Geothermal resource dynamic exploration method, system and equipment and storage medium

Abstract

The invention discloses a geothermal resource dynamic investigation method, system and equipment and a storage medium, and relates to the field of geothermal resource dynamic investigation, comprising the steps of multi-source data acquisition, preprocessing, analysis modeling and dynamic monitoring updating, acquisition of multiple data such as temperature and pressure, adaptive weighted filtering and the like; a distribution probability graph is obtained through deep learning modeling and is dynamically monitored, the system comprises an acquisition unit, a preprocessing unit and the like for cooperative work, and equipment comprises a processor and a memory for executing related steps and storing a memory program for realizing the method steps. According to the method, multiple technical means are integrated, data are comprehensively and accurately collected and analyzed, geothermal resource distribution and changes are accurately predicted, auxiliary decisions are visually displayed, geothermal resource exploration efficiency and development rationality are effectively improved, and sustainable utilization of the geothermal resources is promoted.

Description

Technical Field

[0001] The present invention relates to the technical field of dynamic exploration of cross-border geothermal resources, and in particular to a method, system, equipment and storage medium for dynamic exploration of geothermal resources. Background Art

[0002] As the global demand for clean energy grows, geothermal resources, as a clean and sustainable energy source, have attracted much attention for their development and utilization. However, geothermal resources are buried deep underground, and their distribution and reserves are complex and changeable, so traditional exploration methods have many limitations.

[0003] Previous surveys mostly relied on limited borehole sampling, and the information points obtained were few, making it difficult to fully and accurately reflect the geothermal conditions of the entire target area. In terms of data collection, sensor technology is not advanced enough, and the collected parameters are single, such as only focusing on temperature or pressure. The collection accuracy and frequency are insufficient, and the dynamic changes of geothermal resources cannot be captured. The data processing methods are simple, lacking effective denoising and data analysis methods, making it difficult to extract valuable information from massive and complex data.

[0004] The analysis of the relationship between geological structure and geothermal resources is not in-depth enough, and the combined influence of multiple factors such as stratum structure, rock characteristics, and hydrogeology is not fully considered. In terms of modeling and prediction, the accuracy and adaptability of the model are poor, and it cannot be updated in time to reflect the dynamic evolution of geothermal resources. In addition, the lack of an effective monitoring system makes it difficult to detect abnormal changes in geothermal resources in a timely manner and issue early warnings, which is not conducive to the sustainable development and rational use of geothermal resources. This patent aims to overcome these defects and provide a comprehensive, efficient, and dynamic geothermal resource exploration method, system, equipment, and storage medium. Summary of the invention

[0005] The geothermal resource dynamic exploration method, system, equipment and storage medium proposed in the present invention are intended to solve the problems mentioned in the above-mentioned prior art.

[0006] In order to achieve the above object, the present invention adopts the following technical solution: a method for dynamic exploration of geothermal resources, comprising the following steps:

[0007] Data collection step: multiple sensor arrays are set up in the target survey area, and the sensor arrays include temperature sensors, pressure sensors, and seismic wave sensors, which are used to collect temperature data, pressure data, and seismic wave data at different depths underground; at the same time, satellite remote sensing technology is used to obtain surface thermal infrared data of the target area; and stratigraphic structure data and rock property data are extracted from the local geological data database;

[0008] Data preprocessing steps: Clean all types of collected data and remove outliers. The basis for judging outliers is that the temperature data deviates from the historical average temperature of the area by more than ±10°C, the pressure data deviates from the historical average pressure by more than ±0.1MPa, the seismic wave data has a frequency and amplitude that are inconsistent with the formation characteristics, and the surface thermal infrared data has noise points or is different from the thermal infrared data in the surrounding area; the cleaned data is processed using an adaptive weighted smoothing filter method. For temperature data and pressure data, the weighting coefficient is dynamically adjusted according to the local fluctuation of the data. The filter window is 5 to 10 data points, and the formula is: where x′ i is the filtered data, w j is the weighting coefficient, x j is the original data, k is the half width of the filter window, and The improved wavelet transform method is used for seismic wave data filtering. The seismic wave data is first decomposed into multiple scales with 3 to 5 decomposition layers. Then, the wavelet coefficients at each scale are processed by threshold shrinkage method. The threshold is selected by Stein unbiased risk estimation SURE method. Then, wavelet reconstruction is performed to obtain the filtered data. The processed data is standardized and mapped to the [0,1] interval. The formula is: where x norm is the standardized data, x is the original data, and x min and x max are the minimum and maximum values ​​in the original data;

