Method for detecting source measurement heat balance of ground source heat pump system

Abstract

The invention discloses a method for detecting source measurement heat balance of a ground source heat pump system, and particularly relates to the technical field of heat balance detection. Comprising the steps of heat exchange plane boundary determination, temperature monitoring well arrangement, source measurement heat exchange object mass calculation, initial working condition temperature determination, real-time heat balance state monitoring, repair capacity evaluation and repair efficiency curve generation. The heat balance state of the ground source heat pump system can be evaluated in real time through accurate temperature monitoring well layout and a heat flow distribution simulation model, and real-time temperature data of the ground source side are obtained by setting a plurality of temperature monitoring points in the boundary of the source side heat exchange plane and combining a high-precision temperature probe. And data errors are eliminated through a Kalman filtering algorithm, the real-time temperature is combined with the initial working condition temperature for comparison, the heat absorption or release condition of the system is accurately judged, and therefore the heat balance state is accurately monitored, and a basis is provided for follow-up repair capacity evaluation.

Description

Technical Field

[0001] The present invention relates to the technical field of thermal balance detection, and more particularly to a method for detecting thermal balance of a ground source heat pump system. Background Art

[0002] As an efficient and environmentally friendly way of utilizing energy, the ground source heat pump system has been widely used in the fields of heating and cooling. The system achieves efficient energy utilization through heat exchange between the ground source well and the underground soil. Since the operation effect of the ground source heat pump system is affected by multiple factors, such as groundwater level, soil type, and ground source well depth, the stability of its heat exchange process and the thermal balance state of the system are directly related to the energy efficiency and long-term stable operation of the system. However, most existing ground source heat pump systems rely on simple temperature monitoring methods to evaluate the thermal balance of the system. This method has problems such as insufficient accuracy and limited monitoring range, which makes it easy for heat imbalance to occur during the operation of the system, thereby affecting the overall efficiency of the heat pump system. Especially when the load of the ground source heat pump system changes greatly, traditional detection methods are often unable to reflect the dynamic changes of thermal balance in real time and accurately, and lack the ability to predict and respond quickly to potential problems.

[0003] At present, although some ground-source heat pump systems have adopted temperature sensors for monitoring, the layout of temperature monitoring wells often lacks scientific planning, resulting in insufficient representativeness of data collection and the inability to accurately reflect the heat absorbed or released by the ground source side in real time, which in turn affects the accuracy of thermal balance state judgment and cannot effectively track the self-recovery ability and repair efficiency of the ground source side working conditions. In addition, traditional thermal balance assessment methods only stay at the static calculation level and cannot fully consider the dynamic changes of multi-source data such as soil temperature, ambient temperature, and supply and return water temperature. Summary of the Invention

[0004] In order to overcome the above-mentioned defects of the prior art, an embodiment of the present invention provides a method for detecting the source-measured thermal balance of a ground-source heat pump system. By optimizing the layout of temperature monitoring wells, dynamically calculating heat absorption and release, and adaptively adjusting the thermal balance repair efficiency, it not only improves the system operation efficiency and heat exchange capacity, but also effectively reduces the risk of thermal balance imbalance and extends the service life of the equipment. Through the evaluation and prediction of the repair capacity, it can provide early warning of potential problems, optimize the system's operation and control strategies, reduce energy consumption and maintenance costs, and enhance the adaptability and long-term stability of the ground-source heat pump system in complex environments, so as to solve the problems raised in the above-mentioned background technology.

[0005] To achieve the above-mentioned object, the present invention provides the following technical solutions: a method for detecting source-measured heat balance of a ground-source heat pump system, comprising:

[0006] According to the plane distribution of the ground source wells in the ground source heat pump system, the source measurement heat exchange plane boundary is determined to clarify the monitoring area of ​​heat exchange;

[0007] At least two monitoring points are selected on the boundary of the source-measurement heat exchange plane as the locations of temperature monitoring wells. The depth of the temperature monitoring wells is half the depth of the ground source wells. A temperature probe is installed at the bottom of each temperature monitoring well to collect real-time temperature data on the ground source side.

