A distributed predictive control method for a solar district heating system

Abstract

The present application belongs to the technical field of solar heating, and relates to a distributed predictive control method of a solar regional heating system. First, a constitution and a physical model of the regional heating system are established. Second, a predictive control subsystem of a heat collection field, a heat storage device, an auxiliary heat source and a user terminal is constructed. Third, a distributed predictive control structure among the subsystems is analyzed, and an optimal signal distribution scheme among the subsystems is solved. Finally, the overall heating system is coordinated by the distributed controllers according to the characteristics among the subsystems, and finally the distributed control is realized. The present application sets a predictive control system at the source end, comprehensively considers the relationship among the heat source end, the heat storage section and the terminal, fully utilizes the solar energy, and makes the solar regional heating system have the characteristics of lower energy consumption, more accurate regulation and control and better indoor thermal comfort.

Description

Technical Field

[0001] This invention belongs to the field of solar heating technology and relates to a distributed predictive control method for a solar district heating system. Background Technology

[0002] Solar district heating systems refer to systems that use solar energy to provide centralized heating to a specific area or areas of a city or building. Current technologies typically employ traditional fixed control strategies such as temperature difference control, constant temperature control, and proportional-integral control (PID), with optimization of system design and the development of new control strategies based on these. While intelligent control methods are also being applied to solar-related systems, very few are suitable for large-scale solar district heating.

[0003] However, existing solar heating systems employ traditional fixed control strategies such as temperature difference control, constant temperature control, and proportional-integral control (PID), which cannot effectively balance supply and demand. This results in high energy consumption, poor precision control, and insufficient indoor thermal comfort. Therefore, a control method for solar district heating systems that is energy-efficient, precisely controllable, and provides good comfort is needed to address these issues. Summary of the Invention

[0004] The technical solution adopted by this invention to solve the technical problem is to coordinate the control among four subsystems using a distributed predictive control approach. In the solar collector section, flow rate control is primarily based on collector-related temperature parameters. The average temperature of the fluid in the solar collector depends on the inlet temperature, ambient temperature, solar radiation, and flow velocity (flow rate). Under given weather conditions and inlet temperature, the faster the fluid flow rate, the faster the collector temperature rises as it flows through the collector, resulting in a lower outlet temperature. A lower outlet temperature means lower heat loss in the collector, improving its efficiency to some extent. However, the flow velocity should be controlled within a reasonable range to ensure uniform fluid distribution throughout the collector. The flow velocity should only be controlled to provide enough heat to the centralized heating system to reach the required temperature. Higher flow velocities increase pump power consumption. Therefore, for large-scale solar systems, variable flow rate regulation and intelligent control systems are highly effective.

[0005] The solution to the technical problem of this invention is: a distributed predictive control method for a solar district heating system, comprising the following steps:

[0006] Step S1: Analyze the form of the solar district heating system, the parameters of each system equipment, and establish a physical model and a nonlinear black box model for controller design;

[0007] Step S2: Construct four subsystems: heat collection subsystem, heat storage subsystem, auxiliary heat source subsystem, and terminal subsystem;

[0008] The solar collector subsystem includes: a solar collector field, valves on each row of collectors, and water pumps for the collector field. The parameters contained in the solar collector subsystem include: inlet and outlet temperatures of each row of collectors, flow rates of each row of collectors, overall inlet and outlet temperatures and flow rates of the collector field, heat collection capacity of the collector field, heat collection efficiency of the collector field, and solar energy guarantee rate. The control object of the solar collector subsystem is the flow rate of each row of collectors and the overall collector field, which is achieved by controlling the frequency of the solar collector field circulation pump and the opening degree of each valve (including the heat exchange pump on the water tank side of the indirect system).

[0009] The thermal storage subsystem includes: a solar thermal storage device, which is connected to a plate heat exchanger via a heat exchange circulation pump. The solar thermal storage device stores the heat collected by the solar thermal generator through heat exchange. The solar thermal storage device includes: a thermal storage unit, a thermal storage water supply system, a heat exchange circulation pump for the thermal storage unit, water circulation valves for the thermal storage unit, and water supply valves for the thermal storage unit. The solar thermal storage unit contains relevant parameters including: average thermal storage temperature, stratified temperature of the thermal storage unit, inlet temperature and flow rate of the thermal storage unit, water storage capacity, stored heat and released heat. The control objects of the solar thermal storage unit are the inlet and outlet flow rates of the thermal storage unit, the stored and released heat, the water supply capacity, and the relevant temperatures of the thermal storage unit.

[0010] The auxiliary heat source subsystem includes: auxiliary heat source equipment; the auxiliary heat source equipment includes: auxiliary heat source device, auxiliary heat source side circulation pump or valve, buffer water tank; the auxiliary heat source equipment contains relevant parameters including: auxiliary heat source operating energy consumption, auxiliary heat source inlet and outlet temperature and flow rate, auxiliary heat; the auxiliary heat source equipment controls the start and stop of the auxiliary heat source and the auxiliary heat source inlet and outlet flow rates;

[0011] The terminal subsystem includes: a district heating system, which encompasses heating for residential areas, heating for one or more building complexes, or buildings within a larger area; the specific system form of the end users is not considered here, and the terminal district heating network and user load are simplified as a whole; the district heating system includes a simplified overall model of the target users and the main water pump or valve of the network; the district heating system contains relevant parameters including: total heat load, total pipeline length, number of heat users (or number of heat meters if installed), rated supply water temperature (or supply water temperature range), and rated return water temperature (or return water temperature range); the control object of the district heating system is the frequency regulation of the terminal main water pump to regulate the flow rate of the terminal main pipeline;

[0012] Step S3: Analyze the distributed predictive control structure among the subsystems and solve for the optimal signal allocation scheme among the subsystems;

[0013] Step S4: Based on the characteristics of each subsystem, coordinate the entire heating system using a distributed controller to ultimately achieve distributed control;

[0014] Step S3 includes the following sub-steps:

[0015] Step S3-1: The meteorological parameter detection module receives temperature and solar radiation measurements, sends the collected meteorological parameters to the computer through sensors, and predicts the future distribution of solar radiation and ambient temperature meteorological parameters through a constructed neural network model; the neural network model can adopt a sequence-to-sequence long short-term memory seq2seq-LSTM prediction model to realize a multi-step forward prediction strategy.

