Underground aquifer energy storage system optimization method and device based on proxy model

Abstract

The invention discloses an underground aquifer energy storage system optimization method and device based on an agent model, and relates to the technical field of artificial intelligence, and the method comprises the steps: determining an optimization variable and constraint information according to a preset optimization target; based on the optimization variable and the constraint information, constructing a high-fidelity heat-fluid-solid multi-field coupling numerical model; inputting the operation parameters into the high-fidelity heat-fluid-solid multi-field coupling numerical model to obtain prediction response parameters corresponding to the operation parameters; constructing an agent model based on the operation parameters and the prediction response parameters; and performing target optimization on the agent model to obtain optimized operation parameters, so that the underground aquifer energy storage system operates based on the optimized operation parameters. On the basis of considering the influence of the underground aquifer energy storage system on the geological environment in the operation process, the operation scheme closest to the optimization target can be obtained through optimization, and the optimization effect of the underground aquifer energy storage system is improved.

Description

Technical Field

[0001] The present application relates to the field of artificial intelligence technology, and in particular to an optimization method and device for an underground aquifer energy storage system based on an agent model. Background Art

[0002] Aquifer Thermal Energy Storage (ATES) technology is one of the important ways to develop and utilize shallow geothermal energy. During the operation of the aquifer thermal storage system, heat is extracted from groundwater through a storage heat pump, and the tailwater is then injected into the ground. However, seasonal extraction and reinjection of groundwater will disturb the original pressure field, which may cause problems such as ground subsidence, excessive drop in groundwater level, heat penetration, and system efficiency degradation.

[0003] In recent years, with the rapid development of artificial intelligence, there have been solutions for optimizing underground aquifer energy storage systems with the assistance of agent models, which effectively establish the relationship between input and output. However, this method mainly considers the heat recovery capacity during the optimization process, but ignores the geological environment safety, especially the ground subsidence problem in soft soil areas. It can be seen that the existing optimization methods are not ideal for optimizing underground aquifer energy storage systems. Summary of the invention

[0004] The purpose of this application is to provide an underground aquifer energy storage system optimization method and device based on an agent model, which can improve the optimization effect of the underground aquifer energy storage system.

[0005] To achieve the above objectives, this application provides the following solutions:

[0006] In a first aspect, the present application provides an optimization method for an underground aquifer energy storage system based on an agent model, comprising:

[0007] Determine the optimization variables and constraint information according to the preset optimization target; wherein the optimization variables include at least operation parameters and response parameters, the operation parameters include at least single well injection rate, single well extraction rate, cold well injection temperature and hot well injection temperature, the response parameters include at least maximum ground subsidence, maximum water level drop, influence radius and average heat recovery rate; the constraint information is used to constrain the response parameters;

[0008] Based on the optimization variables and the constraint information, a high-fidelity thermal-fluid-solid multi-field coupling numerical model is constructed;

[0009] The operating parameters are input into the high-fidelity thermal-fluid-solid multi-field coupling numerical model to obtain the predicted response parameters corresponding to the operating parameters; wherein the predicted response parameters at least include the predicted maximum ground subsidence, the predicted maximum water level drop, the predicted influence radius, and the predicted average heat recovery rate;

[0010] Based on the operating parameters and the predicted response parameters, a proxy model is constructed; wherein the proxy model is used to represent the functional relationship between the operating parameters and the predicted response parameters;

[0011] The proxy model is optimized by target search to obtain optimized operating parameters so that the underground aquifer energy storage system can operate based on the optimized operating parameters.

[0012] Optionally, the constraint information includes at least a critical value of the maximum ground subsidence, a critical value of the maximum water level drop, and a critical value of the impact radius.

[0013] Optionally, constructing a high-fidelity thermal-fluid-solid multi-field coupling numerical model based on the optimization variables and the constraint information specifically includes:

[0014] Obtaining a real response parameter corresponding to the operating parameter;

[0015] Inputting the operating parameters into a pre-built initial numerical model to obtain training response parameters;

[0016] Calculating the true response parameter and the training response parameter based on the loss function to obtain a loss value;

[0017] If the loss value is greater than a preset loss threshold, the initial numerical model is trained, and the steps from obtaining the real response parameter corresponding to the operating parameter to calculating the real response parameter and the training response parameter based on the loss function to obtain the loss value are performed;

[0018] If the loss value is less than or equal to the preset loss threshold, the initial numerical model is determined as a trained high-fidelity thermal-fluid-solid multi-field coupling numerical model.

[0019] Optionally, the proxy model is a multiple regression proxy model, and the expression of the multiple regression proxy model is:

[0020]

[0021] in, is the predicted response parameter, is the operating parameter, S is the maximum ground subsidence, Δh is the maximum water level drop, R is the impact radius, η is the average heat recovery rate, S 1is the critical value of the maximum land subsidence, Δh 1 is the critical value of the maximum water level drawdown, R 1 is the critical value of the influence radius, q is the single well injection rate or the single well extraction rate, T c-in is the cold well injection temperature and T h-in is the water injection temperature of the hot well.

[0022] Optionally, the performing target optimization on the proxy model to obtain optimized operating parameters specifically includes:

[0023] Optimizing the agent model to obtain multiple optimization operation indicators;

[0024] Non-dimensionalizing the plurality of optimization operation indicators based on the desirability function to obtain a dimensionless value corresponding to each of the optimization operation indicators;

[0025] Based on the optimization algorithm and the dimensionless value corresponding to each of the optimization operation indicators, multiple optimization operation indicators are optimized to obtain optimization operation parameters.