[0008] According to the area of ​​the source measurement heat exchange plane boundary and the depth of the source well, the mass of the source measurement heat exchange object is determined to provide basic parameters for the calculation of heat change;

[0009] Based on the temperature data collected from the temperature monitoring wells, the average temperature is calculated as the baseline operating condition before the ground source heat pump system is put into operation;

[0010] During the operation of the ground source heat pump system, the thermal balance state of the ground source side is judged by comparing the average temperature collected in real time by the temperature monitoring well;

[0011] In the case of heat balance imbalance, the self-repair evaluation index of the ground source side is calculated, which is the imbalance heat divided by the time required for repair;

[0012] The distribution of the average temperature collected in real time by the temperature monitoring wells during the repair time is used to generate a repair efficiency curve, which is used to dynamically analyze the self-repair evaluation index at different temperatures and evaluate the operating status of the ground source side and its demand for cold and heat sources.

[0013] In a preferred embodiment, based on the horizontal arrangement of the ground source wells and the geological conditions, a ground source heat pump heat flow distribution simulation model is used to calculate the area of ​​the source-to-source heat exchange plane boundary. The model is established based on the two-dimensional heat conduction equation, and the ground source heat pump heat flow distribution simulation model is expressed as follows:

[0014]

[0015] Where T(x,y) is the temperature field distribution at position (x,y); t is time; α is the thermal diffusion coefficient;

[0016] Through numerical simulation calculation, the temperature field T(x,y) is solved, and the boundary area of ​​the heat exchange plane is determined by combining the following formula:

[0017]

[0018] Where A is the boundary area of ​​the heat exchange plane; x1 and x2 are the boundary ranges of the heat exchange plane in the x-axis direction; y1 and y2 are the boundary ranges of the heat exchange plane in the y-axis direction; T(x,y) is the temperature field distribution at position (x,y); T threshold is the temperature threshold of the heat exchange plane; dx and dy are the infinitesimal areas.

[0019] In a preferred embodiment, the location of the temperature monitoring well is selected based on a finite element analysis model, and the arrangement of the monitoring points is optimized by establishing a heat conduction and convection coupling equation, which is expressed as follows:

[0020]

[0021] Where Q is the internal heat source term; ρ is the soil density; c is the soil specific heat capacity; T is the temperature; t is the time; and k is the soil thermal conductivity. The location of the temperature monitoring well is selected based on the finite element analysis model. The layout of the monitoring points is optimized by establishing the heat conduction and convection coupling equation. The heat conduction and convection coupling equation is expressed as follows:

[0022]

[0023] Where Q is the internal heat source term; ρ is the soil density; c is the soil specific heat capacity; T is the temperature; t is the time; and k is the soil thermal conductivity.

[0024] In a preferred embodiment, the real-time average temperature of the temperature well is collected by a temperature probe, and the Kalman filter algorithm is used to remove data errors. Based on the comparison between the initial temperature and the real-time average temperature, the thermal equilibrium state of the ground source side is judged, and the heat absorption or release value is calculated. The thermal equilibrium state prediction formula of the ground source side is:

[0025]

[0026] in, is the current predicted state value; is the corrected temperature value at the k-1th time point; A is the state transfer matrix; B is the control input matrix; u k is the control input; its thermal equilibrium state judgment condition is:

[0027] When T0-T=0℃, the source-measurement thermal balance is stable; when Δt1=T0-T>0℃, the source-measurement absorbs too much heat; when Δt2=T-T0>0℃, the source-measurement releases too much heat. The heat calculation formula is:

[0028] Absorbed heat Q 吸 =C 土壤 m Δt1; release of heat Q 放 =C 土壤 m·Δt2; dynamically eliminate the error of the collected data through the Kalman filter algorithm, combine the thermal balance state judgment conditions and heat calculation formula to provide high-precision real-time working condition monitoring; predict the current predicted state value through the covariance prediction formula, and its covariance prediction formula is:

[0029] P k|k-1=A·P k-1|k-1 ·A T +Q

[0030] Among them, P k|k-1 is the prediction covariance, which indicates the accuracy of the prediction results; P k-1|k-1 is the covariance matrix of the previous moment; A is the state transfer matrix; A T is the transpose of the state transfer matrix, which is used for symmetry processing; Q is the process error.

