Geothermal energy development decision-making method and system based on big data analysis

Abstract

The invention discloses a geothermal energy development decision-making method and system based on big data analysis, and the method comprises the steps: constructing a geothermal energy prediction model through employing a machine learning model, training the geothermal energy prediction model through employing a geothermal energy correlation feature sample and a geothermal label, and obtaining a trained geothermal energy prediction model, then, in the geothermal energy development process, geothermal energy associated feature data corresponding to a target site can be collected, the trained geothermal energy prediction model is adopted for geothermal energy prediction, a geothermal energy prediction result corresponding to the target site is obtained, and finally, workers are assisted in making development decisions according to the geothermal energy prediction result. The geothermal energy development efficiency can be effectively improved, and the geothermal energy development cost can be effectively reduced.

Description

Technical Field

[0001] The present invention belongs to the technical field of big data, and particularly relates to a geothermal energy development decision-making method and system based on big data analysis. Background Art

[0002] As a clean and renewable energy source, geothermal energy has broad application prospects in China. However, there are many uncertain factors in the process of geothermal energy development, such as geological conditions, resource distribution, and market demand, resulting in a relatively high development risk. However, due to the lack of a scientific understanding of shallow geothermal energy, for example, the endowment of shallow geothermal energy resources and the geological conditions in which it is located have not been concerned, which has led to a large waste of human and material resources in the past development and utilization of shallow geothermal energy. Summary of the Invention

[0003] The present invention provides a geothermal energy development decision-making method and system based on big data analysis to solve the problems existing in the prior art.

[0004] In a first aspect, the present invention provides a geothermal energy development decision-making method based on big data analysis, including:

[0005] Collecting big data of geothermal energy samples stored in advance or input by staff; wherein, the big data of geothermal energy samples includes geothermal energy associated feature samples and geothermal temperature labels;

[0006] Using a machine learning model to construct a geothermal energy prediction model, and training the geothermal energy prediction model with the geothermal energy associated feature samples and geothermal temperature labels to obtain a trained geothermal energy prediction model;

[0007] Collecting geothermal associated feature data corresponding to the target location, and using the trained geothermal energy prediction model to predict geothermal energy to obtain a geothermal energy prediction result corresponding to the target location;

[0008] Providing the geothermal energy prediction result corresponding to the target location to the staff so that the staff can make a geothermal energy development decision based on the geothermal energy prediction result.

[0009] In a possible implementation manner, the geothermal energy associated feature samples include: longitude, latitude, intra-regional tectonic conditions, average annual rainfall, average annual temperature, distance from the nearest fault layer, and drilling elevation; the geothermal temperature label represents the geothermal temperature data at 100 m underground.

[0010] In a possible implementation manner, using a machine learning model to construct a geothermal energy prediction model, and training the geothermal energy prediction model with the geothermal energy associated feature samples and geothermal temperature labels to obtain a trained geothermal energy prediction model, including:

[0011] A machine learning model is used to construct a geothermal energy prediction model, and the model parameters of the geothermal energy prediction model are initialized to obtain a plurality of individuals; wherein, each individual includes all hyperparameters to be optimized of the geothermal energy prediction model.

[0012] Using the geothermal energy correlation feature samples as the input of the geothermal energy prediction model and the geothermal temperature labels as the expected output, the loss function value corresponding to each individual is obtained.

[0013] Based on the loss function value corresponding to each individual, the optimal individual is determined.

[0014] Based on the optimal individual, information fusion update is performed on each individual to obtain the individuals after information fusion update.

[0015] Based on the optimal individual, optimal information fusion update is performed on each individual after information fusion update to obtain the individuals after optimal information fusion update.

[0016] Mutual exclusion information fusion update is performed on each individual after optimal information fusion update to obtain the individuals after mutual exclusion information fusion update.

[0017] A global adaptive random information exploration strategy is used to perform global update on the optimal individual to obtain the optimal individual after global update.

[0018] It is judged whether the current training times reach the maximum training times. If so, based on the individuals after information fusion update, the individuals after optimal information fusion update, the individuals after mutual exclusion information fusion update, and the optimal individual after global update, the individual with the minimum loss function value is re-obtained as the optimal individual, and the hyperparameters included in the re-obtained optimal individual are used as the final hyperparameters of the geothermal energy prediction model to obtain the geothermal energy prediction model after training; otherwise, return to the step of obtaining the loss function value.