[0026] Optionally, the step of non-dimensionalizing a current optimization operation index based on the desirability function to obtain a dimensionless value corresponding to the current optimization operation index specifically includes:

[0027] Obtaining maximum and minimum values ​​of the predicted response variables corresponding to the plurality of optimization operation indicators;

[0028] Obtaining a target value of a predicted response variable corresponding to the current optimization operation index;

[0029] Based on the desirability function, the maximum value, the minimum value and the target value are dimensionless to obtain a dimensionless value corresponding to the current optimization operation index;

[0030] And, the calculation formula of the dimensionless value corresponding to the current optimization operation index is:

[0031]

[0032] Among them, d i is the dimensionless value corresponding to the current optimization operation index, is the minimum value, is the maximum value, is the target value, w i The weight of the current optimization running indicator.

[0033] In a second aspect, the present application provides an underground aquifer energy storage system optimization device based on an agent model, comprising:

[0034] A determination unit, configured to determine optimization variables and constraint information according to a preset optimization objective; wherein the optimization variables include at least operating parameters and response parameters, the operating parameters include at least single well injection rate, single well extraction rate, cold well injection temperature and hot well injection temperature, and the response parameters include at least maximum ground subsidence, maximum water level drop, influence radius and average heat recovery rate; the constraint information is used to constrain the response parameters;

[0035] A first construction unit is used to construct a high-fidelity thermal-fluid-solid multi-field coupling numerical model based on the optimization variables and the constraint information;

[0036] An input unit, used to input the operating parameters into the high-fidelity thermal-fluid-solid multi-field coupling numerical model to obtain prediction response parameters corresponding to the operating parameters; wherein the prediction response parameters at least include predicted maximum ground subsidence, predicted maximum water level drop, predicted influence radius, and predicted average heat recovery rate;

[0037] A second construction unit is used to construct a proxy model based on the operating parameters and the predicted response parameters; wherein the proxy model is used to represent the functional relationship between the operating parameters and the predicted response parameters;

[0038] The optimization unit is used to perform target optimization on the proxy model to obtain optimized operating parameters so that the underground aquifer energy storage system can operate based on the optimized operating parameters.

[0039] In a third aspect, the present application provides a computer device comprising: a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the computer program to implement the steps of any of the above-mentioned methods for optimizing an underground aquifer energy storage system based on an agent model.

[0040] In a fourth aspect, the present application provides a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the steps of any of the above-mentioned methods for optimizing an underground aquifer energy storage system based on an agent model.

[0041] In a fifth aspect, the present application provides a computer program product, including a computer program, which, when executed by a processor, implements the steps of any of the above-mentioned agent-based model-based underground aquifer energy storage system optimization methods.

[0042] According to the specific embodiments provided in this application, this application discloses the following technical effects:

[0043] The present application provides an optimization method and device for an underground aquifer energy storage system based on a proxy model, which determines optimization variables and constraint information based on preset optimization objectives, and constructs a high-fidelity thermal-fluid-solid multi-field coupled numerical model based on the optimization variables and constraint information, so that the constructed high-fidelity thermal-fluid-solid multi-field coupled numerical model takes into account the impact of the underground aquifer energy storage system on the geological environment during operation; and constructs a proxy model based on operating parameters and predicted response parameters, and performs target optimization on the proxy model to obtain the best operating plan for the underground aquifer energy storage system. It can be seen that by implementing this optimization method, while considering the impact of the underground aquifer energy storage system on the geological environment during operation, it is possible to optimize the optimized operating parameters closest to the optimization objectives, thereby improving the optimization effect of the underground aquifer energy storage system. BRIEF DESCRIPTION OF THE DRAWINGS

[0044] In order to more clearly illustrate the embodiments of the present application or the technical solutions in the prior art, the drawings required for use in the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0045] Figure 1 A schematic flow chart of an underground aquifer energy storage system optimization method based on an agent model in an embodiment of the present application;

[0046] Figure 2 A schematic diagram of water level drop and impact radius provided in an embodiment of the present application;

[0047] Figure 3 A schematic diagram of the distribution of exemplary locations and monitoring facilities provided in an embodiment of the present application;

[0048] Figure 4 A schematic diagram of a high-fidelity thermal-fluid-solid multi-field coupling numerical model provided in one embodiment of the present application;

[0049] Figure 5 A schematic diagram of initial and boundary conditions of a high-fidelity thermal-fluid-solid multi-field coupling numerical model provided in one embodiment of the present application;

[0050] Figure 6 A schematic diagram of functional modules of an underground aquifer energy storage system optimization device based on a proxy model provided in one embodiment of the present application;

[0051] Figure 7 A schematic diagram of the structure of a computer device provided in one embodiment of the present application. DETAILED DESCRIPTION

[0052] The following will be combined with the drawings in the embodiments of the present application to clearly and completely describe the technical solutions in the embodiments of the present application. Obviously, the described embodiments are only part of the embodiments of the present application, not all of the embodiments. Based on the embodiments in the present application, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of this application.

[0053] In order to make the above-mentioned purposes, features and advantages of the present application more obvious and easy to understand, the present application is further described in detail below with reference to the accompanying drawings and specific implementation methods.

[0054] In an exemplary embodiment, Figure 1 As shown, a method for optimizing an underground aquifer energy storage system based on an agent model is provided. The method is executed by a computer device, specifically, it can be executed by a computer device such as a terminal or a server alone, or it can be executed by a terminal and a server together. The method includes the following steps 101 to 105. Among them:

[0055] Step 101, determining optimization variables and constraint information according to a preset optimization goal.

