Real-time dynamic accurate regulation and control method and system for ground source heat pump central air conditioner

Abstract

The invention relates to a real-time dynamic accurate regulation and control method and system for a ground source heat pump central air conditioner, and the regulation and control method comprises the following steps: a pipeline heat change calculation step: constructing a pipeline heat change calculation model, and calculating the heat change of an output pipeline; an indoor temperature and humidity change trend prediction step: constructing an indoor temperature and humidity change trend prediction model, and predicting and outputting the indoor temperature and humidity change trend; an air conditioning system AI intelligent regulation and control strategy generation step: constructing an air conditioning system AI intelligent regulation and control model, and generating a dynamic control strategy based on the pipeline heat change and the indoor temperature and humidity change trend by taking the total energy consumption of the system and the indoor temperature control deviation as a joint optimization target; and the ground source heat pump central air conditioner is regulated and controlled in real time, wherein the ground source heat pump central air conditioner is regulated and controlled on the basis of the dynamic control strategy. The regulation and control system comprises a pipeline heat change calculation module, an indoor temperature and humidity change trend prediction module, an air conditioning system AI intelligent regulation and control strategy generation module and a ground source heat pump central air conditioner real-time regulation and control module.

Description

Technical Field

[0001] This application belongs to the technical field of heating, ventilation and air conditioning control, and particularly relates to a real-time dynamic precise regulation method and system for a ground-source heat pump central air conditioner. Background Technique

[0002] Currently, for the control strategy of ground-source heat pump central air conditioners, most are controlled by manually setting parameters or using simple PID algorithms and energy consumption optimization models, often ignoring the impact of pipeline heat changes on the system's thermal balance and the indoor temperature and humidity changes caused by external temperature and humidity changes, resulting in a lag in indoor temperature regulation, low temperature control accuracy, and serious energy consumption waste.

[0003] In view of the existence of the above technical defects, this application is proposed.

[0004] It should be noted that the disclosure of the above background technical content is only for assisting in understanding the inventive concept and technical solution of the present invention, and it does not necessarily belong to the prior art of this patent application. Without clear evidence indicating that the above content was publicly available on the filing date of this application, the above background technology should not be used to evaluate the novelty and inventiveness of this application. Summary of the Invention

[0005] The purpose of this application is to provide a real-time dynamic precise regulation method and system for a ground-source heat pump central air conditioner to overcome or mitigate at least one aspect of the known existing technical defects.

[0006] The technical solution of this application is as follows:

[0007] On the one hand, a real-time dynamic precise regulation method for a ground-source heat pump central air conditioner is provided, including:

[0008] Pipeline heat change calculation step: Construct a pipeline heat change calculation model and calculate and output the pipeline heat change;

[0009] Indoor temperature and humidity change trend prediction step: Construct an indoor temperature and humidity change trend prediction model and predict and output the indoor temperature and humidity change trend;

[0010] Air-conditioning system AI intelligent regulation strategy generation step: Construct an air-conditioning system AI intelligent regulation model, and based on the pipeline heat change and the indoor temperature and humidity change trend, with the total system energy consumption and the indoor temperature control deviation as the joint optimization target, generate a dynamic control strategy;

[0011] Real-time regulation step of the ground-source heat pump central air conditioner: Based on the dynamic control strategy, implement the regulation of the ground-source heat pump central air conditioner.

[0012] Optionally, in the above real-time dynamic precise control method for a ground-source heat pump central air conditioner, in the calculation step of the pipeline heat change, the constructed pipeline heat change calculation model takes the water temperature difference between the supply / return water pipelines, the thermal conductivity of the pipeline material, and the temperature of the external environment as inputs, and calculates and outputs the pipeline heat change in real time.

[0013] Optionally, in the above real-time dynamic precise control method for a ground-source heat pump central air conditioner, in the calculation step of the pipeline heat change, the constructed indoor temperature and humidity change trend prediction model takes the indoor temperature and humidity, outdoor temperature and humidity, season, and weather type in the past period of time as inputs, and predicts and outputs the indoor temperature and humidity change trend in the future period of time.

