Intelligent regulation and control method and system for geothermal well group distributed calculation and electronic equipment

Abstract

The invention belongs to the technical field of geothermal well group intelligent control, and discloses an intelligent regulation and control method and system for geothermal well group distributed calculation and electronic equipment. The intelligent regulation and control method for geothermal well group distributed calculation comprises the steps that data perception of a geothermal well group is obtained, a perception data set is established according to the data perception of the geothermal well group, individual output time sequence prediction of individuals in the geothermal well group is configured through the perception data set, energy supply requirements are obtained, and granularity decomposition is conducted on the energy supply requirements. Establishing a geothermal well group operation task according to a granularity decomposition result, establishing a task scheduling evaluation model, performing scheduling optimization on individual output time sequence prediction and the geothermal well group operation task through the task scheduling evaluation model, and performing distributed regulation and control management on the geothermal well group according to a scheduling optimization result; energy consumption can be reduced, heat efficiency can be improved, and heat storage and heat supply stability can be guaranteed.

Description

Technical Field

[0001] The present invention relates to the technical field of intelligent control of geothermal well groups, and in particular to an intelligent control method and system for distributed computing of geothermal well groups, and electronic equipment. Background Art

[0002] The efficient development and utilization of geothermal energy is of great significance to the optimization of energy structure and sustainable development. The rational regulation of geothermal well clusters is the key to ensuring their efficient energy supply. Currently, the main approach to solving the problem of geothermal well cluster regulation is centralized scheduling and management based on experience or simple rules.

[0003] However, due to the lack of accurate perception and prediction of the individual characteristics of geothermal well groups, as well as comprehensive consideration of multi-objective optimization scheduling, the current methods have problems such as high energy consumption, low thermal efficiency, and insufficient heat storage and heating stability in actual operation.

[0004] Among the current related technologies, the regulation of geothermal well groups has technical problems such as high energy consumption, low thermal efficiency, and insufficient heat storage and heat supply stability, which need to be solved. Summary of the Invention

[0005] The purpose of the present invention is to provide an intelligent control method and system for distributed computing of geothermal well groups, and electronic equipment. The present invention solves the technical problems of high energy consumption, low thermal efficiency, and insufficient heat storage and heat supply stability in existing geothermal well group control by obtaining accurate data, configuring individual output timing predictions, creating well group operation tasks, and establishing a multi-objective task scheduling evaluation model and performing scheduling optimization. The technical effects of reducing energy consumption, improving thermal efficiency, and ensuring heat storage and heat supply stability are achieved.

[0006] In order to achieve the above object, the technical solution adopted by the present invention is as follows:

[0007] In a first aspect, the present invention provides an intelligent control method for distributed computing of a geothermal well cluster, comprising the following steps:

[0008] Deploy geothermal well cluster sensing equipment, use the geothermal well cluster sensing equipment to obtain geothermal well cluster data perception, and establish a perception data set based on the geothermal well cluster data perception;

[0009] Individual output time series prediction of individuals in a geothermal well cluster is configured through a perception dataset;

[0010] Obtain energy supply demand, decompose it into granularity, and create geothermal well cluster operation tasks based on the results of the granularity decomposition;

[0011] Establish a task scheduling evaluation model;

[0012] The task scheduling evaluation model is used to optimize the scheduling of individual output time series predictions and geothermal well group operation tasks, and the geothermal well group is managed in a distributed manner based on the scheduling optimization results.

[0013] Furthermore, configuring individual output time series prediction of individuals in the geothermal well group through the perception data set specifically includes:

[0014] Establish standardized weak classifier;

[0015] Collect individual historical data of all individuals in the geothermal well group, conduct reinforcement learning of standardized weak classifiers based on the collected individual historical data, and configure individual prediction classifiers based on the standardized weak classifiers after reinforcement learning;

[0016] Input the perception data set into the corresponding individual prediction classifier to perform time series prediction and establish individual output time series prediction;

[0017] Furthermore, the granular decomposition of energy supply demand specifically includes:

[0018] Decompose the energy demand into dimensional granularity based on spatial partition, time scale, demand type, and response capability to obtain the dimensional granularity decomposition results. Then, perform adaptive task clustering on the dimensional granularity decomposition results by calling the task classifier, and establish the granularity decomposition results based on the results of the adaptive task clustering.

[0019] Furthermore, the scheduling evaluation objectives of the task scheduling evaluation model include minimizing energy consumption objectives, maximizing thermal efficiency objectives, thermal storage stability objectives, and heating supply stability objectives;

[0020] Furthermore, the task scheduling evaluation model is used to optimize the scheduling of individual output time series predictions and geothermal well group operation tasks, specifically including:

[0021] Read the scheduling log of the geothermal well group and generate scheduling constraints based on the scheduling log;

[0022] Create a solution space, and perform task allocation of geothermal well cluster operation tasks within the solution space through scheduling constraints to generate an initial solution;

[0023] Perform scheduling adaptation analysis on the initial solution and individual output timing prediction through the task scheduling evaluation model to generate evaluation results;

[0024] Based on the evaluation results, iterative updates are performed under the solution space and scheduling constraints to complete the scheduling optimization;

[0025] Furthermore, performing iterative updates based on the evaluation results under the solution space and scheduling constraints to complete the scheduling optimization specifically includes:

[0026] Each geothermal well cluster is treated as a scheduling game individual. Based on the game scheduling mechanism, several rounds of strategy evolution between scheduling game individuals are carried out under the solution space and scheduling constraints. The strategy evolution is jointly driven by its own benefits and evaluation results.

[0027] When the strategy evolution between individuals in several rounds of scheduling game converges, the scheduling optimization is completed;

[0028] Furthermore, the method of establishing a perception data set based on data perception of the geothermal well group further includes, after establishing the perception data set based on the data perception of the geothermal well group, performing status analysis of individual geothermal well groups based on the perception data set, establishing an initial status warning, and reporting abnormalities of individual geothermal well groups based on the initial status warning;

[0029] Furthermore, the distributed control and management of the geothermal well group according to the results of the scheduling optimization also includes:

[0030] After distributed control and management of the geothermal well group is performed based on the results of the scheduling optimization, continuous sensing data of the geothermal well group is obtained, and a continuous sensing data set is established based on the continuous sensing data of the geothermal well group;

[0031] Deviation identification is performed through continuous perception data sets and individual output time series prediction. Deviation warnings are created based on the results of deviation identification and the deviation warnings are issued.

[0032] In a second aspect, the present invention provides an intelligent control system for distributed computing of a geothermal well cluster, comprising:

[0033] A geothermal well group data perception module is used to obtain data perception of the geothermal well group and establish a perception data set based on the data perception of the geothermal well group;

[0034] An individual output time series prediction module is used to configure the individual output time series prediction of individuals in the geothermal well group through the perception data set;

[0035] The geothermal well group operation task creation module is used to obtain energy supply demand, decompose the energy supply demand into granularity, and create geothermal well group operation tasks based on the results of the granularity decomposition;

[0036] Task scheduling evaluation model establishment module, used to establish a task scheduling evaluation model;

[0037] The scheduling optimization and distributed control management module is used to perform scheduling optimization on individual output time series prediction and geothermal well group operation tasks through the task scheduling evaluation model, and to perform distributed control management of the geothermal well group based on the results of the scheduling optimization.

