Ground source heat pump heat supplementing method and system based on multi-source complementation and dynamic optimization control

Abstract

This invention discloses a ground source heat pump heat compensation method and system based on multi-source complementarity and dynamic optimization control, belonging to the field of ground source heat pump heat compensation technology. The method includes the following steps: S1, constructing a mechanism-data dual-driven system prediction model, using hourly meteorological parameters, regional thermal inertia coefficient, and the difference in influence factors between the inner surface temperature of the building envelope and the water supply temperature of the underfloor heating coil as input parameters of the system prediction model; S2, constructing a multi-constraint dynamic optimization problem, clarifying hourly control decision variables, establishing an optimization objective centered on minimizing operating costs, and setting multi-dimensional constraints such as heat balance, anti-condensation, equipment safety, and operational energy efficiency; S3, based on a rolling time-domain optimization framework, solving the dynamic optimization problem online to generate the optimal control sequence within the future optimization time domain; S4, executing only the optimal control command for the current time period, updating the system's full-state data in the next control cycle, and repeating steps S1 to S3.

Description

Technical Field

[0001] This invention relates to the field of ground source heat pump heat compensation technology, and in particular to a ground source heat pump heat compensation method and system based on multi-source complementarity and dynamic optimization control. Background Technology

[0002] The statements in this section are merely background information related to the present invention and do not necessarily constitute prior art.

[0003] After years of development, ground source heat pump technology has played a significant role in building energy conservation and carbon emission reduction. However, in northern my country, on the one hand, a number of projects, due to insufficient planning and understanding in the early stages, have built single-condition ground source heat pump systems that only provide heating in winter; on the other hand, residential buildings generally have a large heating load in winter and a small cooling load in summer. Both scenarios result in the system extracting heat from underground soil and rock for a long time with insufficient heat replenishment, causing the soil and rock temperature to drop continuously year by year. This ultimately leads to a significant decrease in the energy efficiency of ground source heat pump units, a deterioration in heating effect, and even problems such as the system failing to operate normally. This has become a core industry pain point restricting the promotion and application of ground source heat pump technology.

[0004] To address the thermal imbalance in underground rock and soil, existing heating methods include electric boilers, gas boilers, and conventional air source heat pumps. However, these methods generally suffer from high energy consumption and operating costs, severely limiting their application in practical engineering. Furthermore, existing technologies still have significant shortcomings in low-cost heating pathways such as air source heat pumps and indoor waste heat recovery, hindering large-scale implementation.

[0005] A search revealed that existing technologies disclose a ground source heat pump composite system for buildings in cold regions and its control method. This system uses a partitioned heat exchanger to heat the soil and formulates a control strategy based on parameters such as temperature and temperature difference. However, this solution can only utilize the latent heat of outdoor air for heat replenishment, and is only suitable for high humidity conditions in July and August. The heat replenishment time is highly concentrated, and it cannot achieve effective heat replenishment under low humidity conditions. At the same time, short-term high-power heat replenishment can easily cause a sudden rise in the temperature of the circulating water in the buried pipe, resulting in a significant decrease in the efficiency of the partitioned heat exchanger and failing to achieve sufficient and stable heat replenishment.

[0006] In addition, existing technologies also disclose a heat exchange system to solve the thermal imbalance of ground source heat pump buried pipe heat exchange systems. This system obtains supplementary heat through the indoor system and avoids the risk of condensation by controlling the supply and return water temperatures. However, this solution relies on the radiant air conditioning system and the heat pump unit to operate in coordination throughout the year, which cannot be adapted to ground source heat pump projects that only provide heating. In projects that provide both heating and cooling, it also has problems such as high energy consumption and complex control parameters. If the building is not equipped with a fresh air dehumidification system, once the control fails, it is very easy to cause indoor condensation and damage to the building structure, resulting in serious insufficiency in the reliability of supplementary heat.

[0007] While the aforementioned solutions alleviate the thermal imbalance problem to some extent, they still suffer from the following unresolved core technical shortcomings: the management of building waste heat utilization is crude, the precision of heat storage management is severely insufficient, and the utilization rate of building waste heat is generally low. Furthermore, existing heating systems mostly employ a single heat source series architecture, failing to achieve dynamic and coordinated allocation of air source heat pumps and building waste heat, and lacking a globally optimized model that considers heating targets, operational safety, condensation prevention, and economic efficiency. Summary of the Invention

[0008] In view of this, the purpose of the present invention is to provide a ground source heat pump heat replenishment method and system based on multi-source complementarity and dynamic optimization control that can solve or at least alleviate the above-mentioned problems, so as to effectively solve the problem of underground soil and rock thermal imbalance in ground source heat pump systems, significantly improve the synergistic efficiency and control accuracy of multi-source heat replenishment, and the utilization rate of building waste heat.

