Optimization control method for coupling of ground source heat pump and PVT system

Abstract

The invention relates to the technical field of heating, in particular to an optimal control method for coupling of a ground source heat pump and a PVT system. According to the technical scheme, the optimal control method for coupling of the ground source heat pump and the PVT system is applied to a coupling system, and the coupling system at least comprises a PVT assembly, a heat storage water tank, a ground source heat pump main machine, a ground heat exchanger, a user terminal, a plurality of circulating water pumps, a plurality of electric valves and a central controller. The method comprises the following steps that S1, system operation parameters are collected in real time through a sensor network, and the parameters at least comprise the PVT assembly outlet temperature (T1), the heat storage water tank layering temperature, the ground heat exchanger outlet temperature T3, the user side water supply and return temperature, the environment temperature Ta, the total solar irradiance G, the loop flow and the system power consumption. Intelligent and self-adaptive operation of the system is realized, and the system can respond to internal and external state changes in real time and dynamically adjust an operation strategy through a closed-loop control architecture.

Description

Technical Field

[0001] This invention relates to the field of heating technology, and in particular to an optimized control method for coupling a ground source heat pump with a PVT system. Background Technology

[0002] Ground source heat pump systems are highly energy efficient due to their utilization of underground constant-temperature thermal energy, while photovoltaic thermal (PVT) systems can simultaneously generate electricity and heat, representing a highly efficient way to utilize solar energy. Theoretically, coupling the two can complement each other's advantages and improve the utilization rate of renewable energy. However, in current technologies, the coupled operation of ground source heat pump and PVT systems often employs relatively simple control strategies, such as on / off control based on a single temperature threshold or mode switching relying on human experience. These control methods often only focus on local or instantaneous states, lacking coordinated optimization and forward-looking planning for the overall energy flow of the system. As a result, the system's operating mode is rigid, making it difficult to dynamically adapt to complex changes in external weather and user loads. This can lead to problems such as energy supply and demand mismatch, soil heat accumulation or heat deficit (i.e., soil thermal imbalance), limiting further improvement in the overall energy efficiency of the coupled system and failing to fully realize its energy-saving potential.

[0003] Existing technologies cannot proactively adjust control strategies based on future climate and load changes, resulting in poor operational economy. Mode switching logic is simplistic or relies on manual intervention, making it impossible to automatically select the most energy-efficient operating state in real time in complex systems with multiple energy inputs and multiple operating modes. The lack of effective closed-loop feedback and adaptive mechanisms means that the system cannot continuously optimize its control strategy based on actual operating results, and its energy efficiency may gradually decline over long-term operation due to equipment performance degradation or changes in the external environment. Summary of the Invention

[0004] This invention proposes an optimized control method for coupling ground source heat pumps and PVT systems, which solves the problem that the control methods of ground source heat pumps and PVT coupled systems in the prior art lack intelligent coordination and dynamic optimization capabilities, resulting in the overall energy efficiency not being fully utilized.

[0005] To solve the above-mentioned technical problems, the technical solution adopted by the present invention is as follows: An optimized control method for coupling a ground source heat pump and a PVT system is disclosed. The method is applied to a coupled system, which includes at least a PVT module, a hot water storage tank, a ground source heat pump unit, a buried pipe heat exchanger, user terminals, multiple circulating water pumps, multiple electric valves, and a central controller. The method includes the following steps: S1: Real-time acquisition of system operating parameters via sensor network, including at least: PVT component outlet temperature (T1), hot water storage tank stratified temperatures (T2a top high temperature, T2b middle temperature, T2c bottom low temperature), buried pipe heat exchanger outlet temperature (T3), user-side supply and return water temperatures (supply water temperature is T4, return water temperature is T5), ambient temperature (Ta), total solar irradiance (G), flow rate of each loop, and system power consumption; S2: The central controller executes a multi-layer decision logic based on the real-time operating parameters and predicted parameters to determine the current optimal operating mode and generate a control instruction set; the multi-layer decision logic includes at least: a fast rule-based decision based on real-time parameters and set thresholds, and an optimization-based decision based on the system model and predicted data; S3: The central controller sends the control instruction set to the corresponding actuator. The control instruction set includes at least: switching or regulating instructions for multiple electric valves used to switch fluid paths, frequency conversion instructions for circulating water pumps used to regulate flow, and instructions for controlling the start-up, shutdown, and capacity of the ground source heat pump unit. S4: After executing the control command, return to step S1 to form closed-loop control, and dynamically adjust the operating mode according to the changes in the operating parameters.

[0006] Furthermore, the operating modes in step S2 include at least five core modes, and their switching logic is as follows: Mode 1, PVT direct storage and active soil heating mode: When it is detected that the user has no immediate heating or cooling needs, and G is greater than the first irradiation threshold, and T1 is greater than T3, the PVT circulation pump and the heating valve connected to the buried pipe side are turned on to inject the heat energy collected by PVT directly or after buffering through the water tank into the soil. Mode 2, direct heating mode from the hot water storage tank: When a user's heating demand is detected and T2a is higher than the user's terminal water supply set temperature, the corresponding valve is controlled to form an independent cycle of "hot water storage tank → user terminal → hot water storage tank", and the ground source heat pump unit is shut down. Mode 3, PVT preheating assisted heat pump heating mode: When users have heating demand but T2a is lower than the direct supply threshold and T1 is higher than the current inlet water temperature on the ground source heat pump evaporator side, the control valve makes the PVT loop output and the buried pipe side loop connected in series or in parallel before the heat pump evaporator to improve the quality of the low temperature heat source of the heat pump. Mode 4, Ground Source Heat Pump Standalone Operation Mode: When the conditions of Mode 2 and 3 are not met, but the user has heating or cooling needs, the ground source heat pump unit is started to extract or release heat from the soil. Mode 5, PVT priority cooling and power generation mode: In summer cooling conditions, when T1 is higher than the set heat dissipation threshold, the control valve guides the waste heat generated by the PVT to the auxiliary heat dissipation device to prevent it from entering the soil or hot water storage tank.