[0038] In a third aspect, the present invention provides an electronic device, comprising:

[0039] a memory for storing executable instructions;

[0040] The processor is configured to implement the intelligent control method for distributed computing of a geothermal well group according to any one of claims 1 to 8 when executing the executable instructions stored in the memory.

[0041] The above technical solution has the following advantages or beneficial effects:

[0042] In its first aspect, the present invention provides an intelligent control method for distributed computing of geothermal well clusters. The method first senses geothermal well cluster data and establishes a sensed dataset. The sensed dataset is then used to configure individual output time series predictions for each well cluster. After obtaining energy demand, the method then performs granular decomposition of the energy demand. The granular decomposition results are used to create well cluster operation tasks. A task scheduling evaluation model is then established, where the scheduling evaluation objectives of the task scheduling evaluation model include minimizing energy consumption, maximizing thermal efficiency, stabilizing thermal storage, and stabilizing thermal supply. Finally, the task scheduling evaluation model is used to perform scheduling optimization based on individual output time series predictions and the well cluster operation tasks. Distributed scheduling management is then performed based on the scheduling optimization results. This method achieves the technical effects of reducing energy consumption, improving thermal efficiency, and ensuring thermal storage and thermal supply stability.

[0043] Furthermore, in terms of energy consumption, traditional control methods may result in unnecessary energy loss in geothermal well operation due to the lack of accurate prediction. However, this method establishes a standardized weak classifier, uses individual historical data for reinforcement learning, and configures an individual prediction classifier, which can accurately predict the individual output timing, thereby rationally regulating the operation of each well, avoiding energy waste, and effectively reducing overall energy consumption; in terms of thermal efficiency, accurate timing prediction enables the geothermal well group to operate in the best state according to actual needs, fully tap the potential of geothermal resources, reduce heat loss during transmission and use, significantly improve thermal efficiency, and achieve efficient utilization of geothermal resources; in terms of heat storage and heat supply stability, this method can grasp the output status of each well in advance, make resource allocation and response measures in advance, avoid heat storage fluctuations and heat supply instability caused by insufficient or excessive heat supply, effectively ensure the stability of heat storage and heat supply, provide users with reliable heating services, and improve the operation quality and reliability of the entire geothermal well group system.

[0044] Furthermore, spatial zoning of energy supply demand can accurately control the specific energy supply conditions in different regions, avoid blind resource allocation, enable each region to obtain energy supply support that matches its needs, and improve the rationality and efficiency of energy supply; granular decomposition on the time scale can flexibly adjust the energy supply strategy according to changes in energy supply demand in different time periods; through adaptive task clustering to establish granular decomposition results, the allocation of energy supply tasks can be further optimized, the complexity of regulation can be reduced, the stable and efficient operation of the geothermal well group energy supply system can be ensured, and the optimal configuration and utilization of energy can be achieved.

[0045] Furthermore, by incorporating minimization of energy consumption into the evaluation objectives, the scheduling model is prompted to give priority to low-energy consumption operating plans when allocating tasks, reasonably plan the operating time and intensity of each well, and avoid unnecessary energy waste; the goal of maximizing thermal efficiency ensures the efficient conversion of energy. This goal drives the scheduling model to explore the optimal operating state of the geothermal well group, so that heat can be more fully extracted and utilized, reduce heat loss during transmission and use, improve the efficiency of converting geothermal resources into useful thermal energy, and realize the efficient development of geothermal resources; stable heat storage ensures the sustainable supply of geothermal resources and avoids the depletion of heat storage due to excessive exploitation; stable heat supply provides users with continuous and stable heat supply services, reduces the inconvenience caused to users by heat supply fluctuations, and enhances the stability and reliability of the entire geothermal well group energy supply system.

[0046] Furthermore, the scheduling log is read and scheduling constraints are generated, which converts past experience into specific constraints, providing a reliable basis for subsequent scheduling, avoiding blind scheduling, and ensuring that the scheduling plan conforms to the actual operating rules; creating a solution space and generating an initial solution provides a broad exploration space for scheduling optimization. By allocating tasks according to scheduling constraints in the solution space, a feasible preliminary scheduling plan can be quickly obtained, laying the foundation for subsequent optimization; using the task scheduling evaluation model to perform scheduling adaptation analysis on the initial solution and individual output timing prediction, the degree of matching between the scheduling plan and actual needs can be accurately evaluated, potential problems can be discovered in time, and a clear direction for optimization can be provided; based on the evaluation results, iterative updates under the solution space and scheduling constraints can continuously optimize the scheduling plan and gradually approach the optimal solution, effectively improving the operating efficiency of the geothermal well group, reducing energy consumption, ensuring the stability of heat storage and heat supply, realizing efficient and intelligent regulation of geothermal resources, and providing strong support for the rational use of geothermal energy.

[0047] Furthermore, each individual geothermal well group is regarded as a scheduling game individual, and multiple rounds of strategy evolution are carried out based on the game scheduling mechanism, so that each well can independently adjust its strategy according to its own benefits and evaluation results, which can more truly reflect the actual operation situation and make the scheduling plan more adaptable and flexible; the strategy evolution is jointly driven by its own benefits and evaluation results, which not only takes into account the individual's interests, but also takes into account the overall evaluation standards, and achieves a balance between local optimization and global optimization. Through multiple rounds of evolution, each well continuously adjusts its strategy to achieve the optimal state, avoiding the local optimal trap that a single optimization method may fall into; when the strategy evolution converges and the scheduling optimization is completed, it means that a relatively stable and optimal scheduling plan has been found, which can effectively reduce the overall energy consumption of the geothermal well group, improve thermal efficiency, ensure the stability of heat storage and heat supply, and provide scientific and intelligent decision-making support for the rational and efficient use of geothermal energy.

[0048] Furthermore, by analyzing the individual status of geothermal well clusters based on the perception data set, the operating status of each well can be grasped in real time. Once abnormal signs are found, an initial status warning can be immediately established, allowing staff to detect potential problems in advance and take preventive measures before a failure occurs, avoiding the expansion of the failure, reducing downtime and maintenance costs caused by equipment failure, and ensuring the stable operation of the geothermal well cluster.

[0049] Furthermore, by using continuous perception data sets and individual output time series predictions for deviation identification, it is possible to quickly determine the deviation between the actual operating conditions and the expected ones. Once a deviation is detected, a deviation warning is immediately created and reported, allowing staff to respond at the early stages of the problem development and take timely measures to make adjustments to avoid the problem from worsening, effectively shortening the abnormal response time.