[0009] Data analysis and modeling steps: Use deep learning algorithms to build a geothermal resource prediction model, and use preprocessed temperature data, pressure data, seismic wave data, surface thermal infrared data, topography data, stratigraphic structure data, and rock property data as model inputs. The model output is a distribution probability map of geothermal resources. The deep learning algorithm uses a convolutional neural network and a recurrent neural network architecture. The convolutional neural network is used to extract spatial features, and the recurrent neural network is used to process time series data. The target area is divided into multiple sub-areas. During the model training process, the cross-validation method is used to divide the data into training set, validation set, and test set. The loss function is a composite loss function that combines the mean square error function with the structural similarity index SSIM. The formula is L = α × MSE + (1-α) × (1-SSIM), where n is the number of samples, y i is the true value, is the predicted value, SSIM is used to measure the structural similarity between the predicted result and the true result, α is the weight coefficient, and the model parameters are adjusted through the back propagation algorithm. When the loss function value on the validation set no longer decreases, the training is stopped and the model is saved;

[0010] Dynamic monitoring and updating steps: repeat the data collection step and the data preprocessing step at a predetermined time interval, input the new data into the trained geothermal resource prediction model, and obtain an updated geothermal resource distribution probability map; compare the geothermal resource distribution probability maps before and after the update, analyze the dynamic changes of geothermal resources, and if the change exceeds the set threshold, issue a warning signal, and store the relevant data and analysis results in a database in a structured data format.

[0011] Furthermore, in the data collection step, the arrangement of the sensor array adopts a layered distributed layout, with a layer of sensors set every 100 meters to 500 meters in the vertical direction, and each layer of sensors is distributed in an equally spaced grid in the horizontal direction with a spacing of 500 meters to 1000 meters. The sensors are installed by deep drilling, and the drilling depth error is controlled within ±0.5 meters. The sensors are fixed with a special fixed bracket, and the sensor array is connected to the data acquisition control center through a wired transmission method using optical fiber cables.

[0012] Furthermore, in the data analysis and modeling steps, local hydrogeological data are introduced when constructing the geothermal resource prediction model, and a high-precision liquid level sensor is used. The measurement principle is based on the relationship between the pressure difference and the liquid level height, and the measurement accuracy is improved through temperature compensation and calibration algorithm; the groundwater flow data uses an electromagnetic flowmeter to measure the flow through the linear relationship between the induced electromotive force and the flow; the groundwater quality data includes pH, conductivity, and major ion concentrations, and a multi-parameter water quality sensor is used.

[0013] Furthermore, in the dynamic monitoring and updating step, when an early warning signal is issued, a finite element numerical simulation method is further used to simulate and analyze the changing trend of geothermal resources. The finite element model uses three-dimensional hexahedral units for spatial discretization, and the unit size is adaptively divided according to the geological complexity of the target area and the distribution characteristics of geothermal resources. According to the simulation results, response strategies are proposed, including adjusting the geothermal resource development plan, strengthening the monitoring frequency, etc., to ensure the sustainable development and utilization of geothermal resources.

[0014] Furthermore, a geothermal resource dynamic exploration system includes:

[0015] Data acquisition unit: used for the data acquisition step, including multiple sensor arrays, satellite remote sensing data receiving devices and geological data extraction interfaces. The sensor arrays are connected to a data acquisition control module, which sets acquisition parameters of the sensors and transmits the acquired data to a data preprocessing unit;

[0016] Data preprocessing unit: used to perform the data preprocessing steps, clean, filter and standardize the data sent by the data acquisition unit, and transmit the processed data to the data analysis and modeling unit;

[0017] Data analysis and modeling unit: used to perform the data analysis and modeling steps, build and train the geothermal resource prediction model using the deep learning algorithm, and output the geothermal resource distribution probability map according to the input data. The model architecture and parameters can be adjusted as needed during the model training process. The model training results and related data are stored in the model storage unit;

[0018] Dynamic monitoring and updating unit: used to execute the dynamic monitoring and updating steps, trigger the data acquisition unit and the data preprocessing unit to work at a predetermined time interval, input new data into the trained model of the data analysis and modeling unit, compare and analyze the dynamic changes of geothermal resources, issue a warning signal when the change amplitude exceeds the threshold, and store the data in the data storage unit, which adopts a distributed storage system and has a data redundancy backup function;

[0019] Model storage unit: used to store the geothermal resource prediction model and model-related parameters built and trained by the data analysis and modeling unit.

