Dynamic digital twin modeling method for energy efficiency optimization of ground source heat pumps

Abstract

The present invention discloses a dynamic digital twin modeling method for optimizing the energy efficiency of a ground-source heat pump, which belongs to the field of digital modeling technology, including: obtaining CAD drawings of ground-source heat pump equipment installed in a target area, and constructing a ground-source heat pump simulation model based on the CAD drawings; receiving real-time operation data and equipment status parameters corresponding to the ground-source heat pump equipment and environmental data corresponding to the target area based on the Internet of Things technology, and driving the ground-source heat pump simulation model based on the real-time operation data to generate a digital twin model; generating real-time monitoring data, and updating the digital twin model based on the real-time monitoring data to generate a dynamic digital twin model; the dynamic digital twin model outputs the performance prediction results of the ground-source heat pump equipment, and the multi-objective optimization algorithm dynamically adjusts the equipment status parameters based on the performance prediction results to complete the energy efficiency optimization of the ground-source heat pump equipment. The present invention can improve the accuracy and practicality of the dynamic digital twin model.

Description

Technical Field

[0001] The present invention belongs to the technical field of digital modeling, and specifically relates to a dynamic digital twin modeling method applied to energy efficiency optimization of ground source heat pumps. Background Art

[0002] As an efficient heating and cooling technology, ground source heat pump systems have attracted widespread attention because they can use the relatively stable temperature of underground soil or groundwater to improve the efficiency of heat pumps. However, the energy efficiency of these systems is affected by many factors, including geological conditions, system design, operating parameters, and environmental variables. Traditional ground source heat pump system design and optimization methods often rely on empirical design and static simulation, which have limitations in dealing with complex and dynamically changing environmental conditions. With the development of Internet of Things (IoT), big data, and artificial intelligence (AI) technologies, digital twin technology provides new possibilities for real-time monitoring, fault prediction, and performance optimization of ground source heat pump systems.

[0003] Similar prior art includes a Chinese patent application with publication number CN114757088A, which provides an energy equipment monitoring system and method based on digital twins. The system includes: a basic physical unit, which obtains the monitoring data of the monitored energy equipment; a data processing unit, which performs algorithmic analysis on the monitoring data, obtains the status data of the monitored energy equipment, and outputs the status data to a mechanism mapping unit; a mechanism mapping unit, which constructs a minimum monitoring environment model for the status data through machine learning; a technical performance unit, which obtains a three-dimensional model of the monitored energy equipment through three-dimensional simulation technology based on the minimum monitoring environment model; and an application interaction unit, which instructs to retrieve monitoring data and / or update status data according to the instructions made by the user based on the three-dimensional model. The embodiment of this specification uses digital twin technology to load the monitoring data of the energy equipment on the three-dimensional model for display. The user can indicate the monitored energy equipment based on the three-dimensional model display information. There is also a Chinese patent application with publication number CN113312773A, which discloses a method for refining the energy efficiency of ground source heat pump equipment, which belongs to the field of ground source heat pump technology. The method includes the following steps: obtaining ground source heat pump equipment data, collecting independent variables and dependent variables in the ground source heat pump equipment data to construct a ground source heat pump mathematical model; using a BP neural network prediction method to determine the energy efficiency value of the ground source heat pump mathematical model; using a limited recursive average filtering method to reduce the noise of the energy efficiency value to generate a predicted energy efficiency value. Compared with the prior art, the energy efficiency refinement method of a ground source heat pump equipment in the application first establishes a mathematical model, then uses a BP neural network model to predict the energy efficiency value of the ground source heat pump equipment, and uses a limited recursive average filtering method to reduce the noise of the predicted value, thereby obtaining an energy efficiency value of the ground source heat pump equipment that meets the accuracy requirements.

[0004] However, in the above-mentioned prior art, only the three-dimensional model of the monitored energy equipment is obtained through the three-dimensional simulation technology, and the energy efficiency of the energy equipment is not optimized and analyzed. The energy efficiency value of the ground source heat pump equipment that meets the accuracy requirements is only obtained by constructing a mathematical model of the ground source heat pump, and the ground source heat pump equipment is not regulated to achieve energy efficiency optimization. Summary of the invention

[0005] To solve the above problems, the present invention provides a dynamic digital twin modeling method applied to the energy efficiency optimization of ground source heat pumps to solve the problems in the prior art.

[0006] In order to achieve the above-mentioned purpose of the invention, the present invention proposes a dynamic digital twin modeling method for optimizing the energy efficiency of a ground source heat pump, comprising:

[0007] S1: Obtaining a CAD drawing of a ground source heat pump device installed in a target area, and constructing a ground source heat pump simulation model based on the CAD drawing;

[0008] S2: receiving the real-time operation data and equipment status parameters corresponding to the ground source heat pump equipment and the environmental data corresponding to the target area based on the Internet of Things technology, performing data preprocessing on the environmental data to generate standard environmental data, and driving the ground source heat pump simulation model based on the real-time operation data to generate a digital twin model;

[0009] S3: Monitor the model operation index of the digital twin model based on the real-time operation data. If the model operation index is less than a first preset value, re-expand the real-time operation data to generate real-time monitoring data, and update the digital twin model based on the real-time monitoring data to generate a dynamic digital twin model. Repeat this step until the model operation index corresponding to the dynamic digital twin model is greater than or equal to the first preset value.

