Geological data analysis method based on AI algorithm

Abstract

The invention belongs to the technical field of geological data analysis, and particularly discloses a geological data analysis method based on an AI algorithm. The method comprises the steps of collecting geological data and integrating the data into a uniform format and a uniform structure; preprocessing the data and extracting features or attributes in the data; analyzing and explaining the data according to the extracted features or attributes; selecting a proper AI algorithm to construct an AI model, and then training the AI model by using the preprocessed data; evaluating the AI model by using the test set data, and optimizing the AI model according to an evaluation result; carrying out deep learning and analysis on the data through an NLP technology and / or a data modeling method; interpreting and modeling the geological phenomenon according to the result of the data analysis, and providing decision support; and iteratively improving the geological AI model regularly or as required. The method has the characteristics of efficient and timely complex data processing, less manual intervention and high prediction accuracy.

Description

Technical Field

[0001] The present invention relates to the technical field of geological data analysis, and in particular to a geological data analysis method based on an AI algorithm with high efficiency and timeliness in complex data processing, little manual intervention and high prediction accuracy. Background Art

[0002] In geological work, various means are used to collect geological data, such as field surveys, drilling trenching and other mountain engineering, geophysical detection, geochemical detection, remote sensing, analysis and testing, and comprehensive research, thus forming a multi-disciplinary, multi-scale, multi-resolution, and multi-purpose collection of professional data such as geology, geography, geophysics, geochemistry, and topographic mapping, to provide data support for the research and exploration of the earth's surface and the material composition, structure, and evolution of the earth.

[0003] Geological data analysis refers to the systematic and scientific processing of data obtained in the geological field, as well as a series of technologies and means for analysis and interpretation. It plays a vital role in many fields such as geology, mineral resource exploration, engineering construction, and geological disaster prevention. Since geological data are usually complex, diverse, and massive, most of the existing geological data analysis methods are based on geological analysis models developed on a single machine for calculation. It is necessary to process and convert various geological analysis models and method tools to form valuable geological understanding or discovery. Therefore, for large-scale geological data sets, it is not only slow in calculation speed and unable to quickly generate analysis results, but also difficult to apply to geological data analysis scenarios that require real-time response; and because a lot of manual intervention such as data cleaning, conversion, and standardization is required in the calculation process, the data analysis efficiency is low, and the accuracy and consistency of the data are difficult to effectively guarantee. In addition, most of the existing geological data analysis methods can only produce corresponding results based on the analysis of historical geological data, and cannot predict future geological phenomena. Or, although they can predict future geological phenomena, they cannot perform adaptive iteration and upgrade by themselves, so the accuracy of the prediction results of future geological phenomena is low, and the actual application value is not great.

[0004] Therefore, it is particularly important to seek a geological data analysis method that can process complex data efficiently and timely, with less human intervention and high prediction accuracy. Summary of the invention

[0005] In view of the deficiencies in the prior art, the present invention provides a geological data analysis method based on an AI algorithm, which has the advantages of efficient and timely processing of complex data, less human intervention and high prediction accuracy.

[0006] The present invention is implemented as follows: including data collection and integration, data preprocessing, data analysis and interpretation, AI model construction, model evaluation and optimization, deep learning and analysis, AI analysis results, and AI model iterative improvement steps, the specific contents are: A. Data collection and integration: Collect geological data from different sources and then integrate the collected data into a unified format and structure; B. Data preprocessing: preprocess the aforementioned integrated data and extract features or attributes from the preprocessed data; C. Data analysis and interpretation: Analyze and interpret the pre-processed data based on the extracted features or attributes; D. AI model construction: Select a suitable AI algorithm, build an AI model based on the selected algorithm, and then use the preprocessed data to train the AI ​​model; E. Model evaluation and optimization: Use test set data to evaluate the AI ​​model and optimize the AI ​​model based on the evaluation results; G. Deep learning and analysis: Perform deep learning and analysis on pre-processed data through NLP technology and / or data modeling methods; H. AI analysis results: Based on the results of the aforementioned data analysis, the geological phenomena are interpreted and modeled, and decision support is provided; J. Iterative improvement of AI model: Iteratively improve the geological AI model regularly or as needed.

[0007] Furthermore, the step A comprises the following sub-steps: A1. Determine the type of data: Determine the type of geological data that needs to be collected; A2. Collect data: Collect data from various sources using manual input, automated sensors, and / or database queries; A3. Data Integration I: Use GIS technology to integrate the collected data into a unified platform or database, or use machine learning algorithms to mine and analyze the collected data to discover the correlation and patterns between the data, and then integrate them into a unified format and structure.

