Three-source coupled compound energy heat pump system and control method thereof

Abstract

This invention discloses a three-source coupled composite energy heat pump system and its control method, relating to the field of composite energy utilization and intelligent control. It addresses the problems of existing systems having a single energy coupling dimension, simple control strategies, and an inability to achieve multi-objective dynamic optimization. The system includes a solar heat exchange module, an air source heat exchange module, a geothermal heat exchange module, and a dynamic scheduling control module. The control module collects operating parameters of each module and user load, and calculates the dynamic contribution weights of the three energy sources based on an attention-based deep neural network model. This generates four coupled operating modes: solar-dominated, geothermal-stable, air-assisted, and optimal economical three-source hybrid mode, controlling the coordinated operation of each module. The model is optimized through offline pre-training and online fine-tuning, and the system also achieves dynamic correction of the control strategy through closed-loop feedback. This invention achieves intelligent dynamic coordinated scheduling of the three energy sources, improving the overall energy efficiency and operational economy of the system, and enhancing environmental adaptability and operational stability.

Description

Technical Field

[0001] This invention relates to the field of composite energy utilization technology, and more particularly to a heat pump system. More specifically, this invention relates to a three-source coupled composite energy heat pump system and its control method, which utilizes artificial intelligence algorithms to dynamically optimize the scheduling of solar energy, air energy, and geothermal energy. Background Technology

[0002] With the global energy structure transformation and increasing demands for environmental protection, the development and utilization of renewable energy has become an important development direction in the field of building heating, ventilation, and air conditioning (HVAC). Air source heat pumps and solar thermal collectors, as two mature renewable energy utilization methods, have attracted widespread attention for their combined energy systems.

[0003] Existing technologies have the following shortcomings in practical applications: 1. Limited energy coupling dimension and adaptability: The above solutions are all limited to dual-source coupling of solar and air energy. Solar energy is intermittent and unstable, while the efficiency of air source heat pumps will significantly decrease in extreme low-temperature environments in winter. The lack of a stable and reliable basic energy source (such as geothermal energy) makes it difficult to guarantee the stability and efficiency of the system under continuous rainy days or severe cold climates. 2. Simple control strategy and lack of optimization capability: Existing technologies mostly adopt logic judgment based on fixed thresholds or traditional PID control. This control method has a slow response and cannot adapt to dynamically changing external environments (such as weather and sunlight), internal loads (such as user behavior), and economic factors (such as time-of-use electricity pricing). Its control objective is usually a single temperature control, making it difficult to balance and optimize multiple objectives such as energy efficiency, cost, and comfort, resulting in the overall operating efficiency of the system falling far short of the theoretical optimal level.

[0004] Therefore, there is an urgent need to develop a composite energy heat pump system that can integrate more diverse energy sources and adopt more advanced and intelligent control strategies to achieve dynamic synergistic optimization of multiple energy sources and improve the overall performance of the system. Summary of the Invention

[0005] The purpose of this invention is to provide a three-source coupled composite energy heat pump system, which aims to at least partially solve the technical problems of existing composite energy systems, such as single energy coupling dimension, simple control strategy, and inability to achieve multi-objective dynamic optimization.

[0006] To achieve the above objectives, the present invention provides the following technical solution: In a first aspect, the present invention provides a three-source coupled composite energy heat pump system, comprising: a solar heat exchange module for collecting solar energy and converting the solar energy into heat energy or electrical energy; an air heat exchange module for exchanging heat with ambient air; a geothermal heat exchange module for exchanging heat with underground rock and soil or groundwater; and a dynamic scheduling and control module for acquiring in real time the operating status parameters of the solar heat exchange module, the air heat exchange module, and the geothermal heat exchange module, as well as the user load demand, and determining the dynamic contribution weights of the three energy sources (solar, air, and geothermal) based on a preset intelligent scheduling model to generate a three-source coupled operation mode and control the three modules to work collaboratively.