[0020] Furthermore, the sensor array in the data acquisition unit also includes a chemical sensor for collecting chemical composition data in groundwater, including mineral content and pH. The chemical sensor uses a technology that combines ion-selective electrode method and colorimetric analysis method. The ion-selective electrode is highly selective for different ions. The colorimetric analysis method uses the degree of color change to perform quantitative analysis through the reaction of reagents with the substance to be tested. Color detection uses high-precision photoelectric sensors to provide data support for analyzing the interaction between geothermal resources and geochemical environment.

[0021] Furthermore, the deep learning algorithm in the data analysis and modeling unit adopts transfer learning technology during the training process, using the basic model trained in other similar geothermal resource exploration areas and fine-tuning it in combination with the data of this area. The selection of the basic model is based on its similarity assessment with the local area in terms of geological structure and geothermal resource type. The similarity index adopts the cosine similarity of the geological feature vector. During the fine-tuning process, a learning rate in the range of 0.0001 to 0.001 is used to gradually adjust the fully connected layer of the model, and the connection weights and biases of the neurons are adjusted according to the characteristics of the data in this area. At the same time, the early stopping method is used to prevent overfitting.

[0022] Furthermore, the dynamic monitoring and updating unit also includes a visualization display module, which is used to display the geothermal resource distribution probability map and the dynamic changes of geothermal resources in the form of intuitive charts, including heat maps and line charts. The visualization display module adopts Web-side visualization technology, builds a user interface based on HTML5, CSS3 and JavaScript front-end technologies, and uses the Echarts visualization library for chart drawing. Users access the visualization interface through a browser. The interface supports zooming, panning, and data query interactive operations, and the data update frequency is synchronized with the dynamic monitoring frequency.

[0023] Furthermore, a geothermal resource dynamic exploration device includes a processor and a memory, wherein the memory stores a computer program, and the processor implements the steps of the geothermal resource dynamic exploration method when executing the computer program.

[0024] Furthermore, a computer-readable storage medium is provided, wherein the computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, the steps of the geothermal resource dynamic exploration method are implemented.

[0025] Compared with the prior art, the present invention has the following beneficial effects:

[0026] In data collection, we use a combination of multi-type high-precision sensor arrays, satellite remote sensing and geological data to comprehensively acquire multi-source data such as temperature, pressure, seismic waves, and surface thermal infrared. The sensor layout is scientific and reasonable, and the data collection is accurate and reliable. In the data preprocessing stage, adaptive weighted smoothing filtering and improved wavelet transform effectively remove noise, and the composite loss function optimization modeling improves data quality.

[0027] In terms of analysis and modeling, the deep learning algorithm integrates multiple network architectures, fully considers multiple factors, introduces hydrogeological data and adopts transfer learning, so that the model can more accurately predict the distribution of geothermal resources and better adapt to the characteristics of different regions.

[0028] The dynamic monitoring and updating function can regularly repeat the collection and analysis, and timely detect the changes in geothermal resources. The early warning mechanism combined with finite element numerical simulation can plan the response strategy in advance to ensure the sustainability of development. The visual display module in the system facilitates personnel to intuitively understand the situation and assist in decision-making. The coordinated operation of each module improves the overall exploration efficiency and effect, and effectively promotes the efficient development and utilization of geothermal resources. BRIEF DESCRIPTION OF THE DRAWINGS

[0029] Figure 1 A schematic block diagram of a geothermal resource dynamic exploration method proposed by the present invention;

[0030] Figure 2 This is a schematic block diagram of a geothermal resource dynamic exploration system proposed by the present invention. DETAILED DESCRIPTION

[0031] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0032] In the description of the present invention, it should be understood that the terms "center", "longitudinal", "lateral", "length", "width", "thickness", "up", "down", "front", "back", "left", "right", "vertical", "horizontal", "top", "bottom", "inside", "outside", "clockwise", "counterclockwise" and the like indicate orientations or positional relationships based on the orientations or positional relationships shown in the accompanying drawings, and are only for the convenience of describing the present invention and simplifying the description, rather than indicating or implying that the referred device or element must have a specific orientation, be constructed and operated in a specific orientation, and therefore should not be understood as limiting the present invention.