[0031] In a preferred embodiment, the Kalman gain is used to calculate P k|k-1 The prediction covariance is updated, and its Kalman gain expression formula is:

[0032] K k =P k|k-1 ·H T ·(H·P k|k-1 ·H T +R) -1

[0033] Among them, K k is the Kalman gain; P k|k-1 is the prediction covariance; H is the measurement matrix; H T is the transpose of the measurement matrix; R is the measurement error; the predicted value is adjusted by the Mann gain, and its expression formula is:

[0034]

[0035] in, The updated real status; is the predicted state value; K k is the Kalman gain; H is the measurement matrix; z k is the actual measurement value, the temperature value measured by the sensor at the kth time point; is the difference between the measured value and the predicted value; the confidence at the current time point is recalculated through covariance update, and the covariance update expression formula is:

[0036] P k|k =(IK k ·H)·P k|k-1

[0037] Among them, P k|k is the updated uncertainty. After the predicted value is corrected, the confidence at the current time point is recalculated. I is the unit matrix, which represents a constant that has no effect on the system and is used to calculate the corrected uncertainty. H is the measurement matrix. K k is the Kalman gain; P k|k-1 is the prediction covariance.

[0038] In a preferred embodiment, a support vector machine model is used to construct a dynamic heat balance prediction. The supply and return water temperature, return water temperature, circulating water flow, soil temperature, and ambient temperature are used as input data of the support vector machine model and trained by a radial basis kernel function. The radial basis kernel function expression formula is:

[0039]

[0040] Among them, x is the input feature vector; x i is the sample in the training set; γ is the parameter of the kernel function, which controls the influence range of the sample; is the Euclidean distance between samples; substitute the output predicted thermal equilibrium state, and the thermal equilibrium state expression formula of the support vector machine output prediction is:

[0041]

[0042] Where H(t) is the thermal equilibrium state; α i is the coefficient of the support vector; y i is the label value of the sample; b is the bias term; is the similarity between samples calculated by the radial basis kernel function.

[0043] In a preferred embodiment, a multi-source data fusion model is used to generate a repair efficiency curve. The supply and return water temperature, return water temperature, circulating water flow, soil temperature, and ambient temperature data in the temperature monitoring well are fused with other sensor data to form a new input feature vector. The weighted average method is used to fuse these data. The calculation formula is:

[0044]

[0045] Among them, x i is the input value of different data sources; w i is the weight coefficient of the data source; the repair efficiency calculation formula is used to generate a repair efficiency curve to evaluate the time and heat required for the system to recover thermal balance. The repair efficiency expression formula is:

[0046]

[0047] Where E(t) is the repair efficiency; T current is the current temperature; T0 is the initial temperature; t is the time.

[0048] The technical effects and advantages of the present invention are as follows:

[0049] 1. Through the precise layout of temperature monitoring wells and the heat flow distribution simulation model, the thermal balance state of the ground source heat pump system can be evaluated in real time. By setting up multiple temperature monitoring points within the boundary of the source-measurement heat exchange plane and combining them with high-precision temperature probes, real-time temperature data on the ground source side can be obtained. The Kalman filter algorithm is used to eliminate data errors. By comparing the real-time temperature with the initial operating temperature, the heat absorption or release of the system can be accurately judged, thereby accurately monitoring the thermal balance state and providing a basis for subsequent repair capacity assessment.