[0009] To achieve the above objectives, the present invention adopts the following technical solution: In a first aspect, the present invention provides a ground source heat pump heat compensation method based on multi-source complementarity and dynamic optimization control, for use in a ground source heat pump heat compensation system including a buried pipe heat exchange system, a multi-source heat compensation module, a sensing and detection module, and an intelligent controller, comprising the following steps: S1. Construct a system prediction model driven by both mechanism and data. The system prediction model takes hourly meteorological parameters, regional thermal inertia coefficient, and the difference between the inner surface temperature of the building envelope and the water supply temperature of the underfloor heating coil as the core influencing factors. The regional thermal inertia coefficient is the time required for the excess temperature of the heat storage body relative to the environment to decay from the initial value to 1 / e of the initial value after the water supply to the corresponding underfloor heating coil area is stopped. The excess temperature is the difference between the temperature of the heat storage body and the ambient air temperature. S2. Construct a multi-constraint dynamic optimization problem, clarify the hourly control decision variables, take the minimization of operating costs as the core optimization objective, and set multi-dimensional constraints such as heat balance, anti-condensation, equipment safety and operating energy efficiency. S3. Based on the rolling time-domain optimization framework, solve dynamic optimization problems online and generate the optimal control sequence in the future optimization time domain; S4. Execute only the optimal control command for the current time period, update the system's full state data in the next control cycle, and repeat steps S1 to S3.

[0010] In some embodiments, step S1 includes: S11. Construct a heat gain model for a gas-water heat exchanger: Determine the core influencing parameters based on the ε-NTU (efficiency-number of heat transfer units) heat transfer theory, collect historical operating data under different temperature and humidity conditions, and train a machine learning proxy model after normalization preprocessing until the model's coefficient of determination R² ≥ 0.95. S12. Construct a heat extraction model for underfloor heating coils: Based on the radiation-convection coupled heat transfer theory, determine the core influencing parameters, obtain the thermal inertia coefficient of each region through historical temperature data fitting or building thermal simulation, clarify the total building heat storage constraint, collect historical operating data under different operating conditions, and train the machine learning proxy model until the model's coefficient of determination R² ≥ 0.95; the regional thermal inertia coefficient is the time required for the excess temperature of the heat storage body relative to the environment to decay from the initial value to 1 / e of the initial value after the water supply to the corresponding underfloor heating coil region is stopped; the excess temperature is the difference between the temperature of the heat storage body and the ambient air temperature; S13. Construct a soil and rock heating model: Based on the finite-length heat source theory, calculate the changes in soil and rock temperature field and real-time heating efficiency under intermittent heating conditions through the superposition principle, and complete the construction of the analytical model.

[0011] In some embodiments, in step S2: The decision variables include the start-up and shutdown status and fan frequency of the air-water heat exchanger unit, the start-up and shutdown status of valves in each heating coil area, and the start-up and shutdown status and operating frequency of each circulating pump. The optimization objective is to minimize the total operating cost of the system within the time domain while simultaneously maximizing the synergistic efficiency of the multi-source heat source. The mathematical expression is:

[0012] In the formula, N is the total number of time periods in the optimization time domain, and i is the time period number. This refers to the operating power of the air-water heat exchanger unit. The operating power of the underfloor heating coil circulation system. The operating power of the underground pipe circulation system. Let be the grid electricity price for the i-th time period. The duration of a single time period; The constraints include: Thermal balance constraint: The cumulative heat replenishment during the optimization cycle is greater than or equal to the preset target heat replenishment; Anti-condensation constraint: The water supply temperature of the underfloor heating coils ≥ the indoor air dew point temperature of the corresponding area + preset safety margin; Equipment safety constraints: The water temperature, circulating water flow rate, and equipment start-up and shutdown frequency at each node of the system are all within the preset rated range; Energy efficiency constraint: The system's real-time operating COP is greater than or equal to the preset energy efficiency threshold.

[0013] In some embodiments, in step S3, the control period is set to 1 hour, the optimization time domain is the next 24 hours, and the control time domain is the next 4 hours. At the beginning of each control period, based on the latest hourly weather forecast, system operating status data, and building thermal data, a mixed-integer nonlinear programming solver is used to solve the optimization problem to obtain the optimal control sequence for the next 24 hours. In step S4, only the control instructions for the current 1-hour period in the optimal control sequence are executed. In the next control period, the completed heat replenishment, the latest hourly weather data, and the system operating data are updated synchronously to perform rolling corrections on the control sequence and simultaneously trigger the temperature and humidity linkage start / stop and energy efficiency threshold protection logic.

[0014] Secondly, the present invention provides a ground source heat pump heat replenishment system based on multi-source complementarity and dynamic optimization control, for realizing the above-mentioned ground source heat pump heat replenishment method based on multi-source complementarity and dynamic optimization control, including a buried pipe heat exchange system, and also including a multi-source heat replenishment module, a sensing and detection module and an intelligent controller. The multi-source heat exchange module and the buried pipe heat exchange system are connected in series through pipelines to form a closed heat exchange loop. The multi-source heat replenishment module includes a gas-water heat exchanger unit and a zoned underfloor heating heat exchange coil unit arranged in parallel. The execution components of the gas-water heat exchanger unit, the zoned underfloor heating heat exchange coil unit, and the buried pipe heat exchange system are all electrically connected to the intelligent controller. The sensing and detection module is connected to the intelligent controller and is used to collect hourly system operating status parameters, indoor and outdoor environmental parameters, building envelope temperature parameters and power grid price information. The intelligent controller incorporates a mechanism-data dual-driven system prediction model and a rolling time-domain optimization control module to perform dynamic optimization heat replenishment control.