[0010] S4: The dynamic digital twin model analyzes and predicts the real-time monitoring data based on the standard environmental data, and outputs the performance prediction results of the ground source heat pump equipment, and sets a multi-objective optimization algorithm. The multi-objective optimization algorithm dynamically adjusts the equipment state parameters based on the performance prediction results to complete the energy efficiency optimization of the ground source heat pump equipment.

[0011] Furthermore, constructing a ground source heat pump simulation model based on the CAD drawing comprises the following steps:

[0012] Building a component information library based on the CAD drawing, the component information library includes component types, component coordinates and component shapes of all components, respectively obtaining the operation association between any two components in the component information library and the operation priority of any component based on the component type, setting any of the components as nodes, generating a connection line between any two nodes based on the operation association and the operation priority, and sequentially combining all nodes and the connection lines to generate a node association graph;

[0013] All components are fused and simulated using a simulator based on the node association diagram and the CAD drawing to generate the ground source heat pump simulation model.

[0014] Furthermore, the receiving of the real-time operation data and equipment status parameters corresponding to the ground source heat pump equipment and the environmental data corresponding to the target area based on the Internet of Things technology includes the following steps:

[0015] Based on the node association graph, a plurality of sensors are respectively arranged in the ground source heat pump device and the target area, a data storage unit and a device adjustment unit are arranged for each sensor, a collection method of the sensor is obtained, and the sensor data is stored in a corresponding storage unit based on the collection method;

[0016] A plurality of sensors are randomly selected and set as a sensor group, and sensing data corresponding to the sensor group is obtained. If any sensing data changes, a data request is sent to the storage unit corresponding to the sensor group;

[0017] After receiving the data request, the storage unit divides the sensor data corresponding to the sensor group into multiple data groups based on time characteristics, and transmits all the data groups to the ground source heat pump simulation model in sequence, wherein the data group belonging to the target area is set as the environmental data, and the data group belonging to the ground source heat pump equipment is set as the real-time operation data, and the sensor corresponding to any of the components is obtained based on the node association diagram and set as the component sensor, and the information contained in the equipment adjustment unit of the component sensor is set as the equipment status parameter.

[0018] Furthermore, the generation of the digital twin model comprises the following steps:

[0019] The real-time operation data is integrated and transmitted into the geothermal heat pump simulation model, and the geothermal heat pump simulation model is calibrated using the real-time operation data based on a preset time interval until the confidence level of the geothermal heat pump simulation model is greater than or equal to a second preset value, and the calibrated geothermal heat pump simulation model is set as the digital twin model.

[0020] Furthermore, monitoring the model operation index of the digital twin model based on the real-time operation data includes the following steps:

[0021] All components in the component information library are marked with importance levels based on the corresponding component types, an importance list is generated, the component types corresponding to all the real-time operation data are obtained and set as a first equipment group, the importance levels corresponding to the first equipment group are accumulated based on the importance list to generate a first value, the accuracy of the real-time operation data is checked, and the mean accuracy of all the real-time operation data is calculated based on the accuracy, and the product of the mean accuracy and the first value is set as the model operation index.

[0022] Furthermore, the generating of real-time monitoring data comprises the following steps:

[0023] Extracting components whose importance levels are greater than or equal to a third preset value from the importance list and setting them as a second device group, combining the first device group and the second device group and setting them as a third device group, calculating the coordinate distance between any two components in the third device group based on the component coordinates, and setting the components whose coordinate distance is less than a fourth preset value as the device group to be selected;

[0024] Components of the same component type and with lower operation priority in the to-be-selected equipment group are eliminated, and real-time operation data corresponding to the remaining components in the third equipment group are set as the real-time monitoring data.

[0025] Furthermore, the outputting of the performance prediction result of the ground source heat pump equipment comprises the following steps:

[0026] Respectively obtain the environment tags corresponding to the standard environment data and the component tags corresponding to the real-time monitoring data, collect the standard environment data and the real-time monitoring data based on a preset time period, and set them as the first type of historical data and the second type of historical data;

[0027] Constructing a neural network model and a mathematical analysis model, inputting the first category of historical data into the neural network model for learning and training, and outputting the environmental prediction results corresponding to the environmental labels within the prediction time, combining all environmental prediction results and setting them as the first category of prediction results;

[0028] The mathematical analysis model extracts associated data corresponding to preset performance influencing factors from the first category of historical data and the second category of historical data based on a feature selection algorithm, wherein the performance influencing factors include groundwater variables, output heat, and input power;

[0029] The performance coefficient C of the ground source heat pump device at the i-th time point is calculated based on the first formula i, the first formula is: Among them, Q i is the associated data of the i-th time point corresponding to the output heat, P in For the correlation data of the i-th time point corresponding to the input power, all performance coefficients are summarized based on the preset period to generate a performance sequence, and a response coefficient between the correlation data corresponding to the groundwater variable and the performance sequence is obtained;

[0030] Based on the environmental tag, the variable prediction result corresponding to the groundwater variable is extracted from the first type of prediction results, and based on the response coefficient and the variable prediction result, the numerical result of the performance sequence within the prediction time is generated, and the numerical result is set as the performance prediction result.