[0007] As a preferred embodiment, the dynamic scheduling control module is configured to calculate the first of the solar, air, and geothermal energy sources using a deep neural network model based on an attention mechanism. Energy in time step Dynamic contribution weight The calculation formula is as follows: ,

[0008] in, For the first Energy in time step Dynamic contribution weights; This is an index for energy types, with values ​​ranging from 1 to 3 in the system, corresponding to solar energy, air energy, and geothermal energy, respectively. The total number of coupled energy sources in the system ; This represents the time step of the current control cycle. For the first Energy in time step The source contribution potential vector; For the system at time step The global state vector; and Let be the trainable weight matrix of the deep neural network model; and These are the trainable weight vector and bias vector of the deep neural network model, respectively. It is the hyperbolic tangent activation function; It is a natural exponential function.

[0009] As a preferred embodiment, the source contribution potential vector Including with the At least one real-time environmental parameter related to a certain energy source; the global state vector It includes at least one user load parameter and at least one operating cost parameter.

[0010] As a preferred embodiment, when the energy type is solar energy, the source contributes the potential energy vector. This includes: solar irradiance intensity and collector surface temperature; when the energy type is air energy, the source contribution potential vector. This includes: outdoor ambient temperature and outdoor ambient humidity; when the energy type is geothermal energy, the source contribution potential energy vector. This includes: the inlet and outlet temperature difference of the buried pipe heat exchanger, and the average temperature of the underground soil and rock mass; the global state vector. This includes: user-set temperature, current indoor temperature, time-of-use electricity price, and weather forecast data for the future scheduled time period.

[0011] As a preferred embodiment, the trainable weight matrix and trainable weight vector of the deep neural network model are obtained through offline pre-training using historical running data and online fine-tuning using real-time running data; the goal of the offline pre-training is to minimize a joint loss function that includes total energy consumption and operating cost. : ,

[0012] in, In order to time step The total energy consumption of the system. In order to time step The system operating cost, and The preset balance coefficient, This represents the total step size of the training cycle.

[0013] As a preferred embodiment, the three-source coupled operation mode includes: a solar-dominated heating mode, a geothermal energy stable heating / cooling mode, an air-source-assisted mode, and a three-source hybrid optimal economic mode. Among these, the solar-dominated heating mode: when the dynamic contribution weight... When the first preset condition is met, the solar heat exchange module is used as the primary heat source. Geothermal energy stable heating / cooling mode: when the dynamic contribution weight... When the second preset condition is met, the geothermal heat exchange module is used as the base load unit. Air source heat pump auxiliary mode: When neither solar nor geothermal energy can meet the user's load demand, the air source heat exchange module is activated as a supplementary or peak regulation unit. Three-source hybrid optimal economic mode: Based on the dynamic contribution weight... The real-time value is used to proportionally allocate the output power of each energy module, thereby minimizing the overall operating cost of the system.

[0014] Secondly, the present invention also provides a control method for a three-source coupled composite energy heat pump system disclosed in the first aspect, comprising the following steps:

[0015] Status data acquisition steps: Real-time acquisition of operating status parameters of solar heat exchange modules, air heat exchange modules and geothermal heat exchange modules, as well as user load demand and environmental parameters, to form system status information;

[0016] Dynamic weight calculation steps: Input the system state information into a pre-trained intelligent scheduling model to calculate a dynamic contribution weight for each of the three energy sources: solar energy, air energy, and geothermal energy.

[0017] Energy dispatch execution steps: Based on the dynamic contribution weight, a three-source coupled operation mode is determined, and control commands are generated for the solar heat exchange module, air heat exchange module, geothermal heat exchange module, and compressor module (such as air conditioner).

[0018] Closed-loop feedback adjustment steps: Execute the control command and monitor the system operation effect, using the effect deviation as a feedback signal for dynamic weight calculation in the next control cycle. This application provides an electronic device, including: a processor; and a memory storing program instructions, which, when executed by the processor, cause the electronic device to implement one or more embodiments of the first and second aspects described above.

[0019] This application provides a computer-readable storage medium having computer-readable instructions stored thereon, which, when executed by one or more processors, implement one or more embodiments of the first and second aspects described above.