[0050] Secondly, the present invention provides an intelligent control system for distributed computing of geothermal well groups. The geothermal well group data perception module can accurately obtain geothermal well group data and establish a perception data set, providing a reliable data basis for the entire system, ensuring that subsequent analysis and decision-making have a basis, and helping to fully understand the operating status of the geothermal well group; the individual output timing prediction module uses the perception data set to make predictions, and can grasp the output of each well in advance, facilitating the reasonable planning of resource allocation, improving energy utilization efficiency, reducing resource waste, and providing guarantees for energy supply stability; the geothermal well group operation task creation module granularly decomposes energy supply demand and creates tasks, making task allocation more scientific and reasonable, and fully considering the needs of different spaces, times, types and response capabilities, and improving the quality of energy supply services; the task scheduling evaluation model establishment module provides standards for scheduling optimization, ensuring the scientificity and effectiveness of the scheduling plan; the scheduling optimization and distributed control management module can optimize the operation of the geothermal well group, reduce energy consumption, improve thermal efficiency, ensure stable heat storage and heat supply, realize efficient and intelligent utilization of geothermal resources, and improve the operating quality and economic benefits of the entire system through model optimization and distributed control.

[0051] On the third aspect, the present invention provides an electronic device that relies on a memory to store executable instructions and a processor to execute instructions to implement an intelligent control method for distributed computing of geothermal well groups, thereby realizing intelligent and automated control. On the one hand, it can accurately perceive geothermal well group data, reasonably configure individual output timing predictions, and create scientific tasks based on the granularity of energy supply demand decomposition, thereby greatly improving the accuracy and scientific nature of control, reducing energy consumption, and improving thermal efficiency. On the other hand, through task scheduling evaluation models and distributed control management, it can respond to operational changes in a timely manner to ensure stable heat storage and heat supply. Moreover, the electronic device can be flexibly deployed to facilitate the control of geothermal well groups of different sizes and layouts, reduce labor costs, and improve the economy and reliability of the overall operation. BRIEF DESCRIPTION OF THE DRAWINGS

[0052] Figure 1This is a flow chart of an intelligent control method for distributed computing of a geothermal well cluster according to the present invention;

[0053] Figure 2 This is a schematic diagram of the structure of an intelligent control system for distributed computing of a geothermal well group according to the present invention;

[0054] Figure 3 It is a structural schematic diagram of the computer device of the present invention;

[0055] In the figure, 10 is a geothermal well group data perception module; 20 is an individual output timing prediction module; 30 is a geothermal well group operation task creation module; 40 is a task scheduling evaluation model establishment module; 50 is a scheduling optimization and distributed control management module; 301 is an input device; 302 is a processor; 303 is a memory; 304 is an output device. DETAILED DESCRIPTION

[0056] The above description is only an overview of the technical solution of the present application. In order to more clearly understand the technical means of the present application, it can be implemented in accordance with the contents of the specification. In order to make the above and other purposes, features and advantages of the present application more obvious and easy to understand, the specific implementation methods of the present application are listed below.

[0057] In order to make the purpose, technical solutions and advantages of this application clearer, the application will be further described in detail below with reference to the accompanying drawings. The described embodiments should not be regarded as limiting this application. All other embodiments obtained by ordinary technicians in this field without making creative work are within the scope of protection of this application.

[0058] In the following description, reference is made to “some embodiments” which describe a subset of all possible embodiments, but it will be understood that “some embodiments” may be the same subset or different subsets of all possible embodiments and may be combined with each other without conflict. The terms “including” and “having” and any variations thereof are intended to cover non-exclusive inclusions. For example, a process, method, system, product or server that includes a series of steps or units is not necessarily limited to those steps or units clearly listed, but may include other steps or modules that are not clearly listed or inherent to these processes, methods, products or devices. Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by those skilled in the art to which this application belongs. The terms used herein are for the purpose of describing the embodiments of this application only.

[0059] Example:

[0060] See also Figure 1 The present invention provides an intelligent control method for distributed computing of a geothermal well group, comprising the following steps:

[0061] Step S100: deploying geothermal well group sensing equipment, obtaining data perception of the geothermal well group through the geothermal well group sensing equipment, and establishing a sensing data set based on the data perception of the geothermal well group;

[0062] Preferably, the geothermal well cluster sensing device is a device used to monitor the operating status of geothermal wells, including temperature sensors, pressure sensors, flow sensors, etc.; multiple sensors are deployed at key locations of each geothermal well (such as the wellhead and different depths underground), and the data collected by the sensors are transmitted in real time to a local or cloud server through a data acquisition module (such as a data acquisition card, an industrial Internet of Things gateway); a database management system (such as MySQL, MongoDB) is used to store the collected data and establish a sensing data set for data analysis and modeling;

[0063] Preferably, the temperature sensor adopts a high-precision Pt100 platinum resistance temperature sensor, the pressure sensor adopts a piezoresistive pressure sensor, and the flow sensor adopts an electromagnetic flowmeter; a Siemens S7-1200 PLC is used as a data acquisition module, and the sensor data is uploaded to the server via the Modbus protocol; a MySQL database is used to establish a data table geothermal_data, with fields including timestamp, well_id (well number), temperature, pressure, flow_rate, etc.;

[0064] In one possible implementation, after establishing a perception data set based on data perception of a geothermal well group, the method further includes performing status analysis of individual geothermal well groups based on the perception data set, establishing an initial status warning, and reporting abnormalities of individual geothermal well groups based on the initial status warning.

[0065] Specifically, features related to the status of individual well groups are extracted from the perception data set, such as the rate of change and fluctuation range of parameters such as temperature, pressure, and flow rate. A state analysis model is established, and machine learning algorithms (such as support vector machines and random forests) or statistical analysis methods (such as cluster analysis and principal component analysis) can be used to evaluate the status of individual well groups. For example, the temperature change rate (ΔT / Δt), pressure fluctuation range (maximum pressure - minimum pressure), mean and variance of flow rate, and other features of each geothermal well are calculated. A support vector machine (SVM) is used to establish a state analysis model. The extracted features are used as input, and the model outputs the current status of the individual well group (normal or abnormal).

[0066] Specifically, based on the output of the state analysis model, a threshold for the initial state warning is set. When the state indicator output by the model exceeds the threshold range, the initial state warning is triggered. An early warning mechanism is established to promptly issue an alarm signal when the initial state warning is triggered. For example, the threshold for the temperature change rate is set to ±5°C / h, the threshold for the pressure fluctuation range is set to ±0.5MPa, and the thresholds for the mean and variance of the flow rate are set to ±10% and ±20%, respectively. When the output of the state analysis model exceeds the set threshold, an alarm signal is issued via SMS, email, or the alarm module of the industrial automation system.

[0067] Specifically, further anomaly detection is performed on individual wells that trigger initial status warnings, and the specific causes of the anomalies are analyzed. Abnormal information is reported promptly, and can be displayed using a visual interface or sent to maintenance personnel through an automated system. For example, for geothermal wells that trigger initial status warnings, the changing trends of parameters such as temperature, pressure, and flow are further analyzed to determine whether the anomaly is caused by equipment failure, thermal storage anomalies, or other reasons. The location and type of abnormal well are displayed with a red alarm sign through the industrial automation system's visual interface. At the same time, the abnormality information is sent to maintenance personnel via text message or email for timely processing.