[0015] In some embodiments, the buried pipe heat exchange system is equipped with a first circulating water pump; the air-water heat exchanger unit includes at least one air-water heat exchanger, each air-water heat exchanger is equipped with a variable frequency fan, a spray water pump and an electric regulating valve that can operate independently or in coordination; the zoned underfloor heating heat exchange coil group consists of one plate heat exchanger connected in series and at least two independent underfloor heating coil areas, each underfloor heating coil area has an electric regulating valve installed in its inlet pipe, the electric regulating valve is electrically connected to an intelligent controller and is used to independently control the water flow and flow rate of the corresponding area; the total inlet of the zoned underfloor heating heat exchange coil group is connected to the return pipe of the closed-loop heat replenishment circulation loop, and the total outlet is connected to the inlet of the first circulating water pump.

[0016] In some embodiments, the main outlet pipe of the zoned underfloor heating heat exchange coil group is equipped with a second circulating water pump.

[0017] In some embodiments, the sensing and detection module includes: An outdoor weather station is used to collect outdoor air dry-bulb temperature, humidity, and dew point temperature hourly. Indoor temperature and humidity sensors are installed at least once for each underfloor heating coil area to collect indoor air temperature and relative humidity for the corresponding area; Temperature sensors are installed at the inlet and outlet of the air-water heat exchanger unit, the inlet and outlet of the local heating coil area, the inlet and outlet of the buried pipe heat exchange system, and the inner surface of the building envelope. Flow sensors are installed in the outlet pipes of each variable frequency circulating pump; The power grid communication unit is used to obtain time-of-use electricity price information and electricity consumption information.

[0018] In some embodiments, the mechanism-data dual-driven system prediction model includes a gas-water heat exchanger heat gain model, a floor heating coil heat extraction model, and a soil and rock heat replenishment model. The heat gain model of the air-water heat exchanger is a machine learning surrogate model built based on the ε-NTU heat transfer mechanism. The input parameters include hourly air mass flow rate, circulating water mass flow rate, circulating water inlet water temperature, air dry bulb temperature and air humidity. The output is the real-time heat exchange of the air-water heat exchanger. The heat exchange model of the underfloor heating coil is a machine learning proxy model built on the radiation-convection coupled heat transfer mechanism. The input parameters include water supply temperature, circulating water flow rate, indoor reference temperature, outdoor temperature, regional thermal inertia coefficient, and the difference between the inner surface temperature of the building envelope and the water supply temperature of the underfloor heating coil. The output is the real-time heat exchange of the underfloor heating coil. The soil and rock heating model is an analytical model based on the finite-length heat source theory and the superposition principle. The input parameters include the inlet water temperature of the buried pipe, the circulating water flow rate, the initial temperature of the soil and rock, the thermal properties of the soil and rock, and the cumulative running time. The output is the real-time heat supply of the buried pipe and the change data of the soil and rock temperature field.

[0019] In some embodiments, the rolling time-domain optimization control module is used to update the system state, hourly meteorological data and building thermal data with a fixed control cycle, solve the constrained dynamic optimization problem in a rolling manner, generate the optimal control sequence, and execute the control instructions for the current time period only.

[0020] Compared with the prior art, the present invention has at least the following beneficial effects: 1. The ground source heat pump heat replenishment method based on multi-source complementarity and dynamic optimization control provided by this invention establishes a dynamic optimization problem with minimizing operating costs as the core and taking into account multi-dimensional constraints by using regional thermal inertia, building envelope temperature difference and hourly meteorological parameter influencing factors as input parameters. Adaptive control is achieved through rolling time-domain optimization, which significantly improves the synergistic efficiency and control accuracy of multi-source heat replenishment. It can effectively solve the problem of soil thermal imbalance, avoid the risk of condensation in building waste heat utilization, and significantly improve the building waste heat utilization rate compared with traditional solutions.

[0021] 2. In this invention, a closed-loop system is adopted, in which the gas-water heat exchanger is connected in parallel with the zoned underfloor heating coils and the buried pipes are connected in series. The two heat sources can operate independently or collaboratively, adapting to the dynamic heat replenishment needs under multiple working conditions and providing hardware support for overall optimization.

[0022] 3. In this invention, time-of-use electricity pricing, heat replenishment targets, safety and energy efficiency constraints are incorporated into global optimization. By combining dual-source settings, dynamic and coordinated allocation of air source heat pumps and building waste heat is achieved. While ensuring heat replenishment targets, optimal operating costs are achieved, realizing multi-source coordinated global optimization and significantly reducing operating costs.

[0023] 4. In this invention, the heat storage characteristics are quantified by the regional thermal inertia coefficient. Combined with the independent control architecture of the underfloor heating coil, a zoned heat extraction method under the constraint of total heat storage is established to avoid disorderly consumption of heat storage and realize efficient and controllable utilization of waste heat.

[0024] 5. This invention significantly improves the stability and energy efficiency of heat replenishment through fully closed-loop rolling optimization control, realizing a leap from passive open-loop to active predictive closed-loop control, accurately capturing the efficient heat replenishment window, adapting to hourly meteorological fluctuations, and greatly improving the system's operating efficiency.

[0025] 6. This invention uses the temperature difference between the building envelope and the water supply as the core optimization variable, sets hard constraints on dew point temperature and safety margin, and combines a zoned control architecture to simultaneously maximize waste heat extraction efficiency and achieve zero condensation throughout the process. Attached Figure Description

[0026] The accompanying drawings, which form part of this invention, are used to provide a further understanding of the invention. The illustrative embodiments of the invention and their descriptions are used to explain the invention and do not constitute an improper limitation of the invention.