[0020] Compared with the prior art, the present invention has the following beneficial effects: 1. Enhanced system adaptability and operational stability: This invention couples intermittent solar energy, air energy which is highly susceptible to ambient temperature fluctuations, and stable and reliable geothermal energy to form a complementary energy structure. Geothermal energy can serve as a stable base load unit, ensuring reliable system operation even at night, during continuous rainy days, or under extreme weather conditions. This significantly expands the system's effective operating range and improves its stability and reliability throughout the year.

[0021] 2. Achieved a high degree of intelligence and optimization in energy dispatch: This invention abandons the traditional simple control logic based on fixed thresholds and innovatively introduces a deep neural network model based on an attention mechanism. This model can comprehensively process multi-dimensional, nonlinear dynamic information, including energy availability, user load, environmental parameters, economic costs (time-of-use pricing), and even weather forecasts, and calculate the optimal contribution weight of each energy source in real time. This data-driven intelligent dispatching method enables the system to adaptively find a dynamic balance point among multiple objectives such as maximizing energy efficiency, minimizing costs, or both, thereby achieving globally optimal operation.

[0022] 3. Significantly improves overall system energy efficiency and economy: Through precise and coordinated control of the three energy sources using an intelligent scheduling model, this invention can prioritize and maximize the use of the most efficient and lowest-cost energy source (such as free solar energy during the day) at any given time, and rationally allocate other energy sources as supplements or backups. Compared to traditional dual-source systems or systems using simple control strategies, this invention can effectively reduce the ineffective operation of high-energy-consuming components such as compressor modules while meeting user needs, thereby significantly reducing total system energy consumption and operating costs, and achieving a higher coefficient of performance (COP) and economic benefits. Attached Figure Description

[0023] To more clearly illustrate the technical solutions in the embodiments of the present invention, the accompanying drawings used in the description of the embodiments will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0024] Figure 1 This is a flowchart of a control method for a three-source coupled composite energy heat pump system provided in an embodiment of the present invention.

[0025] Figure 2 This is a detailed flowchart of the dynamic weight calculation steps in an embodiment of the present invention.

[0026] Figure 3 This is a schematic diagram of an exemplary structural framework of a three-source coupled composite energy heat pump system provided in an embodiment of the present invention. Detailed Implementation

[0027] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0028] This invention provides a three-source coupled composite energy heat pump system, the core of which lies in constructing a physical system integrating solar energy, air energy, and geothermal energy, and realizing the intelligent and optimized synergistic utilization of these three energy sources through an artificial intelligence-based dynamic scheduling and control module. At the physical level, this three-source coupled composite energy heat pump system includes: a solar heat exchange module, such as a flat-plate or evacuated tube solar collector array; an air energy heat exchange module, such as a finned tube heat exchanger with a fan; and a geothermal energy heat exchange module, such as underfloor heating coils, fan coil units, radiators, or vertical or horizontal buried pipe underground heat exchangers.

[0029] The key innovation of this three-source coupled composite energy heat pump system lies in its dynamic scheduling and control module. This module is typically implemented by an embedded controller or industrial computer, which runs a pre-set intelligent scheduling model. The system control method dominated by this module is described in detail below with reference to the accompanying drawings.

[0030] Please see Figure 2 The diagram illustrates a control method for a three-source coupled composite energy heat pump system according to an embodiment of the present invention. The method mainly includes the following steps: Step S100: Status data acquisition.

[0031] At the beginning of each control cycle (e.g., every 5 minutes), the dynamic scheduling control module collects a series of status data in real time using sensors deployed throughout the system. This data forms the basis for the intelligent scheduling model's decision-making. Specifically, the collected data includes: Parameters related to solar energy include: solar irradiance, inlet and outlet water temperature of solar collectors, and surface temperature of collectors.

[0032] Parameters related to air source heat pumps include: outdoor ambient temperature, outdoor relative humidity, and air source heat exchanger fin temperature.

[0033] Parameters related to geothermal energy include: inlet and outlet water temperatures of buried pipe heat exchangers, and average temperature of underground rock and soil (which can be obtained through distributed temperature sensors).

[0034] Parameters related to user load: user-set target temperature, current indoor temperature of each room, return water temperature, etc.