[0016] Step S3-2: For the solar collector subsystem, in order to obtain a control model oriented towards the solar collector field, the flow rate of each row of solar collectors is allocated by considering the basic relationships between the states and parameters of each row of solar collectors. The key points of this control strategy are as follows:

[0017] Step S3-2-1: Without global constraints, obtain the flow rate of each row of collectors sequentially; use the local cost function and flow rate value as decision variables:

[0018] Constraints:

[0019] In the formula, x0 and x0+N p For time step, The local cost function is expressed as a measure of the system's performance with respect to the flow sequence, q. l,i (x0) is a vector containing flow values, which can be represented as q l,i (x0)=[q1(x0),q1(x0+1),…,q1(x0+N u -1),q2(x),q2(x0+1),…,q Nl (x0+N u -1)], x0 is the controller time step corresponding to the instant t=x×ΔT, Np is the prediction range, Nu is the prediction range; qmin is the minimum flow rate of each row of collectors, and qmax is the maximum flow rate of each row of collectors, which are selected according to the Reynolds number and maximum pressure drop of each row of collectors, respectively.

[0020] Local cost function:

[0021]

[0022] In the formula, W is within a certain time step. i (x0) represents the heat collection capacity of a single-row solar collector. α(q) represents the maximum outlet temperature of a single-row solar collector. i (x0)-q i (x0-1)) 2 This represents the flow rate on a single row of solar collectors, and α is the tuning parameter.

[0023] Step S3-2-2: Is the obtained flow solution feasible? If feasible, proceed to step S3-3. If not feasible, recalculate the local model predicted control solution for each row of collectors, but the maximum flow limit is changed.

[0024] Constraints:

[0025] Local cost function in the formula As mentioned earlier, the maximum flow limit constraint has been modified to... This provides a feasible solution for traffic allocation, but it may not be the optimal solution.

[0026] Step S3-2-3: Check if the total flow rate of the system is less than or equal to the maximum flow rate. If so, check if the flow rate can be increased by adding a row of solar collectors, and then calculate the increment of the local cost function obtained by increasing the corresponding flow rate.

[0027] If at this time the condition is satisfied This increases the flow rate of one row of solar collectors within the total available flow rate, where... This is represented as an increase of Δq l,i Local cost function after (x0) flow

[0028] If the system's total flow rate is not less than or equal to the maximum flow rate, check if a cost function can be added by increasing the flow rate of one drain while reducing the flow rate of another drain by the same amount.

[0029] calculate That is, redetermine the cost function

[0030] Calculate the increment of the local cost function obtained by increasing the corresponding flow rate of the heat collection cycle j:

[0031] In the formula This is represented as an increase of Δq l,i Local cost function after (x0) flow

[0032] Calculate the reduction in the local cost function obtained by reducing the corresponding flow rate of the i-th row of collectors:

[0033] In the formula This indicates a reduction Local cost function after flow

[0034] Step S3-2-4, Judgment The formula represents the local cost function. Increment and The maximum value of the sum of reductions is used to determine whether it exceeds the total flow limit. If it does not, proceed to step S3-3. If it does, control calculations are used to reduce the flow of one row and increase the flow of another row.

[0035] Output the final flow rate of the heat collection field;

[0036] Step S3-3: Based on the final flow rate of the heat collection field obtained in the previous step, calculate whether the heat collection capacity of the heat collection field and the heat storage, auxiliary heat and load of other subsystems can reach the optimal state after adjusting the heat collection circulation pump or related control valves. If the optimal state of the system as a whole is reached, jump to step S3-7. If not, continue to step S3-4.

[0037] Step S3-4: Update the prediction parameters based on the relevant parameters between each subsystem. The prediction parameters include: heat collection, heat storage, load, auxiliary heat, and meteorological parameters. Reset the optimization variables and constraints of the heat storage subsystem and optimize the local model predictive control solution for each subsystem.

[0038] Step S3-5: Is the solution obtained in the previous step optimal? If it is the optimal solution, jump to step S3-7; if it is not the optimal solution, jump to step S3-6 and continue to adjust the state of other subsystems.

[0039] Step S3-6: Based on the optimization of the heat collection subsystem and the heat storage subsystem, update the prediction parameters again, reset the optimization variables and constraints of the auxiliary heat subsystem, and optimize the calculation of the local model predictive control solution of each subsystem; if the overall system reaches the optimal state, continue to step S3-7.

[0040] Step S3-7: Obtain the signal values ​​of the other subsystems under the optimal state of the system as a whole, with the heat collection subsystem as the dominant system.