[0019] In a possible implementation manner, using the geothermal energy correlation feature samples as the input of the geothermal energy prediction model and the geothermal temperature labels as the expected output, the loss function value corresponding to each individual is obtained, and based on the loss function value corresponding to each individual, the optimal individual is determined, including:

[0020] For any one individual, after applying the hyperparameters included in the individual to the geothermal energy prediction model, using the geothermal energy correlation feature samples as the input of the geothermal energy prediction model and the geothermal temperature labels as the expected output, the root mean square loss function value corresponding to the individual is obtained to obtain the loss function value corresponding to each individual.

[0021] Based on the loss function value corresponding to each individual, the individual with the minimum loss function value is determined as the optimal individual.

[0022] In a possible implementation manner, based on the optimal individual, information fusion update is performed on each individual to obtain the individual after information fusion update, including:

[0023] According to all individuals and the loss function values corresponding to all individuals, the centroid position individual is obtained by using a one-by-one weighting method;

[0024] Based on the current number of training times, a non-linear information fusion factor is obtained;

[0025] For any individual, based on the non-linear information fusion factor, full information fusion is performed on the individual to obtain an information fusion term, and based on the centroid position individual and the information fusion term, information fusion update is performed on the individual to obtain the individual after information fusion update.

[0026] In a possible implementation manner, based on the optimal individual, optimal information fusion update is performed on each individual after information fusion update to obtain the individual after optimal information fusion update, including:

[0027] Based on the current number of training times, an adaptive inertia weight is obtained by using the hyperbolic tangent function;

[0028] For any individual after information fusion update, based on the optimal individual, the individual and the optimal individual are information-fused by using the adaptive inertia weight to obtain the individual after optimal information fusion update.

[0029] In a possible implementation manner, mutually exclusive information fusion update is performed on each individual after optimal information fusion update to obtain the individual after mutually exclusive information fusion update, including:

[0030] The loss function value corresponding to the individual after optimal information fusion update is obtained, and based on the loss function value corresponding to the individual after optimal information fusion update, the optimal individual and the worst individual are re-obtained;

[0031] Based on the centroid position individual and the optimal individual, a learning information term is obtained; based on the centroid position individual and the worst individual, a mutually exclusive information term is obtained;

[0032] For any individual after optimal information fusion update, based on the learning information term and the mutually exclusive information term, mutually exclusive information fusion update is performed on the individual to obtain the individual after mutually exclusive information fusion update.

[0033] In a possible implementation manner, a global adaptive random information exploration strategy is used to globally update the optimal individual to obtain the optimal individual after global update, including:

[0034] Randomly match the optimal individual with the first random individual, the second random individual, and the third random individual;

[0035] Obtain the global exploration value corresponding to the optimal individual according to the first random individual, the second random individual, and the third random individual;

[0036] Judge whether the loss function value corresponding to the global exploration value is less than the loss function value corresponding to the optimal individual. If so, use the global exploration value as the optimal individual after global update; otherwise, directly use the original optimal individual as the optimal individual after global update.

[0037] In a possible implementation manner, collect the geothermal-related feature data corresponding to the target location, and use the trained geothermal energy prediction model to perform geothermal energy prediction to obtain the geothermal energy prediction result corresponding to the target location, including: collect the geothermal-related feature data corresponding to the target location, use the geothermal-related feature data as the input of the trained geothermal energy prediction model, obtain the output of the geothermal energy prediction model, and obtain the geothermal energy prediction result corresponding to the target location.

[0038] In a second aspect, the present invention provides a geothermal energy development decision-making system based on big data analysis, including: a big data acquisition module, a big data learning module, a geothermal prediction module, and an auxiliary development decision-making module;

[0039] The big data acquisition module is used to collect the big data of geothermal energy samples stored in advance or input by staff; wherein, the big data of geothermal energy samples includes geothermal energy-related feature samples and ground temperature labels;

[0040] The big data learning module is used to construct a geothermal energy prediction model by using a machine learning model, and train the geothermal energy prediction model by using the geothermal energy-related feature samples and ground temperature labels to obtain a trained geothermal energy prediction model;

[0041] The geothermal prediction module is used to collect the geothermal-related feature data corresponding to the target location, and use the trained geothermal energy prediction model to perform geothermal energy prediction to obtain the geothermal energy prediction result corresponding to the target location;

[0042] The auxiliary development decision-making module is used to provide the geothermal energy prediction result corresponding to the target location to the staff, so that the staff can make a geothermal energy development decision according to the geothermal energy prediction result.