[0027] Figure 1 A flowchart of the ground source heat pump heat replenishment method based on multi-source complementarity and dynamic optimization control of the present invention is shown.

[0028] Figure 2 A schematic diagram of the structure of the ground source heat pump heat replenishment system based on multi-source complementarity and dynamic optimization control of the present invention is shown.

[0029] Explanation of reference numerals in the attached diagram: 1-Buried pipe heat exchange system; 2-Air-water heat exchanger unit; 3-Districted floor heating heat exchange coil group; 4-First circulating water pump; 5-Plate heat exchanger; 6-Electric regulating valve; 7-Second circulating water pump. Detailed Implementation

[0030] The present invention will be further described below with reference to the accompanying drawings and specific embodiments.

[0031] It should be noted that the following detailed description is illustrative and intended to provide further explanation of the invention. Unless otherwise specified, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this invention pertains.

[0032] Example 1 refer to Figure 1 This invention provides a ground source heat pump heat compensation method based on multi-source complementarity and dynamic optimization control (hereinafter referred to as the method). Preferably, the method adopts a prediction-optimization-execution-feedback fully closed-loop control framework. The method includes the following steps: S1. Construct a system prediction model driven by both mechanism and data. Using hourly meteorological parameters, regional thermal inertia coefficient, and the difference between the inner surface temperature of the building envelope and the water supply temperature of the underfloor heating coil as input parameters, accurate prediction of heat exchange under all operating conditions of multi-source heat replenishment is achieved.

[0033] Preferably, step S1 includes: S11. Construct a heat gain model for the gas-water heat exchanger. Specifically, based on the ε-NTU (efficiency-number of heat transfer units) heat transfer theory, determine the core influencing parameters. Supplement the enthalpy difference correction logic for the wet operating condition (air-side condensation) of the gas-water heat exchanger. Collect historical operating data under different temperature and humidity conditions. After min-max normalization preprocessing, construct a backpropagation (BP) neural network machine learning surrogate model until the model's coefficient of determination R² ≥ 0.95.

[0034] S12. Construct a heat extraction model for underfloor heating coils. Specifically, based on the radiation-convection coupled heat transfer theory, determine the core influencing parameters. Obtain the thermal inertia coefficient of each region through historical temperature data fitting or building thermal simulation, clarify the constraint on the total building heat storage, collect historical operating data under different operating conditions, and train a support vector regression machine learning surrogate model until the model's coefficient of determination R² ≥ 0.95. The regional thermal inertia coefficient is the time required for the excess temperature of the heat storage body relative to the environment to decay from its initial value to 1 / e of its initial value after the water supply to the corresponding underfloor heating coil region is stopped; the excess temperature is the difference between the temperature of the heat storage body and the ambient air temperature, which is a standard characteristic parameter of a first-order unsteady-state heat conduction system.

[0035] S13. Construct a soil and rock heat replenishment model. Specifically, based on the finite-length heat source theory, clarify the heat transfer boundary conditions for heat release to soil and rock under heat replenishment conditions, calculate the changes in soil and rock temperature field and real-time heat replenishment efficiency under intermittent operation conditions through the superposition principle, and complete the construction of the analytical model.

[0036] S2. Construct a multi-constraint dynamic optimization problem. Define the hourly control decision variables, establish the optimization objective with minimizing operating costs as the core, and set multi-dimensional constraints such as thermal balance, anti-condensation, equipment safety, and operating energy efficiency.

[0037] Preferably, in step S2, the hourly control decision variables include the start-up and shutdown status of the gas-water heat exchanger unit and the fan frequency, the start-up and shutdown status of the valves in each heating coil area, and the start-up and shutdown status and operating frequency of each circulating pump.

[0038] The optimization objective is to minimize the total operating cost of the system in the time domain while simultaneously maximizing the synergistic efficiency of multiple heat sources. The mathematical expression is:

[0039] In the formula, N is the total number of time periods in the optimization time domain, and i is the time period number. This refers to the operating power of the air-water heat exchanger unit. The operating power of the underfloor heating coil circulation system. The operating power of the underground pipe circulation system. Let be the grid electricity price for the i-th time period. This refers to the duration of a single time period.

[0040] Multidimensional constraints include the following constraints: (1) Thermal balance constraint: The cumulative heat replenishment during the optimization cycle is greater than or equal to the preset target heat replenishment; (2) Anti-condensation constraint: The water supply temperature of the underfloor heating coil is ≥ the indoor air dew point temperature of the corresponding area + the preset safety margin; (3) Equipment safety constraints: The water temperature, circulating water flow rate, and equipment start-up and shutdown frequency at each node of the system are all within the preset rated range; (4) Energy efficiency constraint: The real-time operating COP of the system is greater than or equal to the preset energy efficiency threshold.