[0008] Furthermore, the step B comprises the following sub-steps: B1. Data preprocessing: remove duplicate data, missing values, and outliers from the data integrated in step A; B2. Data integration II: Integrate the pre-processed data into a unified measurement unit or range; B3. Data extraction: extracting preset features or attributes from the above integrated data.

[0009] Furthermore, the step C comprises the following sub-steps: C1. Feature analysis: Analyze the influence of each feature or attribute extracted in step B on the prediction results of the AI ​​model; C2. Visualization: Use visualization technology to display the results of the above data feature analysis; C3. Anomaly processing: Identify outliers or abnormal conditions in the preprocessed data and explain their geological significance.

[0010] Furthermore, the step D comprises the following sub-steps: D1. Algorithm selection: Based on the above data analysis, select one of the deep learning, machine learning or reinforcement learning algorithms as the AI ​​algorithm; D2. Build AI model: Build an AI model including input layer, hidden layer and output layer according to the selected AI algorithm; D3. AI model training: Use preprocessed data to train the AI ​​model, and optimize model performance by adjusting model parameters based on the training results.

[0011] Furthermore, the step E comprises the following sub-steps: E1. AI model evaluation: First, find a data set with similar geological characteristics and distribution as a test set, then use the test set data to input the AI ​​model, compare the prediction results of the AI ​​model with the actual results, calculate the accuracy, recall rate and / or F1-score indicators of the AI ​​model, and evaluate the performance of the AI ​​model on the test set; E2. AI model optimization: Adjust the parameters or structure of the AI ​​model based on the above evaluation results.

[0012] Furthermore, the G step includes the following optional items: G1. Deep learning: Use the trained AI model to conduct in-depth analysis on pre-processed data, explore the potential patterns and associations in the data, and provide strong support for the interpretation and modeling of geological phenomena; G2, NLP Assisted: Using NLP technology, extract key information from pre-processed text data and identify trends and patterns in text data, and then automatically generate geological reports based on the extracted information and identified trends, providing geologists with fast and accurate data interpretation and analysis results; G3. Image Classification: Identify geological features from pre-processed aerial and satellite images and classify geological phases from images of rocks and minerals using computer vision techniques.

[0013] Furthermore, the G1 step includes the following steps: G11. Directly learn the mapping of seismic data to velocity models through AI models, and improve the accuracy and speed of inversion through end-to-end proxy models or a combination of data-driven and physical-driven methods; G12. Use deep learning technology to automatically identify geomorphic features in geomorphic maps and improve the accuracy of geomorphic classification and feature recognition.

[0014] Furthermore, the H step includes the following sub-steps: H1. Interpretation and modeling: Based on the analysis results of the AI ​​model, the geological phenomena of structural evolution, lithology distribution and / or mineral resource prediction are interpreted and modeled; H2. Decision support: Based on the above interpretation and modeling, provide decision support for geological exploration, resource development and environmental protection, including exploration area selection, resource reserve assessment and / or geological disaster early warning.

[0015] Furthermore, the J step includes the following sub-steps: J1. Application feedback: Collect feedback from geologists and domain experts to understand the performance of the model in practical applications; J2. Model iteration: Iterate and improve the AI ​​model based on feedback and new geological data.

[0016] The beneficial effects of the present invention are: 1. Compared with the existing geological data analysis methods, the geological data analysis method of the present invention can not only efficiently process complex geological data and extract valuable geological information from it by building in a suitable AI model, but also can be widely used in geology, mineral resource exploration, engineering construction, geological disaster prevention and control and other fields; and through data preprocessing and parallel computing and distributed processing of AI models, it can complete large-scale data analysis tasks in a short time, so that data can be processed in real time and analysis results can be generated, providing data support for decision makers to make decisions in a timely manner, so that it can meet the geological data analysis scenarios that require real-time response.

[0017] 2. The geological data analysis method of the present invention can automatically perform data preprocessing tasks such as data cleaning, conversion and standardization through an intelligent AI model, thereby ensuring the accuracy and consistency of the data and reducing the time and cost of manual operations. At the same time, the AI ​​model can also quickly process a large amount of geological data, thereby improving the efficiency of data analysis.