[0041] Step S3-8: For the thermal storage subsystem, in order to obtain a control-oriented thermal storage device model, consider the relevant parameters of the thermal storage system, the thermal storage and release state, and the temperature of the energy storage device to avoid solar system stagnation and achieve optimal utilization of thermal energy reserves. Generally, the supply temperature should be maintained below 95℃ due to the boiling point of water and the temperature limits of the equipment in the solar heat transfer device. For short-term thermal storage, the storage temperature can reach 80-90℃ when the thermal storage device is in a state of moderate heat loss. High temperatures are only used when the storage size of the thermal storage device is limited or when the centralized heating network requires high temperatures to eliminate the need for auxiliary heat sources. With sufficient buffer storage capacity, a lower storage temperature is more conducive to improving the efficiency of the collector. For seasonal thermal energy storage devices, a higher inlet temperature can be used to prevent the disruption of temperature stratification. If the thermal storage device has different inlet temperatures, it can utilize solar energy over a wider temperature range because heat can enter the thermal storage device at different temperatures, thus ensuring temperature stratification. When the temperature of the seasonal heat storage device is too low, the heat cannot be directly utilized by the central heating system. In this case, an auxiliary heat source can be used to raise the temperature to the temperature required for direct heating, or the heat can be stored at the top of the heat storage device to reduce the outlet temperature of the solar collector.

[0042] Step S3-9: Repeat steps S3-2 to S3-7 to finally obtain the signal value required for heat storage and control the temperature and heat storage capacity of the heat storage device.

[0043] Step S3-10: For the auxiliary heating subsystem, in order to obtain a control-oriented auxiliary heat source model, consider the relevant parameters of the auxiliary heating subsystem, the auxiliary heat output, the auxiliary heating water temperature, and maintain the water tank temperature to stabilize the system's heat supply. Its control strategy should minimize the auxiliary heating output and maximize the output of the solar thermal collector system. The auxiliary heating form should also be optimized to utilize other heat sources such as heat pumps, boilers, and waste heat.

[0044] Step S3-11: Repeat steps S3-2 to S3-7 to finally obtain the signal value required for auxiliary heating and control the auxiliary heating temperature and amount of auxiliary heat.

[0045] Step S3-12: For the terminal subsystem, simplify the terminal as a whole, obtain the terminal load and supply and return water temperature and flow parameters, construct a neural network related model, and establish a nonlinear prediction model for predictive regulation based on the user's energy consumption.

[0046] Step S3-13: Through communication between distributed predictive control systems, continuously coordinate the control signals between the four subsystems, explore the relationship between heat collection, heat storage, auxiliary heat, and heat load, continuously adjust the constraint conditions in the MPC of each subsystem to optimize the variables, and finally enable the system as a whole to obtain the best operating state.

[0047] Preferably, in step S1, the solar district heating system includes: a solar collector field, a heat storage device, a heat exchanger, an auxiliary heat source, and a buffer water tank; the solar collector field is used to convert solar energy into heat energy and then store the heat energy in the heat storage device after exchanging it through the heat exchanger; the heat storage device and the auxiliary heat source are used to provide district heating; and the buffer water tank is used to buffer the heating fluctuations of the heat storage device and the auxiliary heat source.

[0048] Preferably, the solar collector field includes: a single-row flow regulation device for multiple rows of collectors, a meteorological parameter detection module, a GPS sensor, a flow and temperature detection module at the collector field, and a flow distribution control module for the collector field; the heat storage device includes: a flow and temperature detection module at the heat storage device, and a heat storage data monitoring module; the auxiliary heat source includes: a flow and temperature detection module at the auxiliary heat source, and an auxiliary heat data monitoring module; the solar district heating system further includes: a terminal overall supply and return water temperature and flow detection module, a heat collection data monitoring module, a user heat load data monitoring module, a distributed control device, a cloud control server, and a water pump control module.

[0049] Preferably, in steps S3-8, the key control points of the thermal storage subsystem are as follows:

[0050] The solar energy storage device is connected to the plate heat exchanger via a heat exchange circulation pump. Sensors receive the current parameters and status of each subsystem of the system, predict relevant parameters in the future, optimize and calculate the control signal of the heat storage subsystem, and transmit the control signal to the heat storage circulation pump to control its flow rate and realize the change of heat exchange temperature. The water supply to the energy storage device is controlled by a float valve. When the water level in the tank is lower than the position of the float valve, tap water is supplied. When the water level reaches the control level of the float valve, the cold water is shut off.

[0051] Preferably, in step S3-10, the key control points of the auxiliary heating subsystem are as follows:

[0052] The auxiliary heat source is designed to supplement the heat of the energy storage system. Sensors receive the current parameters and status of each subsystem, predict relevant future parameters, and optimize the calculation of the control signal for the auxiliary heating subsystem. When the heating water temperature provided by the heat storage device is lower than the set temperature, the control signal is transmitted to the auxiliary heat source and the heat exchange circulation pump of the auxiliary heating subsystem to supplement the system heat until the water temperature in the heat storage device is higher than the set temperature, at which point the supplementary heating stops. In addition, the frequency of the auxiliary heating circulation pump needs to be controlled to control the flow rate. Based on the subsystem signals, the optimal heating outlet temperature of the auxiliary heating subsystem is obtained. The subsystem signals include: heat storage, heat collection, and load.

[0053] The model used in the implementation of the distributed MPC control in the above steps can be transformed into a discrete state-space equation form using a physical model, or it can be a data-driven model (nonlinear black-box model). To improve the versatility of this patented control method in different solar district heating systems, a nonlinear model based on artificial neural networks is used to implement the MPC control scheme.

[0054] 1. Data Acquisition. The data acquisition targets are the relevant parameters of each subsystem in the above steps. In order to obtain a model suitable for all working conditions, the data must cover the entire range of change of the corresponding variables during the working process. The above process is completed using an artificial neural network model, and the relevant data needs to be filtered and normalized.

[0055] 2. Select and establish a suitable nonlinear black-box model. Once established, this model can be expressed as a nonlinear system using the following formula:

[0056]

[0057] In the formula, u is the vector of input variables; y is the output variable; and P is the measurable disturbance vector. Model identification requires finding the system's mapping f such that... nyny and np need to be determined through the design and observation of the corresponding solar heating system.

[0058] 3. Select past signal values ​​as input.