[0035] Parameters related to operating costs: current time-of-use electricity price.

[0036] Predictive parameters: Weather forecast data for a future period of time (e.g., 24 hours) obtained through a network interface, including hourly temperature, weather conditions (sunny, cloudy, rainy), and solar radiation forecasts.

[0037] Step S200: Dynamic weight calculation.

[0038] This step is the core of the method of the present invention. The dynamic scheduling control module preprocesses and organizes the multidimensional data collected in step S100, inputs it into the preset intelligent scheduling model, and calculates the dynamic contribution weights of the three energy sources, namely solar energy, air energy and geothermal energy, at the current time step.

[0039] Please see Figure 3 It illustrates the detailed process of dynamic weight calculation. In this embodiment, the intelligent scheduling model is a deep neural network (DNN) model based on an attention mechanism.

[0040] First, the collected data is constructed into a specific input vector. For the first... Type of energy ( Construct their source contribution potential energy vectors (corresponding to solar energy, air energy, and geothermal energy respectively). .For example, This may include solar irradiance and collector surface temperature. This includes outdoor ambient temperature and outdoor ambient humidity. This includes the temperature difference between the inlet and outlet of the buried pipe and the average temperature of the underground soil and rock mass. Simultaneously, a global state vector is constructed. It contains global information independent of specific energy types, including user-set temperature, current indoor temperature, time-of-use electricity price, and encoded future weather forecast data. These vectors are then fed into an attention network. This network calculates the first... Energy in time step Dynamic contribution weight : ,

[0041] In this formula, The calculation process can be understood as calculating the inherent potential of each energy source ( ) and the current global needs and environment ( The features are then fused together to generate a comprehensive feature representation. It is a non-linear activation function. Its purpose is to calculate an "attention score," which measures the attention level of the first [the subject / group] in the current global state. The importance or suitability of each energy source for meeting the overall system objectives is determined. Finally, the attention scores of all energy sources are converted into a set of probability distributions summing to 1 using the Sofmax function (i.e., fractional exponentiation and normalization in the formula), representing the dynamic contribution weights. .

[0042] A higher weight for an energy source indicates that the model believes utilizing that energy source at that moment offers better overall benefits (energy efficiency, cost, etc.). It's worth noting the parameters in the model... It's not manually set, but rather obtained through machine learning training. Specifically, it can be pre-trained offline using a large amount of historical system data. The training objective is to minimize a joint loss function. ,For example: ,

[0043] in, Is the system at time step Total energy consumption (mainly the power consumption of compressor modules, water pumps, fans, etc.). It is the corresponding operating cost (energy consumption multiplied by the time-of-use electricity price at that time). and This is a balancing coefficient used to prioritize the two objectives of energy saving and cost saving. Through backpropagation, the model parameters are continuously adjusted, enabling the model to learn how to generate a weight allocation strategy that minimizes long-term total cost and total energy consumption based on the input state. In actual system operation, real-time data can be used to fine-tune the model online, allowing it to continuously adapt to new operating conditions and user habits.

[0044] Step S300: Energy dispatch execution.

[0045] The dynamic scheduling control module uses the calculated dynamic contribution weights. The optimal three-source coupling operation mode is determined, and specific control commands are generated.

[0046] For example, the following mode switching logic can be set: Solar-dominated heating mode: When Much larger and When the solar energy level exceeds a certain high threshold (e.g., 0.7), the system determines that the current solar energy resources are sufficient and the utilization efficiency is highest. At this time, the control module will prioritize activating the solar thermal collection cycle, supplying the collected heat directly or via a heat pump to the user terminal. Geothermal energy stable heating / cooling mode: At night or when solar energy is insufficient, if... Significantly higher than The system will prioritize activating the geothermal heat exchange module, utilizing geothermal energy as a stable heat or cold source.

[0047] Air source heat pump assist mode: When neither solar nor geothermal energy can meet the user's load demand, or during the transition season when air source heat pump efficiency is extremely high ( When the temperature is high, the air source heat exchange module is activated as a supplement or peak adjustment unit.