[0050] 2. By combining multi-source data fusion and the generation of repair efficiency curves, the self-repair capability and operational optimization of the ground-source heat pump system are effectively improved. The data from the temperature monitoring wells and other sensor data are fused using the weighted averaging method to form a new input feature vector. The thermal balance recovery speed is dynamically evaluated based on the repair efficiency formula. Combined with the support vector machine model, historical and real-time data are used to predict the future thermal balance state of the system, further optimizing the system operation and control strategy to achieve early warning and efficient response. BRIEF DESCRIPTION OF THE DRAWINGS

[0051] Figure 1 It is a schematic diagram of the process of the present invention.

[0052] Figure 2 It is a horizontal schematic diagram of the source well of the present invention.

[0053] Figure 3 It is a vertical schematic diagram of the source well of the present invention. DETAILED DESCRIPTION

[0054] The following will clearly and completely describe the technical solutions in the embodiments of the present invention in conjunction with the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative efforts are within the scope of protection of the present invention.

[0055] Refer to the instruction manual Figure 1-3 , a method for detecting source-measurement heat balance of a ground-source heat pump system according to an embodiment of the present invention includes:

[0056] Determination of the boundary of the heat exchange plane: According to the plane distribution of the ground source wells in the ground source heat pump system, the boundary of the source-measurement heat exchange plane is determined to clarify the monitoring area of ​​heat exchange. The ground source heat pump system consists of a ground source well, a circulating water pump, a heat pump host, a temperature monitoring well and a data acquisition and control module. The ground source well serves as the core heat exchange equipment to exchange heat with the soil; the circulating water pump drives the circulating water to flow between the ground source well and the heat pump host; the heat pump host transfers heat through the compressor to achieve heating or cooling; the temperature monitoring well collects soil temperature data in real time to determine the thermal equilibrium state; the data acquisition and control module collects parameters such as temperature and flow, runs the control algorithm, and dynamically optimizes the system performance. These modules work together to achieve efficient absorption and release of heat on the ground source side and intelligent control of system operation;

[0057] Layout of temperature monitoring wells: Select at least two monitoring points on the boundary of the source-measurement heat exchange plane as the locations of temperature monitoring wells. The depth of the temperature monitoring wells should be half the depth of the ground source wells. Install a high-precision temperature probe with corrosion resistance at the bottom of each temperature monitoring well to collect real-time temperature data on the ground source side.

[0058] Calculation of the mass of the source-measured heat exchange object: Based on the area of ​​the source-measured heat exchange plane boundary and the depth of the ground source well, the mass of the source-measured heat exchange object is determined to provide basic parameters for heat change calculation;

[0059] Determination of initial operating temperature: Based on the temperature data collected from the temperature monitoring wells, the average temperature is calculated as the baseline operating condition before the ground source heat pump system is put into operation;

[0060] Real-time thermal balance status monitoring: During the operation of the ground source heat pump system, the thermal balance status of the ground source side is judged by comparing the average temperature collected in real time by the temperature monitoring well;

[0061] Evaluation of repair capacity: In the case of thermal imbalance, the self-repair evaluation index of the ground source side is calculated, and its value is the imbalance heat divided by the time required for repair;

[0062] Generation of repair efficiency curve: The average temperature distribution during the repair time collected in real time by temperature monitoring wells is used to generate a repair efficiency curve. This curve is used to dynamically analyze the self-repair evaluation index at different temperatures and evaluate the operating status of the ground source and its demand for cold and heat sources.

[0063] Based on the horizontal arrangement of the ground source wells and the geological conditions, the ground source heat pump heat flow distribution simulation model is used to calculate the area of ​​the source and measurement heat exchange plane boundary. The model is established based on the two-dimensional heat conduction equation, and the ground source heat pump heat flow distribution simulation model expression formula is:

[0064]

[0065] Where T(x,y) is the temperature field distribution at position (x,y), in °C; t is the time; α is the thermal diffusion coefficient, in m 2 / s, the calculation formula is Where k is the thermal conductivity of soil, ρ is the soil density, and c is the specific heat capacity of soil;