[0059] 4. Select the structural parameters to be used in the model. This step is equivalent to obtaining the structure and scale of the artificial neural network. For example, the most widely used neural network models in the model identification process, such as multilayer perceptrons or radial basis function neural networks and their corresponding derivative models, can be used.

[0060] 5. Train the neural network. Select an appropriate training algorithm based on the type of neural network used in different system configurations. For example, constructive / non-constructive algorithms in multilayer perceptrons, and data selection for radial basis function neural networks can be achieved using data-driven methods or trial-and-error approaches.

[0061] 6. Confirm the obtained model and apply it in the control system. A series of evaluation metrics can be used to measure the accuracy of the model's input / output data, such as SSE, R², RMSE, and MAE. Finally, implement a neural network in the MPC scheme. The neural network will serve as the predictive model in the MPC, predicting the future outputs of each subsystem within the solar heating system based on the current inputs and control actions. The MPC will then use these predictions to determine the optimal control actions that minimize the cost function of each subsystem.

[0062] The beneficial effects of this invention are:

[0063] 1. This invention integrates the relationship between the heat source, the heat storage section, and the terminal by setting up a predictive control system at the source end, thus making full use of solar energy. Therefore, the system has the characteristics of lower energy consumption, more precise control, and better indoor thermal comfort.

[0064] 2. This invention is an improvement on the control method of solar district heating system. Compared with the traditional fixed control strategy, it can give full play to the advantages of MPC rolling optimization and real-time control feedback to achieve day-ahead regulation, and to a certain extent make up for the shortcomings of bidirectional dynamic instability of load and solar energy, further improving the energy utilization efficiency. In addition, it realizes multiple functions through distributed coordination communication.

[0065] 3. This invention can optimize the operation of the system based on predicted weather conditions, user needs and other relevant factors, so that the control strategy can be adjusted in real time, thereby making full use of solar energy resources, making up for the problem of strong fluctuations in solar energy, and thus improving the energy efficiency of the system and user comfort.

[0066] 4. This invention can take into account nonlinear dynamics and constraints on system inputs and outputs, which can improve the stability and reliability of the system.

[0067] 5. This invention achieves refined management of the system and improves system performance by dividing the solar centralized heating system into multiple subsystems and designing a distributed model predictive control strategy for each subsystem.

[0068] 6. This invention is implemented using distributed computing programs and computer equipment, which can be easily updated and modified to adapt to changes in the system or environment.

[0069] 7. The method of the present invention has strong versatility and can be applied to different types of solar centralized heating systems, providing an effective way to achieve green and efficient heating. Attached Figure Description

[0070] Figure 1 This is a schematic diagram illustrating the system working principle of a distributed predictive control method for a solar district heating system.

[0071] Figure 2 It is a flowchart of the overall work framework;

[0072] Figure 3 This is a flowchart of the flow control process for the thermal collector subsystem.

[0073] Figure 4 This is a schematic diagram of the principle of distributed information feedback.

[0074] Figure 5 This is a flowchart for building a nonlinear model based on artificial neural networks for MPC. Detailed Implementation

[0075] The related technologies of the present invention will be clearly and completely described below with reference to the accompanying drawings of the embodiments of the present invention. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art without creative effort are within the scope of protection of the present invention.

[0076] refer to Figures 1-5 A distributed predictive control method for a solar district heating system includes the following steps:

[0077] Includes the following steps:

[0078] Step S1: Analyze the form of the solar district heating system, the parameters of each system equipment, and establish a physical model and a nonlinear black box model for controller design;

[0079] Step S2: Construct four subsystems: heat collection subsystem, heat storage subsystem, auxiliary heat source subsystem, and terminal subsystem;

[0080] The solar collector subsystem includes: a solar collector field, valves on each row of collectors, and water pumps for the collector field. The parameters contained in the solar collector subsystem include: inlet and outlet temperatures of each row of collectors, flow rates of each row of collectors, overall inlet and outlet temperatures and flow rates of the collector field, heat collection capacity of the collector field, heat collection efficiency of the collector field, and solar energy guarantee rate. The control object of the solar collector subsystem is the flow rate of each row of collectors and the overall collector field, which is achieved by controlling the frequency of the solar collector field circulation pump and the opening degree of each valve (including the heat exchange pump on the water tank side of the indirect system).

[0081] The thermal storage subsystem includes: a solar thermal storage device, which is connected to a plate heat exchanger via a heat exchange circulation pump. The solar thermal storage device stores the heat collected by the solar thermal generator through heat exchange. The solar thermal storage device includes: a thermal storage unit, a thermal storage water supply system, a heat exchange circulation pump for the thermal storage unit, water circulation valves for the thermal storage unit, and water supply valves for the thermal storage unit. The solar thermal storage unit contains relevant parameters including: average thermal storage temperature, stratified temperature of the thermal storage unit, inlet temperature and flow rate of the thermal storage unit, water storage capacity, stored heat and released heat. The control objects of the solar thermal storage unit are the inlet and outlet flow rates of the thermal storage unit, the stored and released heat, the water supply capacity, and the relevant temperatures of the thermal storage unit.