[0056] In the embodiment of the present application, the preset optimization goal can be to find an optimal combination of operating parameters of the underground aquifer energy storage system so that the maximum ground subsidence, the maximum water level drop and the impact radius are within the critical value range, which can maximize the average heat recovery rate of the system.

[0057] The optimization variables include at least operating parameters and response parameters, and the operating parameters include at least single well injection rate, single well extraction rate, cold well water injection temperature, and hot well water injection temperature. Multiple groups of operating parameters can be preset.

[0058] The response parameters include at least the maximum ground subsidence, the maximum water level drop, the impact radius and the average heat recovery rate; the constraint information is used to constrain the response parameters. The constraint information includes at least the critical value of the maximum ground subsidence, the critical value of the maximum water level drop and the critical value of the impact radius.

[0059] In the embodiment of the present application, the underground aquifer energy storage system (ATES) specifically utilizes the good thermal insulation properties of the underground pressurized aquifer and uses the pipe well recharge method to inject heat into the underground aquifer for storage in summer and extract it in winter for indoor heating of the building, that is, inject in summer and use in winter; inject cold energy into the aquifer for storage in winter and extract it in summer for cooling, that is, inject in winter and use in summer.

[0060] Cold well: also known as a winter-irrigation and summer-use well, in which cold energy is pumped into the aquifer for storage in winter and extracted in summer for indoor cooling of buildings.

[0061] Thermal well: also known as a summer-irrigation and winter-use well, heat is pumped into underground aquifers for storage in the summer and extracted in the winter to provide indoor heating for buildings.

[0062] Heat recovery rate: the ratio of the total amount of heat energy extracted by the system to the total amount of heat energy stored. It is an important indicator to characterize the heat recovery capacity.

[0063]

[0064] In the above formula: η is the average heat recovery rate, η c and η h are the heat recovery rates of the cold and hot wells, Q s is the total amount of heat energy injected into the ground, Q se and Q we are the heat energy extracted from the cold and hot wells, respectively, c w is the specific heat capacity of water, q is the flow rate, T 0 is the initial temperature of the aquifer, T i is the average water injection temperature, T c-out and T h-out are the average pumping temperatures of cold and hot wells, respectively, t b is the heat penetration time, and t is the current accumulated time.

[0065] In the above formula, in summer conditions, the average pumping temperature of the cold well is T c-out Lower than the initial temperature of the aquifer T 0 , T i represents the temperature of hot water recharged in summer. At this time, the heat recovery rate of the cold well can be calculated (η c ). Similarly, in winter conditions, the average pumping temperature of the hot well is T h-out Higher than the initial temperature of the aquifer T 0 , T i represents the temperature of the cold water refilled in winter. At this time, the heat recovery rate of the hot well (η h ). It should be noted that groundwater with a higher or lower recharge temperature than the initial temperature of the aquifer will cause a high or low temperature zone to form in the aquifer near the recharge well. Then, during the pumping and recharge process, the heat of the high or low temperature zone is continuously transferred to the pumping well along the groundwater seepage direction, resulting in T c-out Greater than T 0 or T h-out Lower than T 0 , that is, heat penetration occurs, at this time t is greater than t b Obviously, the occurrence of heat penetration reduces the heat recovery rate.

[0066] Ground subsidence: A local downward movement (or engineering geological phenomenon) in which the surface elevation of the earth's crust is lowered due to the consolidation and compression of loose underground strata under the influence of human engineering and economic activities.

[0067] Please also read Figure 2 , Figure 2 A schematic diagram of a water level drop and an impact radius provided in an embodiment of the present application;

[0068] Drawdown: The amount by which the groundwater level decreases during pumping compared to the initial level.

[0069] Impact radius: the straight-line distance between the water level drop threshold and the pumping well.

[0070] Step 102: construct a high-fidelity thermal-fluid-solid multi-field coupling numerical model based on the optimization variables and the constraint information.

[0071] In the embodiment of the present application, in the process of constructing a high-fidelity thermal-fluid-solid multi-field coupled numerical model, reference can also be made to real geological environment parameters, such as the composition of soil in the real geological environment, the pumping temperature of the energy storage well, the ground temperature, the groundwater level and other parameters.

[0072] As an optional implementation, the method of constructing a high-fidelity thermal-fluid-solid multi-field coupling numerical model based on the optimization variables and the constraint information in step 102 may specifically include:

[0073] Obtaining a real response parameter corresponding to the operating parameter;

[0074] Inputting the operating parameters into a pre-built initial numerical model to obtain training response parameters;

[0075] Calculating the true response parameter and the training response parameter based on the loss function to obtain a loss value;

[0076] If the loss value is greater than a preset loss threshold, the initial numerical model is trained, and the steps from obtaining the real response parameter corresponding to the operating parameter to calculating the real response parameter and the training response parameter based on the loss function to obtain the loss value are performed;

[0077] If the loss value is less than or equal to the preset loss threshold, the initial numerical model is determined as a trained high-fidelity thermal-fluid-solid multi-field coupling numerical model.

[0078] Among them, by implementing this implementation method, the constructed initial numerical model can be trained based on the real response parameters, thereby obtaining a high-fidelity thermal-fluid-solid multi-field coupled numerical model in which the loss between the output predicted influencing parameters and the real required parameters is less than or equal to the preset loss threshold, thereby improving the accuracy of the predicted response parameters.

[0079] Step 103: input the operating parameters into the high-fidelity thermal-fluid-solid multi-field coupling numerical model to obtain predicted response parameters corresponding to the operating parameters.