[0068] Step S200 , configuring individual output time series predictions of individuals in a geothermal well cluster through a perception data set;

[0069] Preferably, individual output time series prediction refers to time series prediction of the output (e.g., heat, flow) of each geothermal well, which is used to plan the operation strategy of the well group in advance. The perception data set is cleaned (noise data is removed and missing values ​​are filled) and normalized. Machine learning or deep learning algorithms (e.g., LSTM, ARIMA) are used to perform time series prediction of the output (e.g., heat, flow) of each geothermal well. The model is trained using historical data, and the performance of the model is evaluated through methods such as cross-validation to ensure the accuracy of the prediction.

[0070] Preferably, the data is cleaned and normalized using Python's Pandas library, and missing values ​​are filled using linear interpolation. An LSTM neural network is used for time series prediction, and an LSTM model is constructed using the TensorFlow framework. Input features include historical temperature, pressure, flow, etc., and the output is a predicted value of heat or flow in the future. The data set is divided into a training set and a test set. The model is trained using the training set, and the prediction performance of the model is verified using the test set. The mean square error (MSE) is used as the evaluation metric.

[0071] In one possible implementation, configuring individual output time series predictions for individuals in a geothermal well cluster using the perception data set, that is, step S200 further includes:

[0072] Step S210, establishing a standardized weak classifier;

[0073] Step S220: collecting individual historical data of all individuals in the geothermal well group, performing reinforcement learning of a standardized weak classifier based on the collected individual historical data, and configuring an individual prediction classifier based on the standardized weak classifier after reinforcement learning;

[0074] Step S230, inputting the perception data set into the individual prediction classifier to perform time series prediction and establish individual output time series prediction;

[0075] Preferably, in step S210, a standardized weak classifier is established, which includes: selecting a weak classifier suitable for time series prediction, such as a decision tree, a linear regression model, etc.; standardizing the weak classifier to ensure that its output is within the same scale range, which is convenient for subsequent ensemble learning. For example, a decision tree with a depth of 3 is selected as a weak classifier. The decision tree model is simple, easy to understand and implement, and suitable for processing time series data; standardizing the output of the decision tree to normalize the output value to the [0,1] interval; using the formula Perform normalization;

[0076] Preferably, step S220 collects individual historical data for all individuals in the geothermal well group, performs reinforcement learning on a standardized weak classifier using the collected individual historical data, and configures an individual prediction classifier using the standardized weak classifier after reinforcement learning, including: extracting historical data of each geothermal well from the perception data set, including parameters such as temperature, pressure, and flow, performing reinforcement learning on the standardized weak classifier using the individual historical data, optimizing the performance of the weak classifier through a reward mechanism, and configuring the individual prediction classifier; for example, extracting temperature, pressure, and flow data of each geothermal well over the past year from the perception data set as individual historical data, performing reinforcement learning using a Q-learning algorithm, and then defining a reward function, giving a positive reward when the prediction error of the weak classifier is less than a certain threshold, otherwise giving a negative reward, and gradually optimizing the weak classifier under the reward mechanism by continuously adjusting the parameters of the weak classifier;

[0077] Preferably, step S230, inputting the sensed dataset into an individual prediction classifier to perform time series prediction and establish an individual output time series prediction, includes: inputting the sensed dataset into an individual prediction classifier optimized by reinforcement learning, using the individual prediction classifier to perform time series prediction, and outputting the future output (e.g., heat, flow) of each geothermal well; for example, inputting real-time collected temperature, pressure, and flow data into the individual prediction classifier. The individual prediction classifier outputs predicted heat and flow values ​​for each geothermal well within the next 24 hours.

[0078] Step S300: Obtain energy supply demand, perform granular decomposition on the energy supply demand, and create geothermal well group operation tasks based on the results of the granular decomposition;

[0079] Preferably, a demand analysis is performed based on the user's energy supply demand (such as heating area and heating time) to determine the total energy supply demand, and the total energy supply demand is decomposed into multiple subtasks, each subtask corresponding to the operation task of one or more geothermal wells; granularity decomposition can be performed based on time (such as hours, days) or space (such as regions); based on the granularity decomposition results, the subtasks are assigned to specific geothermal wells to generate well group operation tasks; for example, a user needs to heat a 10,000 square meter building in winter, with a heating time of 8 hours a day. The heating time is decomposed into hourly energy demand, and based on the hourly heat demand, the task is decomposed into multiple subtasks, each subtask corresponding to the operation task of one or more geothermal wells, and the subtasks are assigned to specific geothermal wells based on the predicted output capacity of each geothermal well. A task allocation algorithm (such as a greedy algorithm) is used for task allocation to ensure that the operation tasks of each geothermal well are reasonable;

[0080] In one possible implementation, the granular decomposition of energy supply demand further includes: performing dimensional granular decomposition of the energy supply demand based on spatial partition, time scale, demand type, and response capability to obtain dimensional granular decomposition results, performing adaptive task clustering on the dimensional granular decomposition results by calling a task classifier, and establishing granular decomposition results based on the results of the adaptive task clustering;

[0081] Preferably, the energy supply demand is decomposed into dimensional granularity of spatial partition, time scale, demand type, and response capability to obtain the dimensional granularity decomposition result, including: defining a multidimensional granularity space, including four dimensions of spatial partition, time scale, demand type, and response capability, and then using a multidimensional decomposition algorithm to decompose the energy supply demand into the granularity space of each dimension, and dividing the energy supply area into multiple sub-areas according to geographic information system (GIS) data; for example, for spatial partition, the energy supply area is divided into multiple sub-areas, such as area A, area B, area C, etc., each area The sub-regions correspond to different geothermal well groups. For the time scale, the energy supply time is divided into multiple time scales such as hours, days, and weeks. For the hourly scale, it can be further subdivided into peak hours (such as 8:00-10:00) and valley hours (such as 22:00-6:00). For the demand type, the energy supply demand is divided into heating, hot water supply, power generation and other types. Different types of energy supply demand have different requirements for the operating parameters of geothermal wells. For the response capability, the output capacity of the geothermal well (such as heat, flow) and the equipment status (such as normal operation) are used to determine the energy supply demand. , partial failure), geothermal wells are divided into three levels of high responsiveness, medium responsiveness, and low responsiveness; for spatial partition decomposition, for example, the K-means clustering algorithm is used to divide geothermal wells into different sub-areas according to their location and energy demand; for time scale decomposition, the energy supply time is divided into different time scales according to historical data and user needs; for example, the time series analysis method is used to analyze the changes in energy demand in different time periods, and the energy supply time is divided into peak period, off-peak period, and valley period; for demand type decomposition, the total energy demand is decomposed into different types of sub-demands according to the type of energy demand; for example, a classification algorithm (such as decision tree, support vector machine) is used to classify energy demand, and heating demand, hot water supply demand, and power generation demand are extracted respectively; for response capability decomposition, geothermal wells are divided into different response capability levels according to their output capacity and equipment status; for example, a performance evaluation model is used to evaluate their response capability based on their real-time data (such as temperature, pressure, flow), and geothermal wells are divided into three levels: high, medium, and low;