[0033] In addition, the terms "first" and "second" are used for descriptive purposes only and cannot be understood as indicating or implying relative importance or implicitly indicating the number of technical features indicated. Therefore, the features defined as "first" and "second" may explicitly or implicitly include one or more of the features. In the description of the present invention, the meaning of "multiple" is two or more, unless otherwise clearly and specifically defined. In addition, the terms "installed", "connected" and "connected" should be understood in a broad sense. For example, it can be a fixed connection, a detachable connection, or an integral connection; it can be a mechanical connection or an electrical connection; it can be a direct connection, or it can be indirectly connected through an intermediate medium, and it can be the internal connection of two elements. For ordinary technicians in this field, the specific meanings of the above terms in the present invention can be understood according to specific circumstances. The present invention will be further described in detail below in conjunction with the accompanying drawings.

[0034] Reference Figure 1-2 :A geothermal resource dynamic exploration method comprises the following steps:

[0035] Data collection steps: multiple sensor arrays are set up in the target survey area, and the sensor arrays include temperature sensors, pressure sensors, and seismic wave sensors, which are respectively used to collect temperature data, pressure data, and seismic wave data at different depths underground. The temperature sensor collects data once every hour to once every 6 hours with an accuracy of ±0.1°C, the pressure sensor collects data once every 3 hours to once every 12 hours with an accuracy of ±0.01MPa, and the seismic wave sensor collects data once every minute to once every 5 minutes, which can detect seismic wave signals with a frequency range of 1Hz to 100Hz; at the same time, satellite remote sensing technology is used to obtain surface thermal infrared data of the target area with a resolution of not less than 10 meters, as well as topographic data; and stratigraphic structure data and rock property data are extracted from the local geological data database.

[0036] Data preprocessing steps: Clean all types of collected data and remove outliers. The basis for judging outliers is that the temperature data deviates from the historical average temperature of the area by more than ±10°C, the pressure data deviates from the historical average pressure by more than ±0.1MPa, the seismic wave data has a frequency and amplitude that are obviously inconsistent with the formation characteristics, and the surface thermal infrared data has obvious noise points or is too different from the thermal infrared data in the surrounding area; the cleaned data is processed using an adaptive weighted smoothing filter method. For temperature data and pressure data, the weighting coefficient is dynamically adjusted according to the local fluctuation of the data. The filter window is 5 to 10 data points, and the formula is: where x′ i is the filtered data, w j is the weighting coefficient, x j is the original data, k is the half width of the filter window, and The improved wavelet transform method is used for seismic wave data filtering. The seismic wave data is first decomposed into multiple scales with 3 to 5 decomposition layers. Then, the wavelet coefficients at each scale are processed by threshold shrinkage method. The threshold is selected by Stein unbiased risk estimation (SURE) method. Then, wavelet reconstruction is performed to obtain the filtered data. The processed data is standardized and mapped to the [0,1] interval. The formula is: where x norm is the standardized data, x is the original data, and x min and x max are the minimum and maximum values ​​in the original data.

[0037] Data analysis and modeling steps: Use deep learning algorithms to build a geothermal resource prediction model, and use preprocessed temperature data, pressure data, seismic wave data, surface thermal infrared data, topography data, stratigraphic structure data, and rock property data as model inputs. The model output is a distribution probability map of geothermal resources. The deep learning algorithm uses a convolutional neural network and a recurrent neural network architecture. The convolutional neural network is used to extract spatial features, and the recurrent neural network is used to process time series data. The target area is divided into multiple sub-areas. During the model training process, the cross-validation method is used to divide the data into training set, validation set, and test set in a ratio of 7:2:1. The loss function is a composite loss function that combines the mean square error function with the structural similarity index (SSIM). The formula is L = α × MSE + (1-α) × (1-SSIM), where n is the number of samples, y i is the true value, is the predicted value, SSIM is used to measure the structural similarity between the predicted result and the true result, α is the weight coefficient, and the value range is 0.5 to 0.8. The model parameters are adjusted through the back propagation algorithm. When the loss function value on the validation set no longer decreases, the training is stopped and the model is saved.

