0046] The total carbon emissions mainly come from thermal power extraction steam turbines and electrode boilers. Thermal storage tanks have no carbon emissions. The values ​​are calculated based on the output of each heat source and the unit carbon emission coefficient to ensure that the optimization results meet environmental protection requirements.

[0047] Pipeline hydraulic imbalance risk Used to measure the stability of pipeline network operation, its value is related to the output distribution of each heat source, pipeline pressure, and temperature distribution, to avoid problems such as excessive local pressure and uneven temperature in the pipeline network due to unreasonable load distribution.

[0048] Finally, intelligent optimization algorithms (such as genetic algorithms and particle swarm optimization) are used, substituting the above constraints and multi-objective collaborative optimization functions, to solve for the actual output of each heat source. Through multiple iterative optimizations, an optimization objective function that satisfies all constraints is obtained. The heat source load distribution scheme with the smallest value, i.e. , , The optimal value.

[0049] By constructing a multi-objective collaborative optimization model with constraints based on actual heating capacity and real-time heat load demand, and solving the optimal load allocation scheme for each heat source, the system can achieve comprehensive minimization of operating costs, carbon emissions, and hydraulic imbalance risks. This solves the technical problems of high energy consumption, poor economic efficiency, and insufficient environmental protection caused by the traditional extensive allocation based solely on the maximum output of heat sources.

[0050] In a preferred embodiment of the present invention, after outputting the optimal load distribution scheme for each heat source, the gradient change of the pipe color can also be dynamically calculated based on the pressure and temperature values ​​of each node in the pipe network. The color calculation expression is as follows: In the formula, The final color to be displayed for the pipes The color of the pipe when the fluid is at low temperature. The color of the pipe when the fluid is at high temperature. This represents the actual temperature of the fluid inside the pipe. , These represent the minimum and maximum design temperatures of the fluid inside the pipe, respectively; simultaneously, a particle system is used to simulate steam flow rate and particle velocity. With the flow velocity inside the pipe They are directly proportional, and their expression is: In the formula, As a proportional coefficient, it allows operators to intuitively see the "flow of heat energy" and local overheated areas, while clearly understanding the load distribution of each heat source, providing support for subsequent command issuance and closed-loop control.

[0051] By dynamically calculating the color gradient changes of the pipeline based on the pressure and temperature at each node of the pipeline network, and using a particle system to simulate the steam flow rate, the operating status of the heating system can be presented intuitively and visually. This can accurately reflect the trend of heat energy flow and local overheating areas within the pipeline network, while providing operators with clear and reliable scheduling basis, facilitating the rapid issuance of control commands and the realization of closed-loop system optimization, effectively improving the safety, real-time performance, and intelligence level of heating scheduling.

[0052] Based on the same inventive concept as the above method, the present invention also proposes a device including a memory and a processor, wherein the memory stores computer instructions that can be executed on the processor, and the processor executes the above-described method for regulating a multi-source heating system when executing the computer instructions.

[0053] Based on the same inventive concept as the above method, the present invention also proposes a computer-readable storage medium storing computer instructions thereon, which can realize the above-mentioned multi-source heating system control method when the computer instructions are executed.

[0054] Any references to memory, storage, database, or other media used in the embodiments provided in this invention may include non-volatile and / or volatile memory. Non-volatile memory may include read-only memory (ROM), programmable ROM (PROM), electrically programmable ROM (EPROM), electrically erasable programmable ROM (EEPROM), or flash memory. Volatile memory may include random access memory (RAM) or external cache memory.

[0055] It should be noted that, in this document, relational terms such as "first" and "second" are used merely to distinguish one entity or operation from another, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Furthermore, 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 a process, method, article, or apparatus. Without further limitations, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the process, method, article, or apparatus that includes said element.

[0056] Although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features; and these modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. A method for regulating a multi-source heating system, characterized in that, include: Acquire operational data of the heating system; wherein the heating system includes at least two heat sources; The operational data is input into a pre-trained hybrid model to obtain the actual heating capacity of each heat source; Obtain the heat load demand, and based on the heat load demand and the actual heating capacity, determine the optimal load allocation scheme for each heat source.

2. The method for regulating a multi-source heating system according to claim 1, characterized in that, The heating system includes: thermal power extraction steam heating, electrode boiler heating and thermal storage tank heating; The operating data for thermal power plant steam extraction heating includes: thermal power plant steam extraction flow rate, steam extraction enthalpy, condensate enthalpy, and thermal power plant heat exchange efficiency; The operating data for electrode boiler heating includes: rated electrical power of the electrode boiler and electrode boiler conversion efficiency; The operating data for heat storage tank heating includes: heat storage tank volume, current heat storage tank charging temperature, final heat release temperature, and heat release duration.