[0066] Through numerical simulation calculation, the temperature field T(x,y) is solved, and the boundary area of ​​the heat exchange plane is determined by combining the following formula:

[0067]

[0068] Where A is the boundary area of ​​the heat exchange plane; x1 and x2 are the boundary ranges of the heat exchange plane in the x-axis direction; y1 and y2 are the boundary ranges of the heat exchange plane in the y-axis direction; T(x,y) is the temperature field distribution at position (x,y); T threshold is the temperature threshold of the heat exchange plane; dx and dy are the infinitesimal areas. The two-dimensional heat conduction equation is used to accurately describe the heat diffusion process, clarify the heat exchange boundary range and optimize the monitoring design. The actual heat exchange capacity of the ground source heat pump is calculated based on the boundary area of ​​the heat exchange plane A. Combined with the temperature field distribution T(x,y) and the size of the boundary area of ​​the heat exchange plane A, the ground source well layout and system design are optimized.

[0069] The location of the temperature monitoring wells is selected based on the finite element analysis model. The layout of the monitoring points is optimized by establishing the heat conduction and convection coupling equation. The heat conduction and convection coupling equation is expressed as follows:

[0070]

[0071] Where Q is the internal heat source term; ρ is the soil density; c is the soil specific heat capacity, T is the temperature; t is the time; k is the soil thermal conductivity; a constant temperature condition T = T is set at the boundary of the source and measurement heat exchange plane. env Ambient temperature, heat flux condition q=h·(TT fluid ), where h is the heat transfer coefficient, the temperature gradient of the heat transfer plane is analyzed based on the numerical solution of the model, the layout of the temperature monitoring wells is optimized by the multi-physics field coupling method, and the areas with significant changes in thermal gradient are selected as monitoring points to ensure the comprehensiveness and representativeness of data collection. The thermal field distribution is simulated and calculated using the finite element analysis model, and the temperature monitoring wells are preferably arranged at the locations where the thermal field changes most significantly, thereby improving the accuracy of the temperature data and the reliability of the monitoring results;

[0072] The real-time average temperature of the temperature well is collected by the temperature probe, and the Kalman filter algorithm is used to remove data errors. Based on the comparison between the initial temperature and the real-time average temperature, the thermal equilibrium state of the ground source side is judged, and the heat absorption or release value is calculated. The thermal equilibrium state prediction formula of the ground source side is:

[0073]

[0074] in, is the current predicted state value; is the corrected temperature value at the k-1th time point; A is the state transfer matrix; B is the control input matrix; u k is the control input; its thermal equilibrium state judgment condition is:

[0075] When T0-T=0℃, the source-measurement thermal balance is stable; when Δt1=T0-T>0℃, the source-measurement absorbs too much heat; when Δt2=T-T0>0℃, the source-measurement releases too much heat. The heat calculation formula is:

[0076] Absorbed heat Q 吸 =C 土壤 m Δt1; release of heat Q 放 =C 土壤 m Δt2; The current predicted state value is predicted using the covariance prediction formula, and the covariance prediction formula is:

[0077] P k|k-1 =A·P k-1|k-1 ·A T +Q

[0078] Among them, P k|k-1 is the prediction covariance, which indicates the accuracy of the prediction result. If the value is large, the prediction result will be biased. A is the state transfer matrix. P k-1|k-1 is the covariance matrix of the previous moment; A T is the transpose of the state transfer matrix, used for symmetry processing; A·P k-1|k-1 ·A T The influence of the last uncertainty is the last uncertainty P k-1|k-1 Transfer to the current time point; Q is the process error; based on the temperature and patterns in the historical data, predict the current temperature value and determine the deviation of the predicted value;

[0079] The Kalman gain is used to calculate P k|k-1 The prediction covariance is updated, and its Kalman gain expression formula is:

[0080] K k =P k|k-1 ·H T ·(H·P k|k-1 ·H T +R) -1

[0081] Among them, K kKalman gain, which represents an adjustment coefficient that determines the weight of the predicted value and the actual measured value. k If it is large, trust the actual measured value; if it is small, trust the predicted value; k|k-1 is the prediction covariance, which indicates the accuracy of the prediction results; H is the measurement matrix; H T is the transpose of the measurement matrix; R is the measurement error, which represents the measurement error of the sensor. If the sensor is not very accurate, the value of R is larger; the predicted value is adjusted by the Mann gain, and its expression formula is:

[0082]

[0083] in, is the updated true state value, which is the final temperature value obtained by combining the predicted value and the actual measured value; is the predicted state value; K k is the Kalman gain; H is the measurement matrix; z k is the actual measurement value, the temperature value measured by the sensor at the kth time point; is the difference between the measured value and the predicted value, indicating the error between the actual measured value and the predicted value; the confidence at the current time point is recalculated through covariance update, and the covariance update expression formula is:

[0084] P k|k =(IK k ·H)·P k|k-1

[0085] Among them, P k|k is the updated uncertainty, and after the predicted value is corrected, the confidence at the current time point is recalculated; I is the unit matrix, which represents a constant that has no effect on the system and is used to calculate the corrected uncertainty; K k is the Kalman gain; P k|k-1 To predict the covariance, the thermal equilibrium state of the ground source side is predicted. First, the current temperature value is predicted based on the historical temperature data and the law of heat conduction, and the uncertainty of the predicted value is evaluated. Then, through the covariance update step, the difference between the actual measured temperature and the predicted value is compared, and the predicted value is adjusted according to the Kalman gain to make it closer to the true value. Finally, the corrected temperature value is obtained and the current uncertainty is updated to prepare for the next prediction. The thermal equilibrium state judgment condition is:

[0086] When T0-T=0℃, the source-measurement thermal balance is stable; when Δt1=T0-T>0℃, the source-measurement absorbs too much heat; when Δt2=T-T0>0℃, the source-measurement releases too much heat. The heat calculation formula is:

[0087] Absorbed heat Q 吸 =C 土壤m Δt1; release of heat Q 放 =C 土壤 m·Δt2; Dynamically eliminate the error of collected data through the Kalman filter algorithm, and combine the thermal equilibrium state judgment conditions and heat calculation formula to provide high-precision real-time working condition monitoring;

[0088] The support vector machine model is used to build a dynamic heat balance prediction, and the supply and return water temperature T in , return water temperature, circulating water flow, soil temperature, and ambient temperature are used as input data for the support vector machine model and trained through the radial basis kernel function. The radial basis kernel function expression formula is:

[0089]

[0090] Among them, x is the input feature vector; x i is the sample in the training set; γ is the parameter of the kernel function, which controls the influence range of the sample; || xx i || is the Euclidean distance between samples; substitute the output predicted thermal equilibrium state, and the thermal equilibrium state expression formula of the support vector machine output prediction is:

[0091]

[0092] Where H(t) is the thermal equilibrium state; α i is the coefficient of the support vector, reflecting the influence of the support vector on the model; i is the label value of the sample, indicating the thermal equilibrium state of the sample in the training set; b is the bias term, which controls the offset of the model; K(x,x i ) is the similarity between samples calculated by the radial basis kernel function; the parameters of the support vector machine are adjusted by the training data set to obtain the optimal support vector coefficient α i , thereby accurately predicting the future thermal equilibrium state H(t), using the support vector machine (SVM) model, based on historical and real-time data, to predict the future thermal equilibrium state, achieve early warning, and optimize the system's operation and control strategy;

[0093] The repair efficiency curve is generated by the multi-source data fusion model. The supply and return water temperature, return water temperature, circulating water flow, soil temperature, and ambient temperature data in the temperature monitoring well are fused with other sensor data to form a new input feature vector. The weighted average method is used to fuse these data. The calculation formula is:

[0094]