[0038] Dynamic monitoring and updating steps: repeat the data collection step and the data preprocessing step at a predetermined time interval, input the new data into the trained geothermal resource prediction model, and obtain an updated geothermal resource distribution probability map; the predetermined time interval is 1 month to 3 months; compare the geothermal resource distribution probability maps before and after the update, and analyze the dynamic changes of geothermal resources. If the change exceeds the set threshold, such as the geothermal resource distribution probability change exceeds 20%, an early warning signal is issued, and the relevant data and analysis results are stored in the database. The database storage format is a structured data format to facilitate data query and retrieval.

[0039] In the present invention, in the data collection step, the arrangement of the sensor array adopts a layered distributed layout, and a layer of sensors is arranged every 100 meters to 500 meters in the vertical direction, and each layer of sensors is distributed in a grid-like manner with equal spacing in the horizontal direction, with a spacing of 500 meters to 1000 meters. The sensor installation adopts a deep-buried drilling method, and the drilling depth error is controlled within ±0.5 meters. The sensor is fixed with a special fixed bracket to ensure the stability of the sensor in the underground environment and reduce the impact of external interference on data collection. The sensor array is connected to the data acquisition control center through a wired transmission method, using optical fiber cable, and the data transmission rate is not less than 100Mbps, so as to ensure the comprehensive coverage of the underground space of the target area, the accuracy of data collection, and the efficiency of data transmission.

[0040] In the present invention, in the data analysis and modeling steps, when constructing a geothermal resource prediction model, local hydrogeological data are also introduced, including groundwater level data, groundwater flow data, and groundwater quality data, wherein the groundwater level data acquisition accuracy is ±0.1 meters, and a high-precision liquid level sensor is used. The measurement principle is based on the relationship between pressure difference and liquid level height, and the measurement accuracy is improved through temperature compensation and calibration algorithm; the groundwater flow data acquisition accuracy is ±0.01 cubic meters per second, and an electromagnetic flowmeter is used, and the magnetic field strength in its measuring tube is stable at ±0.05 Tesla, and the flow is measured through the linear relationship between the induced electromotive force and the flow; the groundwater quality data includes pH value (pH value), conductivity, main ion concentration, etc., and a multi-parameter water quality sensor is used, the pH value measurement accuracy is ±0.02, the conductivity measurement accuracy is ±0.5%, and the main ion concentration measurement accuracy is ±0.1ppm, so as to improve the model's ability to analyze the interaction between geothermal resources and groundwater, thereby improving the accuracy of geothermal resource prediction.

[0041] In the present invention, in the dynamic monitoring and updating step, after the early warning signal is issued, the finite element numerical simulation method is further used to simulate and analyze the changing trend of geothermal resources. The finite element model uses three-dimensional hexahedral units for spatial discretization, and the unit size is adaptively divided according to the geological complexity of the target area and the distribution characteristics of geothermal resources. The unit size is small in the area of ​​​​drastic changes, and the minimum can reach 10 meters. The unit size is large in the relatively stable area, and the maximum is 100 meters. The model boundary conditions are determined according to the geological structure and surrounding environment of the target area, such as setting temperature boundaries, pressure boundaries, and flow boundaries. The time span of the simulation analysis is 1 to 5 years in the future. The time step is dynamically adjusted according to the change rate of geothermal resources. When the change is fast, the time step is small, the minimum is 1 day, and when the change is slow, the time step is large, and the maximum is 1 month. According to the simulation results, a response strategy is proposed, such as adjusting the geothermal resource development plan, strengthening the monitoring frequency, etc., to ensure the sustainable development and utilization of geothermal resources.

[0042] The present invention discloses a geothermal resource dynamic exploration system, comprising:

[0043] Data acquisition unit: used to execute the data acquisition step described in claim 1, including multiple sensor arrays, satellite remote sensing data receiving devices and geological data extraction interfaces. The sensor arrays are connected to a data acquisition control module. The data acquisition control module can set the acquisition parameters of the sensor, such as acquisition frequency, acquisition accuracy, etc., and transmit the collected data to the data preprocessing unit.