[0080] In an embodiment of the present application, the predicted response parameters include at least the predicted maximum ground subsidence, the predicted maximum water level drop, the predicted impact radius, and the predicted average heat recovery rate.

[0081] Since multiple groups of operating parameters can be preset, the predicted response parameters corresponding to each group of operating parameters can be obtained, and the number of groups can be set according to the experimental design method.

[0082] Step 104: constructing a proxy model based on the operating parameters and the predicted response parameters.

[0083] In the embodiment of the present application, the proxy model is used to represent the functional relationship between the operating parameter and the predicted response parameter. In the optimization design, the proxy model can replace the more complex and time-consuming numerical model.

[0084] In the embodiment of the present application, the proxy model may be a multiple regression proxy model. In this case, the expression of the multiple regression proxy model is:

[0085]

[0086] in, is the predicted response parameter, is the operating parameter, S is the maximum ground subsidence, Δh is the maximum water level drop, R is the impact radius, η is the average heat recovery rate, S 1 is the critical value of the maximum land subsidence, Δh 1 is the critical value of the maximum water level drawdown, R 1 is the critical value of the influence radius, q is the single well injection rate or the single well extraction rate, T c-in is the cold well injection temperature and T h-in is the water injection temperature of the hot well.

[0087] It can be seen that by implementing this implementation method, the relationship between the operating parameters and the predicted response parameters is expressed in the form of a multivariate regression proxy model, and the influence of the operating parameters on the average heat recovery rate can be more accurately determined by controlling the maximum ground subsidence, the maximum water level drop and the influence radius in the predicted response parameters.

[0088] In addition, the proxy model can have a second-order polynomial function form. In this case, the general form of the second-order polynomial function of the proxy model can be expressed as:

[0089]

[0090] Where β represents the model parameters of the proxy model (which can be obtained by data analysis software), and p is the number of operating parameters input into the model. The reliability of the proxy model can be measured by the correlation index R2 To estimate, R 2 The value of is between 0 and 1. The closer it is to 1, the higher the fitting accuracy of the surrogate model:

[0091]

[0092] Among them, y i is the true value of the ith response variable, is the predicted value of the ith response variable, is the mean of the true values ​​of the response variable, and k is the number of response variables.

[0093] In addition, when R 2 When it is less than the expected value of the fitting accuracy of the proxy model, first consider increasing the number of preset groups in step 103, and then consider adjusting the functional form of the proxy model.

[0094] Step 105, performing target optimization on the proxy model to obtain optimized operating parameters, so that the underground aquifer energy storage system operates based on the optimized operating parameters.

[0095] As an optional implementation, step 105 performs target optimization on the proxy model to obtain optimized operating parameters, which may specifically include:

[0096] Optimizing the agent model to obtain multiple optimization operation indicators;

[0097] Non-dimensionalizing the plurality of optimization operation indicators based on the desirability function to obtain a dimensionless value corresponding to each of the optimization operation indicators;

[0098] Based on the optimization algorithm and the dimensionless value corresponding to each of the optimization operation indicators, multiple optimization operation indicators are optimized to obtain optimization operation parameters.

[0099] Among them, by implementing this implementation method, on the basis of optimizing multiple operating indicators corresponding to the agent model, each optimized operating indicator can be dimensionless through a consensus function to obtain a dimensionless value of each optimized operating indicator, and then a more accurate optimization result can be obtained based on the dimensionless value of each optimized operating indicator.

[0100] In an embodiment of the present application, multi-objective optimization (MOP) technology can be used to perform target optimization on the proxy model, that is, there are multiple objective functions in a problem, that is, the multiple optimization operation indicators, and it is necessary to weigh between these objective functions to find an optimal solution.

[0101] In the embodiment of the present application, the proxy model is combined with the multi-objective optimization (MOP) technology, which can be described as:

[0102]

[0103] Where x is the vector of input operation parameters, Ω is the parameter space of x, and n is the number of optimization operation indicators.

[0104] In the embodiment of the present application, a current optimization operation index is dimensionless based on the desirability function, and a method for obtaining a dimensionless value corresponding to the current optimization operation index may specifically include:

[0105] Obtaining maximum and minimum values ​​of the predicted response variables corresponding to the plurality of optimization operation indicators;

[0106] Obtaining a target value of a predicted response variable corresponding to the current optimization operation index;

[0107] Based on the desirability function, the maximum value, the minimum value and the target value are dimensionless to obtain a dimensionless value corresponding to the current optimization operation index;

[0108] And, the calculation formula of the dimensionless value corresponding to the current optimization operation index is:

[0109]

[0110] Among them, d i is the dimensionless value corresponding to the current optimization operation index, is the minimum value, is the maximum value, is the target value, w i The weight of the current optimization running indicator.

[0111] It can be seen that by implementing this implementation method, a more accurate dimensionless value corresponding to the current optimization operation index can be obtained through the above calculation formula.

[0112] Optionally, if there is more than one optimal operating index, each for a consensus function, then these individual consensus functions will be integrated into a composite expected value CD, which is defined as the weighted geometric mean of all individual consensus functions, calculated as follows:

[0113]

[0114] In the embodiments of the present application, the optimization algorithm may be an imperialist competitive algorithm (ICA), a genetic algorithm (GA) or a particle swarm optimization algorithm (PSO), etc., and the embodiments of the present application do not limit this.