[0082] Preferably, the dimensional granularity decomposition results are adaptively clustered by calling a task classifier, and the granularity decomposition results are established according to the results of the adaptive task clustering, including: selecting an algorithm suitable for multi-task classification, such as a multi-task learning (MTL) algorithm or a deep learning model (such as a multi-task neural network); inputting the multi-dimensional granularity space decomposition results (including information on dimensions such as spatial partitioning, time scale, demand type, and response capability) into the multi-task classifier; the multi-task classifier automatically clusters the tasks based on the input multi-granularity space decomposition results, and the basis for clustering is the similarity between tasks. For example, tasks with similar spatial partitioning, time scale, demand type, and response capability will be clustered into one category; the multi-task classifier can dynamically adjust the clustering results according to real-time data to ensure the accuracy and adaptability of task clustering, and the multi-task classifier outputs the task clustering results, and each task cluster corresponds to the operating tasks of one or more geothermal wells; according to the task clustering results, the tasks are assigned to specific geothermal wells to establish the granularity decomposition results;

[0083] For example, a multi-task learning algorithm is selected, such as a deep neural network with shared hidden layers. This network structure can learn multiple tasks simultaneously, share underlying features, and improve the generalization ability of the model; the input data includes the spatial partitioning information of each geothermal well (such as area A, area B), time scale information (such as peak hours, off-peak hours), demand type information (such as heating, hot water supply), response capability information (such as high response capability, low response capability), etc.; the multi-task classifier clusters tasks with similar characteristics into one category based on the multi-granularity spatial decomposition results of the input; specifically, if multiple tasks belong to area A, peak hours, heating demand, and high response capability, these tasks will be clustered into a task cluster.

[0084] Step S400, establishing a task scheduling evaluation model;

[0085] Specifically, the scheduling evaluation objectives of the task scheduling evaluation model include minimizing energy consumption, maximizing thermal efficiency, stabilizing heat storage, and stabilizing heat supply.

[0086] Preferably, the task scheduling evaluation model is a model for evaluating and optimizing scheduling schemes, which comprehensively considers multiple objectives such as energy consumption, thermal efficiency, thermal storage stability, and heating supply stability to construct a task scheduling evaluation model, and adopts a multi-objective optimization algorithm (such as genetic algorithm, particle swarm optimization algorithm) to evaluate and optimize the scheduling scheme. According to the actual operation data, the model parameters are adjusted to ensure the accuracy and practicality of the model; for example, the energy consumption target is defined as the amount of electricity or fuel consumed per unit time, the thermal efficiency target is defined as the ratio of output heat to input energy, the thermal storage stability target is defined as the rate of change of the thermal storage temperature, and the heating supply stability target is defined as the fluctuation range of the heating temperature. The task scheduling evaluation model is constructed using a genetic algorithm, and a fitness function is defined. Indicators such as energy consumption, thermal efficiency, thermal storage stability, and heating supply stability are comprehensively considered. The optimal scheduling scheme is found through the crossover, mutation, and selection operations of the genetic algorithm. According to the actual operation data, the crossover probability, mutation probability and other parameters of the genetic algorithm are adjusted to ensure the optimization effect of the model.

[0087] Step S500: Optimizing the scheduling of individual output time series predictions and geothermal well group operation tasks through a task scheduling evaluation model, and performing distributed control and management of the geothermal well group based on the scheduling optimization results;

[0088] Preferably, a task scheduling evaluation model is used, combined with individual output time series prediction and well group operation tasks, to perform scheduling optimization, and a multi-objective optimization algorithm is used to find the optimal scheduling solution. Based on the scheduling optimization results, a distributed control system (such as an industrial automation control system) is used to perform real-time scheduling management of the geothermal well group, monitor the operating status of the geothermal well group in real time, and adjust the scheduling solution based on feedback data to ensure stable operation of the system. For example, a particle swarm optimization algorithm is used for scheduling optimization, the particle swarm is initialized, each particle represents a scheduling solution, and the optimal scheduling solution is found through iterative optimization of the particle swarm. Siemens S7-1500 PLC is used as a distributed control system. Based on the scheduling optimization results, the operating parameters of each geothermal well (such as pump speed, valve opening) are adjusted in real time. The operating status of the geothermal well group is monitored in real time through an industrial Internet of Things platform (such as Siemens MindSphere), and the scheduling solution is adjusted based on feedback data to ensure stable operation of the system.

[0089] In one possible implementation, the task scheduling evaluation model is used to optimize the scheduling of individual output time series predictions and geothermal well group operation tasks, further including:

[0090] Step S510, reading the scheduling log of the geothermal well group, and generating scheduling constraints according to the scheduling log;

[0091] Step S520 , creating a solution space, and performing task allocation of geothermal well group operation tasks within the solution space by scheduling constraints to generate an initial solution;

[0092] Step S530 , performing scheduling adaptation analysis on the initial solution and the individual output timing prediction using a task scheduling evaluation model to generate an evaluation result;

[0093] Step S540, performing iterative updates based on the evaluation results in the solution space and the scheduling constraints to complete the scheduling optimization;

[0094] Preferably, step S510, reading the scheduling log of the geothermal well group, includes: setting a log recording module in the system to record the scheduling operation of the geothermal well group in real time, including information such as scheduling time, scheduling task, and scheduling result, and reading the historical scheduling log from the log storage system through the log reading module; for example, using a relational database (such as MySQL) to store the scheduling log, the log table schedule_log includes the fields: log_id (log number), schedule_time (scheduling time), task_id (task number), well_id (geothermal well number), schedule_result (scheduling result), etc., and reading the scheduling log of the past month from the database through SQL query statements for analyzing the historical scheduling situation;

[0095] Preferably, step S510 generates scheduling constraints based on the scheduling log. Scheduling constraints refer to restrictive conditions obtained from analyzing historical scheduling logs and used to guide the allocation of scheduling tasks for the geothermal well group. The steps include: analyzing the scheduling log to extract scheduling-related constraints, such as task priority, geothermal well operating status, equipment maintenance time, etc., and generating scheduling constraint rules based on the extracted constraints to guide the allocation of scheduling tasks. For example, if it is analyzed from the scheduling log that some tasks have high priority (such as emergency heating tasks) and some geothermal wells require maintenance within a specific time period (such as monthly equipment maintenance time), scheduling constraint rules are generated, such as "high-priority tasks are allocated first" and "geothermal wells during maintenance are not involved in scheduling".