[0044] Data preprocessing unit: used to execute the data preprocessing step described in claim 1, clean, filter and standardize the data transmitted from the data acquisition unit, and transmit the processed data to the data analysis and modeling unit.

[0045] Data analysis and modeling unit: used to execute the data analysis and modeling steps described in claim 1, use deep learning algorithms to build and train geothermal resource prediction models, and output geothermal resource distribution probability maps based on input data. During the model training process, the model architecture and parameters can be adjusted as needed, and the model training results and related data are stored in the model storage unit.

[0046] Dynamic monitoring and updating unit: used to execute the dynamic monitoring and updating steps described in claim 1, trigger the data acquisition unit and the data preprocessing unit to work at a predetermined time interval, input new data into the trained model of the data analysis and modeling unit, compare and analyze the dynamic changes of geothermal resources, issue a warning signal when the change amplitude exceeds a threshold, and store the data in a data storage unit. The data storage unit adopts a distributed storage system and has a data redundancy backup function to ensure the security and integrity of the data.

[0047] Model storage unit: used to store the geothermal resource prediction model and model-related parameters built and trained by the data analysis and modeling unit, so as to quickly call the model for calculation during dynamic monitoring and updating.

[0048] In the present invention, the sensor array in the data acquisition unit also includes a chemical sensor for collecting chemical composition data in groundwater, such as mineral content, pH, etc., and the chemical composition data collection accuracy is ±0.01ppm. The chemical sensor adopts a technology combining ion selective electrode method and colorimetric analysis method. The ion selective electrode is highly selective for different ions, and the response time is between 10 seconds and 30 seconds. The colorimetric analysis method reacts with the substance to be tested through a specific reagent, and uses the degree of color change for quantitative analysis. The color detection adopts a high-precision photoelectric sensor with a resolution of 1 / 1024, which provides data support for analyzing the interaction between geothermal resources and geochemical environment, and further improves the comprehensiveness of data collection.

[0049] In the present invention, the deep learning algorithm in the data analysis and modeling unit adopts transfer learning technology during the training process, using the basic model trained in other similar geothermal resource exploration areas, combined with a small amount of data in this area for fine-tuning. The selection of the basic model is based on its similarity assessment with the local area in terms of geological structure, geothermal resource type, etc. The similarity index adopts the cosine similarity of the geological feature vector, and the threshold is set to be above 0.6. During the fine-tuning process, a smaller learning rate is used, ranging from 0.0001 to 0.001, and the fully connected layer of the model is gradually adjusted. The connection weights and biases of the neurons are adjusted according to the characteristics of the data in this area. At the same time, the early stopping method is used to prevent overfitting. The number of training rounds is determined according to the performance of the validation set, generally 10 to 50 rounds, which speeds up the model training speed and improves the adaptability and accuracy of the model in this area.

[0050] In the present invention, the dynamic monitoring and updating unit also includes a visualization display module, which is used to display the geothermal resource distribution probability map and the dynamic change of geothermal resources in the form of intuitive charts, such as heat maps, line charts, etc. The visualization display module adopts Web-side visualization technology, builds a user interface based on front-end technologies such as HTML5, CSS3 and JavaScript, and uses visualization libraries such as Echarts to draw charts. Users can access the visualization interface through a browser. The interface supports interactive operations such as zooming, panning, and data query. The data update frequency is synchronized with the dynamic monitoring frequency, which is convenient for geological surveyors and geothermal resource developers to quickly understand the status of geothermal resources and assist in decision-making.

[0051] The present invention discloses a geothermal resource dynamic exploration device, comprising a processor and a memory, wherein the memory stores a computer program, and when the processor executes the computer program, the steps of the geothermal resource dynamic exploration method described in the claim are implemented.

[0052] The present invention discloses a computer-readable storage medium, wherein the computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, the steps of the geothermal resource dynamic exploration method are implemented.

[0053] The above are only preferred specific implementation modes of the present invention, but the protection scope of the present invention is not limited thereto. Any technician familiar with the technical field can make equivalent replacements or changes according to the technical solutions and inventive concepts of the present invention within the technical scope disclosed by the present invention, which should be covered by the protection scope of the present invention.