[0031] Furthermore, the multi-objective optimization algorithm dynamically adjusts the device state parameters based on the performance prediction result, including the following steps:

[0032] If the performance prediction results are all less than the fifth preset value within the prediction time, the target variables are set based on the performance prediction results, an optimization model is constructed based on the target variables and the multi-objective optimization algorithm, and the real-time monitoring data and the standard environment data are input into the optimization model, the equipment status parameters are adjusted, and new real-time monitoring data are generated. This step is repeated until the performance prediction results are all greater than or equal to the fifth preset value.

[0033] Compared with the prior art, the beneficial effects of the present invention are at least as follows:

[0034] The present invention first generates a ground source heat pump simulation model through simulation using CAD drawings and node association diagrams, and can accurately establish a three-dimensional model of various components included in the ground source heat pump equipment. Then, the real-time operation data is integrated and transmitted to the ground source heat pump simulation model through the Internet of Things technology to generate a digital twin model. The real-time operation data is expanded through the model operation index to generate a dynamic digital twin model, which can improve the accuracy and practicality of the dynamic digital twin model.

[0035] The present invention also constructs an optimization model by setting a target optimization algorithm, and uses a dynamic digital twin model to predict and analyze the performance coefficient of the ground source heat pump equipment, thereby improving the energy efficiency optimization effect of the ground source heat pump equipment. BRIEF DESCRIPTION OF THE DRAWINGS

[0036] Figure 1 This is a step flow chart of a dynamic digital twin modeling method for optimizing the energy efficiency of a ground source heat pump according to the present invention. DETAILED DESCRIPTION

[0037] In order to make the purpose, technical solution and advantages of the present invention more clearly understood, the present invention is further described in detail below in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present invention and are not intended to limit the present invention.

[0038] It is understood that the terms "first", "second", etc. used in this application may be used herein to describe various elements, but unless otherwise specified, these elements are not limited by these terms. These terms are only used to distinguish a first element from another element. For example, without departing from the scope of this application, a first xx script may be referred to as a second xx script, and similarly, a second xx script may be referred to as a first xx script.

[0039] like Figure 1 As shown, a dynamic digital twin modeling method for optimizing the energy efficiency of a ground source heat pump includes:

[0040] S1: Obtain the CAD drawings of the ground source heat pump equipment installed in the target area, and build a ground source heat pump simulation model based on the CAD drawings.

[0041] Specifically, in this embodiment, the ground source heat pump equipment is a heating and cooling system equipment that uses renewable energy technology of shallow geothermal resources on the earth's surface (usually several meters underground), and its main components include: heat pump unit, underground heat exchanger, fluid circulation system, control unit, cooling tower, terminal equipment, compressor and various valve parts, etc. The target area refers to the actual location where the ground source heat pump equipment is installed, and the design and installation parameters of all necessary components of the ground source heat pump equipment can be obtained in the CAD drawings. The ground source heat pump simulation model refers to a ground source heat pump simulation model established based on the physical information in the CAD drawings and the collected data, and can include a replica simulation of the actual main components of the ground source heat pump equipment.

[0042] S2: Based on the Internet of Things technology, the real-time operation data and equipment status parameters corresponding to the ground source heat pump equipment and the environmental data corresponding to the target area are received, and the environmental data is preprocessed to generate standard environmental data. The ground source heat pump simulation model is driven based on the real-time operation data to generate a digital twin model.

[0043] Specifically, in this embodiment, the data of the ground source heat pump equipment can be synchronized to the ground source heat pump simulation model in real time through the Internet of Things technology, thereby generating a digital twin model. Among them, real-time operation data refers to data collected by setting various sensors when the ground source heat pump equipment is running, such as the water pressure of the underground heat exchanger, the exhaust temperature of the heat pump unit, etc., and the equipment state parameters refer to the state parameters of various main components when the ground source heat pump equipment is running, such as the operating frequency of the compressor and the flow parameters of the control unit, that is, the data passively acquired in the ground source heat pump equipment is divided into real-time operation data, and the data for actively controlling the operation of the equipment is divided into equipment state parameters. Environmental data is data corresponding to groundwater temperature, groundwater flow, soil temperature, load side ambient temperature, soil humidity and solar radiation in the target area. Data preprocessing refers to processing environmental data by methods such as data cleaning and data conversion to generate standardized and accurate data, that is, standard environmental data, which can provide an accurate data basis for the operation of the ground source heat pump equipment. By integrating real-time operation data into the ground source heat pump simulation model and updating the data information contained in the ground source heat pump simulation model in real time, a digital twin model can be constructed and generated.