3. The method for regulating a multi-source heating system according to claim 2, characterized in that, Training the hybrid model specifically includes: Based on the laws of conservation of energy and mass, a mechanism model is constructed; wherein, the mechanism model includes: a thermal power extraction steam supply model, an electrode boiler supply model, and a thermal storage tank supply model; Construct a residual network and obtain historical operating data; wherein, the historical operating data includes: the theoretical heat supply output by the mechanistic model and the actual heat supply of each heat source; Based on the theoretical heat supply output by the aforementioned mechanism model and the actual heat supply of each heat source, the residual corresponding to each set of historical data is determined. The historical running data is used as the input to the residual network, the residual corresponding to each set of historical data is used as the label, and the residual network is trained based on the loss function to obtain the trained residual network. Based on the aforementioned mechanism model and the trained residual network, a trained hybrid model is obtained.

4. The method for regulating a multi-source heating system according to claim 3, characterized in that, The expression for the thermal power plant steam extraction heating model is as follows: In the formula, For the heating capacity of thermal power plants, This refers to the mass flow rate of steam extracted from thermal power plants. This is the enthalpy of extraction. For hydrophobic enthalpy, For the heat exchange efficiency of thermal power plants; The expression for the electrode boiler heating model is as follows: In the formula, This represents the actual heating capacity of the electrode boiler. The rated power of the electrode boiler, For electrode boiler conversion efficiency; The expression for the heat supply model of the heat storage tank is: In the formula, To provide heat supply power to the thermal storage tank, The density of water, For the volume of the heat storage tank, The specific heat capacity of water, This refers to the heat release time of the heat storage tank.

5. The method for regulating a multi-source heating system according to claim 4, characterized in that, The expression for the loss function is: In the formula, For data-driven error terms, For energy conservation constraints, For parameter smoothing constraint terms, For mechanistic model parameters, These are the prior values ​​of the parameters of the mechanistic model. , Weights for constraint terms; The expression for the hybrid model is: In the formula, To determine the actual heating capacity of each heat source output by the hybrid model. The output value of the mechanistic model. In order to obtain the runtime data, For the parameters of the mechanistic model, The output value of the residual network. For the parameters of the residual network, This represents the historical output state of the residual network.

6. The method for regulating a multi-source heating system according to claim 2, characterized in that, The determination of the optimal load allocation scheme for each heat source specifically includes: Determine the constraints; wherein the constraints include: heat load balance constraints, unit operation constraints, and ramp rate constraints; Construct a multi-objective collaborative optimization function; Based on the heat load demand and the actual heating capacity, the optimal load allocation scheme for each heat source is obtained by using the multi-objective collaborative optimization function and constraints.

7. The method for regulating a multi-source heating system according to claim 6, characterized in that, The expression for the heat load balance constraint is: In the formula, Extracting steam from thermal power plants to supply heat. To supply heat to the electrode boiler, For the heat storage tank to release heat, To meet the system's real-time heat load requirements; The expression for the unit operation constraints is: , In the formula, These represent the minimum and maximum available output of steam extracted from thermal power plants, respectively. The actual output of steam extraction for thermal power plants; These are the minimum and maximum heat outputs of the electrode boiler, respectively. This refers to the actual heat supply of the electrode boiler; The expression for the climbing rate constraint is: In the formula, This represents the maximum ramp rate of a thermal power extraction steam turbine unit. This represents the maximum ramp-down rate for thermal power extraction steam turbine units. For the rate of change of output of thermal power extraction steam turbine units; The expression for the multi-objective collaborative optimization function is: In the formula, , , Pre-set weighting coefficients by humans The total operating cost of the heating system, Total carbon emissions Risk of hydraulic imbalance in the pipeline network.

8. A method for regulating a multi-source heating system according to claim 2, characterized in that, After obtaining the heat load demand and determining the optimal load allocation scheme for each heat source based on the heat load demand and the actual heating capacity, the method further includes: Based on the pressure and temperature values ​​at each node of the pipeline network, the gradient change of the pipeline color is calculated, and the steam flow rate is simulated using a particle system.

9. A device comprising a memory and a processor, wherein the memory stores computer instructions executable on the processor, characterized in that, When the processor executes the computer instructions, it performs a multi-source heating system control method according to any one of claims 1 to 8.

10. A computer-readable storage medium storing computer instructions thereon, characterized in that, When computer instructions are executed, a multi-source heating system control method as described in any one of claims 1 to 8 can be implemented.

Images