[0115] For example, the seasonal underground aquifer energy storage project in Chongming Island, Shanghai is used as an example to further describe this application. Figure 3 , Figure 4 as well as Figure 5 , Figure 3 A schematic diagram of the distribution of exemplary locations and monitoring facilities provided in an embodiment of the present application; Figure 4 A schematic diagram of a high-fidelity thermal-fluid-solid multi-field coupling numerical model provided in one embodiment of the present application; Figure 5 A schematic diagram of the initial and boundary conditions of a high-fidelity thermal-fluid-solid multi-field coupling numerical model provided in one embodiment of the present application.

[0116] S1. Determine the basic profile of the object instance, including energy storage wells, optimization objectives and constraints. In this embodiment, the underground aquifer energy storage system mainly consists of two pairs of wells, namely, two cold wells (C1, C2) and two hot wells (H1, H2). The well depth is 104.5m. Various monitoring facilities are also set up in the site. The plane distribution is shown in Figure 3 The optimization goal is to maximize the average heat recovery rate and meet the geological environment response constraints, where the constraints are set as maximum ground settlement ≤5mm / a, maximum water level drop ≤2m, and impact radius ≤100m (the drop threshold is 0.2m).

[0117] S2. First, the TOUGH-FLAC3D platform was used to build and simulate the initial thermal-fluid-solid multi-field coupling numerical model. The size of the numerical model was 1000m(X)×1000m(Y)×110m(Z). The meshing, initial and boundary conditions of the numerical model were shown in Figure 4 and Figure 5 The initial numerical model constructed was verified based on the pumping temperature, ground temperature, groundwater level and settlement monitoring data of the energy storage well, and then a high-fidelity thermal-fluid-solid multi-field coupling numerical model was obtained.

[0118] S3, combining the high-fidelity thermal-fluid-solid multi-field coupling numerical model of S2 with the operating parameters to generate a sample data set (the sample data set contains multiple groups of operating parameters and the predicted response parameters corresponding to each group of operating parameters). The experimental design adopts the "Box-Behnken" design method (BBD), in which each operating parameter uses 3 different level values ​​(minimum value, middle value, maximum value) to reduce the number of simulations. The values ​​of each operating parameter in this embodiment are shown in Table 1.

[0119] Table 1 Value ranges of operating parameters

[0120]

[0121] In this example, two pairs of wells are pumped and recharged seasonally alternately, that is, cold water is extracted from the cold wells in summer for building cooling (153 days), and hot water is extracted from the hot wells in winter for building heating (212 days). The simulation time is set to 5 years, and a sample data set based on BBD is obtained (Table 2).

[0122] Table 2 Sample dataset

[0123]

[0124]

[0125] Where t>t b It represents that heat penetration occurs.

[0126] Based on the above 13 sample data, the proxy model containing 4 response parameters is obtained through multiple regression as follows:

[0127]

[0128] From the above formula, we can know that R 2 =0.93~1, indicating that the proxy model can better predict the relationship between operating parameters and response parameters.

[0129] S4. The agent model is then combined with the imperialist competition (ICA) technique to search for optimal operating parameters. Since the optimization goal of this case is to maximize the η value while ensuring that S, Δh, and R do not exceed their critical values. Therefore, when one of the response functions exceeds its critical value, the underground aquifer energy storage system stops operating. At this time, we set a very large value (1×10 5 ), so the lower the CD value, the closer it is to the expected value (maximizing η). In addition, since this case involves an optimization goal, that is, an optimization operation indicator, the weight of a single consensus function (w i ) can be ignored. The optimization results show that when the injection or extraction rate of a single well is 29.68m 3 / h, when the injection temperatures of hot and cold wells are 33.42℃ and 15℃ respectively, the maximum average heat recovery rate is 0.888, which meets the constraints of ground subsidence, water level drop and influence radius. The actual average heat recovery rate of this case is 0.465~0.715, and it is 0.888 after optimization, which shows that the optimization process proposed by the present invention significantly improves the heat recovery capacity of the system. For each optimization task, the average time consumption is about 2s, and the calculation efficiency is significantly faster than the original numerical simulation (several hours / day). It can be seen that the use of proxy models to replace the original numerical simulation for optimization can greatly reduce the amount of calculation and improve the optimization efficiency. Finally, the optimized operating parameters are input into the numerical model. The results show that the error between the optimized response parameters obtained by the proxy model and the numerical simulation results is 0~3.33%, which further proves the reliability of the proxy model. The case results show that the constructed proxy model has high reliability. Optimization based on the proxy model can improve the heat recovery capacity of the system, ensure the safety of the geological environment, and significantly accelerate the optimization process.

[0130] As an embodiment, the applicability of the present invention is not limited to ATES, but is also expected to solve other complex engineering problems, such as geothermal systems with a single heating or cooling function. At this time, it is necessary to consider the optimization objectives, optimization variables and corresponding constraints of the specific problem, which determine the functional form of the proxy model.

[0131] By implementing the above steps 101 to 105, on the basis of considering the impact of the underground aquifer energy storage system on the geological environment during operation, it is possible to find the optimal operating parameters that are closest to the optimization target and improve the optimization effect of the underground aquifer energy storage system. In addition, since the proxy model in this application replaces the original numerical model, the optimization time can be greatly shortened and the optimization efficiency can be improved. In addition, based on the proxy model, this application can also quickly and accurately determine the impact of operating parameters on ground subsidence, groundwater level drop, impact radius and average heat recovery rate.