[0095] Among them, x i is the input value of different data sources; w i is the weight coefficient of the data source, satisfying Based on the fused data, a repair efficiency curve is generated using the repair efficiency calculation formula to evaluate the time and heat required for the system to recover thermal balance. The repair efficiency expression formula is:

[0096]

[0097] Where E(t) is the restoration efficiency, that is, the speed or effect of the ground source heat pump system restoring thermal balance during the restoration process; T current is the current temperature, which indicates the temperature of the ground source heat pump system at the current moment, usually the data collected in real time through temperature monitoring wells or other sensors; T0 is the initial temperature, which indicates the initial temperature when the system starts running, usually the temperature when no thermal imbalance occurs; t is the time, which indicates the time interval from the occurrence of thermal imbalance to the current moment.

[0098] Finally: The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of the present invention should be included in the scope of protection of the present invention.

Claims

1. A method for detecting heat balance of a ground source heat pump system, characterized in that: include; According to the plane distribution of the ground source wells in the ground source heat pump system, the source measurement heat exchange plane boundary is determined to clarify the monitoring area of ​​heat exchange; At least two monitoring points are selected on the boundary of the source-measurement heat exchange plane as the locations of temperature monitoring wells. The depth of the temperature monitoring wells is half the depth of the ground source wells. A temperature probe is installed at the bottom of each temperature monitoring well to collect real-time temperature data on the ground source side. According to the area of ​​the source measurement heat exchange plane boundary and the depth of the source well, the mass of the source measurement heat exchange object is determined to provide basic parameters for the calculation of heat change; Based on the temperature data collected from the temperature monitoring wells, the average temperature is calculated as the baseline operating condition before the ground source heat pump system is put into operation; During the operation of the ground source heat pump system, the thermal balance state of the ground source side is judged by comparing the average temperature collected in real time by the temperature monitoring well; In the case of heat balance imbalance, the self-repair evaluation index of the ground source side is calculated, which is the imbalance heat divided by the time required for repair; The distribution of the average temperature collected in real time by the temperature monitoring wells during the repair time is used to generate a repair efficiency curve, which is used to dynamically analyze the self-repair evaluation index at different temperatures and evaluate the operating status of the ground source side and its demand for cold and heat sources.

2. The method for detecting source heat balance of a ground source heat pump system according to claim 1, characterized in that: Based on the horizontal arrangement of the ground source wells and the geological conditions, the ground source heat pump heat flow distribution simulation model is used to calculate the area of ​​the source and measurement heat exchange plane boundary. The model is established based on the two-dimensional heat conduction equation, and the ground source heat pump heat flow distribution simulation model expression formula is: Where T(x,y) is the temperature field distribution at position (x,y); t is time; α is the thermal diffusion coefficient; Through numerical simulation calculation, the temperature field T(x,y) is solved, and the boundary area of ​​the heat exchange plane is determined by combining the following formula: Where A is the boundary area of ​​the heat exchange plane; x1 and x2 are the boundary ranges of the heat exchange plane in the x-axis direction; y1 and y2 are the boundary ranges of the heat exchange plane in the y-axis direction; T(x,y) is the temperature field distribution at position (x,y); T threshold is the temperature threshold of the heat exchange plane; dx and dy are the infinitesimal areas.

3. The method for detecting source heat balance of a ground source heat pump system according to claim 2, characterized in that: The location of the temperature monitoring wells is selected based on the finite element analysis model. The layout of the monitoring points is optimized by establishing the heat conduction and convection coupling equation. The heat conduction and convection coupling equation is expressed as follows: Where Q is the internal heat source term; ρ is the soil density; c is the soil specific heat capacity; T is the temperature; t is the time; and k is the soil thermal conductivity.