[0096] Preferably, step S520, creating a solution space, and performing task allocation of geothermal well group operation tasks within the solution space according to scheduling constraints to generate an initial solution, includes: defining a solution space for scheduling tasks, that is, a set of all possible scheduling schemes, and performing task allocation within the solution space according to the scheduling constraints to generate an initial solution; for example, the solution space can be represented as a matrix, where rows represent geothermal wells, columns represent time points, and elements in the matrix represent task allocation status of the geothermal wells at the time points; performing task allocation within the solution space according to the scheduling constraints (e.g., prioritizing high-priority tasks), and the initial solution can be a simple greedy allocation scheme that preferentially allocates high-priority tasks to geothermal wells with high responsiveness;

[0097] Preferably, step S530, performing a scheduling adaptation analysis on the initial solution and the individual output timing prediction through the task scheduling evaluation model to generate an evaluation result, including: inputting the initial solution and the individual output timing prediction into the task scheduling evaluation model to perform a scheduling adaptation analysis. Based on the output of the evaluation model, an evaluation result is generated, including scores for indicators such as energy consumption, thermal efficiency, thermal storage stability, and heating stability; for example, the initial solution and the individual output timing prediction are input into the task scheduling evaluation model based on the genetic algorithm, and the model is evaluated according to preset evaluation indicators (such as minimizing energy consumption and maximizing thermal efficiency). The evaluation results show that the energy consumption of the current initial solution is 100 units, the thermal efficiency is 85%, the thermal storage stability score is 70, and the heating stability score is 80;

[0098] Preferably, step S540 performs iterative updates in the solution space and the scheduling constraints based on the evaluation results to complete the scheduling optimization, including: performing iterative updates in the solution space according to the evaluation results to find a better scheduling solution, and using an optimization algorithm (such as a genetic algorithm or a particle swarm optimization algorithm) to perform iterative optimization until preset optimization conditions (such as the number of iterations and the evaluation index threshold) are met; for example, using a genetic algorithm for iterative updates, selecting a solution with high fitness based on the evaluation results for crossover and mutation operations to generate a new solution, the crossover probability of the genetic algorithm is set to 0.8, and the mutation probability is set to 0.1, and through multiple generations of iterations, the scheduling solution is gradually optimized to eventually find the optimal solution that meets the evaluation index;

[0099] In one possible implementation, step S540, performing iterative updates under the solution space and scheduling constraints based on the evaluation results to complete the scheduling optimization, further includes:

[0100] Step S541: Treat each geothermal well cluster as a scheduling game individual, and conduct several rounds of strategy evolution between the scheduling game individuals based on the game scheduling mechanism, wherein the strategy evolution is jointly driven by its own benefits and evaluation results;

[0101] Step S542: After several rounds of scheduling game strategies have evolved and converged, the scheduling optimization is completed.

[0102] Preferably, in step S541, each geothermal well group individual is regarded as a scheduling game individual, and several rounds of strategy evolution between scheduling game individuals are carried out based on the game scheduling mechanism under the solution space and scheduling constraints, wherein the strategy evolution is jointly driven by its own benefits and evaluation results, including: treating each geothermal well as an independent scheduling game individual, and the goal of each individual is to maximize its own benefits under the premise of satisfying the scheduling constraints; designing the game scheduling mechanism, defining the interaction rules and benefit function between individuals, and the benefit function may include multiple evaluation indicators such as energy consumption, thermal efficiency, thermal storage stability, and heating supply stability; based on the game mechanism, individuals perform strategy evolution according to their own benefits and evaluation results, and the strategy evolution is driven by iteratively updating the individual benefits. The scheduling strategy of each individual can adapt to the strategy changes of other individuals. For example, geothermal wells A, B, and C are regarded as game individuals 1, 2, and 3 respectively. The goal of each individual is to maximize its own benefits while satisfying the scheduling constraints. The interaction rule between individuals is defined as "prioritize high-priority tasks" and the benefit function is benefit = α × thermal efficiency + β × thermal storage stability + γ × heating stability - δ × energy consumption, where α, β, γ, and δ are weight coefficients. According to the current scheduling strategy and benefit function, individual 1 adjusts its own scheduling strategy. If individual 1 finds that adjusting the strategy can improve its own benefits, it will update its strategy. At the same time, individuals 2 and 3 will also adjust their strategies based on their own benefits and evaluation results.

[0103] Preferably, step S542, when the strategy evolution between the individuals of several rounds of scheduling game converges, the scheduling optimization is completed, including: setting the convergence conditions of the game evolution, such as the strategy change is less than a certain threshold, the number of iterations reaches a preset value, etc., in each round of the game, according to the individual strategy adjustment result, the scheduling plan is updated, the game process is repeated until the convergence conditions are met, and when the game evolution converges, the final scheduling plan is output to complete the scheduling optimization; for example, the strategy change threshold is set to 0.01, and the number of iterations is capped at 100 rounds. In each round of the game, the individual adjusts the strategy according to its own benefits and evaluation results. If the strategy change is less than 0.01, it is considered that the game evolution has converged, otherwise, the next round of the game is continued. Assuming that after 30 rounds of game evolution, the individual strategy change is less than 0.01, the game evolution converges, at this time, the final scheduling plan is output to complete the scheduling optimization;

[0104] In one possible implementation, step S500, performing distributed control and management on the geothermal well group according to the results of the scheduling optimization, further includes:

[0105] Step S550, acquiring continuous sensing data of the geothermal well group, and establishing a continuous sensing data set based on the continuous sensing data of the geothermal well group;

[0106] Step S560 , performing deviation identification through continuous sensing data sets and individual output time series prediction, creating a deviation warning based on the deviation identification result, and issuing the deviation warning to achieve distributed control and management of the geothermal well group;

[0107] Preferably, step S550, obtaining continuous sensing data of the geothermal well group, and establishing a continuous sensing data set based on the continuous sensing data of the geothermal well group, includes: during the operation of the geothermal well group, continuously using sensing devices (such as temperature sensors, pressure sensors, flow sensors, etc.) to collect real-time data, storing the collected real-time data in a database, and forming a continuous sensing data set;

[0108] Preferably, step S560 performs deviation identification through continuous sensing data set and individual output time series prediction, creates deviation warning according to the result of deviation identification, and issues deviation warning to realize distributed control and management of geothermal well group, including: constructing a deviation identification model to compare the difference between continuous sensing data and individual output time series prediction, and identifying significant deviation between actual operation data and predicted data through statistical analysis or machine learning methods (such as anomaly detection algorithm); for example, using a statistical analysis method based on Z-score to perform deviation identification, calculating the difference between continuous sensing data and individual output time series prediction, and normalizing it to a Z-score value; for example, setting the Z-score threshold to ±2, and when the Z-score exceeds the threshold, it is considered that there is a deviation. In the event of a significant deviation, specifically, if the difference between the actual temperature of a geothermal well and the predicted temperature exceeds 2 standard deviations, deviation recognition is triggered; based on the deviation recognition result, a deviation warning is created, and the warning mechanism may include the generation and sending of alarm signals, and the deviation warning is reported through various means (such as SMS, email, and the alarm module of the industrial automation system) to notify relevant personnel or systems; for example, when the deviation recognition model detects a significant deviation, an alarm signal is generated, and the alarm signal may include specific information about the deviation, such as the geothermal well number, deviation type (such as temperature anomaly, pressure anomaly), degree of deviation, etc. The deviation information is displayed with a red alarm sign through the visual interface of the industrial automation system. At the same time, the deviation warning is sent to maintenance personnel via SMS or email so that timely measures can be taken.