[0132] The present application also provides an application scenario, which applies the above-mentioned underground aquifer energy storage system optimization method based on a proxy model. Specifically: The underground aquifer energy storage system optimization method based on a proxy model provided in this embodiment can be applied to real geological scenarios. Based on the preset optimization objectives and the real response parameters in the real geological scenario, a high-fidelity thermal-fluid-solid multi-field coupling numerical model is constructed, and a proxy model is constructed based on the operating parameters and the predicted response parameters output by the high-fidelity thermal-fluid-solid multi-field coupling numerical model, so as to obtain the optimized operating parameters of the underground aquifer energy storage system that can achieve the optimization objectives in the real geological scenario based on the proxy model, that is, the optimal operating parameter combination.

[0133] Based on the same inventive concept, the embodiment of the present application also provides an underground aquifer energy storage system optimization device based on a proxy model for implementing the underground aquifer energy storage system optimization method based on a proxy model involved above. The implementation scheme for solving the problem provided by the device is similar to the implementation scheme recorded in the above method, so the specific limitations in one or more embodiments of the underground aquifer energy storage system optimization device based on a proxy model provided below can refer to the limitations of the underground aquifer energy storage system optimization method based on a proxy model above, and will not be repeated here.

[0134] In an exemplary embodiment, Figure 6 As shown, an underground aquifer energy storage system optimization device based on an agent model is provided, comprising:

[0135] The determination unit 601 is used to determine the optimization variables and constraint information according to the preset optimization objectives; wherein the optimization variables include at least operating parameters and response parameters, the operating parameters include at least single well injection rate, single well extraction rate, cold well injection temperature and hot well injection temperature, and the response parameters include at least maximum ground subsidence, maximum water level drop, influence radius and average heat recovery rate; the constraint information is used to constrain the response parameters; the constraint information includes at least the critical value of the maximum ground subsidence, the critical value of the maximum water level drop and the critical value of the influence radius.

[0136] A first construction unit 602 is used to construct a high-fidelity thermal-fluid-solid multi-field coupling numerical model based on the optimization variables and the constraint information;

[0137] An input unit 603 is used to input the operating parameters into the high-fidelity thermal-fluid-solid multi-field coupling numerical model to obtain prediction response parameters corresponding to the operating parameters; wherein the prediction response parameters at least include predicted maximum ground subsidence, predicted maximum water level drop, predicted influence radius, and predicted average heat recovery rate;

[0138] A second construction unit 604 is used to construct a proxy model based on the operating parameters and the predicted response parameters; wherein the proxy model is used to represent the functional relationship between the operating parameters and the predicted response parameters;

[0139] The optimization unit 605 is used to perform target optimization on the proxy model to obtain optimized operating parameters so that the underground aquifer energy storage system can operate based on the optimized operating parameters.

[0140] As an optional implementation manner, the first construction unit 602 constructs a high-fidelity thermal-fluid-solid multi-field coupling numerical model based on the optimization variables and the constraint information in the following manner:

[0141] Obtaining a real response parameter corresponding to the operating parameter;

[0142] Inputting the operating parameters into a pre-built initial numerical model to obtain training response parameters;

[0143] Calculating the true response parameter and the training response parameter based on the loss function to obtain a loss value;

[0144] If the loss value is greater than a preset loss threshold, the initial numerical model is trained, and the steps from obtaining the real response parameter corresponding to the operating parameter to calculating the real response parameter and the training response parameter based on the loss function to obtain the loss value are performed;

[0145] If the loss value is less than or equal to the preset loss threshold, the initial numerical model is determined as a trained high-fidelity thermal-fluid-solid multi-field coupling numerical model.

[0146] Among them, by implementing this implementation method, the constructed initial numerical model can be trained based on the real response parameters, thereby obtaining a high-fidelity thermal-fluid-solid multi-field coupled numerical model in which the loss between the output predicted influencing parameters and the real required parameters is less than or equal to the preset loss threshold, thereby improving the accuracy of the predicted response parameters.

[0147] As an optional implementation, the proxy model is a multiple regression proxy model, and the expression of the multiple regression proxy model is:

[0148]

[0149] in, is the predicted response parameter, is the operating parameter, S is the maximum ground subsidence, Δh is the maximum water level drop, R is the impact radius, η is the average heat recovery rate, S 1 is the critical value of the maximum land subsidence, Δh 1 is the critical value of the maximum water level drawdown, R 1 is the critical value of the influence radius, q is the single well injection rate or the single well extraction rate, T c-in is the cold well injection temperature and T h-in is the water injection temperature of the hot well.

[0150] Among them, by implementing this implementation method, the relationship between the operating parameters and the predicted response parameters is expressed in the form of a multivariate regression proxy model, and the influence of the operating parameters on the average heat recovery rate can be more accurately determined by controlling the maximum ground subsidence, the maximum water level drop and the influence radius in the predicted response parameters.

[0151] As an optional implementation, the optimization unit 605 performs target optimization on the proxy model to obtain the optimized operating parameters in the following manner:

[0152] Optimizing the agent model to obtain multiple optimization operation indicators;

[0153] Non-dimensionalizing the plurality of optimization operation indicators based on the desirability function to obtain a dimensionless value corresponding to each of the optimization operation indicators;

[0154] Based on the optimization algorithm and the dimensionless value corresponding to each of the optimization operation indicators, multiple optimization operation indicators are optimized to obtain optimization operation parameters.

[0155] Among them, by implementing this implementation method, on the basis of optimizing multiple operating indicators corresponding to the agent model, each optimized operating indicator can be dimensionless through a consensus function to obtain a dimensionless value of each optimized operating indicator, and then a more accurate optimization result can be obtained based on the dimensionless value of each optimized operating indicator.