[0082] The auxiliary heat source subsystem includes: auxiliary heat source equipment; the auxiliary heat source equipment includes: auxiliary heat source device, auxiliary heat source side circulation pump or valve, buffer water tank; the auxiliary heat source equipment contains relevant parameters including: auxiliary heat source operating energy consumption, auxiliary heat source inlet and outlet temperature and flow rate, auxiliary heat; the auxiliary heat source equipment controls the start and stop of the auxiliary heat source and the auxiliary heat source inlet and outlet flow rates;

[0083] The terminal subsystem includes: a district heating system, which encompasses heating for residential areas, heating for one or more building complexes, or buildings within a larger area; the specific system form of the end users is not considered here, and the terminal district heating network and user load are simplified as a whole; the district heating system includes a simplified overall model of the target users and the main water pump or valve of the network; the district heating system contains relevant parameters including: total heat load, total pipeline length, number of heat users (or number of heat meters if installed), rated supply water temperature (or supply water temperature range), and rated return water temperature (or return water temperature range); the control object of the district heating system is the frequency regulation of the terminal main water pump to regulate the flow rate of the terminal main pipeline;

[0084] Step S3: Analyze the distributed predictive control structure among the subsystems and solve for the optimal signal allocation scheme among the subsystems;

[0085] Step S4: Based on the characteristics of each subsystem, coordinate the entire heating system using a distributed controller to ultimately achieve distributed control;

[0086] Step S3 includes the following sub-steps:

[0087] Step S3-1: The meteorological parameter detection module receives temperature and solar radiation measurements, sends the collected meteorological parameters to the computer through sensors, and predicts the future distribution of solar radiation and ambient temperature meteorological parameters through a constructed neural network model; the neural network model can adopt a sequence-to-sequence long short-term memory seq2seq-LSTM prediction model to realize a multi-step forward prediction strategy.

[0088] Step S3-2: For the solar collector subsystem, in order to obtain a control model oriented towards the solar collector field, the flow rate of each row of solar collectors is allocated by considering the basic relationships between the states and parameters of each row of solar collectors. The key points of this control strategy are as follows:

[0089] Step S3-2-1: Without global constraints, obtain the flow rate of each row of collectors sequentially; use the local cost function and flow rate value as decision variables:

[0090] Constraints:

[0091] In the formula, x0 and x0+N p For time step, The local cost function is expressed as a measure of the system's performance with respect to the flow sequence, q. l,i (x0) is a vector containing flow values, which can be represented as q l,i (x0)=[q1(x0),q1(x0+1),…,q1(x0+N u-1),q2(x),q2(x0+1),…,q Nl (x0+N u -1)], x0 is the controller time step corresponding to the instant t=x×ΔT, Np is the prediction range, Nu is the prediction range; qmin is the minimum flow rate of each row of collectors, and qmax is the maximum flow rate of each row of collectors, which are selected according to the Reynolds number and maximum pressure drop of each row of collectors, respectively.

[0092] Local cost function:

[0093]

[0094] In the formula, W is within a certain time step. i (x0) represents the heat collection capacity of a single-row solar collector. α(q) represents the maximum outlet temperature of a single-row solar collector. i (x0)-q i (x0-1)) 2 This represents the flow rate on a single row of solar collectors, and α is the tuning parameter.

[0095] Step S3-2-2: Is the obtained flow solution feasible? If feasible, proceed to step S3-3. If not feasible, recalculate the local model predicted control solution for each row of collectors, but the maximum flow limit is changed.

[0096] Constraints:

[0097] Local cost function in the formula As mentioned earlier, the maximum flow limit constraint has been modified to... This provides a feasible solution for traffic allocation, but it may not be the optimal solution.

[0098] Step S3-2-3: Check if the total flow rate of the system is less than or equal to the maximum flow rate. If so, check if the flow rate can be increased by adding a row of solar collectors, and then calculate the increment of the local cost function obtained by increasing the corresponding flow rate.

[0099] If at this time the condition is satisfied This increases the flow rate of one row of solar collectors within the total available flow rate, where... This is represented as an increase of Δq l,i Local cost function after (x0) flow

[0100] If the system's total flow rate is not less than or equal to the maximum flow rate, check if a cost function can be added by increasing the flow rate of one drain while reducing the flow rate of another drain by the same amount.

[0101] calculate That is, redetermine the cost function

[0102] Calculate the increment of the local cost function obtained by increasing the corresponding flow rate of the heat collection cycle j:

[0103] In the formula This is represented as an increase of Δq l,i Local cost function after (x0) flow

[0104] Calculate the reduction in the local cost function obtained by reducing the corresponding flow rate of the i-th row of collectors:

[0105] In the formula This indicates a reduction Local cost function after flow

[0106] Step S3-2-4, Judgment The formula represents the local cost function. Increment and The maximum value of the sum of reductions is used to determine whether it exceeds the total flow limit. If it does not, proceed to step S3-3. If it does, control calculations are used to reduce the flow of one row and increase the flow of another row.

[0107] Output the final flow rate of the heat collection field;

[0108] Step S3-3: Based on the final flow rate of the heat collection field obtained in the previous step, calculate whether the heat collection capacity of the heat collection field and the heat storage, auxiliary heat and load of other subsystems can reach the optimal state after adjusting the heat collection circulation pump or related control valves. If the optimal state of the system as a whole is reached, jump to step S3-7. If not, continue to step S3-4.

[0109] Step S3-4: Update the prediction parameters based on the relevant parameters between each subsystem. The prediction parameters include: heat collection, heat storage, load, auxiliary heat, and meteorological parameters. Reset the optimization variables and constraints of the heat storage subsystem and optimize the local model predictive control solution for each subsystem.

[0110] Step S3-5: Is the solution obtained in the previous step optimal? If it is the optimal solution, jump to step S3-7; if it is not the optimal solution, jump to step S3-6 and continue to adjust the state of other subsystems.

[0111] Step S3-6: Based on the optimization of the heat collection subsystem and the heat storage subsystem, update the prediction parameters again, reset the optimization variables and constraints of the auxiliary heat subsystem, and optimize the calculation of the local model predictive control solution of each subsystem; if the overall system reaches the optimal state, continue to step S3-7.

[0112] Step S3-7: Obtain the signal values ​​of the other subsystems under the optimal state of the system as a whole, with the heat collection subsystem as the dominant system.