[0018] 3. The geological data analysis method of the present invention can be optimized through the AI ​​model, and a prediction model can be constructed based on historical geological data to predict future geological phenomena; and through the iterative upgrade of the AI ​​model, the prediction results are ensured to have high accuracy, which can provide strong support for the fields of geology, mineral resource exploration, engineering construction, geological disaster prevention and control, etc., and can discover potential patterns and associations in geological data, promote the revelation of the inherent laws of geological phenomena, and help geologists to have a deeper understanding of geological processes.

[0019] In summary, the present invention has the characteristics of efficient and timely complex data processing, less manual intervention and high prediction accuracy. BRIEF DESCRIPTION OF THE DRAWINGS

[0020] Figure 1 It is a schematic diagram of the data collection and integration process of the present invention; Figure 2 It is a schematic diagram of the data preprocessing process of the present invention. DETAILED DESCRIPTION

[0021] In order to make the purpose, technical solution and advantages of the present invention more clearly understood, the present invention is further described in detail below in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present invention and are not used to limit the present invention.

[0022] like Figure 1 and 2 As shown, the present invention includes data collection and integration, data preprocessing, data analysis and interpretation, AI model construction, model evaluation and optimization, deep learning and analysis, AI analysis results, and AI model iterative improvement steps, and the specific contents are as follows: A. Data collection and integration: Collect geological data from different sources and then integrate the collected data into a unified format and structure; B. Data preprocessing: preprocess the aforementioned integrated data and extract features or attributes from the preprocessed data; C. Data analysis and interpretation: Analyze and interpret the pre-processed data based on the extracted features or attributes; D. AI model construction: Select a suitable AI algorithm, build an AI model based on the selected algorithm, and then use the preprocessed data to train the AI ​​model; E. Model evaluation and optimization: Use test set data to evaluate the AI ​​model and optimize the AI ​​model based on the evaluation results; G. Deep learning and analysis: Perform deep learning and analysis on pre-processed data through NLP technology and / or data modeling methods; H. AI analysis results: Based on the results of the aforementioned data analysis, the geological phenomena are interpreted and modeled, and decision support is provided; J. Iterative improvement of AI model: Iteratively improve the geological AI model regularly or as needed.

[0023] The A step includes the following sub-steps: A1. Determine the type of data: Determine the type of geological data that needs to be collected; A2. Collect data: Collect data from various sources using manual input, automated sensors, and / or database queries; A3. Data Integration I: Use GIS technology to integrate the collected data into a unified platform or database, or use machine learning algorithms to mine and analyze the collected data to discover the correlation and patterns between the data, and then integrate them into a unified format and structure.

[0024] The geological data types include geological exploration data, satellite images, drilling records, seismic data and geological reports.

[0025] The step B comprises the following sub-steps: B1. Data preprocessing: remove duplicate data, missing values, and outliers from the data integrated in step A; B2. Data integration II: Integrate the pre-processed data into a unified measurement unit or range; B3. Data extraction: extracting preset features or attributes from the above integrated data.

[0026] The step C comprises the following sub-steps: C1. Feature analysis: Analyze the influence of each feature or attribute extracted in step B on the prediction results of the AI ​​model, so as to understand the inherent laws and correlations of geological data; C2. Visualization: Use visualization techniques to display the results of the above data feature analysis to facilitate geologists’ understanding and interpretation; C3. Anomaly processing: Identify outliers or abnormal conditions in the preprocessed data and explain their geological significance.

[0027] The D step includes the following sub-steps: D1. Algorithm selection: Based on the above data analysis, select one of the deep learning, machine learning or reinforcement learning algorithms as the AI ​​algorithm; D2. Build AI model: Build an AI model including input layer, hidden layer and output layer according to the selected AI algorithm; D3. AI model training: Use preprocessed data to train the AI ​​model, and optimize model performance by adjusting model parameters based on the training results.

[0028] The construction of the AI ​​model also includes the selection of constructing an activation function and / or a loss function.

[0029] The E step includes the following sub-steps: E1. AI model evaluation: First, find a data set with similar geological characteristics and distribution as a test set, then use the test set data to input the AI ​​model, compare the prediction results of the AI ​​model with the actual results, calculate the accuracy, recall rate and / or F1-score indicators of the AI ​​model, and evaluate the performance of the AI ​​model on the test set; E2. AI model optimization: Adjust the parameters or structure of the AI ​​model based on the above evaluation results.

[0030] The adjustment of the parameters or structure of the AI ​​model includes adjusting the learning rate, adding hidden layers and using regularization.