[0014] Optionally, in the above real-time dynamic precise control method for a ground-source heat pump central air conditioner, in the calculation step of the pipeline heat change, the constructed indoor temperature and humidity change trend prediction model uses an LSTM neural network. Among the inputs, the season includes four seasons: spring, summer, autumn, and winter, and the weather type includes five types: sunny, cloudy, overcast, rainy, and snowy.

[0015] Optionally, in the above real-time dynamic precise control method for a ground-source heat pump central air conditioner, in the step of generating the AI intelligent control strategy for the air conditioning system, the constructed AI intelligent control model for the air conditioning system uses the DRL deep reinforcement learning algorithm, and the joint optimization objective is designed as:

[0016] min( total system energy consumption + λ·∣T_actual - T_target∣);

[0017] Wherein,

[0018] T_actual is the indoor actual measured temperature, and T_target is the indoor target temperature;

[0019] λ is the importance coefficient of the indoor temperature control deviation relative to the total system energy consumption. The larger its value, the greater the importance of the indoor temperature control deviation relative to the total system energy consumption in the target evaluation. On the contrary, the smaller its value, the smaller the importance of the indoor temperature control deviation relative to the total system energy consumption in the target evaluation;

[0020] Generating the dynamic control strategy includes adjusting the frequency of the ground-source heat pump host, the rotational speed of the water pump, and the opening degree of the valve.

[0021] On the other hand, a real-time dynamic precise control system for a ground-source heat pump central air conditioner is provided, including:

[0022] A pipeline heat change calculation module for constructing a pipeline heat change calculation model and calculating and outputting the pipeline heat change;

[0023] An indoor temperature and humidity change trend prediction module for constructing an indoor temperature and humidity change trend prediction model and predicting and outputting the indoor temperature and humidity change trend;

[0024] An AI intelligent control strategy generation module for the air conditioning system, which is used to build an AI intelligent control model for the air conditioning system. Based on the changes in pipeline heat and the trends of indoor temperature and humidity, with the total system energy consumption and indoor temperature control deviation as the joint optimization objectives, a dynamic control strategy is generated.

[0025] A real-time control module for the ground-source heat pump central air conditioning system, which is used to implement the control of the ground-source heat pump central air conditioning system based on the dynamic control strategy.

[0026] Optionally, in the above real-time dynamic precise control system for the ground-source heat pump central air conditioning system, in the pipeline heat change calculation module, the constructed pipeline heat change calculation model takes the water temperature difference between the input and return water pipelines, the thermal conductivity of the pipeline material, and the temperature of the external environment as inputs, and calculates and outputs the pipeline heat change in real time.

[0027] Optionally, in the above real-time dynamic precise control system for the ground-source heat pump central air conditioning system, in the pipeline heat change calculation module, the constructed indoor temperature and humidity change trend prediction model takes the indoor temperature and humidity, outdoor temperature and humidity, season, and weather type in the past period of time as inputs, and predicts and outputs the indoor temperature and humidity change trend in the future period of time.

[0028] Optionally, in the above real-time dynamic precise control system for the ground-source heat pump central air conditioning system, in the pipeline heat change calculation module, the constructed indoor temperature and humidity change trend prediction model uses an LSTM neural network. Among the inputs, the season includes four seasons: spring, summer, autumn, and winter, and the weather type includes five types: sunny, cloudy, overcast, rainy, and snowy.

[0029] Optionally, in the above real-time dynamic precise control system for the ground-source heat pump central air conditioning system, in the AI intelligent control strategy generation module for the air conditioning system, the constructed AI intelligent control model for the air conditioning system uses a DRL deep reinforcement learning algorithm, and the joint optimization objective is designed as:

[0030] min(total system energy consumption + λ·∣T_actual - T_target∣);

[0031] Where

[0032] T_actual is the indoor actually measured temperature, and T_target is the indoor target temperature;

[0033] λ is the importance coefficient of the indoor temperature control deviation relative to the total system energy consumption. The larger its value, the greater the importance of the indoor temperature control deviation relative to the total system energy consumption in the target evaluation. Conversely, the smaller its value, the smaller the importance of the indoor temperature control deviation relative to the total system energy consumption in the target evaluation.