[0113] Step S3-8: For the thermal storage subsystem, in order to obtain a control-oriented thermal storage device model, consider the relevant parameters of the thermal storage system, the thermal storage and release state, and the temperature of the energy storage device to avoid solar system stagnation and achieve optimal utilization of thermal energy reserves. Generally, the supply temperature should be maintained below 95℃ due to the boiling point of water and the temperature limits of the equipment in the solar heat transfer device. For short-term thermal storage, the storage temperature can reach 80-90℃ when the thermal storage device is in a state of moderate heat loss. High temperatures are only used when the storage size of the thermal storage device is limited or when the centralized heating network requires high temperatures to eliminate the need for auxiliary heat sources. With sufficient buffer storage capacity, a lower storage temperature is more conducive to improving the efficiency of the collector. For seasonal thermal energy storage devices, a higher inlet temperature can be used to prevent the disruption of temperature stratification. If the thermal storage device has different inlet temperatures, it can utilize solar energy over a wider temperature range because heat can enter the thermal storage device at different temperatures, thus ensuring temperature stratification. When the temperature of the seasonal heat storage device is too low, the heat cannot be directly utilized by the central heating system. In this case, an auxiliary heat source can be used to raise the temperature to the temperature required for direct heating, or the heat can be stored at the top of the heat storage device to reduce the outlet temperature of the solar collector.

[0114] Step S3-9: Repeat steps S3-2 to S3-7 to finally obtain the signal value required for heat storage and control the temperature and heat storage capacity of the heat storage device.

[0115] Step S3-10: For the auxiliary heating subsystem, in order to obtain a control-oriented auxiliary heat source model, consider the relevant parameters of the auxiliary heating subsystem, the auxiliary heat output, the auxiliary heating water temperature, and maintain the water tank temperature to stabilize the system's heat supply. Its control strategy should minimize the auxiliary heating output and maximize the output of the solar thermal collector system. The auxiliary heating form should also be optimized to utilize other heat sources such as heat pumps, boilers, and waste heat.

[0116] Step S3-11: Repeat steps S3-2 to S3-7 to finally obtain the signal value required for auxiliary heating and control the auxiliary heating temperature and amount of auxiliary heat.

[0117] Step S3-12: For the terminal subsystem, simplify the terminal as a whole, obtain the terminal load and supply and return water temperature and flow parameters, construct a neural network related model, and establish a nonlinear prediction model for predictive regulation based on the user's energy consumption.

[0118] Step S3-13: Through communication between distributed predictive control systems, continuously coordinate the control signals between the four subsystems, explore the relationship between heat collection, heat storage, auxiliary heat, and heat load, continuously adjust the constraint conditions in the MPC of each subsystem to optimize the variables, and finally enable the system as a whole to obtain the best operating state.

[0119] Furthermore, in step S1, the solar district heating system includes: a solar collector field, a heat storage device, a heat exchanger, an auxiliary heat source, and a buffer water tank; the solar collector field is used to convert solar energy into heat energy and then store the heat energy in the heat storage device after exchanging it through the heat exchanger; the heat storage device and the auxiliary heat source are used to provide district heating; and the buffer water tank is used to buffer the heating fluctuations of the heat storage device and the auxiliary heat source.

[0120] Furthermore, the solar collector field includes: a single-row flow regulation device for multiple rows of collectors, a meteorological parameter detection module, a GPS sensor, a flow and temperature detection module at the collector field, and a flow distribution control module for the collector field; the heat storage device includes: a flow and temperature detection module at the heat storage device, and a heat storage data monitoring module; the auxiliary heat source includes: a flow and temperature detection module at the auxiliary heat source, and an auxiliary heat data monitoring module; the solar district heating system also includes: a terminal overall supply and return water temperature and flow detection module, a heat collection data monitoring module, a user heat load data monitoring module, a distributed control device, a cloud control server, and a water pump control module.

[0121] Furthermore, in steps S3-8, the key control points of the thermal storage subsystem are as follows:

[0122] The solar energy storage device is connected to the plate heat exchanger via a heat exchange circulation pump. Sensors receive the current parameters and status of each subsystem of the system, predict relevant parameters in the future, optimize and calculate the control signal of the heat storage subsystem, and transmit the control signal to the heat storage circulation pump to control its flow rate and realize the change of heat exchange temperature. The water supply to the energy storage device is controlled by a float valve. When the water level in the tank is lower than the position of the float valve, tap water is supplied. When the water level reaches the control level of the float valve, the cold water is shut off.

[0123] Furthermore, in step S3-10, the key control points of the auxiliary heating subsystem are as follows:

[0124] The auxiliary heat source is designed to supplement the heat of the energy storage system. Sensors receive the current parameters and status of each subsystem, predict relevant future parameters, and optimize the calculation of the control signal for the auxiliary heating subsystem. When the heating water temperature provided by the heat storage device is lower than the set temperature, the control signal is transmitted to the auxiliary heat source and the heat exchange circulation pump of the auxiliary heating subsystem to supplement the system heat until the water temperature in the heat storage device is higher than the set temperature, at which point the supplementary heating stops. In addition, the frequency of the auxiliary heating circulation pump needs to be controlled to control the flow rate. Based on the subsystem signals, the optimal heating outlet temperature of the auxiliary heating subsystem is obtained. The subsystem signals include: heat storage, heat collection, and load.

[0125] In summary, this invention, by setting up a predictive control system at the source end, integrates the relationship between the heat source, the heat storage section, and the terminal, making full use of solar energy. This results in a solar district heating system with lower energy consumption, more precise control, and better indoor thermal comfort. Therefore, this invention has broad application prospects.

[0126] It should be emphasized that the above are merely preferred embodiments of the present invention and are not intended to limit the present invention in any way. Any simple modifications, equivalent changes and alterations made to the above embodiments based on the technical essence of the present invention shall still fall within the scope of the technical solution of the present invention.