[0007] Furthermore, the switching between the modes has priority logic. When the central controller determines that a user has a heating demand, it attempts to enter the corresponding mode in the following priority order: it first tries to meet the conditions of mode two; if the conditions are not met, it tries to meet the conditions of mode three; if the conditions are still not met, it enters mode four.

[0008] Furthermore, the optimization layer decision in step S2 is implemented using a model predictive control algorithm, specifically including: S2.1: Establish a comprehensive system model that includes the component model of the coupled system, the building load prediction model, and the electricity price model; S2.2: With the forecast time domain being at least 24 hours in the future, the optimization objective being the lowest total system operating cost or the lowest primary energy consumption, and with constraints such as equipment physical limitations and soil thermal balance constraints, a rolling optimization problem is constructed. S2.3: In each control cycle, solve the optimization problem to obtain the optimal control sequence of each actuator in the future time domain, and apply the instructions of the first control cycle to the system.

[0009] Furthermore, the soil heat balance constraint is that, within a set period, the absolute value of the difference between the total heat extracted from the soil and the total heat discharged to the soil by the buried pipe heat exchanger is not greater than a set threshold; in the optimization objective function, different cost coefficients are set for the behavior of extracting heat from the soil or discharging heat to the soil, so as to guide the controller to actively supplement the soil with heat during periods of low electricity prices or abundant solar energy.

[0010] Furthermore, the method also includes optimized control of the energy generated by the PVT: The electrical energy generated by the PVT is preferentially used through the inverter to drive the compressor, circulating water pump, and central controller of the ground source heat pump unit. When PVT generates a surplus of electricity, the surplus energy is stored in batteries or fed into the grid; when PVT generates insufficient electricity, the grid provides supplementary power. The central controller adjusts the start / stop and power of the auxiliary electric heater in the hot water storage tank according to the time-of-use electricity price signal, so as to store heat during the off-peak electricity price period.

[0011] Furthermore, the hot water storage tank is a layered hot water storage tank. In step S3, the opening degree of the valves connected to different heights of the water tank is controlled to achieve graded storage and retrieval of heat energy: in the direct supply mode, high-temperature water is taken from the top of the water tank; when preheating is required for the heat pump, medium-temperature water is taken from the middle of the water tank; when receiving PVT heat, the appropriate height of the water tank is selected according to the inlet water temperature of the PVT circuit.

[0012] Furthermore, the sensor network also includes soil temperature sensors arranged at different depths and locations in the buried pipeline field to monitor the distribution of the soil temperature field; in the decision-making process of step S2, the central controller combines the soil temperature field data to dynamically adjust the target temperature of soil heating and the selection of the heating circuit to achieve balanced restoration of soil temperature.

[0013] Furthermore, the central controller is pre-installed with a self-learning module, which is used to record historical operating data, mode switching effects and actual energy efficiency, and dynamically correct the judgment threshold of the fast rule layer in step S2 and the parameters of the system comprehensive model in step S4 through machine learning algorithms, so that the control strategy can adapt to changes in local climate and user habits.

[0014] Furthermore, it also includes a feedback correction step: after each rolling optimization cycle of the model predictive control, the actual measured key operating parameters of the system are compared with the corresponding parameters predicted in the previous cycle, and the prediction error is calculated; based on the prediction error, the key parameters in the system integrated model are dynamically corrected or the initial system state in the next prediction time domain is corrected to improve the accuracy of subsequent rolling optimization.

[0015] The positive effects of this invention are: The system achieves intelligent and adaptive operation. Through a closed-loop control architecture of "perception-decision-execution-feedback", the system can respond to changes in internal and external states in real time and dynamically adjust its operating strategy, thus possessing adaptive capabilities and significantly improving stability and applicability under different working conditions.

[0016] The system's overall energy efficiency and economy have been improved. By introducing a multi-layer decision-making logic of "rapid rule layer + optimization layer", the system can not only ensure rapid and reliable response under critical operating conditions, but also optimize energy efficiency from a global and long-term perspective, achieving the best balance between instantaneous response and long-term economy, thereby significantly improving the system's comprehensive energy efficiency.

[0017] This method solves the challenge of coordinated control in complex coupled systems. It transforms the complex energy coupling problem into a structured, multi-layered decision-making problem and automatically selects the optimal operating mode through clear logic. This effectively solves the coordination problem between ground source heat pumps and PVT systems, as well as between energy supply and demand, ensuring the precise and efficient use of energy. Detailed Implementation

[0018] The technical solutions of the present invention will be clearly and completely described below with reference to 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.