[0041] Preferably, COP (Coefficient of Performance) refers to the ratio of the total heat supply power output by the ground source heat pump heat supply system to the buried pipe heat exchange system under real-time operation to the total electrical power input of all power-consuming equipment in the system. All power-consuming equipment includes at least the power consumption of multi-source heat supply modules (e.g., gas-water heat exchanger units), each circulating water pump, and the intelligent controller. More preferably, the COP value is dynamically calculated by the intelligent controller based on real-time flow rate, inlet and outlet water temperature difference, and electrical power data collected by the sensor detection module, and is used to characterize the overall energy efficiency level of the system under current operating conditions.

[0042] S3. Perform rolling optimization. Based on the rolling time-domain optimization framework, solve the dynamic optimization problem online to generate the optimal control sequence within the future optimization time domain. Specifically, set the control period to 1 hour, the optimization time domain to the next 24 hours, and the control time domain to the next 4 hours; at the beginning of each control period, based on the latest hourly weather forecasts, system operating status data, and building thermal data, use a mixed-integer nonlinear programming solver (such as the Gurobi solver) to solve the optimization problem and obtain the optimal control sequence for the next 24 hours.

[0043] S4. Execution and Feedback Correction. Only the optimal control command for the current time period is executed. In the next control cycle, the system's full-state data is updated, and steps S1 to S3 are repeated to achieve fully closed-loop adaptive dynamic optimization control. Specifically, only the control command for the current 1-hour time period in the optimal control sequence is executed. In the next control cycle, the completed heat replenishment, the latest hourly meteorological data, and the system pressure difference and flow operation data are updated synchronously to perform rolling corrections on the control sequence. Temperature and humidity linkage start / stop and energy efficiency threshold protection logic are triggered simultaneously to achieve stable and coordinated operation of multi-source heat replenishment branches.

[0044] Example 2 like Figure 2 As shown, the present invention also provides a ground source heat pump heat replenishment system (hereinafter referred to as the system) based on multi-source complementarity and dynamic optimization control, which can execute the methods in the above embodiments. Preferably, the system adopts a closed architecture in which the multi-source heat replenishment module and the buried pipe heat exchange system are directly connected in series. The system is a closed heat replenishment circulation loop, including a multi-source heat replenishment module and a buried pipe heat exchange system 1 connected in series through pipelines. The system is equipped with a sensing and detection module and an intelligent controller. The heat replenishment circulation water is first heated by the multi-source heat replenishment module, and then sent to the buried pipe heat exchange system 1 through the first circulation water pump 4 to replenish the underground soil and rock. After the heat replenishment is completed, the circulation water flows back to the multi-source heat replenishment module to be heated again, forming a closed loop.

[0045] Preferably, the multi-source heat replenishment module employs a parallel arrangement of air-water heat exchanger unit 2 and zoned underfloor heating heat exchange coil group 3, achieving dual-source complementarity of air energy and building waste heat. The two heat replenishment sources can operate independently or collaboratively, adapting to heat replenishment needs under different operating conditions and providing a hardware foundation for dynamic optimization control. The air-water heat exchanger unit 2 includes at least one air-water heat exchanger, each equipped with an independent variable frequency fan, spray water pump, and electric regulating valve. It can operate independently or collaboratively, absorbing sensible and latent heat from the outdoor air through circulating water to provide an air energy heat replenishment source for the system. The zoned underfloor heating heat exchange coil group 3 consists of a plate heat exchanger 5 connected in series and at least two independent underfloor heating coil zones. The plate heat exchanger 5 achieves hydraulic decoupling between the heat replenishment system and the underfloor heating terminal system, preventing the heat replenishment circulation from affecting the indoor heating system. Each underfloor heating coil area has an independent electric regulating valve 6 for its inlet water pipe. The electric regulating valve 6 is electrically connected to the intelligent controller and is used to independently control the water flow and on / off status of the corresponding area. The main return water pipe of the sub-area underfloor heating heat exchange coil group 3 is equipped with a second circulating water pump 7.

[0046] The underground pipe heat exchange system 1 is a U-shaped vertical underground pipe heat exchange system, equipped with a first circulating water pump 4, which is used to provide hydraulic drive for the closed-loop heat replenishment circulation, and send the circulating water heated by the multi-source heat replenishment module into the underground soil and rock, and after completing the heat replenishment, it flows back to the multi-source heat replenishment module.

[0047] The sensing module communicates with the intelligent controller to collect hourly system operating status parameters, indoor and outdoor environmental parameters, building envelope temperature parameters, and grid electricity price information, providing comprehensive data support for the intelligent controller. The sensing module includes an outdoor weather station, indoor temperature and humidity sensors, a temperature sensor, a flow sensor, and a grid communication unit. The outdoor weather station collects hourly outdoor air dry-bulb temperature, humidity, and dew point temperature. The indoor temperature and humidity sensors collect indoor air temperature and relative humidity for the corresponding area. Preferably, at least one set of indoor temperature and humidity sensors is installed for each underfloor heating coil area. Temperature sensors are installed at the inlet and outlet of the air-to-water heat exchanger unit, the inlet and outlet of each underfloor heating coil area, the inlet and outlet of the buried pipe heat exchange system, and on the inner surface of the building envelope. Flow sensors are installed on the outlet pipes of each variable frequency circulating pump. The grid communication unit is used to obtain time-of-use electricity price information and electricity consumption information.

[0048] Preferably, the intelligent controller adopts an industrial-grade PLC controller, is equipped with a high-performance computing module, and has a built-in system prediction model driven by both mechanism and data and a rolling time-domain optimization control module. All actuators such as electric regulating valves, variable frequency fans, and circulating water pumps are electrically connected to the intelligent controller to execute complete dynamic optimization heat replenishment control logic.