[0031] The G step includes the following optional items: G1. Deep learning: Use the trained AI model to conduct in-depth analysis on pre-processed data, explore the potential patterns and associations in the data, and provide strong support for the interpretation and modeling of geological phenomena; G2, NLP Assisted: Using NLP technology, extract key information from pre-processed text data and identify trends and patterns in text data, and then automatically generate geological reports based on the extracted information and identified trends, providing geologists with fast and accurate data interpretation and analysis results; G3. Image Classification: Identify geological features from pre-processed aerial and satellite images and classify geological phases from images of rocks and minerals using computer vision techniques.

[0032] The G1 step includes the following steps: G11. Directly learn the mapping of seismic data to velocity models through AI models, and improve the accuracy and speed of inversion through end-to-end proxy models or a combination of data-driven and physical-driven methods; G12. Use deep learning technology to automatically identify geomorphic features in geomorphic maps and improve the accuracy of geomorphic classification and feature recognition.

[0033] The deep learning technology includes convolutional neural network.

[0034] The H step comprises the following sub-steps: H1. Interpretation and modeling: Based on the analysis results of the AI ​​model, the geological phenomena of structural evolution, lithology distribution and / or mineral resource prediction are interpreted and modeled; H2. Decision support: Based on the above interpretation and modeling, provide decision support for geological exploration, resource development and environmental protection, including exploration area selection, resource reserve assessment and / or geological disaster early warning.

[0035] The J step includes the following sub-steps: J1. Application feedback: Collect feedback from geologists and domain experts to understand the performance of the model in practical applications; J2. Model iteration: Iterate and improve the AI ​​model based on feedback and new geological data.

[0036] Example 1 like Figure 1 and 2As shown in the figure, the seismic data is inverted and optimized. The specific process is as follows: S100: Determine the types of geological data that need to be collected, such as geological exploration data, satellite images, drilling records, seismic data, and geological reports, and then use manual input, automated sensors, and / or database queries to collect data from various sources. The collected data is then integrated into a unified platform or database using GIS technology, or machine learning algorithms are used to mine and analyze the collected data to discover the correlations and patterns between the data, and then integrated into a unified format and structure.

[0037] S200: Remove duplicate, missing and anomalous seismic data, convert data from different sources into a unified unit of measurement, and extract features related to seismic wave velocity.

[0038] S300: Analyze and interpret the pre-processed data according to the features extracted above. The specific process is as follows: S310: Analyze the influence of the aforementioned extracted features on the prediction results of the AI ​​model, so as to understand the inherent laws and correlations of the seismic data.

[0039] S320: Use visualization techniques to present the results of the aforementioned data feature analysis to facilitate geologists’ understanding and interpretation.

[0040] S330: Identify outliers or abnormal conditions in the preprocessed data and explain their geological significance.

[0041] S400: Select a deep learning algorithm and design a neural network model (i.e., AI model) that includes an input layer, multiple hidden layers, and an output layer. Use preprocessed seismic data to train the model and optimize model performance by adjusting model parameters based on the training results.

[0042] S500: Use the test set data to evaluate the neural network model, and adjust the neural network model parameters according to the evaluation results to improve the inversion accuracy and speed. The specific process is as follows: S510: first, a data set with similar geological characteristics and distribution is found as a test set, and then the test set data is used to input the neural network model, and the prediction results of the neural network model are compared with the actual results, and the accuracy, recall rate and / or F1-score index of the neural network model are calculated to evaluate the performance of the neural network model on the test set; S520: Adjust the parameters or structure of the neural network model according to the aforementioned evaluation results.

[0043] S520: According to the aforementioned evaluation results, the neural network model is modified by adjusting the learning rate, adding hidden layers, and using regularization parameters to improve the inversion accuracy and speed.

[0044] S600: Perform deep learning and analysis on the pre-processed data through data modeling methods. The specific process is as follows: S610: Use the trained neural network model to conduct in-depth analysis of the preprocessed data, explore the potential patterns and associations in the data, and provide strong support for the interpretation and modeling of geological phenomena.

[0045] S700: Based on the results of the above data analysis, interpret and model the earthquake phenomenon and provide decision support. The specific process is as follows: S710: Based on the analysis results of the neural network model, the geological phenomena of structural evolution and lithology distribution prediction are explained and modeled.

[0046] S720: Provide decision support for geological disaster warning based on the above interpretation and modeling.

[0047] S800: Regularly or as needed, iteratively improve the neural network model. The specific process is as follows: S810: Collect feedback from geologists and earthquake experts on how the model performs in real-world applications.