[0019] An optimized control method for coupling a ground source heat pump and a PVT system is disclosed. The method is applied to a coupled system, which includes at least a PVT component 1, a hot water storage tank 2, a ground source heat pump unit 3, a buried pipe heat exchanger 4, a user terminal 5, multiple circulating water pumps, multiple electric valves, and a central controller. The method includes the following steps: S1: Collect system operating parameters in real time through a sensor network. The parameters include at least: PVT component outlet temperature (T1), hot water storage tank stratified temperature (T2a top high temperature, T2b middle temperature, T2c bottom low temperature), buried pipe heat exchanger outlet temperature (T3), user-side supply and return water temperature (T4, T5), ambient temperature (Ta), total solar irradiance (G), flow rate of each loop, and system power consumption. S2: The central controller executes a multi-layer decision logic based on the real-time operating parameters and predicted parameters to determine the current optimal operating mode and generate a control instruction set; the multi-layer decision logic includes at least: a fast rule-based decision based on real-time parameters and set thresholds, and an optimization-based decision based on the system model and predicted data; S3: The central controller sends the control instruction set to the corresponding actuator. The control instruction set includes at least: switching or regulating instructions for multiple electric valves used to switch fluid paths, frequency conversion instructions for circulating water pumps used to regulate flow, and instructions for controlling the start-up, shutdown and capacity of the ground source heat pump host 3. S4: After executing the control command, return to step S1 to form closed-loop control, and dynamically adjust the operating mode according to the changes in the operating parameters.

[0020] The core of this solution is the closed-loop control process of the central controller. In step S1, the specific implementation of the sensor network includes: installing a temperature sensor on the outlet pipe of the PVT component to collect temperature T1; vertically installing three temperature sensors at the top, middle, and bottom of the hot water storage tank to collect the high-temperature zone temperature T2a, the medium-temperature zone temperature T2b, and the low-temperature zone temperature T2c, respectively; installing a temperature sensor on the outlet pipe of the buried pipe heat exchanger to collect temperature T3; installing temperature sensors on the user-side supply and return water pipes to collect temperature T4 and T5, respectively; installing an ambient temperature sensor (Ta) and a total solar irradiance sensor (G) outdoors; installing flow sensors on each circulation loop to collect flow rates; and installing an energy meter on the system's main power supply line to collect system power consumption. These sensors transmit real-time data to the central controller via wired or wireless means.

[0021] In step S2, the multi-layer decision-making logic is implemented as follows: The rapid rule-based decision-making layer involves the controller instantly comparing the collected real-time parameters (such as T1, T2a, T3, T4, T5, G, etc.) with preset fixed thresholds (such as the first irradiance threshold, direct supply threshold, heat dissipation threshold, etc.), and directly generating a preliminary control intention based on the comparison result. The optimization-based decision-making layer involves the controller calling upon the built-in system comprehensive model (including component characteristic curves, building thermal parameters, etc.) and combining it with externally acquired predictive parameters (such as 24-hour weather forecast data and time-of-use electricity price information) to perform forward-looking optimization calculations and output a better control strategy. The results of the two layers of decision-making are fused through weighted or logical judgment to ultimately determine the current optimal operating mode and generate a control instruction set containing specific action instructions.

[0022] In step S3, the issuance of the control command set is specifically manifested as follows: the controller sends instructions to the designated electric valve to open, close, or adjust the opening degree through the digital output module; sends the target frequency instruction to the variable frequency circulating water pump through the analog output module or communication protocol; and sends start, stop, or capacity adjustment instructions to the ground source heat pump host through the dedicated communication interface.

[0023] The specific implementation of step S4 to form closed-loop control is as follows: After executing step S3, the controller does not stop, but immediately returns to step S1 to start the next control cycle, collects the latest system operating parameters again, and re-executes steps S2 and S3 based on the new parameter values, thereby realizing adaptive closed-loop control that continuously adjusts the control strategy according to the dynamic changes in the system operating conditions.

[0024] By establishing a complete closed-loop control architecture of "perception-decision-execution-feedback," the system has undergone a fundamental transformation from static, open-loop operation to dynamic, adaptive operation. Its beneficial effects are as follows: First, the comprehensive collection of operating parameters through a sensor network provides a precise data foundation for intelligent decision-making, overcoming the blindness of experience-based control. Second, the design of multi-layered decision logic (rapid rule layer + optimization layer) enables the system to respond instantly and reliably to sudden operating conditions while also optimizing energy efficiency from a long-term perspective, balancing real-time performance and economy. Finally, closed-loop control ensures that the system continuously tracks its own state changes and dynamically adjusts, significantly improving its adaptability and overall energy efficiency under different weather conditions and load demands, achieving automated and intelligent operation of the system.

[0025] The operating modes in step S2 include at least five core modes, and the switching logic is as follows: Mode 1, PVT direct storage and active soil heating mode: When it is detected that the user has no immediate heating or cooling needs, and G is greater than the first irradiation threshold, and T1 is greater than T3, the PVT circulation pump and the heating valve connected to the buried pipe side are turned on to inject the heat energy collected by PVT directly or after buffering through the water tank into the soil. Mode 2, direct heating mode of hot water storage tank: When a user's heating demand is detected and T2a is higher than the user's terminal water supply set temperature, the corresponding valve is controlled to form an independent cycle of "hot water storage tank 2 → user terminal 5 → hot water storage tank 2", and the ground source heat pump unit 3 is shut down. Mode 3, PVT preheating assisted heat pump heating mode: When users have heating demand but T2a is lower than the direct supply threshold and T1 is higher than the current inlet water temperature on the ground source heat pump evaporator side, the control valve makes the PVT loop output and the buried pipe side loop connected in series or in parallel before the heat pump evaporator to improve the quality of the low temperature heat source of the heat pump. Mode 4, Ground Source Heat Pump Standalone Operation Mode: When the conditions of Mode 2 and 3 are not met, but the user has heating or cooling needs, the ground source heat pump unit 3 is started to extract or exhaust heat from the soil. Mode 5, PVT priority cooling and power generation mode: In summer cooling conditions, when T1 is higher than the set heat dissipation threshold, the control valve guides the waste heat generated by the PVT to the auxiliary heat dissipation device to prevent it from entering the soil or hot water storage tank.