Claims

1. A distributed predictive control method for a solar district heating system, characterized in that, Includes the following steps: Step S1: Analyze the form of the solar district heating system, the parameters of each system equipment, and establish a physical model and a nonlinear black box model for controller design; Step S2: Construct four subsystems: heat collection subsystem, heat storage subsystem, auxiliary heat source subsystem, and terminal subsystem; The solar collector subsystem includes: a solar collector field, valves on each row of collector fields, and a collector field water pump. The parameters contained in the solar collector subsystem include: inlet and outlet temperatures of each row of collectors, flow rates of each row of collectors, overall inlet and outlet temperatures and flow rates of the solar collector field, heat collection capacity of the solar collector field, solar collector field efficiency, and solar energy guarantee rate. The control object of the solar collector subsystem is the flow rate of each row of collectors and the overall solar collector field, which is achieved by controlling the frequency of the solar collector field circulation pump and the opening degree of each valve. The thermal storage subsystem includes: a solar thermal storage device, which is connected to a plate heat exchanger via a heat exchange circulation pump. The solar thermal storage device stores the heat collected by the solar thermal generator through heat exchange. The solar thermal storage device includes: a thermal storage unit, a thermal storage water supply system, a heat exchange circulation pump for the thermal storage unit, a water circulation valve for the thermal storage unit, and a water supply valve for the thermal storage unit. The solar thermal storage unit contains relevant parameters including: average thermal storage temperature, stratified temperature of the thermal storage unit, inlet temperature and flow rate of the thermal storage unit, water storage capacity, stored heat and released heat. The control objects of the solar thermal storage unit are the inlet and outlet flow rates of the thermal storage unit, the stored and released heat, the water supply capacity, and the relevant temperatures of the thermal storage unit. The auxiliary heat source subsystem includes: auxiliary heat source equipment; the auxiliary heat source equipment includes: auxiliary heat source device, auxiliary heat source side circulation pump or valve, and buffer water tank; the auxiliary heat source equipment contains relevant parameters including: auxiliary heat source operating energy consumption, auxiliary heat source inlet and outlet temperature and flow rate, and auxiliary heat; the auxiliary heat source equipment controls the start and stop of the auxiliary heat source and the auxiliary heat source inlet and outlet flow rates. The terminal subsystem includes: a district heating system, which comprises community heating and heating for one or more building complexes; the district heating system includes a simplified overall model of the target users and a main water pump or valve in the pipeline network; the district heating system contains relevant parameters including: total heat load, total pipeline length, number of heat users, rated supply water temperature, and rated return water temperature; the control object of the district heating system is the frequency regulation of the main water pump at the terminal to adjust the flow rate of the main pipeline at the terminal; Step S3: Analyze the distributed predictive control structure among the subsystems and solve for the optimal signal allocation scheme among the subsystems; Step S4: Based on the characteristics of each subsystem, coordinate the entire heating system using a distributed controller to ultimately achieve distributed control; Step S3 includes the following sub-steps: Step S3-1: The meteorological parameter detection module receives temperature and solar radiation measurements, sends the collected meteorological parameters to the computer through sensors, and predicts the future distribution of solar radiation and ambient temperature meteorological parameters through a constructed neural network model; the neural network model adopts a sequence-to-sequence long short-term memory seq2seq-LSTM prediction model to realize a multi-step forward prediction strategy. Step S3-2: For the solar collector subsystem, in order to obtain a control model for the solar collector field, the flow rate of each row of solar collectors is allocated by considering the basic relationships between the states and parameters of each row of solar collectors. The key points of the control strategy are as follows: Step S3-2-1: Without global constraints, obtain the flow rate of each row of collectors sequentially; use the local cost function and flow rate value as decision variables: In the formula and For time step, The local cost function is used to measure the system's performance with respect to a flow sequence. The vector containing the flow value is represented as , It is instant The corresponding controller time step, N p It is the prediction range. N u It is the prediction range; q min It is the minimum flow rate of each row of solar collectors. q max It is the maximum flow rate of each row of collectors, which is selected based on the Reynolds number and maximum pressure drop of each row of collectors; Local cost function: In the formula, within a certain time step This represents the heat collection capacity of a single-row solar collector. This represents the maximum outlet temperature of a single-row solar collector. Expressed as the flow rate on a single row of solar collectors. These are tuning parameters; Step S3-2-2: Is the obtained flow solution feasible? If feasible, proceed to step S3-3. If not feasible, recalculate the local model predicted control solution for each row of collectors, but the maximum flow limit is changed. Local cost function in the formula As mentioned earlier, the maximum flow limit constraint has been modified to... This provides a feasible solution for traffic allocation, but it may not be the optimal solution. Step S3-2-3: Check if the total system flow rate is less than or equal to the maximum flow rate. If so, check if adding a row of solar collectors will increase the flow rate, and then calculate the increment of the local cost function obtained by increasing the corresponding flow rate. If at this time the conditions are met This increases the flow rate of one row of solar collectors within the total available flow rate, where... This indicates that an increase has been made. Local cost function after flow If the system's total flow rate is not less than or equal to the maximum flow rate, check if the cost function should be increased by adding flow rate to one drain and reducing flow rate to another drain by the same amount. calculate That is, redetermine the cost function Calculate the increment of the local cost function obtained by increasing the corresponding flow rate of the heat collection cycle j: In the formula This indicates that an increase has been made. Local cost function after flow Calculate the reduction in the local cost function obtained by reducing the corresponding flow rate of the i-th row of collectors: In the formula This indicates a reduction Local cost function after flow Step S3-2-4, Judgment The formula represents the local cost function. Increment and The maximum value of the sum of reductions is used to determine whether it exceeds the total flow limit. If it does not, proceed to step S3-3. If it does, control calculations are used to reduce the flow of one row and increase the flow of another row. Output the final flow rate of the heat collection field; Step S3-3: Based on the final flow rate of the heat collection field obtained in the previous step, calculate whether the heat collection capacity of the heat collection field and the heat storage, auxiliary heat and load of other subsystems can reach the optimal state after adjusting the heat collection circulation pump or related control valves. If the optimal state of the system as a whole is reached, jump to step S3-7. If not, continue to step S3-4. Step S3-4: Update the prediction parameters based on the relevant parameters between each subsystem. The prediction parameters include: heat collection, heat storage, load, auxiliary heat, and meteorological parameters. Reset the optimization variables and constraints of the heat storage subsystem and optimize the calculation of the local model predictive control solution for each subsystem. Step S3-5: Is the solution obtained in the previous step optimal? If it is the optimal solution, jump to step S3-7; if it is not the optimal solution, jump to step S3-6 and continue to adjust the state of other subsystems. Step S3-6: Based on the optimization of the heat collection subsystem and the heat storage subsystem, update the prediction parameters again, reset the optimization variables and constraints of the auxiliary heat subsystem, and optimize the calculation of the local model prediction control solution of each subsystem; if the overall optimal state of the system is reached, continue to step S3-7. Step S3-7: Obtain the signal values ​​of the other subsystems under the optimal state of the system as a whole, with the heat collection subsystem as the dominant system. Step S3-8: For the thermal storage subsystem, in order to obtain a control-oriented thermal storage device model, consider the relevant parameters of the thermal storage system, the thermal storage and release state and the temperature of the energy storage device, avoid the solar energy system from shutting down and achieve the best utilization of thermal energy storage. Step S3-9: Repeat steps S3-2 to S3-7 to finally obtain the signal value required for heat storage and control the temperature and heat storage capacity of the heat storage device. Step S3-10: For the auxiliary heating subsystem, in order to obtain a control-oriented auxiliary heat source model, consider the relevant parameters of the auxiliary heating subsystem, the auxiliary heat output, the auxiliary heating water temperature, and maintain the water tank temperature to stabilize the system's heat supply. Its control strategy should reduce the auxiliary heating output and increase the output of the solar thermal collector system. The auxiliary heating form should optimize the use of other heat sources. Step S3-11: Repeat steps S3-2 to S3-7 to finally obtain the signal value required for auxiliary heating and control the auxiliary heating temperature and amount of auxiliary heat. Step S3-12: For the terminal subsystem, simplify the terminal as a whole, obtain the terminal load and supply and return water temperature and flow parameters, construct a neural network related model, and establish a nonlinear prediction model for predictive regulation based on the user's energy consumption. Step S3-13: Through communication between distributed predictive control systems, continuously coordinate the control signals between the four subsystems, explore the relationship between heat collection, heat storage, auxiliary heat, and heat load, continuously adjust the constraint conditions in the MPC of each subsystem to optimize the variables, and finally enable the system as a whole to obtain the best operating state.