[0043] A geothermal energy development decision-making method and system based on big data analysis provided by the present invention constructs a geothermal energy prediction model by using a machine learning model, and trains the geothermal energy prediction model by using geothermal energy correlation feature samples and geothermal temperature labels to obtain the trained geothermal energy prediction model. Then, during the geothermal energy development process, geothermal energy correlation feature data corresponding to the target location can be collected, and the trained geothermal energy prediction model is used for geothermal energy prediction to obtain the geothermal energy prediction result corresponding to the target location. Finally, it assists the staff to make development decisions based on the geothermal energy prediction result, which can effectively improve the geothermal energy development efficiency and reduce the geothermal energy development cost. BRIEF DESCRIPTION OF THE DRAWINGS

[0044] The accompanying drawings herein are incorporated into the specification and form a part of the specification, showing embodiments consistent with the present invention, and are used together with the specification to explain the principles of the present invention.

[0045] Figure 1 It is a flowchart of a geothermal energy development decision-making method based on big data analysis provided by the present invention.

[0046] Figure 2 It is a schematic structural diagram of a geothermal energy development decision-making system based on big data analysis provided by the present invention.

[0047] In the drawings, 21 - big data acquisition module, 22 - big data learning module, 23 - geothermal prediction module, 24 - auxiliary development decision-making module.

[0048] Through the above-mentioned accompanying drawings, specific embodiments of the present invention have been shown, and there will be more detailed descriptions hereinafter. These drawings and text descriptions are not intended to limit the scope of the inventive concept in any way, but to illustrate the concept of the present invention to those skilled in the art by referring to specific embodiments. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0049] Here, the exemplary embodiments will be described in detail, and the examples are shown in the accompanying drawings. When the following description refers to the accompanying drawings, unless otherwise indicated, the same numbers in different drawings represent the same or similar elements. The embodiments described in the following exemplary embodiments do not represent all embodiments consistent with the present invention. On the contrary, they are merely examples of devices and methods consistent with some aspects of the present invention as detailed in the appended claims.

[0050] The embodiments of the present invention will be described in detail below with reference to the accompanying drawings.

[0051] Generally speaking, the most common method to obtain the ground temperature is to dig a temperature measurement hole. However, the drilling cost is very high. If this method is used to obtain the large-scale ground temperature distribution, it will greatly increase the cost of shallow geothermal energy utilization. Therefore, this invention will adopt a machine learning method to predict the distribution of the ground temperature at 100m in the study area, which can effectively reduce the cost consumption in the process of geothermal energy development and improve the development efficiency of geothermal energy.

[0052] As Figure 1 shown, this invention provides a geothermal energy development decision-making method based on big data analysis, including:

[0053] S11. Collect the big data of geothermal energy samples pre-stored or input by staff; among them, the big data of geothermal energy samples includes geothermal energy-related feature samples and ground temperature labels;

[0054] The big data of geothermal energy samples can be the data measured during the previous geothermal energy development process. The geothermal energy-related feature samples can be some feature data related to geothermal energy, and then the ground temperature label is the ground temperature data at 100 meters underground. By learning the relationship between these data, geothermal prediction can be effectively realized to assist in making geothermal energy development decisions.

[0055] S12. Use a machine learning model to construct a geothermal energy prediction model, and use the geothermal energy-related feature samples and ground temperature labels to train the geothermal energy prediction model to obtain the trained geothermal energy prediction model;

[0056] Using a machine learning model to construct a geothermal energy prediction model can include: using a BP neural network or a convolutional neural network to construct a geothermal energy prediction model. However, it should be noted that other neural networks can also be used to construct a geothermal energy prediction model to achieve prediction.

[0057] After constructing the geothermal energy prediction model, it cannot be directly used. The gradient descent intelligent optimization algorithm can be used to train the geothermal energy prediction model to achieve big data learning.

[0058] S13. Collect the geothermal energy-related feature data corresponding to the target location, and use the trained geothermal energy prediction model to predict geothermal energy to obtain the geothermal energy prediction result corresponding to the target location;

[0059] The data structures and data types of the geothermal energy-related feature samples and the geothermal energy-related feature data should be the same. Whether it is the geothermal energy-related feature samples or the geothermal energy-related feature data, they can be normalized before being input into the geothermal energy prediction model to reduce the data complexity.