[0034] Generating a dynamic control strategy includes adjusting the frequency of the ground-source heat pump main unit, the rotational speed of the water pump, and the opening degree of the valve.

[0035] The present application has at least the following beneficial technical effects:

[0036] Provided is a real-time dynamic precise regulation method for a ground-source heat pump central air conditioner. Through real-time monitoring and modeling of multi-dimensional data, integrating the heat change in the pipeline and the indoor temperature and humidity trend, an AI model adaptively generates a dynamic control strategy to implement the regulation of the ground-source heat pump central air conditioner, enabling real-time dynamic precise regulation of the ground-source heat pump central air conditioner, greatly reducing energy consumption, improving the temperature control accuracy, and having characteristics such as intelligent, real-time, unmanned, and dynamic operation. BRIEF DESCRIPTION OF THE DRAWINGS

[0037] Figure 1 is a schematic diagram of the real-time dynamic precise regulation method for a ground-source heat pump central air conditioner provided by an embodiment of the present application;

[0038] Figure 2 is a schematic diagram of the real-time dynamic precise regulation system for a ground-source heat pump central air conditioner provided by an embodiment of the present application.

[0039] For better illustration of this embodiment, some components in the drawings are omitted, enlarged, or reduced, which do not represent the dimensions of the actual product. In addition, the drawings are only for illustrative purposes and cannot be construed as a limitation to this patent. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0040] To make the technical solutions and their advantages of the present application clearer, the technical solutions of the present application will be further clearly and completely described in detail below with reference to the drawings. It can be understood that the specific embodiments described herein are only part of the embodiments of the present application, which are only used to explain the present application and not to limit the present application. It should be noted that for the convenience of description, only the parts related to the present application are shown in the drawings, and other related parts can refer to the general design. Without conflict, the embodiments in the present application and the technical features in the embodiments can be combined with each other to obtain new embodiments.

[0041] In addition, unless otherwise defined, the technical terms or scientific terms used in the description of this application should have the ordinary meanings understood by those of ordinary skill in the art to which this application pertains. The words indicating directions such as "upper", "lower", "left", "right", "center", "vertical", "horizontal", "inner", "outer", etc. used in the description of this application are only used to indicate relative directions or positional relationships, rather than implying that the device or component must have a specific orientation, be constructed and operated in a specific orientation. When the absolute position of the object being described changes, its relative positional relationship may also change accordingly. Therefore, it should not be construed as a limitation to this application. The terms "first", "second", "third", and similar terms used in the description of this application are only for descriptive purposes to distinguish different components, and cannot be construed as indicating or implying relative importance. The similar words such as "a", "an", or "the" used in the description of this application should not be construed as an absolute limitation on the quantity, but should be understood as having at least one. The similar words such as "including" or "comprising" used in the description of this application are intended to mean that the elements or objects appearing before this word cover the elements or objects listed after this word and their equivalents, without excluding other elements or objects.

[0042] In addition, it should also be noted that, unless otherwise clearly specified and limited, the similar words such as "installed", "connected", "linked" used in the description of this application should be understood in a broad sense. For example, the connection can be a fixed connection, a detachable connection, or an integral connection; it can be a mechanical connection or an electrical connection; it can be directly connected or indirectly connected through an intermediate medium, and can also be the communication inside two components. Those skilled in the art can understand its specific meaning in this application according to the specific situation.

[0043] A real-time dynamic precise regulation method for a ground-source heat pump central air conditioner, as Figure 1 shown.

[0044] Steps for calculating the pipeline heat change: Construct a calculation model for the pipeline heat change and calculate and output the pipeline heat change.

[0045] Research shows that the pipeline heat change is related to the water temperature of the water supply pipeline, the water temperature of the return water pipeline, the thermal conductivity of the pipeline material, and the temperature of the external environment.

[0046] For the water temperature of the water supply pipeline and the water temperature of the return water pipeline, they can be measured by setting temperature sensors on the water supply pipeline and the return water pipeline, and data is collected every 10 minutes, and can be characterized by the temperature difference between the supply / return water pipelines.