[0048] The optimal economic model for a three-source hybrid system: In most cases, the weights of the three energy sources will be in an intermediate state. At this point, the system will proceed according to... The system coordinates and controls the output power of the three energy modules in a proportional manner. For example, by controlling the operating frequency of the compressor module, adjusting the speed of the water pumps in each heat exchange module, or adjusting the opening of the electronic expansion valve, the three energy sources share the load according to their optimal weight ratio, thereby minimizing the overall operating cost of the system. This invention abandons the traditional simple control logic based on fixed thresholds and innovatively introduces a deep neural network model based on an attention mechanism. This deep neural network model can comprehensively process multi-dimensional, nonlinear dynamic information, including energy availability, user load, environmental parameters, economic costs (time-of-use pricing), and even weather forecasts, and calculate the optimal contribution weight of each energy source in real time. This data-driven intelligent scheduling method enables the system to adaptively find a dynamic balance point among multiple objectives such as maximum energy efficiency, minimum cost, or both, thereby achieving globally optimal operation.

[0049] The control commands are ultimately implemented as specific operation signals for each actuator in the system (such as compressor module start / stop and frequency, water pump speed, three-way valve switching status, fan speed, etc.).

[0050] Step S400: Closed-loop feedback adjustment.

[0051] After executing control commands, the system continuously monitors the operational performance, such as whether the indoor temperature stabilizes near the set value. The actual operating status (e.g., the deviation between the actual room temperature and the set room temperature) is used as a feedback signal. This signal is used for fine-tuning of traditional PID and other low-level control loops, and also serves as a basis for the next control cycle. One of the state inputs participates in a new round of dynamic weight calculation. This closed-loop feedback mechanism ensures that the system can continuously correct its control strategy to cope with various disturbances and model prediction deviations, guaranteeing the robustness and accuracy of control. Through the precise and coordinated control of the three energy sources by the intelligent scheduling model, this invention can prioritize and maximize the use of the most efficient and lowest-cost energy source (such as free solar energy during the day) at any time, and rationally allocate other energy sources as supplements or reserves. Compared with traditional dual-source systems or systems using simple control strategies, this invention can effectively reduce the ineffective operation of high-energy-consuming components such as compressor modules while meeting user needs, thereby significantly reducing the total system energy consumption and operating costs, and achieving a higher coefficient of performance (COP) and economic benefits.

[0052] The electronic device provided by this invention can be a dedicated controller for implementing the aforementioned dynamic scheduling control module. It internally includes a processor and a memory, with the memory containing a computer program that implements the above-described method. When the processor executes this program, it can perform intelligent control of the three-source coupled composite energy heat pump system. The computer-readable storage medium provided by this invention, such as a USB flash drive, optical disc, or solid-state drive, stores a computer program implementing the above-described method. This program can be loaded into a general-purpose or special-purpose computing device with a corresponding hardware interface for execution.

[0053] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, and not to limit them; although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features; and these modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. A three-source coupled composite energy heat pump system, characterized in that, include: A solar heat exchange module is used to collect solar energy and convert it into heat energy or electrical energy; An air source heat exchange module is used for heat exchange with ambient air; a geothermal heat exchange module is used for heat exchange with underground rock and soil or groundwater; and a dynamic scheduling and control module is used to acquire the operating status parameters of the solar heat exchange module, air source heat exchange module and geothermal heat exchange module and user load demand in real time, and determine the dynamic contribution weight of the three energy sources (solar, air and geothermal) based on a preset intelligent scheduling model to generate a three-source coupled operation mode, and control the coordinated operation of the solar heat exchange module, air source heat exchange module and geothermal heat exchange module.

2. The three-source coupled composite energy heat pump system according to claim 1, characterized in that, The dynamic scheduling control module is configured to calculate the first of the solar, air, and geothermal energy sources using a deep neural network model based on an attention mechanism. Energy in time step Dynamic contribution weight The calculation formula is as follows: , in, For the first Energy in time step Dynamic contribution weights; This is an index for energy types, with values ​​ranging from 1 to 3 in the system, corresponding to solar energy, air energy, and geothermal energy, respectively. The total number of coupled energy sources in the system ; This represents the time step of the current control cycle. For the first Energy in time step The source contribution potential vector; For the system at time step The global state vector; and Let be the trainable weight matrix of the deep neural network model; and These are the trainable weight vector and bias vector of the deep neural network model, respectively; It is the hyperbolic tangent activation function; It is a natural exponential function.