[0060] S14. Provide the geothermal energy prediction result corresponding to the target location to the staff so that the staff can make geothermal energy development decisions based on the geothermal energy prediction result.

[0061] By providing the geothermal energy prediction result corresponding to the target location to the staff, the staff can develop geothermal energy, effectively avoiding the problems of high cost and low efficiency caused by excavating temperature measurement holes, thereby improving the effect of geothermal energy development.

[0062] In a possible implementation manner, the geothermal energy related feature samples include: longitude, latitude, in-area tectonic conditions, average annual rainfall, average annual temperature, distance from the nearest fault layer, and borehole elevation; the geothermal temperature label represents the geothermal temperature data at 100 m underground.

[0063] It should be noted that the specific data in the above geothermal energy related feature samples are only the preferred implementation manners of the embodiments of the present invention, and other data can also be used to construct the geothermal energy related feature samples, so as to make the geothermal temperature prediction more accurate.

[0064] In a possible implementation manner, a machine learning model is used to construct a geothermal energy prediction model, and the geothermal energy related feature samples and the geothermal temperature labels are used to train the geothermal energy prediction model to obtain the trained geothermal energy prediction model, including:

[0065] Use a machine learning model to construct a geothermal energy prediction model, and initialize the model parameters of the geothermal energy prediction model to obtain multiple individuals; where each individual includes all the hyperparameters to be optimized of the geothermal energy prediction model.

[0066] For example, when using a BP neural network to construct a geothermal energy prediction model, random initialization can be performed between the upper and lower limits of its weights, and the initialized weights are combined into an individual. After repeating multiple times, multiple different individuals are obtained.

[0067] Use the geothermal energy related feature samples as the input of the geothermal energy prediction model, and the geothermal temperature labels as the expected output to obtain the loss function value corresponding to each individual.

[0068] Determine the optimal individual according to the loss function value corresponding to each individual.

[0069] Based on the optimal individual, perform information fusion update on each individual to obtain the individuals after information fusion update.

[0070] Based on the optimal individual, perform optimal information fusion update on each individual after information fusion update to obtain the individuals after optimal information fusion update.

[0071] Perform mutually exclusive information fusion update on each individual after optimal information fusion update to obtain the individuals after mutually exclusive information fusion update.

[0072] The optimal individual is globally updated using a global adaptive random information exploration strategy to obtain the optimal individual after global update;

[0073] Determine whether the current number of training times has reached the maximum number of training times. If so, based on the individual after information fusion update, the optimal individual after optimal information fusion update, the individual after mutually exclusive information fusion update, and the optimal individual after global update, re-obtain the individual with the smallest loss function value as the optimal individual, and use the hyperparameters included in the re-obtained optimal individual as the final hyperparameters of the geothermal energy prediction model to obtain the trained geothermal energy prediction model. Otherwise, return to the step of obtaining the loss function value.

[0074] Since the existing technology will fall into local optimality during the process of training a machine learning model, resulting in poor final geothermal prediction ability. Therefore, the embodiments of the present invention provide a training strategy, which can not only effectively improve the training speed of the algorithm, but also effectively improve the global exploration ability of the algorithm, ultimately improving the geothermal prediction accuracy and assisting the staff to improve the geothermal energy development efficiency.

[0075] In a possible implementation manner, using the geothermal energy associated feature samples as the input of the geothermal energy prediction model and the geothermal temperature label as the expected output, obtain the loss function value corresponding to each individual, and determine the optimal individual according to the loss function value corresponding to each individual, including:

[0076] For any one individual, after applying the hyperparameters included in the individual to the geothermal energy prediction model, using the geothermal energy associated feature samples as the input of the geothermal energy prediction model and the geothermal temperature label as the expected output, obtain the root mean square loss function value corresponding to the individual to obtain the loss function value corresponding to each individual;

[0077] According to the loss function value corresponding to each individual, determine the individual with the smallest loss function value as the optimal individual.