[0047] The thermal conductivity of the pipeline material is an inherent property of the pipeline, which can be obtained by referring to a material handbook or obtained through experiments.

[0048] The temperature of the external environment specifically includes the temperature of the spaces through which the pipes pass, such as stairs, corridors, rooms, etc., which can be obtained by setting temperature sensors, and data is collected every 10 minutes.

[0049] The pipeline heat change calculation model takes the water temperature difference between the supply / return water pipes, the thermal conductivity of the pipeline material, and the temperature of the external environment as inputs, and calculates and outputs the pipeline heat change in real time, which can be used to compensate for pipeline heat loss.

[0050] For the change in pipeline heat, it can be calculated by constructing a physical model or obtained by mathematical fitting based on actual data.

[0051] Attention can be focused only on the water temperature difference between the supply / return water pipes of each room, the thermal conductivity of the pipeline material, the temperature of the external environment, and the change in pipeline heat.

[0052] Steps for predicting the indoor temperature and humidity change trend: Construct an indoor temperature and humidity change trend prediction model and predict and output the indoor temperature and humidity change trend.

[0053] Research shows that the indoor temperature change trend is related to indoor temperature and humidity, outdoor temperature and humidity, seasons, and weather types.

[0054] The indoor temperature and humidity specifically include the temperature and humidity of the spaces through which the pipes pass, such as stairs, corridors, rooms, etc., which can be obtained by setting temperature and humidity sensors, and data is collected every 10 minutes.

[0055] The outdoor temperature and humidity are the temperature of the outdoor environment, that is, the temperature and humidity of the outdoor atmospheric environment, which can be obtained by setting temperature and humidity sensors, and data is collected every 10 minutes.

[0056] Seasons include spring, summer, autumn, and winter, which can be further divided into twelve solar terms, namely early spring, mid-spring, late spring, early summer, mid-summer, late summer, early autumn, mid-autumn, late autumn, early winter, mid-winter, and late winter, or twenty-four solar terms, and are determined once a day.

[0057] Weather types include sunny, cloudy, overcast, rainy, and snowy, and are determined once a day.

[0058] The indoor temperature and humidity change trend prediction model takes the indoor temperature and humidity, outdoor temperature and humidity, seasons, and weather types in the past period of time as inputs, and predicts and outputs the indoor temperature and humidity change trend in the future period of time. Specifically, it can use the data of indoor temperature and humidity, outdoor temperature and humidity, seasons, and weather types in the past 6 hours to predict and output the indoor temperature and humidity change trend in the next 1 hour, so as to provide conditions for pre-adjusting the indoor temperature.

[0059] The indoor temperature and humidity change trend prediction model can adopt an LSTM neural network and be trained using historical data.

[0060] It is only necessary to focus on the indoor temperature and humidity, outdoor temperature and humidity, season, weather type, and the changing trend of indoor temperature and humidity in each room.

[0061] Steps for generating the AI intelligent control strategy of the air conditioning system: Build an AI intelligent control model for the air conditioning system. Based on the changes in pipeline heat and the changing trend of indoor temperature and humidity, with the total system energy consumption and indoor temperature control deviation as the joint optimization objectives, generate a dynamic control strategy.

[0062] Knowing the changes in pipeline heat can compensate for pipeline heat loss in a targeted manner. Knowing the changing trend of indoor temperature and humidity can provide conditions for the pre-regulation of indoor temperature, eliminate the hysteresis of temperature regulation. The AI intelligent control model of the air conditioning system can generate a dynamic control strategy based on this, including the regulation of aspects such as the frequency of the ground source heat pump host, the rotational speed of the water pump, and the opening degree of the valve, so as to reduce energy consumption and improve the accuracy of temperature control.

[0063] The joint optimization objective can be designed as:

[0064] min(total system energy consumption + λ·∣T_actual - T_target∣);

[0065] Among them,

[0066] T_actual is the indoor actually measured temperature, and T_target is the indoor target temperature. The average value can be calculated by weighting according to the size of each room.