[0044] S3: Monitor the model operation index of the digital twin model based on the real-time operation data. If the model operation index is less than the first preset value, re-expand the real-time operation data to generate real-time monitoring data, and update the digital twin model based on the real-time monitoring data to generate a dynamic digital twin model. Repeat this step until the model operation index corresponding to the dynamic digital twin model is greater than or equal to the first preset value.

[0045] Specifically, in this embodiment, the model operation index refers to the necessity of integrating the acquired data in the digital twin model. In order to more accurately monitor the operation of the digital twin model and reduce the load of data integration, the model operation index can be used to regulate the integration type of real-time operation data and reduce the transmission of unnecessary real-time operation data. When the model operation index is less than the first preset value, it means that the necessity of the real-time operation data integrated by the digital twin model is low. Therefore, it is necessary to re-expand the real-time operation data, set the expanded real-time operation data as real-time monitoring data, and integrate the real-time monitoring data into the digital twin model to generate a dynamic digital twin model. Among them, the model operation index of the dynamic digital twin model is greater than or equal to the first preset value, and the setting of the first preset value is determined by the type of real-time operation data.

[0046] S4: The dynamic digital twin model analyzes and predicts the real-time monitoring data based on standard environmental data, and outputs the performance prediction results of the ground-source heat pump equipment. A multi-objective optimization algorithm is set up. The multi-objective optimization algorithm dynamically adjusts the equipment state parameters based on the performance prediction results to complete the energy efficiency optimization of the ground-source heat pump equipment.

[0047] Specifically, in this embodiment, the performance prediction result refers to the energy efficiency change result of the ground source heat pump equipment generated by the dynamic digital twin model combined with the change trend of the standard environmental data. The multi-objective optimization algorithm refers to an algorithm that considers multiple objectives or criteria in the optimization problem at the same time and attempts to find a solution that achieves the best trade-off between these objectives. It can be used to solve the energy efficiency optimization needs of ground source heat pump equipment. Among them, adjusting the equipment state parameters of the ground source heat pump equipment to achieve energy efficiency optimization, for example, adjusting the flow value of the circulating fluid in the fluid circulation system can reduce energy consumption.

[0048] As a preferred technical solution of the present invention, building a ground source heat pump simulation model based on CAD drawings includes the following steps:

[0049] A component information library is constructed based on CAD drawings. The component information library includes the component types, component coordinates and component shapes of all components. Based on the component type, the operation correlation between any two components in the component information library and the operation priority of any component are obtained respectively. Any component is set as a node. Based on the operation correlation and operation priority, a connection line between any two nodes is generated. All nodes and connection lines are combined in sequence to generate a node association graph.

[0050] All components are integrated and simulated using a simulator based on the node association diagram and CAD drawings to generate a ground source heat pump simulation model.

[0051] Specifically, in this embodiment, the component information library refers to a list library established by CAD drawings containing all component information in the ground source heat pump equipment, wherein the component refers to the constituent parts, for example, various sensors for collecting data, and components of the ground source heat pump equipment, the component type refers to the component name and component performance, the component coordinates refer to the spatial position of the component installed in the target area corresponding to the ground source heat pump equipment, and the component shape refers to the structural shape of the component. Operation relevance refers to whether there is coordinated progressive work between any two components, and the operation priority refers to the operation order of the components. For example, after component A in the ground source heat pump equipment is operated, component B immediately receives the fluid of component A and then performs operation processing, then there is an operation relevance between component A and component B, and the operation priority of component A is higher than that of component B. Since the ground source heat pump equipment is a complex linkage device, each component can be set as a node, and then any two nodes are connected by operation relevance, and the direction of the connecting line is set by the operation priority, thereby generating a node association diagram, wherein the node association diagram can clearly locate the operation relevance of each component, which is convenient for the subsequent establishment of the intelligent regulation of the digital twin model as a basis.

[0052] A simulator refers to simulation software, such as ANSYS, SolidWorks Simulation and other simulation software. A simulator can be used to generate a ground source heat pump simulation model from component shapes, node association diagrams and CAD drawings.

[0053] As a preferred technical solution of the present invention, receiving the real-time operation data and equipment status parameters corresponding to the ground source heat pump equipment and the environmental data corresponding to the target area based on the Internet of Things technology includes the following steps:

[0054] Based on the node association diagram, multiple sensors are respectively set in the ground source heat pump equipment and the target area, a data storage unit and a device adjustment unit are set for each sensor, the collection method of the sensor is obtained, and the sensor data is stored in the corresponding storage unit based on the collection method.

[0055] A plurality of sensors are randomly selected and set as a sensor group, and sensing data corresponding to the sensor group is obtained. If any sensing data changes, a data request is sent to the storage unit corresponding to the sensor group.

[0056] After receiving the data request, the storage unit divides the sensor data corresponding to the sensor group into multiple data groups based on the time characteristics, and transmits all the data groups to the ground source heat pump simulation model in turn, wherein the data group belonging to the target area is set as environmental data, and the data group belonging to the ground source heat pump equipment is set as real-time operation data. Based on the node association diagram, the sensor corresponding to any component is obtained and set as the component sensor, and the information contained in the equipment adjustment unit of the component sensor is set as the equipment status parameter.