Claims

1. A method for dynamic exploration of geothermal resources, characterized in that: The following steps are involved: Data collection step: multiple sensor arrays are set up in the target survey area, and the sensor arrays include temperature sensors, pressure sensors, and seismic wave sensors, which are used to collect temperature data, pressure data, and seismic wave data at different depths underground respectively; at the same time, satellite remote sensing technology is used to obtain surface thermal infrared data of the target area; And extract stratigraphic structure data and rock property data from the local geological data database; Data preprocessing steps: Clean all types of collected data and remove outliers. The basis for judging outliers is that the temperature data deviates from the historical average temperature of the area by more than ±10°C, the pressure data deviates from the historical average pressure by more than ±0.1MPa, the seismic wave data has a frequency and amplitude that are inconsistent with the formation characteristics, and the surface thermal infrared data has noise points or is different from the thermal infrared data in the surrounding area; the cleaned data is processed using an adaptive weighted smoothing filter method. For temperature data and pressure data, the weighting coefficient is dynamically adjusted according to the local fluctuation of the data. The filter window is 5 to 10 data points, and the formula is: where x′ i is the filtered data, w j is the weighting coefficient, x j is the original data, k is the half width of the filter window, and The improved wavelet transform method is used for seismic wave data filtering. The seismic wave data is first decomposed into multiple scales with 3 to 5 decomposition layers. Then the wavelet coefficients at each scale are processed by threshold shrinkage method. The threshold is selected by Stein unbiased risk estimation SURE method. Then wavelet reconstruction is performed to obtain the filtered data. The processed data is standardized and mapped to the [0,1] interval. The formula is where x norm is the standardized data, x is the original data, and x min and x max are the minimum and maximum values ​​in the original data; Data analysis and modeling steps: Use deep learning algorithms to build a geothermal resource prediction model, and use preprocessed temperature data, pressure data, seismic wave data, surface thermal infrared data, topography data, stratigraphic structure data, and rock property data as model inputs. The model output is a distribution probability map of geothermal resources. The deep learning algorithm uses a convolutional neural network and a recurrent neural network architecture. The convolutional neural network is used to extract spatial features, and the recurrent neural network is used to process time series data. The target area is divided into multiple sub-areas. During the model training process, the cross-validation method is used to divide the data into training set, validation set, and test set. The loss function is a composite loss function that combines the mean square error function with the structural similarity index SSIM. The formula is L = α × MSE + (1-α) × (1-SSIM), where n is the number of samples, y i is the true value, is the predicted value, SSIM is used to measure the structural similarity between the predicted result and the true result, α is the weight coefficient, and the model parameters are adjusted through the back propagation algorithm. When the loss function value on the validation set no longer decreases, the training is stopped and the model is saved; Dynamic monitoring and updating steps: repeat the data collection step and the data preprocessing step at a predetermined time interval, input the new data into the trained geothermal resource prediction model, and obtain an updated geothermal resource distribution probability map; compare the geothermal resource distribution probability maps before and after the update, analyze the dynamic changes of geothermal resources, and if the change exceeds the set threshold, issue a warning signal, and store the relevant data and analysis results in a database in a structured data format.

2. The geothermal resource dynamic exploration method according to claim 1, characterized in that: In the data collection step, the arrangement of the sensor array adopts a layered distributed layout, with a layer of sensors set every 100 meters to 500 meters in the vertical direction, and each layer of sensors is distributed in an equally spaced grid in the horizontal direction with a spacing of 500 meters to 1000 meters. The sensor is installed by deep drilling, and the drilling depth error is controlled within ±0.5 meters. The sensor is fixed with a special fixed bracket, and the sensor array is connected to the data acquisition control center through a wired transmission method using optical fiber cable.

3. The geothermal resource dynamic exploration method according to claim 1, characterized in that: In the data analysis and modeling steps, local hydrogeological data are introduced when constructing the geothermal resource prediction model, and a high-precision liquid level sensor is used. The measurement principle is based on the relationship between pressure difference and liquid level height, and the measurement accuracy is improved through temperature compensation and calibration algorithm; the groundwater flow data uses an electromagnetic flowmeter to measure the flow through the linear relationship between the induced electromotive force and the flow; the groundwater quality data includes pH, conductivity, and major ion concentrations, and a multi-parameter water quality sensor is used.