[0026] The implementation of the five core modes relies on the monitoring of specific parameter combinations and the corresponding coordinated control of valves and pumps. The implementation details of Mode 1 (PVT direct storage and active soil heating mode) are as follows: The controller continuously monitors the user-end signal. When the heating / cooling valve is closed and the flow rate is extremely low, it is determined that there is no immediate demand. Simultaneously, it monitors that G is greater than the set first irradiance threshold (e.g., 300W / m²). 2When T1 is consistently higher than T3 by a certain difference, the controller generates a command to start the PVT-side circulation pump and open the dedicated electric valve for heat replenishment connected between the PVT loop and the underground pipe loop. Simultaneously, it may close the valves flowing to the water tank or user terminal, allowing the heat generated by the PVT to be directly pumped into the underground pipe. The implementation details of Mode Two (direct heating mode from the hot water storage tank) are as follows: When an activation signal is detected at the user terminal (e.g., thermostat activation), and T2a is higher than the set water supply temperature (e.g., 40℃) for the user terminal (e.g., underfloor heating), the controller generates a command to open the electric valves connecting the top of the water tank to the user's water supply pipe and the user's return water pipe to the bottom of the water tank. Simultaneously, it closes the valves of the ground source heat pump unit and its connected pipes, and starts the user-side circulation pump, forming a circulation independent of the heat pump. The implementation details of Mode 3 (PVT preheating assisted heat pump heating mode) are as follows: When there is a heating demand but T2a is lower than the direct supply threshold, the controller simultaneously compares T1 with the current fluid temperature entering the heat pump evaporator (which may be T3 or the temperature in the middle of the water tank). If T1 is higher, the controller controls the three-way valve or parallel valve to mix the PVT hot fluid with the ground source return water before entering the heat pump evaporator, thus increasing the inlet temperature. Mode 4 (Ground source heat pump standalone operation mode) is the default action when the conditions of Modes 2 and 3 are not met. The implementation details of Mode 5 (PVT priority cooling and power generation mode) are as follows: During the cooling season, when T1 is higher than the set heat dissipation threshold (e.g., 50°C), the controller operates the valve to switch the PVT loop to connect with the cooling tower or air radiator, preventing high-temperature fluid from entering the system core.

[0027] By defining five core operating modes, the system comprehensively covers optimal operating strategies under various conditions. Its beneficial effects include precise energy scheduling and efficient utilization: Mode 1 actively replenishes soil heat when solar energy is abundant and there is no load, effectively solving the soil thermal imbalance problem caused by long-term operation of the ground source heat pump and ensuring its long-term high performance. Mode 2 utilizes high-temperature water from the tank for direct heating when conditions permit, completely avoiding the need to operate the high-energy-consuming heat pump compressor and maximizing instantaneous energy efficiency. Mode 3 utilizes PVT heat to increase the source-side temperature of the heat pump, directly improving the heat pump's coefficient of performance. Mode 4 serves as a basic guarantee, while Mode 5 avoids the adverse effects of high-temperature PVT waste heat on the soil thermal environment in summer. Overall, this design achieves dynamic optimal matching between solar energy, soil energy, electrical energy, and user load.

[0028] The switching between modes has a priority logic. When the central controller determines that a user has a heating demand, it attempts to enter the corresponding mode in the following priority order: it first tries to meet the conditions of mode two; if the conditions are not met, it tries to meet the conditions of mode three; if the conditions are still not met, it enters mode four.

[0029] The priority logic is implemented in the controller program using a sequential judgment structure. When the controller detects a user's heating demand (e.g., receiving a heating start signal), its internal program first calls the "Mode Two Condition Judgment Subroutine," which checks if T2a is greater than or equal to the set temperature. If the condition is true, the program immediately jumps to the "Mode Two Execution Subroutine" and skips subsequent mode judgments. If the condition is false, the program then calls the "Mode Three Condition Judgment Subroutine," checking if T1 is higher than the heat pump evaporator inlet water temperature, etc. If true, it jumps to the "Mode Three Execution Subroutine." If none of the aforementioned conditions are met, the program finally executes the "Mode Four Execution Subroutine." This sequential logic is implemented in the controller using if-else if-else or case statements, ensuring that the highest priority available mode is executed first.

[0030] By setting explicit mode-switching priority logic, the direct benefit is that it forces the system to automatically select the most energy-efficient operating mode under any heating demand. This "energy efficiency first" decision-making mechanism eliminates the possibility of the system operating in an inefficient state from a control logic perspective, ensuring the tiered and rational use of energy. For example, it prioritizes direct heating, which consumes no electricity, followed by heat pump heating, which consumes electricity but has improved efficiency, and finally conventional heat pump heating, thus ensuring the continuous optimization of overall energy efficiency from the system control strategy level.

[0031] The optimization layer decision in step S2 is implemented using a model predictive control algorithm, specifically including: S2.1: Establish a comprehensive system model that includes the component model of the coupled system, the building load prediction model, and the electricity price model; S2.2: With the forecast time domain being at least 24 hours in the future, the optimization objective being the lowest total system operating cost or the lowest primary energy consumption, and with constraints such as equipment physical limitations and soil thermal balance constraints, a rolling optimization problem is constructed. S2.3: In each control cycle, solve the optimization problem to obtain the optimal control sequence of each actuator in the future time domain, and apply the instructions of the first control cycle to the system.