[0057] Specifically, in this embodiment, the Internet of Things (IoT) technology is a network concept composed of multiple networks such as the Internet, traditional telecommunications networks, and sensor networks. Objects in the real world are connected to the network through embedded systems to achieve intelligent control and management. Based on the node association graph, multiple sensors are set up respectively, wherein the sensors include but are not limited to temperature sensors, intelligent controllers, liquid level sensors, etc., and have the functions of data acquisition and IoT control response. The data storage unit refers to the storage of data collected by the sensor, and the equipment adjustment unit has the function of storing and adjusting the equipment state parameters of the components corresponding to the sensors in the ground source heat pump equipment. For example, the sensor corresponding to the intelligent controller installed on the compressor can adjust the operating frequency of the compressor. The acquisition mode refers to the working mode, sampling frequency, data type, etc. of the sensor, and the sensor data refers to the data collected and generated by the sensor by the corresponding acquisition mode.

[0058] A plurality of sensors are extracted to form a sensor group. If any corresponding sensor data changes, it indicates that there is an operation change of the ground source heat pump equipment in the sensor group. Therefore, a data request is sent to the storage unit where all the sensor data in the sensor group are located.

[0059] After receiving the data request, the storage unit divides the corresponding sensor data into multiple data groups according to the time feature, and finally transmits all the data groups in sequence. Among them, the time feature refers to the time point of the corresponding collection of the sensor data. Dividing the sensor data at the same time point into one data group can avoid the delay caused by the simultaneous transmission of all sensor data. The ground source heat pump simulation model receives all the data groups accordingly. And the sensors corresponding to the data groups can classify all the data groups into environmental data and real-time operation data.

[0060] The sensor of the component used to control the ground source heat pump equipment, that is, the component sensor, can be extracted through the node association diagram, and the data information of its equipment adjustment unit at the current time point can be set as the equipment state parameter. For example, if the equipment adjustment unit corresponding to the component sensor used to control the compressor operating frequency contains information P1, it means that the equipment state parameter of the compressor is the operating frequency P1.

[0061] As a preferred technical solution of the present invention, generating a digital twin model includes the following steps:

[0062] The real-time operation data is integrated and transmitted to the ground source heat pump simulation model, and the ground source heat pump simulation model is calibrated using the real-time operation data based on a preset time interval until the confidence of the ground source heat pump simulation model is greater than or equal to a second preset value, and the calibrated ground source heat pump simulation model is set as a digital twin model.

[0063] Specifically, in this embodiment, the real-time operation data is transmitted to the ground source heat pump simulation model by the Internet of Things technology, and the preset time interval means that at each preset interval, the model needs to be calibrated by the confidence of the ground source heat pump simulation model, wherein the confidence can be evaluated by comparing the similarity between the data received and displayed by the corresponding ground source heat pump simulation model at the same time point and the real-time operation data. If the confidence is greater than or equal to the second preset value, it means that the calibrated ground source heat pump simulation model can accurately receive and display the real-time operation data, and the model can be set as a digital twin model.

[0064] As a preferred technical solution of the present invention, monitoring the model operation index of the digital twin model based on real-time operation data includes the following steps:

[0065] All components in the component information library are marked with importance levels based on the corresponding component types, an importance list is generated, the component types corresponding to all real-time operation data are obtained and set as the first equipment group, the importance levels corresponding to the first equipment group are accumulated based on the importance list to generate a first value, the accuracy of the real-time operation data is tested, and the mean accuracy of all real-time operation data is calculated based on the accuracy, and the product of the mean accuracy and the first value is set as the model operation index.

[0066] Specifically, in this embodiment, the component information library contains all component information in the ground source heat pump device, including sensor information. The importance level is the importance of each component. For example, components of the same component type have different importance levels in different locations. The importance list refers to a new list formed by adding the importance levels of all components in the component information library.

[0067] The component type corresponding to all currently received real-time operation data is set to the first device group, because, in the digital twin model, it is not necessary to receive and display the sensor data corresponding to all components, and the load of data transmission can be reduced. The accuracy can be tested by the proportion of outliers in the received real-time operation data. For example, the standard deviation method is used to extract outliers in the real-time operation data, and the accuracy G is calculated based on the second formula. The second formula is: Among them, g1 is the total number of real-time operation data, and g2 is the number of outliers. The mean accuracy refers to the average value of the accuracy of the real-time operation data corresponding to the first device group. For example, the first device group contains component types including components A1, A2 and A3, and the corresponding importance levels are 1, 2 and 3 respectively. Then the first value is: 1+2+3=6, and the accuracy of the real-time operation data corresponding to components A1, A2 and A3 are 0.7, 0.8 and 0.9 respectively. Then the mean accuracy is: (0.7+0.8+0.9) / 3=0.8, and the model operation index is: 0.8×6=4.8.