[0067] λ is the importance coefficient of the indoor temperature control deviation relative to the total system energy consumption. The larger its value, the greater the importance of the indoor temperature control deviation relative to the total system energy consumption in the target evaluation. Conversely, the smaller its value, the smaller the importance of the indoor temperature control deviation relative to the total system energy consumption in the target evaluation. Its specific value can be designed according to the specific actual situation.

[0068] The AI intelligent control model of the air conditioning system can adopt the DRL deep reinforcement learning algorithm, be trained using historical data, and the response delay < 30 seconds.

[0069] Real-time control steps of the ground source heat pump central air conditioning: Based on the dynamic control strategy, implement real-time control of the ground source heat pump central air conditioning to achieve unmanned closed-loop control.

[0070] In a specific example, on a sunny winter day in an office building, the monitored outdoor temperature is 2°C and the indoor target temperature is 22°C. With the control strategy of the existing PID algorithm, the system energy consumption is 32 kW·h and the temperature deviation is ±1.2°C. Applying the real-time dynamic precise regulation method for the ground-source heat pump central air-conditioning disclosed in the above embodiment, the model predicts that the heat loss in the pipeline will cause the temperature to drop by 1.5°C, and the predicted heat value to be compensated is 8,500 kJ. The dynamic control strategy adjusts the frequency of the ground-source heat pump main unit to 45 Hz and the pump speed to 1,200 rpm. The system energy consumption is 23 kW·h and the temperature deviation is ±0.4°C, which can greatly reduce the energy consumption and improve the temperature control accuracy.

[0071] It is verified that the real-time dynamic precise regulation method for the ground-source heat pump central air-conditioning disclosed in the above embodiment can autonomously make decisions for more than 90% of non-preset working conditions, can achieve a regulation response delay of <30 seconds for the ground-source heat pump central air-conditioning, overall reduce the system energy consumption by 15% - 30%, and control the temperature control deviation within the range of ±0.5°C.

[0072] The real-time dynamic precise regulation method for the ground-source heat pump central air-conditioning disclosed in the above embodiment, through real-time monitoring and modeling of multi-dimensional data, fuses the heat change in the pipeline and the indoor temperature and humidity trend, and adaptively generates a dynamic control strategy with an AI model to implement regulation on the ground-source heat pump central air-conditioning, can achieve real-time dynamic precise regulation of the ground-source heat pump central air-conditioning, can greatly reduce the energy consumption, improve the temperature control accuracy, and has characteristics such as intelligent, real-time, unmanned, and dynamic operation.

[0073] A real-time dynamic precise regulation system for a ground-source heat pump central air-conditioning, as Figure 2 shown.

[0074] A pipeline heat change calculation module, used to construct a pipeline heat change calculation model and calculate and output the pipeline heat change.

[0075] The pipeline heat change calculation model takes the water temperature difference between the supply / return water pipelines, the thermal conductivity of the pipeline material, and the outdoor environment temperature as inputs, and calculates and outputs the pipeline heat change in real time, which can be used to compensate for the pipeline heat loss.

[0076] An indoor temperature and humidity change trend prediction module, used to construct an indoor temperature and humidity change trend prediction model and predict and output the indoor temperature and humidity change trend.

[0077] The indoor temperature and humidity change trend prediction model takes the indoor temperature and humidity, outdoor temperature and humidity, season, and weather type in the past period of time as inputs, and predicts and outputs the indoor temperature and humidity change trend in the future period of time. Specifically, it can use the indoor temperature and humidity, outdoor temperature and humidity, season, and weather type data in the past 6 hours to predict and output the indoor temperature and humidity change trend in the next 1 hour, so as to provide conditions for pre-adjusting the indoor temperature.

[0078] The indoor temperature and humidity change trend prediction model can adopt the LSTM neural network and be trained using historical data.

[0079] Seasons include spring, summer, autumn, and winter, and weather types include sunny, cloudy, overcast, rainy, and snowy.

[0080] The AI intelligent control strategy generation module of the air conditioning system is used to construct the AI intelligent control model of the air conditioning system. Based on the pipeline heat change and the indoor temperature and humidity change trend, with the total system energy consumption and the indoor temperature control deviation as the joint optimization objective, a dynamic control strategy is generated.