[0078] In a possible implementation manner, based on the optimal individual, perform information fusion update on each individual to obtain the individual after information fusion update, including:

[0079] According to all individuals and the loss function values corresponding to all individuals, use the one-by-one weighting method to obtain the centroid position individual as:

[0080]

[0081]

[0082] Wherein, represents the jth individual in the tth training, j = 1, 2,..., K, and K represents the total number of individuals, represents the individual The corresponding weighting coefficient, X * Indicates the centroid position individual Indicates an individual The corresponding fitness Indicates an individual The corresponding fitness, and the fitness = 1 / (loss function value + 0.001);

[0083] Based on the current training times, obtain the non - linear information fusion factor as:

[0084] ξ = ξ max -e -(1-t / T) *(ξ max -ξ min )

[0085] Among them, ξ represents the non - linear information fusion factor, ξ max Represents the preset maximum value of the non - linear information fusion factor, ξ min Represents the preset minimum value of the non - linear information fusion factor, e represents the natural constant, and T represents the maximum number of training times;

[0086] For any individual, perform full - information fusion on the individual according to the non - linear information fusion factor to obtain an information fusion term, and based on the centroid position individual and the information fusion term, perform information fusion update on the individual to obtain the individual after information fusion update as:

[0087]

[0088] Among them, Represents the d - th hyperparameter of the i - th individual in the t - th training, i = 1, 2,..., K, Represents the d - th hyperparameter of the h - th individual in the t - th training, d = 1, 2,..., D, and D represents the total number of hyperparameters in the individual, Represents the d - th hyperparameter of the centroid position individual, dist ih Represents the Euclidean distance between the i - th individual and the h - th individual in the t - th training, Represents the d - th hyperparameter of the h - th individual after information fusion update, U d Represents the upper limit of the d - th hyperparameter, L d Represents the lower limit of the d - th hyperparameter, α 1 Represents the first fusion coefficient, and is set to 0.5; α 2 Represents the second fusion coefficient, and is set to 1.5.

[0089] The information fusion update provided by the embodiments of the present invention can effectively fuse the information corresponding to all individuals, avoid collisions during the search process, and can improve the global search ability in the early stage of the algorithm and the search accuracy in the later stage.

[0090] In a possible implementation manner, based on the optimal individual, perform optimal information fusion update on each individual after information fusion update to obtain the individual after optimal information fusion update, including:

[0091] Based on the current training times, use the hyperbolic tangent function to obtain the adaptive inertia weight as:

[0092]

[0093] where ω represents the inertia weight, ω max represents the maximum value of the inertia weight, ω min represents the minimum value of the inertia weight, tanh represents the hyperbolic tangent function, and T represents the preset maximum number of training times;

[0094] For any individual after information fusion update, based on the optimal individual, use the adaptive inertia weight to perform information fusion on the individual and the optimal individual to obtain the individual after optimal information fusion update as:

[0095]

[0096] where represents the m-th individual after information fusion update in the t-th training process, represents the individual after optimal information fusion update r 1 represents a random number between (0, 1), represents the optimal individual.

[0097] The optimal information fusion update provided by the embodiments of the present invention can effectively fuse the information of the optimal individual, effectively improve the search efficiency of the algorithm, and improve the search accuracy in the later stage of the algorithm.

[0098] In a possible implementation manner, perform mutually exclusive information fusion update on each individual after optimal information fusion update to obtain the individual after mutually exclusive information fusion update, including:

[0099] Obtain the loss function value corresponding to the individual after optimal information fusion update, and re-obtain the optimal individual and the worst individual according to the loss function value corresponding to the individual after optimal information fusion update;

[0100] Obtain the learning information item according to the centroid position individual and the optimal individual as: where represents the optimal individual, η 1 represents the first learning rate;

[0101] Based on the centroid position individual and the worst individual, the mutually exclusive information item obtained is: Wherein, represents the worst individual, η 2 represents the second learning rate;

[0102] For any individual after optimal information fusion update, based on the learning information item and the mutually exclusive information item, perform mutually exclusive information fusion update on the individual, and the individual after mutually exclusive information fusion update obtained is:

[0103]

[0104] Wherein, represents the kth individual after optimal information fusion update in the tth training process, represents the individual after mutually exclusive information fusion update represents a random information adjustment factor between (0, 1).

[0105] Based on the optimal individual and the worst individual, the embodiment of the present invention searches the local area, so that more local unfamiliar areas can be searched. While approaching a better position, it is far from a worse position, which can improve the ability of the algorithm to escape from the local optimal solution to a certain extent.