[0109] The embodiments of the present application adopt technical means such as deploying sensing devices to obtain accurate data, configuring individual output timing predictions, creating well group operation tasks, and establishing a multi-objective task scheduling evaluation model and performing scheduling optimization, which solves the technical problems of high energy consumption, low thermal efficiency, and insufficient heat storage and heat supply stability in the existing geothermal well group control, and achieves the technical effects of reducing energy consumption, improving thermal efficiency, and ensuring heat storage and heat supply stability.

[0110] In the above, refer to Figure 1The intelligent control method for geothermal well group distributed computing according to the embodiment of the present invention is described in detail. Figure 2 An intelligent control system for distributed computing of a geothermal well cluster according to an embodiment of the present invention is described.

[0111] In one embodiment of the present invention, a geothermal well cluster distributed computing intelligent control system is provided. The geothermal well cluster distributed computing intelligent control system according to the embodiment of the present invention is used to solve the technical problems of high energy consumption, low thermal efficiency, and insufficient heat storage and heat supply stability in existing geothermal well cluster control, thereby achieving the technical effects of reducing energy consumption, improving thermal efficiency, and ensuring heat storage and heat supply stability. The system includes: a geothermal well cluster data perception module 10, an individual output time series prediction module 20, a geothermal well cluster operation task creation module 30, a task scheduling evaluation model establishment module 40, and a scheduling optimization and distributed control management module 50.

[0112] Preferably, the geothermal well group data perception module 10 is used to perceive the geothermal well group data using the geothermal well group perception equipment after the geothermal well group perception equipment is deployed, and establish a perception data set; the individual output time series prediction module 20 is used to configure the individual output time series prediction of the well group individuals using the perception data set; the geothermal well group operation task creation module 30 is used to perform granular decomposition of the energy supply demand after obtaining the energy supply demand, and create the well group operation task using the granular decomposition result; the task scheduling evaluation model establishment module 40 is used to establish a task scheduling evaluation model, and the scheduling evaluation objectives of the task scheduling evaluation model include minimizing the energy consumption objective, maximizing the thermal efficiency objective, the heat storage stability objective, and the heat supply stability objective; the scheduling optimization and distributed control management module 50 is used to use the task scheduling evaluation model to perform scheduling optimization based on the individual output time series prediction and the well group operation task, and perform distributed scheduling management according to the scheduling optimization result;

[0113] Preferably, in the geothermal well group data perception module 10, after establishing the perception data set, the system may further include: a state analysis module for performing state analysis of individual well groups based on the perception data and establishing an initial state warning; an abnormality reporting module for reporting abnormalities of individual well groups based on the initial state warning;

[0114] The specific configuration of the individual output time series prediction module 20 will be described in detail below. As described above, the individual output time series prediction of the well group individuals is configured using the perception data set. The individual output time series prediction module 20 may further include: a standardized weak classifier establishment unit for establishing a standardized weak classifier; a reinforcement learning unit for collecting individual historical data of all well group individuals, and using the individual historical data collection results to perform reinforcement learning of the standardized weak classifier to configure an individual prediction classifier; a time series prediction unit for inputting the perception data set into the corresponding individual prediction classifier to perform time series prediction and establish an individual output time series prediction;

[0115] The specific configuration of the geothermal well group operation task creation module 30 will be described in detail below. As described above, after obtaining the energy supply demand, granular decomposition of the energy supply demand is performed, and the well group operation task is created using the granular decomposition result. The geothermal well group operation task creation module 30 may further include: a multi-dimensional granular space decomposition unit for performing multi-dimensional granular space decomposition on the energy supply demand, where the decomposition axes of the multi-dimensional granular space include spatial partitioning, time scale, demand type, and response capability; an adaptive task clustering unit for calling a multi-task classifier, using the multi-task classifier to perform adaptive task clustering based on the multi-granular space decomposition result, and establishing a granular decomposition result based on the adaptive task clustering result;

[0116] The specific configuration of the scheduling optimization and distributed control management module 50 will be described in detail below. As described above, the task scheduling evaluation model is used to perform scheduling optimization based on individual output time series prediction and the well group operation task, and distributed scheduling management is performed according to the scheduling optimization result. The scheduling optimization and distributed control management module 50 may further include: a scheduling log reading unit for reading the scheduling log of the geothermal well group; a scheduling constraint generation unit for generating scheduling constraints according to the scheduling log; an initial solution creation unit for performing task allocation of the well group operation task in the solution space with the scheduling constraints after creating the solution space, and creating an initial solution; a scheduling adaptation analysis unit for performing scheduling adaptation analysis based on the initial solution and the individual output time series prediction using the task scheduling evaluation model to generate an evaluation result; an iterative update unit for performing iterative update in the solution space and the scheduling constraints based on the evaluation result to complete the scheduling optimization;

[0117] Preferably, based on the evaluation results, iterative updates are performed under the solution space and the scheduling constraints to complete the scheduling optimization. The iterative update unit may further include: a strategy evolution subunit for treating each well group individual as a scheduling game individual, and performing strategy evolution between scheduling game individuals based on a game scheduling mechanism, wherein the strategy evolution is jointly driven by its own benefits and the evaluation results; a convergence subunit for completing the scheduling optimization after multiple rounds of game evolution converge.

[0118] Preferably, after the scheduling optimization and distributed control management module 50 performs distributed control management on the geothermal well group according to the results of the scheduling optimization, the system may further include: a continuous perception data set establishment module for using the geothermal well group perception equipment to perform continuous perception of the geothermal well group and establish a continuous perception data set; a deviation identification module for using the continuous perception data set and the individual output timing prediction to perform deviation identification, create a deviation warning with the deviation identification result, and sound the deviation warning.

[0119] The intelligent control system for distributed computing of geothermal well clusters provided in an embodiment of the present invention can execute the intelligent control method for distributed computing of geothermal well clusters provided in any embodiment of the present invention, and has functional modules and beneficial effects corresponding to the execution method.

[0120] Although the present application makes various references to certain modules in the system according to the embodiments of the present application, any number of different modules may be used and run on the user terminal and / or server, and the various units and modules included are only divided according to functional logic, but are not limited to the above division, as long as the corresponding functions can be achieved; in addition, the specific names of the functional units are only for the convenience of distinguishing each other and are not used to limit the scope of protection of the present invention.