3. The three-source coupled composite energy heat pump system according to claim 2, characterized in that, The source contribution potential energy vector Including with the At least one real-time environmental parameter related to a certain energy source; the global state vector It includes at least one user load parameter and at least one operating cost parameter.

4. The three-source coupled composite energy heat pump system according to claim 3, characterized in that: When the energy type is solar energy, the source contributes the potential energy vector. This includes: solar irradiance intensity and collector surface temperature; when the energy type is air energy, the source contribution potential vector. This includes: outdoor ambient temperature and outdoor ambient humidity; when the energy type is geothermal energy, the source contribution potential energy vector. This includes: the inlet and outlet temperature difference of the buried pipe heat exchanger, and the average temperature of the underground soil and rock mass; the global state vector. This includes: user-set temperature, current indoor temperature, time-of-use electricity price, and weather forecast data for the future scheduled time period.

5. A three-source coupled composite energy heat pump system according to claim 2, characterized in that, The trainable weight matrix and trainable weight vector of the deep neural network model are obtained through offline pre-training using historical running data and online fine-tuning using real-time running data; the goal of the offline pre-training is to minimize a joint loss function that includes total energy consumption and operating cost. : , in, In order to time step The total energy consumption of the system. In order to time step The system operating cost, and The preset balance coefficient, This represents the total step size of the training cycle.

6. A three-source coupled composite energy heat pump system according to claim 1, characterized in that, The three-source coupling operation mode includes: solar-dominated heating mode: when the dynamic contribution weight... When the first preset condition is met, the solar heat exchange module is used as the main heat source; geothermal energy stable heating / cooling mode: when the dynamic contribution weight When the second preset condition is met, the geothermal heat exchange module is used as the base load unit first; Air source heat pump auxiliary mode: when neither solar nor geothermal energy can meet the user's load demand, the air source heat exchange module is activated as a supplement or peak regulation unit; Three-source hybrid optimal economic mode: based on the dynamic contribution weight. The real-time value is used to proportionally allocate the output power of each energy module, thereby minimizing the overall operating cost of the system.

7. A control method for a three-source coupled composite energy heat pump system according to any one of claims 1-6, characterized in that, The process includes the following steps: Status data acquisition step: Real-time acquisition of operating status parameters of solar heat exchange module, air heat exchange module and geothermal heat exchange module, as well as user load demand and environmental parameters, to form system status information; Dynamic weight calculation step: Inputting the system status information into a pre-trained intelligent scheduling model to calculate a dynamic contribution weight for each of the three energy sources, namely solar, air and geothermal. Energy dispatch execution steps: Based on the dynamic contribution weight, a three-source coupled operation mode is determined, and control commands are generated for the solar heat exchange module, air heat exchange module, geothermal heat exchange module, and compressor module; Closed-loop feedback adjustment steps: The control commands are executed, and the system operation effect is monitored. The effect deviation is used as a feedback signal for dynamic weight calculation in the next control cycle.

8. The control method for a three-source coupled composite energy heat pump system according to claim 7, characterized in that, The dynamic weight calculation step specifically includes: calculating the weight of solar energy, air energy, and geothermal energy using the following formula. Energy in time step Dynamic contribution weight : , in, To contribute weight dynamically, For energy type indexing, For total energy, For time step, For the first The potential energy vector contributed by the source of this energy source. This is the system's global state vector. These are the preset parameters within the intelligent scheduling model.

9. An electronic device comprising a memory and a processor, wherein the memory stores a computer program, characterized in that, When the processor executes the computer program, it performs the operation of the system according to any one of claims 1-6 or the method according to any one of claims 7-8.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that, When the computer program is executed by the processor, it performs the operations performed by the system according to any one of claims 1-6 or the method according to any one of claims 7-8.

Images