[0106] In a possible implementation manner, a global adaptive random information exploration strategy is adopted to globally update the optimal individual, and the globally updated optimal individual obtained includes:

[0107] Randomly match a first random individual, a second random individual, and a third random individual for the optimal individual;

[0108] Based on the first random individual, the second random individual, and the third random individual, the global exploration value corresponding to the optimal individual obtained is:

[0109]

[0110] Wherein, represents the global exploration value, represents the first random individual, represents the second random individual, represents the third random individual, r 2 represents a random number between (0, 1).

[0111] Judge whether the loss function value corresponding to the global exploration value is less than the loss function value corresponding to the optimal individual. If so, use the global exploration value as the globally updated optimal individual, otherwise directly use the original optimal individual as the globally updated optimal individual.

[0112] The global adaptive random information exploration strategy provided by the embodiments of the present invention can effectively assist the algorithm in jumping out of the local optimum. Combining the above several search strategies can effectively achieve fast global optimum search and improve the accuracy of geothermal prediction.

[0113] In a possible implementation manner, geothermal correlation feature data corresponding to a target location is collected, and the trained geothermal energy prediction model is used for geothermal energy prediction to obtain the geothermal energy prediction result corresponding to the target location, including: collecting the geothermal correlation feature data corresponding to the target location, using the geothermal correlation feature data as the input of the trained geothermal energy prediction model, obtaining the output of the geothermal energy prediction model, and obtaining the geothermal energy prediction result corresponding to the target location.

[0114] A geothermal energy development decision-making method based on big data analysis provided by the present invention constructs a geothermal energy prediction model by using a machine learning model, and trains the geothermal energy prediction model by using geothermal energy correlation feature samples and geothermal temperature labels to obtain a trained geothermal energy prediction model. Then, during the geothermal energy development process, geothermal correlation feature data corresponding to a target location can be collected, and the trained geothermal energy prediction model is used for geothermal energy prediction to obtain the geothermal energy prediction result corresponding to the target location. Finally, it assists the staff in making development decisions based on the geothermal energy prediction result, which can effectively improve the geothermal energy development efficiency and reduce the geothermal energy development cost.

[0115] As Figure 2 shown, the present invention provides a geothermal energy development decision-making system based on big data analysis, including: a big data acquisition module 21, a big data learning module 22, a geothermal prediction module 23, and an auxiliary development decision-making module 24;

[0116] The big data acquisition module 21 is used to collect big data of geothermal energy samples stored in advance or input by the staff; wherein, the big data of geothermal energy samples includes geothermal energy correlation feature samples and geothermal temperature labels;

[0117] The big data learning module 22 is used to construct a geothermal energy prediction model by using a machine learning model, and train the geothermal energy prediction model by using geothermal energy correlation feature samples and geothermal temperature labels to obtain a trained geothermal energy prediction model;

[0118] The geothermal prediction module 23 is used to collect geothermal correlation feature data corresponding to a target location, and use the trained geothermal energy prediction model for geothermal energy prediction to obtain the geothermal energy prediction result corresponding to the target location;

[0119] The auxiliary development decision-making module 24 is configured to provide the geothermal energy prediction result corresponding to the target location to the staff, so that the staff can make a geothermal energy development decision based on the geothermal energy prediction result.

[0120] The present invention provides a geothermal energy development decision-making system based on big data analysis, which can execute the above method technical solution, and its principle and beneficial effects are similar, so details will not be described here.

[0121] Those skilled in the art should understand that the embodiments of the present invention can be provided as a method, a system, or a computer program product. Therefore, the present invention can take the form of a complete hardware embodiment, a complete software embodiment, or an embodiment combining software and hardware aspects. Moreover, the present invention can take the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk memory, CD-ROM, optical memory, etc.) containing computer-usable program code.

[0122] The present invention is described with reference to the flowcharts and / or block diagrams of methods, apparatuses (systems), and computer program products according to embodiments of the present invention. It should be understood that each process and / or block in the flowchart and / or block diagram, and the combination of processes and / or blocks in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to the processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing devices to generate a machine, so that the instructions executed by the processor of the computer or other programmable data processing devices generate means for implementing the functions specified in Figure 1 one process or multiple processes and / or blocks Figure 1 one block or multiple blocks.