[0049] Example 3 This embodiment 3 implements a ground source heat pump supplementary heating method based on the system provided in embodiment 2. The control cycle is 1 hour, the optimization time domain is the next 24 hours, the control time domain is the next 4 hours, the anti-condensation safety margin is set to 2°C, and the system energy efficiency threshold is set to COP ≥ 3.5. The control process is as follows: Figure 1 As shown, it includes the following steps: S1. Construct a system prediction model driven by both mechanism and data, including: S11. Constructing a heat gain model for the air-water heat exchanger: Based on the ε-NTU heat transfer theory, the core input parameters are determined to be hourly air mass flow rate, circulating water mass flow rate, circulating water inlet temperature, air dry-bulb temperature, and air humidity, with the output being real-time heat exchange. 1200 sets of historical operating data under different temperature and humidity conditions were collected for the unit. After min-max normalization preprocessing, a BP neural network model with two hidden layers (10 neurons per layer) was constructed. After training, the model's coefficient of determination R² = 0.97, meeting the accuracy requirements. Model verification shows that the higher the atmospheric temperature and relative humidity, the higher the COP of the air-water heat exchanger, providing a quantitative basis for subsequent optimization.

[0050] S12. Constructing a heat exchange model for underfloor heating coils: Based on the radiation-convection coupled heat transfer theory, the core input parameters are determined as supply water temperature, circulating water flow rate, indoor reference temperature, outdoor temperature, regional thermal inertia coefficient, and the difference between the inner surface temperature of the building envelope and the supply water temperature of the underfloor heating coils. The output is the real-time heat exchange. The thermal inertia coefficients of each region are obtained by fitting historical temperature data, where τ=8h for two south-facing regions and τ=4h for one north-facing region, clarifying the overall building heat storage constraint and providing boundary conditions for zonal heat exchange optimization. 800 sets of historical operating data under different working conditions are collected to train a support vector regression model. After training, the model's coefficient of determination R²=0.96, meeting the accuracy requirements.

[0051] S13. Construct a soil and rock heating model: Based on the finite-length heat source theory, the change of soil and rock temperature field under intermittent operation is calculated by superposition principle. The input parameters are the inlet water temperature of the buried pipe, the circulating water flow rate, the initial temperature of soil and rock, the thermal properties of soil and rock, and the cumulative operating time. The output is the real-time heating amount and soil and rock temperature data. The analytical model is completed to achieve accurate prediction of soil and rock temperature changes during the heating process.

[0052] S2. Construct a multi-constraint dynamic optimization problem. Define the hourly control decision variables, establish an optimization objective centered on minimizing operating costs, and set multi-dimensional constraints including thermal balance, anti-condensation, equipment safety, and operational energy efficiency. Preferably, in step S2, the decision variables, optimization objective, and constraints are set as follows.

[0053] Decision variables: Hourly control parameter u(i) = [start / stop status of gas-water heat exchanger, fan frequency, start / stop status of valves in underfloor heating area, start / stop status of each circulating pump, and frequency of circulating pump], where i is the time period number from 1 to 24 hours.

[0054] Optimization objective: Minimize the total operating cost within 24 hours, expressed as:

[0055] The constraints include the following: Thermal balance constraint: Cumulative heat replenishment during the optimization cycle ≥ 800,000 kWh; Anti-condensation constraint: The water supply temperature of the underfloor heating coils must be ≥ the indoor air dew point temperature of the corresponding area + 2℃; Equipment safety constraints: water temperature 5-40℃, circulation pump frequency 30-50Hz, number of start-ups and shutdowns per unit per hour ≤2; Energy efficiency constraint: The system's real-time operating COP must be ≥3.5; otherwise, frequency reduction or shutdown protection will be triggered.

[0056] S3. Perform rolling optimization solution.

[0057] At the start of the control cycle (00:00 on the same day), the intelligent controller obtains hourly meteorological parameters for the next 24 hours, time-of-use electricity price information for the day, as well as the current system status, building envelope temperature, and differential pressure flow data through the sensing and detection module. It then starts the Gurobi solver to solve the optimization problem and obtain the optimal control sequence for the next 24 hours.

[0058] In this embodiment, the time-of-use electricity price is as follows: off-peak period 2:00-6:00, price is 0.3 yuan / kWh; flat period 8:00-22:00, price is 0.55 yuan / kWh; peak period 18:00-20:00, price is 0.85 yuan / kWh; and peak period 22:00-24:00, price is 1.0 yuan / kWh.

[0059] S4. Execution and feedback correction, including: S41. Only execute the optimal control command for the current 1-hour period. Based on multi-source collaborative optimization logic, and combined with time-of-use electricity pricing, hourly weather, and building envelope temperature dynamic adjustment strategy, the system hydraulic stability is ensured through frequency coordination adjustment of variable frequency water pumps: During the off-peak period (2:00-6:00): Only the first circulating water pump is started at a frequency of 30Hz, utilizing the low-temperature characteristics of the underground pipe return water at night to increase the heat exchanger temperature difference; During the plateau period (10:00-12:00): When the outdoor temperature reaches the daytime high, the air-water heat exchanger has the best heat exchanger efficiency, and the air-water heat exchanger is started at full power. During the peak period (14:00-16:00): outdoor temperature and building envelope temperature reach their daily peak, the temperature difference between the water supply and the building envelope is the largest, and the building waste heat extraction efficiency is the highest. The underfloor heating coils in the south-facing area with strong heat storage capacity are turned on, and the air-water heat exchanger units are turned on at full power to achieve dual-source coordinated heat supplementation. The hydraulic balance of the two parallel branches is ensured by differential pressure linkage variable frequency regulation. During the peak period (22:00-24:00): electricity prices are high, and all heat source equipment is turned off.