[0068] As a preferred technical solution of the present invention, generating real-time monitoring data includes the following steps:

[0069] Components whose importance levels are greater than or equal to the third preset value are extracted from the importance list and set as the second equipment group; the first equipment group and the second equipment group are combined and set as the third equipment group; the coordinate distance between any two components in the third equipment group is calculated based on the component coordinates, and components whose coordinate distances are less than the fourth preset value are set as the selected equipment group.

[0070] Components with the same component type and lower operation priority in the device group to be selected are eliminated, and the real-time operation data corresponding to the remaining components in the third device group are set as real-time monitoring data.

[0071] Specifically, in this embodiment, the real-time operation data can be expanded according to the importance level to generate real-time monitoring data. A plurality of components with a larger importance level are extracted as the second device group, and are merged with the first device group to form a third device group. In the third device group, the components with a smaller coordinate distance are set as the device group to be selected, wherein the coordinate distance can be calculated by substituting the component coordinates of the two components into the distance formula.

[0072] If the component types are the same and the operating priority is lower, it means that the real-time operating data corresponding to the component is not necessary. Therefore, the real-time operating data corresponding to the component is eliminated in the digital twin model, and the real-time operating data corresponding to the remaining components in the third equipment group are set as real-time monitoring data for display in the digital twin model to generate a dynamic digital twin model.

[0073] As a preferred technical solution of the present invention, outputting the performance prediction results of the ground source heat pump equipment includes the following steps:

[0074] The environmental tags corresponding to the standard environmental data and the component tags corresponding to the real-time monitoring data are obtained respectively, and the standard environmental data and the real-time monitoring data are collected based on a preset time period and set as the first category of historical data and the second category of historical data.

[0075] Construct a neural network model and a mathematical analysis model, input the first category of historical data into the neural network model for learning and training, and output the environmental prediction results corresponding to the environmental labels within the prediction time, combine all environmental prediction results and set them as the first category of prediction results.

[0076] The mathematical analysis model extracts associated data corresponding to preset performance influencing factors from the first category of historical data and the second category of historical data based on a feature selection algorithm, wherein the performance influencing factors include groundwater variables, output heat, and input power.

[0077] Calculate the performance coefficient C of the ground source heat pump equipment at the i-th time point based on the first formula i , the first formula is: Among them, Q i is the associated data of the i-th time point corresponding to the output heat, P in The correlation data at the i-th time point corresponding to the input power is summarized based on the preset period to generate a performance sequence, and the response coefficient between the correlation data corresponding to the groundwater variable and the performance sequence is obtained.

[0078] The variable prediction results corresponding to the groundwater variables are extracted from the first type of prediction results based on the environmental labels, and the numerical results of the performance series within the prediction time are generated based on the response coefficient and the variable prediction results, and the numerical results are set as the performance prediction results.

[0079] Specifically, in this embodiment, the environment tag and the component tag refer to the type information of the corresponding data, for example, the type of the sensor corresponding to the standard environment data collected is set as the environment tag, and the component type of the component corresponding to the real-time monitoring data collected is set as the component tag. The data corresponding to all tags are collected within a preset period to generate the first type of historical data and the second type of historical data.

[0080] The neural network model refers to a recurrent neural network model of the Long Short-Term Memory (LSTM) network. LSTM is a neural network architecture that is particularly suitable for processing time series data. It can capture long-term dependencies in time series, which makes it very suitable for predicting environmental data such as groundwater flow. The environmental prediction results corresponding to the environmental labels are output after the first type of historical data is trained in the neural network model.

[0081] Mathematical analysis model refers to a comprehensive analysis model used to evaluate the operating efficiency of the system. Feature selection algorithm is a series of methods used to select the most important features (also called variables, attributes or indicators), including but not limited to cluster analysis, correlation analysis and principal component analysis. The preset performance influencing factors refer to the relevant variables used to evaluate the energy efficiency of ground source heat pump equipment, among which groundwater variables include but are not limited to groundwater flow, groundwater temperature, soil temperature and soil moisture, etc. Output heat refers to the heat or cold provided by the ground source heat pump equipment to the load side, and input power refers to the electrical power consumed by the ground source heat pump equipment during operation. Associated data refers to the data corresponding to various performance influencing factors in the first category of historical data and the second category of historical data.

[0082] The first formula can be used to calculate the performance coefficient change trend of the ground source heat pump system within a preset time period, that is, the performance sequence. Statistical methods can be used to calculate the Pearson correlation coefficient between the associated data corresponding to the groundwater variable and the performance sequence, and set it as the response coefficient.

[0083] Then, the variable prediction results and response coefficients corresponding to the groundwater variables can be used to infer the changing trend of the performance sequence within the prediction time, which is the performance prediction result.