[0032] The specific implementation of the model predictive control algorithm is divided into three steps. In step S2.1, establishing a comprehensive system model refers to storing or dynamically calculating the mathematical relationships describing the system characteristics within the controller, including: the relationship between the heat / electricity efficiency of the PVT components and irradiance and temperature (model); the variation curves of the cooling / heating coefficient of performance of the ground source heat pump host with load rate and source / user side temperature (model); the heat transfer characteristic parameters of the building envelope (such as thermal resistance and heat capacity) used to construct the load prediction model; and the time-of-use electricity price data model for the next 24 hours. In step S2.2, constructing a rolling optimization problem means that when the algorithm starts at each control cycle (e.g., 15 minutes), it takes the current moment as the starting point, the next 24 hours as the prediction time domain, sets the objective function to minimize the total operating electricity cost (or primary energy consumption) during that period, and uses the maximum and minimum power of the equipment, the upper limit of the water tank capacity, and the soil thermal balance constraints as described in claim 5 as inequality constraints to form a standard optimization problem. In step S2.3, solving the optimization problem and applying instructions means that the controller uses its built-in optimization solver (such as a linear programming or quadratic programming solver) to calculate the optimal setpoint sequence for each actuator (valve, water pump, heat pump) for each sub-time period (e.g., every 15 minutes) within the next 24 hours, but only issues the setpoint instruction corresponding to the current moment (i.e., the first sub-time period) to the actuator. In the next control cycle, the algorithm re-acquires the latest system state and external prediction data, and performs optimization calculations again, achieving "rolling".

[0033] The core benefit of introducing model predictive control algorithms is that they elevate control strategies from passive response to proactive optimization. Based on forecasts of future weather, load, and electricity prices, these algorithms can proactively formulate optimal operating plans, achieving "preemptive" energy management. For example, they can predict cloudy or rainy weather tomorrow, allowing for pre-emptive heat storage or soil heating during periods of abundant solar energy today to meet tomorrow's heating demands. This proactive optimization capability significantly smooths system load, reduces peak power, and fully utilizes off-peak electricity and solar energy, thereby minimizing the system's total daily operating costs or primary energy consumption while ensuring comfort.

[0034] The soil heat balance constraint is that within a set period, the absolute value of the difference between the total heat taken from the soil and the total heat discharged to the soil by the buried pipe heat exchanger (4) is not greater than a set threshold; in the optimization objective function, different cost coefficients are set for the behavior of taking heat from the soil or discharging heat to the soil, so as to guide the controller to actively supplement the soil with heat during periods of low electricity prices or abundant solar energy.

[0035] In the optimization problem of the aforementioned model predictive control algorithm, the specific implementation of the soil heat balance constraint is to transform a long-term physical requirement into a mathematical constraint. For example, a period (such as one year) is set, and the constraint condition is expressed as |ΣQ_extract - ΣQ_reject| ≤ Q_max, where ΣQ_extract is the predicted total heat extraction from the soil within the predicted time domain (but the model will consider the long-term heat accumulation effect), ΣQ_reject is the predicted total heat discharge to the soil, and Q_max is the maximum allowed net heat extraction. This constraint must be satisfied in each rolling optimization. Simultaneously, in the objective function, a positive unit cost coefficient C1 is assigned to the "heat extraction from the soil" behavior, while a zero or negative cost coefficient C2 is assigned to the "heat discharge to the soil" behavior (especially during periods of low electricity prices, C2 can be negative, representing a reward). Thus, when minimizing the total cost, the optimization algorithm will naturally tend to schedule more heat discharge behavior during periods with lower (or negative) C2 to balance the heat extraction during periods with higher C1, thereby achieving automatic adjustment of the soil heat balance through economic incentives.

[0036] By specifying soil thermal balance constraints and differentiated cost coefficients in the model predictive control algorithm, the beneficial effect is that the long-term, abstract environmental goal of maintaining soil thermal balance is transformed into a concrete, actionable optimization problem that is synergistic with economic objectives. While pursuing the lowest operating cost, the system automatically and intelligently arranges supplementary heating activities (such as during periods of low electricity prices or solar energy surplus), thereby effectively preventing a continuous drop or rise in soil temperature without sacrificing or even improving economic efficiency. This ensures the long-term, efficient, and stable operation of the ground source heat pump system, achieving a balance between economic benefits and sustainability.

[0037] The method also includes optimized control of the power generated by the PVT: The electrical energy generated by the PVT is preferentially used by the inverter to drive the compressor, circulating water pump and central controller of the ground source heat pump host (3); When PVT generates a surplus of electricity, the surplus energy is stored in batteries or fed into the grid; when PVT generates insufficient electricity, the grid provides supplementary power. The central controller adjusts the start / stop and power of the auxiliary electric heater in the hot water storage tank (2) according to the time-of-use electricity price signal, so as to store heat during the low electricity price period.