[0060] S42. At 13:00 on the same day, the intelligent controller re-collects system status data, confirms that the heat replenishment has been completed, and synchronously updates the latest hourly weather forecast, building envelope temperature, and differential pressure flow data. It then re-solves the optimization problem and performs rolling corrections on the control sequence from 13:00 to 24:00 to ensure that the heat replenishment target is achieved on schedule.

[0061] Based on the concept of multi-source complementarity, this invention combines global optimization of the heat compensation control method with coordinated adaptation of the system architecture, which can effectively solve the problem of underground soil and rock thermal imbalance in ground source heat pump systems.

[0062] The above description is merely a preferred embodiment of the present invention and is not intended to limit the invention. Various modifications and variations can be made to the present invention by those skilled in the art. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the scope of protection of the present invention.

Claims

1. A ground source heat pump heat supplementing method based on multi-source complementation and dynamic optimization control, for a ground source heat pump heat supplementing system comprising a ground buried pipe heat exchange system, a multi-source heat supplementing module, a sensing detection module and an intelligent controller, characterized in that, Includes the following steps: S1. Construct a system prediction model driven by both mechanism and data. The system prediction model takes hourly meteorological parameters, regional thermal inertia coefficient, and the difference between the inner surface temperature of the building envelope and the water supply temperature of the underfloor heating coil as input parameters. The regional thermal inertia coefficient is the time required for the excess temperature of the heat storage body relative to the environment to decay from the initial value to 1 / e of the initial value after the water supply to the corresponding underfloor heating coil area is stopped. The excess temperature is the difference between the temperature of the heat storage body and the ambient air temperature. S2. Construct a multi-constraint dynamic optimization problem, clarify the hourly control decision variables, take the minimization of operating costs as the optimization objective, and set multi-dimensional constraints such as heat balance, anti-condensation, equipment safety and operating energy efficiency. S3. Based on the rolling time-domain optimization framework, solve dynamic optimization problems online and generate the optimal control sequence in the future optimization time domain; S4. Execute only the optimal control command for the current time period, update the system's full state data in the next control cycle, and repeat steps S1 to S3.

2. The ground-source heat pump heat supplementing method based on multi-source complementation and dynamic optimization control according to claim 1, characterized in that, Step S1 includes: S11. Construct a heat gain model for a gas-water heat exchanger: Determine the core influencing parameters based on the ε-NTU (efficiency-number of heat transfer units) heat transfer theory, collect historical operating data under different temperature and humidity conditions, and train a machine learning proxy model after normalization preprocessing until the model's coefficient of determination R² ≥ 0.

95. S12. Construct a heat extraction model for underfloor heating coils: Determine the core influencing parameters based on the radiation-convection coupled heat transfer theory, obtain the thermal inertia coefficient of each region through historical temperature data fitting or building thermal simulation, clarify the total building heat storage constraint, collect historical operating data under different working conditions, and train the machine learning proxy model until the model determination coefficient R²≥0.

95. S13. Construct a soil and rock heating model: Based on the finite-length heat source theory, calculate the changes in soil and rock temperature field and real-time heating efficiency under intermittent heating conditions through the superposition principle, and complete the construction of the analytical model.

3. The ground-source heat pump heat supplementing method based on multi-source complementation and dynamic optimization control according to claim 1, characterized in that, In step S2: The decision variables include the start-up and shutdown status and fan frequency of the air-water heat exchanger unit, the start-up and shutdown status of valves in each heating coil area, and the start-up and shutdown status and operating frequency of each circulating pump. The optimization objective is to minimize the total operating cost of the system within the time domain while simultaneously maximizing the synergistic efficiency of the multi-source heat source. The mathematical expression is: wherein N is the total number of time periods in the optimization time horizon, i is the time period number, is the operating power of the air-water heat exchanger unit, is the operating power of the floor heating coil circulating system, is the operating power of the ground pipe circulating system, is the electricity price of the i-th time period, is the length of a single time period; The constraints include: Thermal balance constraint: The cumulative heat replenishment during the optimization cycle is greater than or equal to the preset target heat replenishment; Anti-condensation constraint: The water supply temperature of the underfloor heating coils ≥ the indoor air dew point temperature of the corresponding area + preset safety margin; Equipment safety constraints: The water temperature, circulating water flow rate, and equipment start-up and shutdown frequency at each node of the system are all within the preset rated range; Energy efficiency constraint: The system's real-time operating COP is greater than or equal to the preset energy efficiency threshold.