[0156] As an optional implementation, the optimization unit 605 performs dimensionless transformation on a current optimization operation index based on the desirability function, and obtains the dimensionless value corresponding to the current optimization operation index in the following manner:

[0157] Obtaining maximum and minimum values ​​of the predicted response variables corresponding to the plurality of optimization operation indicators;

[0158] Obtaining a target value of a predicted response variable corresponding to the current optimization operation index;

[0159] Based on the desirability function, the maximum value, the minimum value and the target value are dimensionless to obtain a dimensionless value corresponding to the current optimization operation index;

[0160] And, the calculation formula of the dimensionless value corresponding to the current optimization operation index is:

[0161]

[0162] Among them, d i is the dimensionless value corresponding to the current optimization operation index, is the minimum value, is the maximum value, is the target value, w i The weight of the current optimization running indicator.

[0163] Among them, by implementing this implementation method, a more accurate dimensionless value corresponding to the current optimization operation index can be obtained through the above calculation formula.

[0164] By implementing the above-mentioned implementation mode, on the basis of considering the impact of the underground aquifer energy storage system on the geological environment during operation, it is possible to find the optimal operating parameters that are closest to the optimization target, thereby improving the optimization effect of the underground aquifer energy storage system. In addition, since the proxy model in this application replaces the original numerical model, the optimization time can be greatly shortened and the optimization efficiency can be improved. In addition, based on the proxy model, this application can also quickly and accurately determine the impact of operating parameters on ground subsidence, groundwater level drop, impact radius and average heat recovery rate.

[0165] In an exemplary embodiment, a computer device is provided. The computer device may be a server or a terminal. The internal structure diagram thereof may be as follows: Figure 7 As shown. The computer device includes a processor, a memory, an input / output interface (Input / Output, referred to as I / O) and a communication interface. The processor, the memory and the input / output interface are connected through a system bus, and the communication interface is connected to the system bus through the input / output interface. The processor of the computer device is used to provide computing and control capabilities. The memory of the computer device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system, a computer program and a database. The internal memory provides an environment for the operation of the operating system and the computer program in the non-volatile storage medium. The database of the computer device is used to store video tag processing data. The input / output interface of the computer device is used to exchange information between the processor and an external device. The communication interface of the computer device is used to communicate with an external terminal through a network connection. When the computer program is executed by the processor, a method for optimizing an underground aquifer energy storage system based on an agent model is implemented.

[0166] Those skilled in the art will understand that Figure 7 The structure shown in the figure is only a block diagram of a part of the structure related to the solution of the present application, and does not constitute a limitation on the computer device to which the solution of the present application is applied. The specific computer device may include more or fewer components than those shown in the figure, or combine certain components, or have a different arrangement of components.

[0167] In an exemplary embodiment, a computer device is further provided, including a memory and a processor, wherein a computer program is stored in the memory, and the processor implements the steps in the above-mentioned method embodiments when executing the computer program.

[0168] In an exemplary embodiment, a computer-readable storage medium is provided, storing a computer program, and when the computer program is executed by a processor, the steps in the above method embodiments are implemented.

[0169] In an exemplary embodiment, a computer program product is provided, including a computer program, and when the computer program is executed by a processor, the steps in the above method embodiments are implemented.

[0170] It should be noted that the user information (including but not limited to user device information, user personal information, etc.) and data (including but not limited to data used for analysis, stored data, displayed data, etc.) involved in this application are all information and data authorized by the user or fully authorized by all parties, and the collection, use and processing of relevant data must comply with relevant regulations.

[0171] Those of ordinary skill 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, and the computer program can be stored in a non-volatile computer-readable storage medium. When the computer program is executed, it can include the processes of the embodiments of the above-mentioned methods. Among them, any reference to the memory, database or other medium used in the embodiments provided in the present application can include at least one of non-volatile and volatile memory. Non-volatile memory can include read-only memory (ROM), magnetic tape, floppy disk, flash memory, optical memory, high-density embedded non-volatile memory, resistive random access memory (ReRAM), magnetic random access memory (MRAM), ferroelectric random access memory (FRAM), phase change memory (PCM), graphene memory, etc. Volatile memory can include random access memory (RAM) or external cache memory, etc. By way of illustration and not limitation, RAM may be in various forms, such as static random access memory (SRAM) or dynamic random access memory (DRAM).

[0172] The database involved in each embodiment provided in this application may include at least one of a relational database and a non-relational database. The non-relational database may include a distributed database based on blockchain, etc., but is not limited thereto. The processor involved in each embodiment provided in this application may be a general-purpose processor, a central processing unit, a graphics processor, a digital signal processor, a programmable logic device, a data processing logic device based on quantum computing, etc., but is not limited thereto.

[0173] The technical features of the above embodiments may be combined arbitrarily. To make the description concise, not all possible combinations of the technical features in the above 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.

[0174] This article uses specific examples to illustrate the principles and implementation methods of this application. The description of the above embodiments is only used to help understand the method and core ideas of this application. At the same time, for those skilled in the art, according to the ideas of this application, there will be changes in the specific implementation methods and application scope. In summary, the content of this specification should not be understood as limiting this application.