[0123] These computer program instructions can also be stored in a computer-readable memory that can direct a computer or other programmable data processing device to work in a specific manner, so that the instructions stored in the computer-readable memory generate a manufactured article including instruction means, and the instruction means implements the functions specified in Figure 1 one process or multiple processes and / or blocks Figure 1 one block or multiple blocks.

[0124] These computer program instructions can also be loaded onto a computer or other programmable data processing device, so that a series of operation steps are executed on the computer or other programmable device to generate a computer-implemented process, and thus the instructions executed on the computer or other programmable device provide steps for implementing the functions specified in Figure 1 one process or multiple processes and / or blocks Figure 1 one block or multiple blocks.

[0125] Those of ordinary skill in the art can understand that all or part of the steps in implementing the above facts and methods can be completed by instructing relevant hardware through a program. The program involved or the said program can be stored in a computer-readable storage medium. When the program is executed, it includes the following steps: At this time, the corresponding method steps are introduced. The storage medium can be ROM / RAM, magnetic disk, optical disc, etc.

[0126] The specific embodiments described above further elaborate on the purpose, technical solutions, and beneficial effects of the present invention. It should be understood that the above are only specific embodiments of the present invention and are not used to limit the protection scope of the present invention. Any modifications, equivalent replacements, improvements, etc. made within the spirit and principle of the present invention shall be included in the protection scope of the present invention.

Claims

1. A geothermal energy development decision-making method based on big data analysis, characterized in that: include: Collecting geothermal energy sample big data that is pre-stored or input by staff; wherein the geothermal energy sample big data includes geothermal energy related feature samples and ground temperature tags; A geothermal energy prediction model is constructed using a machine learning model, and the geothermal energy related feature samples and geothermal temperature labels are used to train the geothermal energy prediction model to obtain a trained geothermal energy prediction model; Collect geothermal related characteristic data corresponding to the target location, and use the trained geothermal energy prediction model to perform geothermal energy prediction to obtain geothermal energy prediction results corresponding to the target location; The geothermal energy prediction results corresponding to the target location are provided to the staff so that the staff can make geothermal energy development decisions based on the geothermal energy prediction results.

2. The geothermal energy development decision-making method based on big data analysis according to claim 1 is characterized in that: The geothermal energy related feature samples include: longitude, latitude, structural conditions in the area, average annual rainfall, average annual temperature, distance to the nearest fault layer and borehole elevation; the geothermal label represents the geothermal data 100m underground.

3. The geothermal energy development decision-making method based on big data analysis according to claim 1 is characterized in that: A geothermal energy prediction model is constructed using a machine learning model, and the geothermal energy prediction model is trained using geothermal energy-related feature samples and geothermal temperature labels to obtain a geothermal energy prediction model after training, including: A geothermal energy prediction model is constructed by using a machine learning model, and model parameters of the geothermal energy prediction model are initialized to obtain a plurality of individuals; wherein each individual includes all hyperparameters to be optimized of the geothermal energy prediction model; The geothermal energy related feature samples are used as the input of the geothermal energy prediction model, and the ground temperature label is used as the expected output to obtain the loss function value corresponding to each individual; Determine the optimal individual based on the loss function value corresponding to each individual; Based on the optimal individual, information fusion and updating are performed on each individual to obtain an individual after information fusion and updating; Based on the optimal individual, performing optimal information fusion update on each individual after information fusion update to obtain the individual after optimal information fusion update; Perform mutually exclusive information fusion update on each individual after optimal information fusion update to obtain the individual after mutually exclusive information fusion update; The global adaptive random information exploration strategy is used to globally update the optimal individual, and the optimal individual after global update is obtained; Determine whether the current number of training times has reached the maximum number of training times. If so, re-obtain the individual with the smallest loss function value as the optimal individual based on the individual after information fusion update, the individual after optimal information fusion update, the individual after mutually exclusive information fusion update, and the optimal individual after global update, and use the hyperparameters contained in the re-obtained optimal individual as the final hyperparameters of the geothermal energy prediction model to obtain the geothermal energy prediction model after training. Otherwise, return to the step of obtaining the loss function value.