[0048] S820: Iterate and improve the neural network model based on feedback and new geological data.

[0049] Example 2 like Figure 1 and 2 As shown in the figure, the geological report is automatically generated. The specific process is as follows: S100: Determine the types of geological data that need to be collected, such as geological exploration data, satellite images, drilling records, seismic data, and geological reports, and then use manual input, automated sensors, and / or database queries to collect data from various sources. The collected data is then integrated into a unified platform or database using GIS technology, or machine learning algorithms are used to mine and analyze the collected data to discover the correlations and patterns between the data, and then integrated into a unified format and structure.

[0050] S200: Remove duplicate, missing and invalid data, convert data from different sources into a unified unit of measurement, and extract key geological information and features.

[0051] S300: Analyze and interpret the pre-processed data according to the features extracted above. The specific process is as follows: S310: Analyze the influence of the aforementioned extracted features on the prediction results of the AI ​​model, so as to understand the inherent laws and correlations of geological data.

[0052] S320: Use visualization techniques to present the results of the aforementioned data feature analysis to facilitate geologists’ understanding and interpretation.

[0053] S330: Identify outliers or abnormal conditions in the preprocessed data and explain their geological significance.

[0054] S400: Select the NLP algorithm, design a neural network model (i.e., AI model) consisting of a text input layer, a feature extraction layer, and an output layer, use the preprocessed geological literature data to train the model, and optimize the model performance by adjusting the model parameters based on the training results.

[0055] S500: Use the test set data to evaluate the neural network model, and adjust the neural network model parameters based on the evaluation results to improve the accuracy of data extraction and report generation. The specific process is as follows: S510: First, find a data set with similar geological characteristics and distribution as a test set, then use the test set data to input the neural network model, compare the prediction results of the neural network model with the actual results, calculate the accuracy, recall rate and / or F1-score indicators of the neural network model, and evaluate the performance of the neural network model on the test set.

[0056] S520: Based on the aforementioned evaluation results, the neural network model is modified by adjusting the learning rate, adding hidden layers, and using regularization parameters to improve the accuracy of data extraction and report generation.

[0057] S600: Use NLP technology to conduct deep learning and analysis on the pre-processed data. The specific process is as follows: S620: Use NLP technology to extract key information from pre-processed text data and identify trends and patterns in text data. Then, based on the extracted information and identified trends, automatically generate geological reports, thereby greatly reducing the time and workload of geological report writing, improving the accuracy and readability of the report, and providing geologists with fast and accurate data interpretation and analysis results.

[0058] S700: Based on the results of the above data analysis, the geological phenomena are interpreted and modeled, and decision support is provided. The specific process is as follows: S710: Based on the analysis results of the neural network model, interpret and model the geological phenomena of structural evolution, lithology distribution and / or mineral resource prediction.

[0059] S720: Based on the above interpretation and modeling, provide decision support for geological exploration, resource development and environmental protection, including exploration area selection, resource reserve assessment and / or geological disaster warning.

[0060] S800: Regularly or as needed, iteratively improve the neural network model. The specific process is as follows: S810: Collect feedback from geologists and earthquake experts on how the model performs in real-world applications.

[0061] S820: Iterate and improve the neural network model based on feedback and new geological data.

[0062] The above is only a preferred specific embodiment of the present invention, but the protection scope of the present invention is not limited thereto. Any changes or substitutions that can be easily thought of by any technician familiar with the technical field within the technical scope disclosed by the present invention should be included in the protection scope of the present invention. Therefore, the protection scope of the present invention should be based on the protection scope of the claims.

Claims

1. A geological data analysis method based on AI algorithm, characterized in that It includes data collection and integration, data preprocessing, data analysis and interpretation, AI model construction, model evaluation and optimization, deep learning and analysis, AI analysis results, and AI model iterative improvement steps. The specific contents are as follows: A. Data collection and integration: Collect geological data from different sources and then integrate the collected data into a unified format and structure; B. Data preprocessing: preprocess the aforementioned integrated data and extract features or attributes from the preprocessed data; C. Data analysis and interpretation: Analyze and interpret the pre-processed data based on the extracted features or attributes; D. AI model construction: Select a suitable AI algorithm, build an AI model based on the selected algorithm, and then use the preprocessed data to train the AI ​​model; E. Model evaluation and optimization: Use test set data to evaluate the AI ​​model and optimize the AI ​​model based on the evaluation results; G. Deep learning and analysis: Perform deep learning and analysis on pre-processed data through NLP technology and / or data modeling methods; H. AI analysis results: Based on the results of the aforementioned data analysis, the geological phenomena are interpreted and modeled, and decision support is provided; J. Iterative improvement of AI model: Iteratively improve the geological AI model regularly or as needed.