[0084] As a preferred technical solution of the present invention, the multi-objective optimization algorithm dynamically adjusts the equipment state parameters based on the performance prediction results, including the following steps:

[0085] If the performance prediction results within the prediction time are all less than the fifth preset value, the target variables are set based on the performance prediction results, an optimization model is constructed based on the target variables and the multi-objective optimization algorithm, and the real-time monitoring data and standard environment data are input into the optimization model, the equipment status parameters are adjusted, and new real-time monitoring data is generated. Repeat this step until the performance prediction results are all greater than or equal to the fifth preset value.

[0086] Specifically, in this embodiment, if the performance prediction results are all less than the fifth preset value, it means that the energy efficiency of the ground source heat pump equipment within the prediction time is low, and it is necessary to build an optimization model to regulate the equipment state parameters of the ground source heat pump equipment. Among them, the variable corresponding to the performance prediction result is set as the target variable, for example, the output heat or input power is set as the target variable. Then, the real-time monitoring data and the standard environmental data are input into the optimization model, the optimization model identifies the key parameters that affect the target variable, and adjusts the equipment operating parameters through a multi-objective optimization algorithm to find the optimal solution, so that the performance prediction results corresponding to the target variable are all greater than or equal to the fifth preset value.

[0087] It should be understood that, although each step in the flow chart of each embodiment of the present invention is shown in sequence according to the indication of the arrow, these steps are not necessarily performed in sequence according to the order indicated by the arrow. Unless there is a clear explanation in this article, the execution of these steps does not have a strict order restriction, and these steps can be performed in other orders. Moreover, at least a portion of the steps in each embodiment may include a plurality of sub-steps or a plurality of stages, and these sub-steps or stages are not necessarily performed at the same time, but can be performed at different times, and the execution order of these sub-steps or stages is not necessarily performed in sequence, but can be performed in turn or alternately with at least a portion of other steps or sub-steps or stages of other steps.

[0088] Those skilled in the art can understand that all or part of the processes in the above-mentioned embodiment methods can be completed by instructing the relevant hardware through a computer program. The above-mentioned program can be stored in a non-volatile computer-readable storage medium. When the program is executed, it can include the processes of the embodiments of the above-mentioned methods. Among them, any reference to memory, storage, database or other media used in the embodiments provided in this application can include non-volatile and / or volatile memory. Non-volatile memory may include read-only memory (ROM), programmable ROM (PROM), electrically programmable ROM (EPROM), electrically erasable programmable ROM (EEPROM) or flash memory. Volatile memory may include random access memory (RAM) or external cache memory. As an illustration and not limitation, RAM is available in various forms, such as static RAM (SRAM), dynamic RAM (DRAM), synchronous DRAM (SDRAM), double data rate SDRAM (DDRSDRAM), enhanced SDRAM (ESDRAM), synchronous link (Synchlink) DRAM (SLDRAM), memory bus (Rambus) direct RAM (RDRAM), direct memory bus dynamic RAM (DRDRAM), and memory bus dynamic RAM (RDRAM).

[0089] The technical features of the above-mentioned embodiments can be combined arbitrarily. In order to make the description concise, not all possible combinations of the technical features in the above-mentioned embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0090] The above embodiments only express several implementation methods of the present invention, and the descriptions thereof are relatively specific and detailed, but they cannot be understood as limiting the scope of the patent of the present invention. It should be pointed out that, for those of ordinary skill in the art, several variations and improvements can be made without departing from the concept of the present invention, and these all belong to the protection scope of the present invention. Therefore, the protection scope of the patent of the present invention shall be subject to the attached claims.

[0091] The above are only preferred embodiments of the present invention and are not intended to limit the present invention. Any modifications, equivalent substitutions and improvements made within the spirit and principles of the present invention should be included in the protection scope of the present invention.

Claims

1. A dynamic digital twin modeling method for optimizing the energy efficiency of ground source heat pumps, characterized in that: The method comprises the following steps: S1: Obtaining a CAD drawing of a ground source heat pump device installed in a target area, and constructing a ground source heat pump simulation model based on the CAD drawing; S2: receiving the real-time operation data and equipment status parameters corresponding to the ground source heat pump equipment and the environmental data corresponding to the target area based on the Internet of Things technology, performing data preprocessing on the environmental data to generate standard environmental data, and driving the ground source heat pump simulation model based on the real-time operation data to generate a digital twin model; S3: Monitor the model operation index of the digital twin model based on the real-time operation data. If the model operation index is less than a first preset value, re-expand the real-time operation data to generate real-time monitoring data, and update the digital twin model based on the real-time monitoring data to generate a dynamic digital twin model. Repeat this step until the model operation index corresponding to the dynamic digital twin model is greater than or equal to the first preset value. S4: The dynamic digital twin model analyzes and predicts the real-time monitoring data based on the standard environmental data, and outputs the performance prediction result of the ground source heat pump equipment, and sets a multi-objective optimization algorithm. The multi-objective optimization algorithm dynamically adjusts the equipment state parameters based on the performance prediction result to complete the energy efficiency optimization of the ground source heat pump equipment; Wherein, the outputting of the performance prediction result of the ground source heat pump equipment comprises the following steps: Respectively obtain the environment tags corresponding to the standard environment data and the component tags corresponding to the real-time monitoring data, collect the standard environment data and the real-time monitoring data based on a preset time period, and set them as the first type of historical data and the second type of historical data; Constructing a neural network model and a mathematical analysis model, inputting the first category of historical data into the neural network model for learning and training, and outputting the environmental prediction results corresponding to the environmental labels within the prediction time, combining all environmental prediction results and setting them as the first category of prediction results; The mathematical analysis model extracts associated data corresponding to preset performance influencing factors from the first category of historical data and the second category of historical data based on a feature selection algorithm, wherein the performance influencing factors include groundwater variables, output heat, and input power; Calculate the performance coefficient of the ground source heat pump device at the i-th time point based on the first formula , the first formula is: ,in, is the associated data of the i-th time point corresponding to the output heat, For the correlation data of the i-th time point corresponding to the input power, all performance coefficients are summarized based on the preset time period to generate a performance sequence, and a response coefficient between the correlation data corresponding to the groundwater variable and the performance sequence is obtained; Based on the environmental tag, the variable prediction result corresponding to the groundwater variable is extracted from the first type of prediction results, and based on the response coefficient and the variable prediction result, the numerical result of the performance sequence within the prediction time is generated, and the numerical result is set as the performance prediction result.