[0038] The specific implementation of optimized utilization control of PVT-generated power in hardware and software is as follows: The DC power generated by the PVT is connected to a bidirectional inverter, which prioritizes supplying power to the AC bus connecting the ground source heat pump compressor, circulating water pump, and controller. The controller reads the inverter's output power and the system's total power consumption in real time via communication. If the power generation exceeds the power consumption, the controller instructs the inverter to send the surplus power to the battery for storage or invert it into AC power that meets grid requirements and feed it into the grid; if the power generation is insufficient, the controller instructs the inverter to draw power from the grid to supplement it. Simultaneously, the controller has a built-in clock and time-of-use electricity meter. During the set off-peak electricity price period (e.g., 23:00 to 7:00 the next day), if T2a is detected to be lower than a certain set value, a signal is output to start the auxiliary electric heater in the hot water storage tank, and its heating power can be adjusted according to the temperature difference; during peak electricity price periods, a signal is output to forcibly shut down the electric heater.

[0039] The optimized utilization and control of PVT power generation has two main benefits: First, by adhering to the principle of "self-consumption and surplus power fed into the grid," the local consumption rate of photovoltaic energy is greatly improved, reducing dependence on the external power grid and lowering the electricity costs of system operation. Second, by combining time-of-use pricing with the control of electric heaters, "peak shaving and valley filling" are achieved. This involves converting electricity into heat energy for storage during off-peak hours and releasing it during peak hours. This not only further reduces operating costs but also helps alleviate peak load pressure on the power grid, improving the system's economic efficiency and user-friendliness.

[0040] The hot water storage tank 2 is a layered hot water storage tank. In step S3, the opening degree of the valves connected to the water tank at different heights is controlled to achieve the graded storage and retrieval of heat energy: in the direct supply mode, high-temperature water is taken from the top of the water tank; when it is necessary to provide preheating for the heat pump, medium-temperature water is taken from the middle of the water tank; when receiving PVT heat, the appropriate height of the water tank is selected according to the inlet water temperature of the PVT circuit.

[0041] The implementation of tiered hot water storage and stratification relies on the tank structure and precise valve control. The tank has inlets with independent electric valves at different vertical heights (top, middle, and bottom). In direct supply mode, the controller only opens the outlet valve at the top and the return valve at the bottom, thus using only the hottest water from the top. When preheating for the heat pump is required, the controller closes the top valve and opens the outlet valve in the middle of the tank, using medium-temperature water. When the PVT loop injects heat into the tank, the controller compares the PVT return water temperature with the temperatures T2a, T2b, and T2c in the tank. If the return water temperature is higher than T2a, the top inlet valve is opened to inject hot water into the top layer; if the return water temperature is between T2b and T2a, the middle inlet valve is opened; and if the return water temperature is lower, the bottom inlet valve is opened. This control strategy utilizes the natural stratification of hot water, reducing heat loss caused by mixing water of different temperatures.

[0042] By controlling tiered hot water storage tanks to achieve tiered thermal energy storage and retrieval, the beneficial effect is a significant improvement in the system's energy efficiency (usable energy efficiency). Adhering to the principle of "high-quality, high-use; low-quality, low-use" (i.e., direct supply of high-temperature water, preheating of medium-temperature water, and injection according to temperature), the loss of usable energy caused by mixing water of different temperatures is effectively reduced. This allows the stored thermal energy to exert its maximum work capacity, thereby meeting the same demand with less energy and improving the overall quality and efficiency of the energy system. This represents a significant improvement in efficiency, going beyond the traditional focus solely on the "quantity" of energy.

[0043] The sensor network also includes soil temperature sensors deployed at different depths and locations in the buried pipeline field to monitor the distribution of the soil temperature field. In the decision-making process of step S2, the central controller combines the soil temperature field data to dynamically adjust the target temperature for soil reheating and the selection of the reheating circuit to achieve balanced restoration of soil temperature.

[0044] The specific implementation of the sensor network involves installing multiple soil temperature sensors at different depths (e.g., 20 meters, 50 meters, 100 meters) and horizontal positions within representative boreholes of the underground pipeline site. These sensors continuously transmit temperature data to the central controller. During decision-making (especially when determining whether to activate Mode 1 or calculate soil thermal balance), the controller not only uses a single underground pipe outlet temperature T3, but also comprehensively processes readings from all soil temperature sensors to generate a two-dimensional or three-dimensional distribution map of the soil temperature field. Based on this temperature field data, the controller can identify areas with abnormally low temperatures (cold spots) and dynamically adjust the target temperature setpoint for soil reheating (e.g., setting a higher target temperature for low-temperature areas). When the system has multiple independent underground pipe branch loops, the controller prioritizes controlling the valve opening of loops leading to these low-temperature areas for targeted reheating, thereby achieving balanced soil temperature recovery and preventing localized overcooling.

[0045] Utilizing soil temperature field data for decision-making offers the advantage of moving from point-based temperature monitoring to volumetric temperature management. This upgrades soil heating strategies from a uniform, extensive approach to precise, targeted heating for areas with temperature anomalies. This precise thermal management can more effectively correct uneven soil temperature fields caused by geographical or operational differences, preventing reduced heat exchange efficiency due to localized overcooling. This, in turn, improves the average heat exchange efficiency of the entire buried pipe system, ensuring the long-term stability and high efficiency of the ground source heat pump system.

[0046] The central controller is equipped with a self-learning module, which is used to record historical operating data, mode switching effects and actual energy efficiency, and dynamically corrects the judgment threshold of the fast rule layer in step S2 and the parameters of the system comprehensive model in step S4 through machine learning algorithms, so that the control strategy can adapt to changes in local climate and user habits.