[0081] The AI intelligent control model of the air conditioning system generates a dynamic control strategy, including the adjustment of the ground source heat pump host frequency, water pump speed, valve opening, etc., to reduce energy consumption and improve the temperature control accuracy.

[0082] The joint optimization objective can be designed as:

[0083] min(total system energy consumption + λ·∣T_actual - T_target∣);

[0084] Where,

[0085] T_actual is the indoor actual measured temperature, and T_target is the indoor target temperature, which can be weighted and averaged according to the size of each room.

[0086] λ is the importance coefficient of the indoor temperature control deviation relative to the total system energy consumption. The larger its value, the greater the importance of the indoor temperature control deviation relative to the total system energy consumption in the target evaluation. Conversely, the smaller its value, the smaller the importance of the indoor temperature control deviation relative to the total system energy consumption in the target evaluation. Its specific value can be designed according to the specific actual situation.

[0087] The AI intelligent control model of the air conditioning system can adopt the DRL deep reinforcement learning algorithm, be trained using historical data, and the response delay < 30 seconds.

[0088] The real-time control module of the ground source heat pump central air conditioning is used to implement the control of the ground source heat pump central air conditioning based on the dynamic control strategy, realizing unmanned closed-loop control.

[0089] For the ground source heat pump central air conditioning real-time dynamic precise control system disclosed in the above embodiments, since it corresponds to the ground source heat pump central air conditioning real-time dynamic precise control method disclosed in the above embodiments, the description is relatively simple. For specific related parts, reference can be made to the relevant descriptions in the part of the ground source heat pump central air conditioning real-time dynamic precise control method, and its technical effects can also be referred to the relevant technical effects in the relevant part of the ground source heat pump central air conditioning real-time dynamic precise control method, which will not be elaborated here.

[0090] In addition, those skilled in the art should also be able to realize that each module of the real-time dynamic precise control system of the ground-source heat pump central air conditioner disclosed in the embodiments of the present application can be implemented by electronic hardware, computer software, or a combination of the two. To clearly illustrate the interchangeability of hardware and software, in the present application, it is generally described according to functions. Whether these functions are executed in a hardware or software manner depends on the specific application and design constraints of the technical solution. Those skilled in the art can choose different methods to implement the described functions for each specific application and its actual constraints, but such implementation should not be considered to exceed the scope of the present application.

[0091] So far, the technical solution of the present application has been described in combination with the preferred embodiments shown in the drawings. Those skilled in the art should understand that the protection scope of the present application is obviously not limited to these specific embodiments. Without departing from the principle of the present application, those skilled in the art can make equivalent changes or substitutions to the relevant technical features, and the technical solutions after these changes or substitutions will all fall within the protection scope of the present application.

Claims

1. A real-time dynamic and precise control method for a ground source heat pump central air conditioner, characterized in that: include: Pipeline heat change calculation steps: build a pipeline heat change calculation model to calculate the output pipeline heat change; Indoor temperature and humidity change trend prediction steps: construct an indoor temperature and humidity change trend prediction model to predict and output the indoor temperature and humidity change trend; Steps for generating AI intelligent control strategy for air conditioning system: Build AI intelligent control model for air conditioning system, generate dynamic control strategy based on pipeline heat change, indoor temperature and humidity change trend, and take total system energy consumption and indoor temperature control deviation as joint optimization targets; Real-time control steps of ground source heat pump central air conditioning: Based on dynamic control strategy, implement control of ground source heat pump central air conditioning.

2. The real-time dynamic and precise control method of ground source heat pump central air conditioner according to claim 1 is characterized in that: In the pipeline heat change calculation step, a pipeline heat change calculation model is constructed, which takes the water temperature difference of the supply / return water pipeline, the thermal conductivity of the pipeline material and the temperature of the external environment as input, and calculates the output pipeline heat change in real time.

3. The real-time dynamic and precise control method of ground source heat pump central air conditioner according to claim 2 is characterized in that: In the pipeline heat change calculation step, the indoor temperature and humidity change trend prediction model is constructed with the indoor temperature and humidity, outdoor temperature and humidity, season and weather type in the past period as input, and predicts and outputs the indoor temperature and humidity change trend in the future period.