4. The geothermal resource dynamic exploration method according to claim 1, characterized in that: In the dynamic monitoring and updating step, after the early warning signal is issued, the finite element numerical simulation method is further used to simulate and analyze the changing trend of geothermal resources. The finite element model uses three-dimensional hexahedral units for spatial discretization, and the unit size is adaptively divided according to the geological complexity of the target area and the distribution characteristics of geothermal resources. According to the simulation results, response strategies are proposed, including adjusting the geothermal resource development plan, strengthening the monitoring frequency, etc., to ensure the sustainable development and utilization of geothermal resources.

5. A geothermal resource dynamic exploration system, characterized in that: include: Data acquisition unit: used for the data acquisition step, including multiple sensor arrays, satellite remote sensing data receiving devices and geological data extraction interfaces. The sensor arrays are connected to a data acquisition control module, which sets acquisition parameters of the sensors and transmits the acquired data to a data preprocessing unit; Data preprocessing unit: used to perform the data preprocessing steps, clean, filter and standardize the data sent by the data acquisition unit, and transmit the processed data to the data analysis and modeling unit; Data analysis and modeling unit: used to perform the data analysis and modeling steps, build and train the geothermal resource prediction model using the deep learning algorithm, and output the geothermal resource distribution probability map according to the input data. The model architecture and parameters can be adjusted as needed during the model training process. The model training results and related data are stored in the model storage unit; Dynamic monitoring and updating unit: used to execute the dynamic monitoring and updating steps, trigger the data acquisition unit and the data preprocessing unit to work at a predetermined time interval, input new data into the trained model of the data analysis and modeling unit, compare and analyze the dynamic changes of geothermal resources, issue a warning signal when the change amplitude exceeds the threshold, and store the data in the data storage unit, which adopts a distributed storage system and has a data redundancy backup function; Model storage unit: used to store the geothermal resource prediction model and model-related parameters built and trained by the data analysis and modeling unit.

6. The geothermal resource dynamic exploration system according to claim 5, characterized in that: The sensor array in the data acquisition unit also includes a chemical sensor for collecting chemical composition data in groundwater, including mineral content and pH. The chemical sensor uses a technology that combines ion-selective electrode method and colorimetric analysis method. The ion-selective electrode is highly selective for different ions. The colorimetric analysis method uses the degree of color change to perform quantitative analysis through the reaction of reagents with the substance to be tested. Color detection uses high-precision photoelectric sensors to provide data support for analyzing the interaction between geothermal resources and geochemical environment.

7. The geothermal resource dynamic exploration system according to claim 5, characterized in that: The deep learning algorithm in the data analysis and modeling unit adopts transfer learning technology during the training process, and uses the basic model trained in other similar geothermal resource exploration areas, and fine-tunes it in combination with the data of this area. The selection of the basic model is based on its similarity assessment with the local area in terms of geological structure and geothermal resource type. The similarity index adopts the cosine similarity of the geological feature vector. During the fine-tuning process, a learning rate in the range of 0.0001 to 0.001 is used to gradually adjust the fully connected layer of the model, and the connection weights and biases of the neurons are adjusted according to the characteristics of the data in this area. At the same time, the early stopping method is used to prevent overfitting.

8. The geothermal resource dynamic exploration system according to claim 5, characterized in that: The dynamic monitoring and updating unit also includes a visualization display module, which is used to display the geothermal resource distribution probability map and the dynamic changes of geothermal resources in the form of intuitive charts, including heat maps and line charts. The visualization display module adopts Web-side visualization technology, builds a user interface based on HTML5, CSS3 and JavaScript front-end technologies, and uses the Echarts visualization library to draw charts. Users access the visualization interface through a browser. The interface supports zooming, panning, and data query interactive operations, and the data update frequency is synchronized with the dynamic monitoring frequency.

9. A geothermal resource dynamic exploration device, characterized in that: The method comprises a processor and a memory, wherein the memory stores a computer program, and the processor implements the steps of the geothermal resource dynamic exploration method according to any one of claims 1 to 4 when executing the computer program.

10. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, the steps of the geothermal resource dynamic exploration method described in any one of claims 1 to 4 are implemented.

Images