[0047] The self-learning module is implemented as a background software process within the controller. This process continuously records timestamped historical operating data, including: external environmental parameters (Ta, G), mode switching times and system energy efficiency (such as COP) before and after the switch, deviations between predicted and actual loads, and soil temperature changes. Periodically (e.g., weekly), this process invokes built-in machine learning algorithms (e.g., using regression analysis to establish the relationship between parameters and energy efficiency, or using reinforcement learning to explore better thresholds) to analyze the historical data. The analysis results are used to automatically adjust various judgment thresholds in the fast rule layer (e.g., fine-tuning the T2a activation threshold for mode two and the G threshold for mode one) and correct parameters in the system integrated model (e.g., updating the thermal inertia coefficient in the building load prediction model and correcting the efficiency parameters of the PVT model). This allows the control strategy to adapt to changes in the actual operating environment and continuously optimize.

[0048] By introducing a self-learning module, the beneficial effect is that it endows the control system with the ability to self-optimize and continuously evolve. It can overcome the performance degradation caused by factors such as inaccurate initial models, differences in climate at the installation site, and changes in user habits. By continuously learning from historical operating data and automatically correcting thresholds and model parameters, the control strategy becomes increasingly aligned with the actual operating environment, allowing the system to maintain optimal or near-optimal operation in the long term. This represents a leap from static optimization based on "initial settings" to dynamic optimization based on "continuous self-adaptation," improving the system's robustness and long-term performance.

[0049] It also includes a feedback correction step: after each rolling optimization cycle of the model predictive control, the actual measured key operating parameters of the system are compared with the corresponding parameters predicted in the previous cycle, and the prediction error is calculated; based on the prediction error, the key parameters in the system integrated model are dynamically corrected or the initial system state in the next prediction time domain is corrected to improve the accuracy of subsequent rolling optimization.

[0050] The feedback correction step is a component of the model predictive control algorithm and is executed at the end of each rolling optimization cycle. Specifically, the controller extracts key operating parameters actually measured during the last cycle from the database, such as the actual power consumption of the heat pump, the actual heat exchange on the user side, and the actual change in soil temperature. These actual values ​​are then compared with the corresponding values ​​predicted by the model at the beginning of the previous cycle to calculate a prediction error vector. This error vector is then fed into a state estimator (such as a Kalman filter). Based on the error information, the estimator dynamically corrects key parameters in the integrated system model (such as adjusting the building's equivalent heat capacity and correcting the heat transfer coefficient of the buried pipes), or corrects the initial system state (such as the current heat storage of the soil and the real-time indoor temperature of the building) upon which the next cycle's optimization calculation depends. The corrected model and initial state, closer to the real system, are used in the next rolling optimization cycle, thus forming a closed-loop feedback loop that gradually improves prediction accuracy and control precision.

[0051] By adding a feedback correction stage to the Model Predictive Control (MPC) algorithm, its core benefit is a significant improvement in the accuracy and reliability of MPC during long-term operation in real-world environments. By continuously correcting the model using the error between prediction and observation, the "model mismatch" problem (i.e., the difference between the theoretical model and the real physical system) can be effectively compensated, allowing the predictive model to continuously approximate the dynamics of the real system. This acts like a constantly calibrated compass, ensuring that the optimization decisions based on the model always point in the right direction, thereby avoiding the degradation of control quality caused by the accumulation of model errors and guaranteeing the continuity and stability of the optimized control effect.

[0052] The above-described embodiments are detailed and specific, illustrating preferred embodiments of the present invention. They are only used to illustrate the technical ideas and features of the present invention, with the aim of enabling those skilled in the art to understand the content of the present invention and implement it accordingly. However, they are not limited to the present invention, and the patent scope of the present invention cannot be limited by this embodiment alone. That is, any equivalent changes or modifications made to the spirit disclosed in the present invention, without departing from the structure of the present invention, such as local improvements within the system and modifications or transformations between subsystems, are still within the patent scope of the present invention.

Claims

1. An optimized control method for coupling a ground source heat pump with a PVT system, characterized in that, The method is applied to a coupled system, which includes at least a PVT component (1), a hot water storage tank (2), a ground source heat pump unit (3), a buried pipe heat exchanger (4), a user terminal (5), multiple circulating water pumps, multiple electric valves, and a central controller; the method includes the following steps: S1: The system operating parameters are collected in real time through a sensor network. The parameters include at least the PVT component outlet temperature T1, the hot water storage tank stratified temperature, the buried pipe heat exchanger outlet temperature T3, the user-side supply and return water temperatures T4 and T5, the ambient temperature Ta, the total solar irradiance G, the flow rate of each loop and the system power consumption. In the hot water storage tank stratified temperature, T2a is the top high temperature, T2b is the middle temperature, and T2c is the bottom low temperature. S2: The central controller executes a multi-layer decision-making logic based on the real-time operating parameters and predicted parameters to determine the current optimal operating mode and generate a control instruction set; the multi-layer decision-making logic includes at least: a fast rule-based decision-making layer based on real-time parameters and set thresholds, and an optimization-based decision-making layer based on the system model and predicted data; S3: The central controller sends the control instruction set to the corresponding actuator. The control instruction set includes at least: switching or regulating instructions for multiple electric valves used to switch fluid paths, frequency conversion instructions for circulating water pumps used to regulate flow, and instructions for controlling the start-up, shutdown and capacity of the ground source heat pump host (3). S4: After executing the control command, return to step S1 to form closed-loop control, and dynamically adjust the operating mode according to the changes in the operating parameters.