2. The geological data analysis method based on AI algorithm according to claim 1 is characterized in that: The A step includes the following sub-steps: A1. Determine the type of data: Determine the type of geological data that needs to be collected; A2. Collect data: Collect data from various sources using manual input, automated sensors, and / or database queries; A3. Data Integration I: Use GIS technology to integrate the collected data into a unified platform or database, or use machine learning algorithms to mine and analyze the collected data to discover the correlation and patterns between the data, and then integrate them into a unified format and structure.

3. The geological data analysis method based on AI algorithm according to claim 1 is characterized in that: The step B comprises the following sub-steps: B1. Data preprocessing: remove duplicate data, missing values, and outliers from the data integrated in step A; B2. Data integration II: Integrate the pre-processed data into a unified measurement unit or range; B3. Data extraction: extracting preset features or attributes from the above integrated data.

4. The geological data analysis method based on AI algorithm according to claim 1 is characterized in that: The step C comprises the following sub-steps: C1. Feature analysis: Analyze the influence of each feature or attribute extracted in step B on the prediction results of the AI ​​model; C2. Visualization: Use visualization technology to display the results of the above data feature analysis; C3. Anomaly processing: Identify outliers or abnormal conditions in the preprocessed data and explain their geological significance.

5. The geological data analysis method based on AI algorithm according to claim 1 is characterized in that: The D step includes the following sub-steps: D1. Algorithm selection: Based on the above data analysis, select one of the deep learning, machine learning or reinforcement learning algorithms as the AI ​​algorithm; D2. Build AI model: Build an AI model including input layer, hidden layer and output layer according to the selected AI algorithm; D3. AI model training: Use preprocessed data to train the AI ​​model, and optimize model performance by adjusting model parameters based on the training results.

6. The geological data analysis method based on AI algorithm according to claim 1, characterized in that: The E step includes the following sub-steps: E1. AI model evaluation: First, find a data set with similar geological characteristics and distribution as a test set, then use the test set data to input the AI ​​model, compare the prediction results of the AI ​​model with the actual results, calculate the accuracy, recall rate and / or F1-score indicators of the AI ​​model, and evaluate the performance of the AI ​​model on the test set; E2. AI model optimization: Adjust the parameters or structure of the AI ​​model based on the above evaluation results.

7. The geological data analysis method based on AI algorithm according to claim 1, characterized in that: The G step includes the following optional items: G1. Deep learning: Use the trained AI model to conduct in-depth analysis on pre-processed data, explore the potential patterns and associations in the data, and provide strong support for the interpretation and modeling of geological phenomena; G2, NLP Assisted: Using NLP technology, extract key information from pre-processed text data and identify trends and patterns in text data, and then automatically generate geological reports based on the extracted information and identified trends, providing geologists with fast and accurate data interpretation and analysis results; G3. Image Classification: Identify geological features from pre-processed aerial and satellite images and classify geological phases from images of rocks and minerals using computer vision techniques.

8. The geological data analysis method based on AI algorithm according to claim 7 is characterized in that: The G1 step includes the following steps: G11. Directly learn the mapping of seismic data to velocity models through AI models, and improve the accuracy and speed of inversion through end-to-end proxy models or a combination of data-driven and physical-driven methods; G12. Use deep learning technology to automatically identify geomorphic features in geomorphic maps and improve the accuracy of geomorphic classification and feature recognition.

9. The geological data analysis method based on AI algorithm according to claim 1, characterized in that: The H step comprises the following sub-steps: H1. Interpretation and modeling: Based on the analysis results of the AI ​​model, the geological phenomena of structural evolution, lithology distribution and / or mineral resource prediction are interpreted and modeled; H2. Decision support: Based on the above interpretation and modeling, provide decision support for geological exploration, resource development and environmental protection, including exploration area selection, resource reserve assessment and / or geological disaster early warning.

10. The geological data analysis method based on AI algorithm according to any one of claims 1 to 9, characterized in that: The J step includes the following sub-steps: J1. Application feedback: Collect feedback from geologists and domain experts to understand the performance of the model in practical applications; J2. Model iteration: Iterate and improve the AI ​​model based on feedback and new geological data.

Images