4. The ground-source heat pump heat supplementing method based on multi-source complementation and dynamic optimization control according to claim 1, characterized in that, In step S3, the control period is set to 1 hour, the optimization time domain is the next 24 hours, and the control time domain is the next 4 hours. At the beginning of each control period, based on the latest hourly weather forecast, system operation status data, and building thermal data, a mixed integer nonlinear programming solver is used to solve the optimization problem to obtain the optimal control sequence for the next 24 hours. In step S4, only the control instructions for the current 1-hour period in the optimal control sequence are executed. In the next control period, the completed heat replenishment, the latest hourly weather data, and the system operation data are updated synchronously to perform rolling corrections on the control sequence and simultaneously trigger the temperature and humidity linkage start / stop and energy efficiency threshold protection logic.

5. A ground-source heat pump heat supplementing system based on multi-source complementation and dynamic optimization control, used for implementing the ground-source heat pump heat supplementing method based on multi-source complementation and dynamic optimization control in any one of claims 1 to 4, comprising a ground heat exchanger system, characterized in that, It also includes a multi-source heating module, a sensor detection module, and an intelligent controller; The multi-source heat exchange module and the buried pipe heat exchange system are connected in series through pipelines to form a closed heat exchange loop. The multi-source heat replenishment module includes a gas-water heat exchanger unit and a zoned underfloor heating heat exchange coil unit arranged in parallel. The execution components of the gas-water heat exchanger unit, the zoned underfloor heating heat exchange coil unit, and the buried pipe heat exchange system are all electrically connected to the intelligent controller. The sensing and detection module is connected to the intelligent controller and is used to collect hourly system operating status parameters, indoor and outdoor environmental parameters, building envelope temperature parameters and power grid price information. The intelligent controller incorporates a mechanism-data dual-driven system prediction model and a rolling time-domain optimization control module to perform dynamic optimization heat replenishment control.

6. The ground source heat pump supplementary heating system based on multi-source complementarity and dynamic optimization control according to claim 5, characterized in that, The buried pipe heat exchange system is equipped with a first circulating water pump; the air-water heat exchanger unit includes at least one air-water heat exchanger, each of which is equipped with a variable frequency fan, a spray water pump, and an electric regulating valve that can operate independently or in coordination; the zoned underfloor heating heat exchange coil group consists of one plate heat exchanger connected in series and at least two independent underfloor heating coil areas, with an electric regulating valve installed on the inlet pipe of each underfloor heating coil area. The electric regulating valve is electrically connected to an intelligent controller and is used to independently control the water flow and on / off of the corresponding area; the total inlet of the zoned underfloor heating heat exchange coil group is connected to the return water pipe of the closed-loop heat replenishment circulation loop, and the total outlet is connected to the inlet of the first circulating water pump.

7. The ground source heat pump supplementary heating system based on multi-source complementarity and dynamic optimization control according to claim 5, characterized in that, The main outlet pipe of the regional underfloor heating heat exchange coil group is equipped with a second circulating water pump.

8. The ground source heat pump supplementary heating system based on multi-source complementarity and dynamic optimization control according to claim 5, characterized in that, The sensing and detection module includes: An outdoor weather station is used to collect outdoor air dry-bulb temperature, humidity, and dew point temperature hourly. Indoor temperature and humidity sensors are installed at least once for each underfloor heating coil area to collect indoor air temperature and relative humidity for the corresponding area; Temperature sensors are installed at the inlet and outlet of the air-water heat exchanger unit, the inlet and outlet of the local heating coil area, the inlet and outlet of the buried pipe heat exchange system, and the inner surface of the building envelope. Flow sensors are installed in the outlet pipes of each variable frequency circulating pump; The power grid communication unit is used to obtain time-of-use electricity price information and electricity consumption information.

9. The ground source heat pump supplementary heating system based on multi-source complementarity and dynamic optimization control according to claim 5, characterized in that, The mechanism-data dual-driven system prediction model includes a gas-water heat exchanger heat gain model, a floor heating coil heat extraction model, and a soil and rock heat replenishment model. The heat gain model of the air-water heat exchanger is a machine learning surrogate model built based on the ε-NTU heat transfer mechanism. The input parameters include hourly air mass flow rate, circulating water mass flow rate, circulating water inlet water temperature, air dry bulb temperature and air humidity. The output is the real-time heat exchange of the air-water heat exchanger. The heat exchange model of the underfloor heating coil is a machine learning proxy model built on the radiation-convection coupled heat transfer mechanism. The input parameters include water supply temperature, circulating water flow rate, indoor reference temperature, outdoor temperature, regional thermal inertia coefficient, and the difference between the inner surface temperature of the building envelope and the water supply temperature of the underfloor heating coil. The output is the real-time heat exchange of the underfloor heating coil. The soil and rock heating model is an analytical model based on the finite-length heat source theory and the superposition principle. The input parameters include the inlet water temperature of the buried pipe, the circulating water flow rate, the initial temperature of the soil and rock, the thermal properties of the soil and rock, and the cumulative running time. The output is the real-time heat supply of the buried pipe and the change data of the soil and rock temperature field.

10. The ground source heat pump supplementary heating system based on multi-source complementarity and dynamic optimization control according to claim 5, characterized in that, The rolling time-domain optimization control module is used to update the system state, hourly meteorological data and building thermal data with a fixed control cycle, solve the constrained dynamic optimization problem in a rolling manner, generate the optimal control sequence, and execute the control instructions only for the current time period.

Images