2. The distributed predictive control method for a solar district heating system according to claim 1, characterized in that, In step S1, the solar district heating system includes: a solar collector field, a heat storage device, a heat exchanger, an auxiliary heat source, and a buffer water tank; the solar collector field is used to convert solar energy into heat energy and then store the heat energy in the heat storage device after exchanging it through the heat exchanger; the heat storage device and the auxiliary heat source are used to provide district heating; and the buffer water tank is used to buffer the heating fluctuations of the heat storage device and the auxiliary heat source.

3. The distributed predictive control method for a solar district heating system according to claim 2, characterized in that, The solar collector field includes: a multi-row collector field single-row flow regulation device, a meteorological parameter detection module, a GPS sensor, a flow and temperature detection module at the collector field, and a collector field flow distribution control module; the heat storage device includes: a heat storage device flow and temperature detection module and a heat storage data monitoring module; the auxiliary heat source includes: an auxiliary heat source flow and temperature detection module and an auxiliary heat data monitoring module; the solar district heating system also includes: a terminal overall supply and return water temperature and flow detection module, a heat collection data monitoring module, a user heat load data monitoring module, a distributed control device, a cloud control server, and a water pump control module.

4. The distributed predictive control method for a solar district heating system according to claim 1, characterized in that, In steps S3-8, the key control points of the thermal storage subsystem are as follows: The solar energy storage device is connected to the plate heat exchanger via a heat exchange circulation pump. Sensors receive the current parameters and status of each subsystem of the system, predict relevant parameters in the future, optimize and calculate the control signal of the heat storage subsystem, and transmit the control signal to the heat storage circulation pump to control its flow rate and realize the change of heat exchange temperature. The water supply to the energy storage device is controlled by a float valve. When the water level in the tank is lower than the position of the float valve, tap water is supplied. When the water level reaches the control level of the float valve, the cold water is shut off.

5. The distributed predictive control method for a solar district heating system according to claim 1, characterized in that, In step S3-10, the key control points of the auxiliary heating subsystem are as follows: The auxiliary heat source is designed to supplement the heat of the energy storage system. The sensor receives the current parameters and status of each subsystem of the system, predicts the relevant parameters in the future, and optimizes the calculation of the control signal of the auxiliary heat subsystem. When the heating water temperature provided by the heat storage device is lower than the set temperature, the control signal is transmitted to the auxiliary heat source and the heat exchange circulation pump of the auxiliary heat subsystem to supplement the system heat until the water temperature in the heat storage device is higher than the set temperature and the heat supplementation stops. In addition, the frequency of the auxiliary heat circulation pump needs to be controlled to control the flow rate, and the optimal heating outlet temperature of the auxiliary heat subsystem can be obtained based on the subsystem signal. The subsystem signals include: heat storage, heat collection, and load.

Images