2. The method according to claim 1, characterized in that The construction of the ground source heat pump simulation model based on the CAD drawing comprises the following steps: Building a component information library based on the CAD drawing, the component information library includes component types, component coordinates and component shapes of all components, respectively obtaining the operation association between any two components in the component information library and the operation priority of any component based on the component type, setting any of the components as nodes, generating a connection line between any two nodes based on the operation association and the operation priority, and sequentially combining all nodes and the connection lines to generate a node association graph; All components are fused and simulated using a simulator based on the node association diagram and the CAD drawing to generate the ground source heat pump simulation model.

3. The method according to claim 2, characterized in that The receiving of the real-time operation data and equipment status parameters corresponding to the ground source heat pump equipment and the environmental data corresponding to the target area based on the Internet of Things technology comprises the following steps: Based on the node association graph, a plurality of sensors are respectively arranged in the ground source heat pump device and the target area, a data storage unit and a device adjustment unit are arranged for each sensor, a collection method of the sensor is obtained, and the sensor data is stored in a corresponding storage unit based on the collection method; A plurality of sensors are randomly selected and set as a sensor group, and sensing data corresponding to the sensor group is obtained. If any sensing data changes, a data request is sent to the storage unit corresponding to the sensor group; After receiving the data request, the storage unit divides the sensor data corresponding to the sensor group into multiple data groups based on time characteristics, and transmits all the data groups to the ground source heat pump simulation model in sequence, wherein the data group belonging to the target area is set as the environmental data, and the data group belonging to the ground source heat pump equipment is set as the real-time operation data, and the sensor corresponding to any of the components is obtained based on the node association diagram and set as the component sensor, and the information contained in the equipment adjustment unit of the component sensor is set as the equipment status parameter.

4. The method according to claim 2, characterized in that: The generation of the digital twin model includes the following steps: The real-time operation data is integrated and transmitted into the geothermal heat pump simulation model, and the geothermal heat pump simulation model is calibrated using the real-time operation data based on a preset time interval until the confidence level of the geothermal heat pump simulation model is greater than or equal to a second preset value, and the calibrated geothermal heat pump simulation model is set as the digital twin model.

5. The method according to claim 4, characterized in that The step of monitoring the model operation index of the digital twin model based on the real-time operation data comprises the following steps: All components in the component information library are marked with importance levels based on the corresponding component types, an importance list is generated, the component types corresponding to all the real-time operation data are obtained and set as a first equipment group, the importance levels corresponding to the first equipment group are accumulated based on the importance list to generate a first value, the accuracy of the real-time operation data is checked, and the mean accuracy of all the real-time operation data is calculated based on the accuracy, and the product of the mean accuracy and the first value is set as the model operation index.

6. The method according to claim 5, characterized in that Generating real-time monitoring data comprises the following steps: Extracting components whose importance levels are greater than or equal to a third preset value from the importance list and setting them as a second device group, combining the first device group and the second device group and setting them as a third device group, calculating the coordinate distance between any two components in the third device group based on the component coordinates, and setting the components whose coordinate distance is less than a fourth preset value as the device group to be selected; Components of the same component type and with lower operation priority in the to-be-selected equipment group are eliminated, and real-time operation data corresponding to the remaining components in the third equipment group are set as the real-time monitoring data.

7. The method according to claim 1, characterized in that The multi-objective optimization algorithm dynamically adjusts the device state parameters based on the performance prediction result, comprising the following steps: If the performance prediction results are all less than the fifth preset value within the prediction time, the target variables are set based on the performance prediction results, an optimization model is constructed based on the target variables and the multi-objective optimization algorithm, and the real-time monitoring data and the standard environment data are input into the optimization model, the equipment status parameters are adjusted, and new real-time monitoring data are generated. This step is repeated until the performance prediction results are all greater than or equal to the fifth preset value.

Images