4. The geothermal energy development decision-making method based on big data analysis according to claim 3 is characterized in that: Taking geothermal energy related feature samples as the input of geothermal energy prediction model and geothermal label as the expected output, the loss function value corresponding to each individual is obtained, and the optimal individual is determined according to the loss function value corresponding to each individual, including: For any individual, after applying the hyperparameters contained in the individual to the geothermal energy prediction model, the geothermal energy-related feature samples are used as the input of the geothermal energy prediction model, and the ground temperature label is used as the expected output to obtain the root mean square loss function value corresponding to the individual, and the loss function value corresponding to each individual is obtained; According to the loss function value corresponding to each individual, the individual with the smallest loss function value is determined as the optimal individual.

5. The geothermal energy development decision-making method based on big data analysis according to claim 3 is characterized in that: Based on the optimal individual, each individual is updated by information fusion to obtain an individual after information fusion update, including: According to all individuals and the loss function values ​​corresponding to all individuals, the centroid position individuals are obtained by weighting one by one; Based on the current number of training times, obtain the nonlinear information fusion factor; For any individual, full information fusion is performed on the individual according to the nonlinear information fusion factor to obtain an information fusion item, and information fusion update is performed on the individual according to the centroid position individual and the information fusion item to obtain the individual after information fusion update.

6. The geothermal energy development decision-making method based on big data analysis according to claim 5 is characterized in that: Based on the optimal individual, each individual after information fusion update is subjected to optimal information fusion update to obtain the individual after optimal information fusion update, including: Based on the current number of training times, the hyperbolic tangent function is used to obtain the adaptive inertia weight; For any individual after information fusion and updating, based on the optimal individual, adaptive inertia weight is used to fuse the information of the individual with the optimal individual to obtain the individual after optimal information fusion and updating.

7. The geothermal energy development decision-making method based on big data analysis according to claim 6 is characterized in that: Perform mutually exclusive information fusion update on each individual after optimal information fusion update to obtain the individuals after mutually exclusive information fusion update, including: Obtain the loss function value corresponding to the individual after the optimal information fusion update, and re-obtain the best individual and the worst individual according to the loss function value corresponding to the individual after the optimal information fusion update; According to the centroid position individual and the best individual, a learning information item is obtained; according to the centroid position individual and the worst individual, a mutually exclusive information item is obtained; For any individual after optimal information fusion and updating, mutually exclusive information fusion and updating are performed on the individual according to the learning information item and the mutually exclusive information item to obtain the individual after mutually exclusive information fusion and updating.

8. The geothermal energy development decision-making method based on big data analysis according to claim 7 is characterized in that: The global adaptive random information exploration strategy is used to globally update the optimal individual, and the optimal individual after global update is obtained, including: Randomly matching the first random individual, the second random individual, and the third random individual for the optimal individual; Obtaining a global exploration value corresponding to the optimal individual according to the first random individual, the second random individual, and the third random individual; Determine whether the loss function value corresponding to the global exploration value is less than the loss function value corresponding to the optimal individual. If so, the global exploration value is used as the optimal individual after global update. Otherwise, the original optimal individual is directly used as the optimal individual after global update.

9. The geothermal energy development decision-making method based on big data analysis according to claim 1 is characterized in that: Collect geothermal-related characteristic data corresponding to the target location, and use the trained geothermal energy prediction model to perform geothermal energy prediction to obtain the geothermal energy prediction result corresponding to the target location, including: collecting geothermal-related characteristic data corresponding to the target location, using the geothermal-related characteristic data as the input of the trained geothermal energy prediction model, obtaining the output of the geothermal energy prediction model, and obtaining the geothermal energy prediction result corresponding to the target location.

10. A geothermal energy development decision-making system based on big data analysis, characterized in that: include: Big data acquisition module, big data learning module, geothermal prediction module and auxiliary development decision module; The big data acquisition module is used to collect geothermal energy sample big data that is pre-stored or input by staff; wherein the geothermal energy sample big data includes geothermal energy related feature samples and ground temperature tags; The big data learning module is used to construct a geothermal energy prediction model using a machine learning model, and train the geothermal energy prediction model using geothermal energy related feature samples and geothermal temperature labels to obtain a trained geothermal energy prediction model; The geothermal prediction module is used to collect geothermal related characteristic data corresponding to the target location, and use the trained geothermal energy prediction model to perform geothermal energy prediction to obtain the geothermal energy prediction result corresponding to the target location; The auxiliary development decision module is used to provide the geothermal energy prediction results corresponding to the target location to the staff, so that the staff can make geothermal energy development decisions based on the geothermal energy prediction results.

Images