Claims

1. An optimization method for underground aquifer energy storage system based on agent model, characterized in that: The underground aquifer energy storage system optimization method based on the agent model includes: Determine the optimization variables and constraint information according to the preset optimization target; wherein the optimization variables include at least operation parameters and response parameters, the operation parameters include at least single well injection rate, single well extraction rate, cold well injection temperature and hot well injection temperature, the response parameters include at least maximum ground subsidence, maximum water level drop, influence radius and average heat recovery rate; the constraint information is used to constrain the response parameters; Based on the optimization variables and the constraint information, a high-fidelity thermal-fluid-solid multi-field coupling numerical model is constructed; Inputting the operating parameters into the high-fidelity thermal-fluid-solid multi-field coupling numerical model to obtain the predicted response parameters corresponding to the operating parameters; wherein the predicted response parameters at least include the predicted maximum ground subsidence, the predicted maximum water level drop, the predicted influence radius, and the predicted average heat recovery rate; Based on the operating parameters and the predicted response parameters, a proxy model is constructed; wherein the proxy model is used to represent the functional relationship between the operating parameters and the predicted response parameters; The proxy model is optimized by target search to obtain optimized operating parameters so that the underground aquifer energy storage system can operate based on the optimized operating parameters.

2. The underground aquifer energy storage system optimization method based on the proxy model according to claim 1 is characterized in that: The constraint information includes at least a critical value of the maximum ground subsidence, a critical value of the maximum water level drop, and a critical value of the impact radius.

3. The underground aquifer energy storage system optimization method based on the proxy model according to claim 1 or 2, characterized in that: The constructing of a high-fidelity thermal-fluid-solid multi-field coupling numerical model based on the optimization variables and the constraint information specifically includes: Obtaining a real response parameter corresponding to the operating parameter; Inputting the operating parameters into a pre-built initial numerical model to obtain training response parameters; Calculating the true response parameter and the training response parameter based on the loss function to obtain a loss value; If the loss value is greater than a preset loss threshold, the initial numerical model is trained, and the steps from obtaining the real response parameter corresponding to the operating parameter to calculating the real response parameter and the training response parameter based on the loss function to obtain the loss value are performed; If the loss value is less than or equal to the preset loss threshold, the initial numerical model is determined as a trained high-fidelity thermal-fluid-solid multi-field coupling numerical model.

4. The underground aquifer energy storage system optimization method based on the proxy model according to claim 2 is characterized in that: The proxy model is a multiple regression proxy model, and the expression of the multiple regression proxy model is: in, is the predicted response parameter, is the operating parameter, S is the maximum ground subsidence, Δh is the maximum water level drop, R is the impact radius, η is the average heat recovery rate, S1 is the critical value of the maximum ground subsidence, Δh1 is the critical value of the maximum water level drop, R1 is the critical value of the impact radius, q is the single well injection rate or the single well extraction rate, T c-in is the cold well injection temperature and T h-in is the water injection temperature of the hot well.

5. The underground aquifer energy storage system optimization method based on the agent model according to claim 1, characterized in that: The target optimization of the proxy model to obtain optimized operating parameters specifically includes: Optimizing the agent model to obtain multiple optimization operation indicators; Non-dimensionalizing the plurality of optimization operation indicators based on the desirability function to obtain a dimensionless value corresponding to each of the optimization operation indicators; Based on the optimization algorithm and the dimensionless value corresponding to each of the optimization operation indicators, multiple optimization operation indicators are optimized to obtain optimization operation parameters.

6. The underground aquifer energy storage system optimization method based on the agent model according to claim 5 is characterized in that: The step of performing dimensionless transformation on a current optimization operation index based on the desirability function to obtain a dimensionless value corresponding to the current optimization operation index specifically includes: Obtaining maximum and minimum values ​​of the predicted response variables corresponding to the plurality of optimization operation indicators; Obtaining a target value of a predicted response variable corresponding to the current optimization operation index; Based on the desirability function, the maximum value, the minimum value and the target value are dimensionless to obtain a dimensionless value corresponding to the current optimization operation index; And, the calculation formula of the dimensionless value corresponding to the current optimization operation index is: Among them, d i is the dimensionless value corresponding to the current optimization operation index, is the minimum value, is the maximum value, is the target value, w i The weight of the current optimization running indicator.

7. An underground aquifer energy storage system optimization device based on an agent model, characterized in that: The underground aquifer energy storage system optimization device based on the agent model includes: A determination unit, configured to determine optimization variables and constraint information according to a preset optimization objective; wherein the optimization variables include at least operating parameters and response parameters, the operating parameters include at least single well injection rate, single well extraction rate, cold well injection temperature and hot well injection temperature, and the response parameters include at least maximum ground subsidence, maximum water level drop, influence radius and average heat recovery rate; the constraint information is used to constrain the response parameters; A first construction unit is used to construct a high-fidelity thermal-fluid-solid multi-field coupling numerical model based on the optimization variables and the constraint information; An input unit, used to input the operating parameters into the high-fidelity thermal-fluid-solid multi-field coupling numerical model to obtain prediction response parameters corresponding to the operating parameters; wherein the prediction response parameters at least include predicted maximum ground subsidence, predicted maximum water level drop, predicted influence radius, and predicted average heat recovery rate; A second construction unit is used to construct a proxy model based on the operating parameters and the predicted response parameters; wherein the proxy model is used to represent the functional relationship between the operating parameters and the predicted response parameters; The optimization unit is used to perform target optimization on the proxy model to obtain optimized operating parameters so that the underground aquifer energy storage system can operate based on the optimized operating parameters.

8. A computer device comprising: A memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the computer program to implement the steps of the underground aquifer energy storage system optimization method based on the agent model described in any one of claims 1-6.

9. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the steps of the underground aquifer energy storage system optimization method based on the agent model described in any one of claims 1-6 are implemented.

10. A computer program product, comprising a computer program, characterized in that When the computer program is executed by a processor, the steps of the underground aquifer energy storage system optimization method based on the agent model described in any one of claims 1-6 are implemented.

Images