4. The real-time dynamic and precise control method of ground source heat pump central air conditioner according to claim 3 is characterized in that: In the pipeline heat change calculation step, the indoor temperature and humidity change trend prediction model constructed uses an LSTM neural network. The input seasons include spring, summer, autumn, and winter, and the weather types include sunny, cloudy, overcast, rainy, and snowy.

5. The real-time dynamic and precise control method of ground source heat pump central air conditioner according to claim 4 is characterized in that: In the step of generating the AI ​​intelligent control strategy for the air conditioning system, the constructed AI intelligent control model for the air conditioning system adopts the DRL deep reinforcement learning algorithm, and the joint optimization objective is designed as follows: min(total system energy consumption + λ · |T_actual - T_target|); in, T_actual is the actual measured indoor temperature, T_target is the indoor target temperature; λ is the importance coefficient of indoor temperature control deviation relative to the total energy consumption of the system. The larger its value is, the greater the importance of indoor temperature control deviation relative to the total energy consumption of the system in the target evaluation. Conversely, the smaller its value is, the smaller the importance of indoor temperature control deviation relative to the total energy consumption of the system in the target evaluation. The generated dynamic control strategy includes the adjustment of the ground source heat pump host frequency, water pump speed and valve opening.

6. A real-time dynamic and precise control system for ground source heat pump central air conditioning, characterized in that: include: The pipeline heat change calculation module is used to build a pipeline heat change calculation model and calculate the output pipeline heat change; Indoor temperature and humidity change trend prediction module, used to build indoor temperature and humidity change trend prediction model, predict and output indoor temperature and humidity change trend; The AI ​​intelligent control strategy generation module for the air conditioning system is used to build an AI intelligent control model for the air conditioning system. Based on the change in pipeline heat and the trend of indoor temperature and humidity, the total energy consumption of the system and the indoor temperature control deviation are used as joint optimization targets to generate a dynamic control strategy. The ground source heat pump central air conditioning real-time control module is used to control the ground source heat pump central air conditioning based on dynamic control strategies.

7. The real-time dynamic and precise control system for ground source heat pump central air conditioning according to claim 6 is characterized in that: In the pipeline heat change calculation module, a pipeline heat change calculation model is constructed, which takes the water temperature difference of the supply / return water pipeline, the thermal conductivity of the pipeline material and the temperature of the external environment as input, and calculates the output pipeline heat change in real time.

8. The real-time dynamic and precise control system for ground source heat pump central air conditioning according to claim 7 is characterized in that: In the pipeline heat change calculation module, the indoor temperature and humidity change trend prediction model constructed takes the indoor temperature and humidity, outdoor temperature and humidity, season and weather type in the past period as input, and predicts and outputs the indoor temperature and humidity change trend in the future period.

9. The real-time dynamic and precise control system for ground source heat pump central air conditioning according to claim 8 is characterized in that: In the pipeline heat change calculation module, the indoor temperature and humidity change trend prediction model constructed uses an LSTM neural network. The input seasons include spring, summer, autumn, and winter, and the weather types include sunny, cloudy, overcast, rainy, and snowy.

10. The real-time dynamic and precise control system for ground source heat pump central air conditioning according to claim 9, characterized in that: In the air conditioning system AI intelligent control strategy generation module, the constructed air conditioning system AI intelligent control model adopts the DRL deep reinforcement learning algorithm, and the joint optimization objective is designed as: min(total system energy consumption + λ · |T_actual - T_target|); in, T_actual is the actual measured indoor temperature, T_target is the indoor target temperature; λ is the importance coefficient of indoor temperature control deviation relative to the total energy consumption of the system. The larger its value is, the greater the importance of indoor temperature control deviation relative to the total energy consumption of the system in the target evaluation. Conversely, the smaller its value is, the smaller the importance of indoor temperature control deviation relative to the total energy consumption of the system in the target evaluation. The generated dynamic control strategy includes the adjustment of the ground source heat pump host frequency, water pump speed and valve opening.

Images