2. The optimized control method for coupling a ground source heat pump and a PVT system according to claim 1, characterized in that, The operating modes in step S2 include at least five core modes, and the switching logic is as follows: Mode 1, PVT direct storage and active soil heating mode: When it is detected that the user has no immediate heating or cooling needs, and G is greater than the first irradiation threshold, and T1 is greater than T3, the PVT circulation pump and the heating valve connected to the buried pipe side are turned on to inject the heat energy collected by PVT directly or after buffering through the water tank into the soil. Mode 2, direct heating mode of hot water storage tank: When it is detected that the user has a heating demand and T2a is higher than the user terminal water supply set temperature, the corresponding valve is controlled to form an independent cycle of "hot water storage tank (2) to user terminal (5) and then to hot water storage tank (2)", and the ground source heat pump host (3) is shut down; Mode 3, PVT preheating assisted heat pump heating mode: When users have heating demand but T2a is lower than the direct supply threshold and T1 is higher than the current inlet water temperature on the ground source heat pump evaporator side, the control valve makes the PVT loop output and the buried pipe side loop connected in series or in parallel before the heat pump evaporator to improve the quality of the low temperature heat source of the heat pump. Mode 4, Ground source heat pump stand-alone operation mode: When the conditions of Mode 2 and 3 are not met, but the user has heating or cooling needs, the ground source heat pump host (3) is started to extract or exhaust heat from the soil. Mode 5, PVT priority cooling and power generation mode: In summer cooling conditions, when T1 is higher than the set heat dissipation threshold, the control valve guides the waste heat generated by the PVT to the auxiliary heat dissipation device to prevent it from entering the soil or hot water storage tank.

3. The optimized control method for coupling a ground source heat pump and a PVT system according to claim 2, characterized in that, The switching between the modes has priority logic. When the central controller determines that a user has a heating demand, it tries to enter the corresponding mode in the following priority order: it first tries to meet the conditions of mode two. If the conditions are not met, try to meet the conditions of mode three; if they are still not met, proceed to mode four.

4. The optimized control method for coupling a ground source heat pump and a PVT system according to claim 1, characterized in that, The optimization layer decision in step S2 is implemented using a model predictive control algorithm, specifically including: S2.1: Establish a comprehensive system model that includes the component model of the coupled system, the building load prediction model, and the electricity price model; S2.2: With the forecast time domain being at least 24 hours in the future, the optimization objective being the lowest total system operating cost or the lowest primary energy consumption, and with constraints such as equipment physical limitations and soil thermal balance constraints, a rolling optimization problem is constructed. S2.3: In each control cycle, solve the optimization problem to obtain the optimal control sequence of each actuator in the future time domain, and apply the instructions of the first control cycle to the system.

5. The optimized control method for coupling a ground source heat pump and a PVT system according to claim 4, characterized in that, The soil heat balance constraint is that within a set period, the absolute value of the difference between the total heat taken from the soil and the total heat discharged to the soil by the buried pipe heat exchanger (4) is not greater than a set threshold; in the optimization objective function, different cost coefficients are set for the behavior of taking heat from the soil or discharging heat to the soil, so as to guide the controller to actively supplement the soil with heat during periods of low electricity prices or abundant solar energy.

6. The optimized control method for coupling a ground source heat pump and a PVT system according to claim 1, characterized in that, The method also includes optimized control of the power generated by the PVT: The electrical energy generated by the PVT is preferentially used by the inverter to drive the compressor, circulating water pump and central controller of the ground source heat pump host (3); When PVT generates a surplus of electricity, the surplus energy is stored in batteries or fed into the grid; when PVT generates insufficient electricity, the grid provides supplementary power. The central controller adjusts the start / stop and power of the auxiliary electric heater in the hot water storage tank (2) according to the time-of-use electricity price signal, so as to store heat during the low electricity price period.

7. The optimized control method for coupling a ground source heat pump and a PVT system according to claim 1, characterized in that, The hot water storage tank (2) is a layered hot water storage tank. In step S3, the opening degree of the valves connected to the tank at different heights is controlled to achieve the graded storage and retrieval of heat energy: in the direct supply mode, high temperature water is taken from the top of the tank; when it is necessary to provide preheating for the heat pump, medium temperature water is taken from the middle of the tank; when receiving PVT heat, the appropriate height of the water tank is selected according to the inlet water temperature of the PVT circuit.

8. The optimized control method for coupling a ground source heat pump and a PVT system according to claim 1, characterized in that, The sensor network also includes soil temperature sensors arranged at different depths and locations in the buried pipeline to monitor the distribution of the soil temperature field. In the decision-making process of step S2, the central controller combines the soil temperature field data to dynamically adjust the target temperature for soil reheating and the selection of the reheating circuit to achieve balanced restoration of soil temperature.

9. The optimized control method for coupling a ground source heat pump and a PVT system according to claim 1, characterized in that, The central controller is pre-installed with a self-learning module, which is used to record historical operating data, mode switching effects and actual energy efficiency, and dynamically corrects the judgment threshold of the fast rule layer in step S2 and the parameters of the system comprehensive model in step S4 through machine learning algorithms, so that the control strategy can adapt to changes in local climate and user habits.

10. An optimized control method for coupling a ground source heat pump and a PVT system according to claim 4 or 5, characterized in that, It also includes a feedback correction step: after each rolling optimization cycle of the model predictive control, the actual measured key operating parameters of the system are compared with the corresponding parameters predicted in the previous cycle, and the prediction error is calculated; based on the prediction error, the key parameters in the system integrated model are dynamically corrected or the initial system state in the next prediction time domain is corrected to improve the accuracy of subsequent rolling optimization.