4. The method for detecting source heat balance of a ground source heat pump system according to claim 3, characterized in that: The real-time average temperature of the temperature well is collected by the temperature probe, and the Kalman filter algorithm is used to remove data errors. Based on the comparison between the initial temperature and the real-time average temperature, the thermal equilibrium state of the ground source side is judged, and the heat absorption or release value is calculated. The thermal equilibrium state prediction formula of the ground source side is: in, is the current predicted state value; is the corrected temperature value at the k-1th time point; A is the state transfer matrix; B is the control input matrix; u k is the control input; its thermal equilibrium state judgment condition is: When T0-T=0℃, the source-measurement thermal balance is stable; when Δt1=T0-T>0℃, the source-measurement absorbs too much heat; when Δt2=T-T0>0℃, the source-measurement releases too much heat. The heat calculation formula is: Absorb heat Release heat The Kalman filter algorithm is used to dynamically eliminate the error of the collected data. Combined with the thermal balance state judgment conditions and the heat calculation formula, high-precision real-time working condition monitoring is provided. The covariance prediction formula is used to predict the current predicted state value. The covariance prediction formula is as follows: P k|k-1 =A·P k-1|k-1 ·A T +Q Among them, P k|k-1 is the prediction covariance, which indicates the accuracy of the prediction results; P k-1|k-1 is the covariance matrix of the previous moment; A is the state transfer matrix; A T is the transpose of the state transfer matrix, which is used for symmetry processing; Q is the process error.

5. The method for detecting source heat balance of a ground source heat pump system according to claim 4, characterized in that: The Kalman gain is used to calculate P k|k-1 The prediction covariance is updated, and its Kalman gain expression formula is: K k =P k|k-1 ·H T ·(H·P k|k-1 ·H T +R) -1 Among them, K k is the Kalman gain; P k|k-1 is the prediction covariance; H is the measurement matrix; H T is the transpose of the measurement matrix; R is the measurement error; the predicted value is adjusted by the Mann gain, and its expression formula is: in, The updated real status; is the predicted state value; K k is the Kalman gain; H is the measurement matrix; z k is the actual measurement value, the temperature value measured by the sensor at the kth time point; is the difference between the measured value and the predicted value; the confidence at the current time point is recalculated through covariance update, and the covariance update expression formula is: P k|k =(I-K k ·H)·P k|k-1 Among them, P k|k is the updated uncertainty. After the predicted value is corrected, the confidence at the current time point is recalculated. I is the unit matrix, which represents a constant that has no effect on the system and is used to calculate the corrected uncertainty. H is the measurement matrix. K k is the Kalman gain; P k|k-1 is the prediction covariance.

6. The method for detecting source heat balance of a ground source heat pump system according to claim 5, characterized in that: The support vector machine model is used to construct dynamic heat balance prediction. The supply and return water temperature, return water temperature, circulating water flow, soil temperature, and ambient temperature are used as input data of the support vector machine model and trained by the radial basis kernel function. The radial basis kernel function expression formula is: K(x,x i )=exp(-γ||x-x i || 2 ) Among them, x is the input feature vector; x i is the sample in the training set; γ is the parameter of the kernel function, which controls the influence range of the sample; || xx i || is the Euclidean distance between samples; substitute the output predicted thermal equilibrium state, and the thermal equilibrium state expression formula of the support vector machine output prediction is: Where H(t) is the thermal equilibrium state; α i is the coefficient of the support vector; y i is the label value of the sample; b is the bias term; K(x,x i ) is the similarity between samples calculated by the radial basis kernel function.

7. The method for detecting source heat balance of a ground source heat pump system according to claim 6, characterized in that: The repair efficiency curve is generated by the multi-source data fusion model. The supply and return water temperature, return water temperature, circulating water flow, soil temperature, and ambient temperature data in the temperature monitoring well are fused with other sensor data to form a new input feature vector. The weighted average method is used to fuse these data. The calculation formula is: Among them, x i is the input value of different data sources; w i is the weight coefficient of the data source; the repair efficiency calculation formula is used to generate a repair efficiency curve to evaluate the time and heat required for the system to recover thermal balance. The repair efficiency expression formula is: Where E(t) is the repair efficiency; T current is the current temperature; T0 is the initial temperature; t is the time.

Images