[0121] Based on the above embodiments, the present application also provides an electronic device, Figure 3 1 is a schematic structural diagram of an electronic device provided by an embodiment of the present invention, showing a block diagram of an exemplary electronic device suitable for implementing an embodiment of the present invention. Figure 3 The electronic device shown is merely an example and should not limit the functionality and scope of use of the embodiments of the present invention. The electronic device is in the form of a general-purpose computing device, and its components may include, but are not limited to, an input device 301, a processor 302, a memory 303, and an output device 304. The processor 302 may be one or more; the memory 303 may include a computer-readable medium and at least one program product, which has a set (at least one) of program modules configured to perform the functions of the various embodiments of the present application.

[0122] The memory 303 shown in the embodiment of the present invention can adopt any combination of one or more computer-readable media; the computer-readable storage medium can be but not limited to infrared, semiconductor systems, devices or components, or any combination of the above, used to store software programs, computer executable programs and modules, such as the program instructions / modules corresponding to the intelligent control method for distributed computing of geothermal well clusters in the embodiment of the present invention. The processor 302 executes various functional applications and data processing of the computer device by running the software programs, instructions and modules stored in the memory 303, thereby realizing the above-mentioned intelligent control method for distributed computing of geothermal well clusters.

[0123] The above specific embodiments do not constitute a limitation to the scope of protection of this application. It should be understood by those skilled in the art that various modifications, combinations and substitutions can be made according to design requirements and other factors. Any modifications, equivalent replacements and improvements made within the spirit and principles of this application should be included in the scope of protection of this application. In some cases, the actions or steps recorded in this application can be performed in an order different from that in the embodiments and can still achieve the desired results. In addition, the processes depicted in the accompanying drawings do not necessarily require the specific order or continuous order shown to achieve the desired results. In some embodiments, multitasking and parallel processing are also possible or may be advantageous.

[0124] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit it. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or replace some or all of the technical features therein with equivalents. However, these modifications or replacements do not cause the essence of the corresponding technical solutions to deviate from the scope of the technical solutions of the embodiments of the present invention.

Claims

1. An intelligent control method for distributed computing of geothermal well groups, characterized in that: The following steps are involved: Acquire data perception of geothermal well clusters and establish a perception dataset based on the data perception of geothermal well clusters; Individual output time series prediction of individuals in a geothermal well cluster is configured through a perception dataset; Obtain energy supply demand, decompose it into granularity, and create geothermal well cluster operation tasks based on the results of the granularity decomposition; Establish a task scheduling evaluation model; The task scheduling evaluation model is used to optimize the scheduling of individual output time series predictions and geothermal well group operation tasks, and the geothermal well group is distributedly controlled and managed based on the results of the scheduling optimization.

2. The intelligent control method for distributed computing of geothermal well clusters according to claim 1, characterized in that: The individual output time series prediction of individuals in the geothermal well group configured by the perception data set specifically includes: Establish standardized weak classifier; Collect individual historical data of all individuals in the geothermal well group, conduct reinforcement learning of standardized weak classifiers based on the collected individual historical data, and configure individual prediction classifiers based on the standardized weak classifiers after reinforcement learning; The perception dataset is input into the individual prediction classifier to perform time series prediction and establish the individual output time series prediction.

3. The intelligent control method for distributed computing of geothermal well clusters according to claim 1, characterized in that: The granular decomposition of energy supply demand specifically includes: The energy supply demand is decomposed into dimensional granularity based on spatial partition, time scale, demand type, and response capability to obtain the dimensional granularity decomposition results. The dimensional granularity decomposition results are adaptively clustered into tasks by calling the task classifier, and the granularity decomposition results are established based on the results of the adaptive task clustering.

4. The intelligent control method for distributed computing of a geothermal well cluster according to claim 1, characterized in that: The scheduling evaluation objectives of the task scheduling evaluation model include minimizing energy consumption objectives, maximizing thermal efficiency objectives, thermal storage stability objectives, and heating supply stability objectives.

5. The intelligent control method for distributed computing of geothermal well clusters according to claim 1, characterized in that: The task scheduling evaluation model is used to optimize the scheduling of individual output time series prediction and geothermal well group operation tasks, specifically including: Read the scheduling log of the geothermal well group and generate scheduling constraints based on the scheduling log; Create a solution space, and perform task allocation of geothermal well cluster operation tasks within the solution space through scheduling constraints to generate an initial solution; Perform scheduling adaptation analysis on the initial solution and individual output timing prediction through the task scheduling evaluation model to generate evaluation results; Based on the evaluation results, iterative updates are performed under the solution space and scheduling constraints to complete the scheduling optimization.

6. The intelligent control method for distributed computing of geothermal well clusters according to claim 5, characterized in that: The iterative update is performed based on the evaluation results under the solution space and scheduling constraints to complete the scheduling optimization, specifically including: Each geothermal well cluster is treated as a scheduling game individual. Based on the game scheduling mechanism, several rounds of strategy evolution between scheduling game individuals are carried out under the solution space and scheduling constraints. The strategy evolution is jointly driven by its own benefits and evaluation results. When the strategy evolution between individuals in several rounds of scheduling game converges, the scheduling optimization is completed.

7. The intelligent control method for distributed computing of geothermal well clusters according to claim 1, characterized in that: The method of establishing a perception data set based on data perception of a geothermal well group further includes, after establishing the perception data set based on data perception of the geothermal well group, performing status analysis of individual geothermal well groups based on the perception data set, establishing an initial status warning, and reporting abnormalities of individual geothermal well groups based on the initial status warning.

8. The intelligent control method for distributed computing of geothermal well clusters according to claim 1, characterized in that: The distributed control and management of the geothermal well group according to the results of the scheduling optimization also includes: After distributed control and management of the geothermal well group is performed based on the results of the scheduling optimization, continuous sensing data of the geothermal well group is obtained, and a continuous sensing data set is established based on the continuous sensing data of the geothermal well group; Deviation identification is performed through continuous perception data sets and individual output time series prediction. Deviation warnings are created based on the results of deviation identification and the deviation warnings are issued.

9. An intelligent control system for distributed computing of geothermal well groups, characterized in that: The intelligent control method for distributed computing of a geothermal well group according to any one of claims 1 to 8 comprises: A geothermal well group data perception module is used to obtain data perception of the geothermal well group and establish a perception data set based on the data perception of the geothermal well group; An individual output time series prediction module is used to configure the individual output time series prediction of individuals in the geothermal well group through the perception data set; The geothermal well group operation task creation module is used to obtain energy supply demand, decompose the energy supply demand into granularity, and create geothermal well group operation tasks based on the results of the granularity decomposition; Task scheduling evaluation model establishment module, used to establish a task scheduling evaluation model; The scheduling optimization and distributed control management module is used to perform scheduling optimization on individual output time series prediction and geothermal well group operation tasks through the task scheduling evaluation model, and to perform distributed control management of the geothermal well group based on the results of the scheduling optimization.

10. An electronic device, characterized in that: The electronic device comprises: a memory for storing executable instructions; The processor is configured to implement the intelligent control method for distributed computing of a geothermal well group according to any one of claims 1 to 8 when executing the executable instructions stored in the memory.

Images