Geological environment situation analysis method based on multi-scale observation and three-dimensional field model

Abstract

The present invention relates to the field of data processing technology, and more specifically to a method for geological environment situation analysis based on multi-scale observation and three-dimensional field models. The method comprises: S1: collecting geological data of a target area; S2: constructing a three-dimensional model of the target area based on remote sensing images, and fusing basic geographic data with geological environment data to obtain a dynamic three-dimensional field model; S3: constructing a geological environment database to obtain factor data affecting each geological event; S4: creating a geological event prediction model, and training the geological event prediction model using historical factor data and corresponding geological events; S5: predicting the occurrence of geological events after a predetermined time period based on the trained geological event prediction model and factor data before a predetermined time period. The present invention solves the problems of poor dynamics of geological environment observation and inaccurate situation analysis, realizes multi-scale and dynamic observation of the geological environment, and improves the accuracy of situation analysis.

Description

Technical Field

[0001] The present invention relates to the field of data processing technology, and more specifically to a geological environment situation analysis method based on multi-scale observation and three-dimensional field model. Background Art

[0002] In the past, geological environment monitoring often required in-person staff, which not only posed significant safety risks but was also extremely time-consuming and labor-intensive. The scope and depth of geological environment monitoring services are limited and have yet to fully meet social needs. To accurately understand the spatiotemporal distribution patterns, changes, and development trends of the geological environment and support its prevention, protection, and emergency decision-making, my country has carried out a series of geological environment surveys, monitoring, and assessments. Similar prior art includes Chinese patent publication number CN115587155A, which proposes a GIS-based geological data presentation method and system. This method uses regular drone patrols to acquire image data in a timely and accurate manner, extracting in-depth geological data. Regional classification is performed, and two-dimensional geological models are sequentially constructed. Related two-dimensional geological models are then linked, with the models continuously corrected and tested, ultimately presenting the geological data in the form of three-dimensional images. This approach features low labor costs, real-time data acquisition, high accuracy, and comprehensive presentation, enabling the monitoring and study of geological conditions over a wide area. It can also generate accurate early warning signals, enabling early warning before disasters occur, facilitating proactive protective measures and minimizing losses. Similar prior art includes U.S. Patent Publication No. US8024123B2, which discloses a method for predicting underground formation characteristics of a well site. The method comprises: obtaining seismic data for an area of ​​interest; using the seismic data to obtain an initial seismic cube; and using the seismic data to obtain a shifted seismic cube, wherein each shifted seismic cube is shifted from the initial seismic cube. Seismic cubes using the seismic data, wherein each shifted seismic cube is shifted from the initial seismic cube. The method also includes generating a neural network using the initial seismic cube, the shifted seismic cube, and well logging data, and applying the neural network to the seismic data to obtain a model of the area of ​​interest, wherein the model is used to adjust well site operations. Both of these patent documents address the problem of predicting and presenting geological hazards in the geological environment. However, the two-dimensional geological models mentioned above cannot intuitively and dynamically present the specific details of the geological environment, and the prediction accuracy of the geological hazards is also insufficient. Based on this, the present invention proposes a geological environment situation analysis method based on multi-observation humidity and three-dimensional field models. Summary of the Invention

[0003] In order to better solve the above problems, the present invention provides a geological environment situation analysis method based on multi-scale observation and three-dimensional field model, the method comprising:

[0004] Collecting geological data of the target area through a data collection unit, wherein the geological data includes basic geographical data and geological environment data, and the geological environment data includes a variety of environmental data;

[0005] constructing a three-dimensional model of the target area based on the remote sensing image in the basic geographic data, fusing the basic geographic data, the geological environment data, and the geological environment dynamic image corresponding to the geological environment data to obtain a dynamic three-dimensional field model of the target area, and periodically updating the geological data and the geological environment dynamic image;

[0006] Constructing a geological environment database based on the dynamic three-dimensional field model, and obtaining historical geological data within a predetermined time period before each geological event occurs based on the geological environment database, and obtaining factor data affecting each geological event by analyzing the difference between the historical geological data and normal geological data;

[0007] Creating a geological event prediction model by using historical element data corresponding to the geological event in the geological database, and training the geological event prediction model by using the historical element data;

[0008] The occurrence of the geological event after the predetermined time period is predicted based on the trained geological event prediction model and the element data before the predetermined time period.

[0009] As a preferred technical solution of the present invention, obtaining the dynamic three-dimensional model of the target area includes:

[0010] The target area is divided into grids, and the remote sensing images in the basic geographic data are mapped to construct a three-dimensional model of the target area according to the position correspondence between the remote sensing images and the grids;

[0011] Standardizing the other basic geographic data and the environmental data except the remote sensing image, and marking them in the three-dimensional model, and interpolating at locations where the geological data are discontinuous to obtain continuous geological data;

[0012] The geological environment data corresponding to each grid is integrated with the basic geographic data marked in the three-dimensional model, and the geological environment data within a set time period is input into a corresponding existing neural network model to obtain dynamic images of the geological environment from multiple visual angles;

[0013] The dynamic geological environment image corresponding to each geological environment data is mapped and fused based on the three-dimensional model according to a visual angle to obtain a dynamic three-dimensional field model of the target area.

[0014] As a preferred technical solution of the present invention, obtaining element data of each geological event includes:

[0015] Constructing a geological database within the target area based on the dynamic three-dimensional field model, wherein the geological database includes a basic geographic database and multiple geological environment databases, and the environmental database includes continuous geological environment data and dynamic images of the geological environment from multiple visual angles;

[0016] Obtaining N identical geological events based on the geological database, and obtaining historical geological data within a predetermined time period before each geological event occurs, performing a difference operation between a value corresponding to each data type in the N sets of historical geological data and a normal value corresponding to the data type in the corresponding normal geological data, and obtaining a difference result corresponding to each data type in the historical geological data;

[0017] Clustering all the difference results corresponding to N groups of geological data according to the size of the difference, and when the number of difference results corresponding to the same data type in the clusters where each difference result is greater than a set threshold is greater than or equal to N, taking the data corresponding to the data type in the clusters as the element data of the geological event;

[0018] Similarly, the element data of each other geological event is obtained, wherein the value range of N is a positive integer greater than or equal to 2.

[0019] As a preferred technical solution of the present invention, the training of the geological event prediction model includes:

[0020] Creating the geological event prediction model, obtaining historical element data corresponding to each geological event through the geological database, dividing the historical element data into a test data set and a training data set, and training the geological event prediction model using the training data set;

[0021] The test data set is divided into multiple test data subsets according to the similarity with the data in the training data set, and each test data subset is input into the geological event prediction model, and an output result is obtained. The accuracy rate corresponding to the test data subset is calculated based on the output result and the geological event corresponding to the test data subset, wherein the number of data in the test data set is less than the number of data in the training data set.

[0022] As a preferred technical solution of the present invention, when the accuracy rate corresponding to the test data subset is less than the set accuracy rate, a first similarity between the test data set and the training data is obtained, and the training data is used as the target training data, and the geological event corresponding to the target training data is used as the target geological event. When the data collection unit collects new geological data of the target area, target element data of the target geological event is extracted from the new geological data, and a second similarity between the target element data and the corresponding target training data is calculated. When the difference between the second similarity and the first similarity is less than or equal to the set difference, the target element data is input into the geological event prediction model, and a prediction result is obtained. The first geological environment dynamic image of the geological environment corresponding to the target geological event in the new geological data and K second geological environment dynamic images before the first geological environment dynamic image are analyzed and compared with the historical geological environment dynamic images corresponding to the occurrence of the target geological event in chronological order to obtain a similarity sequence, and whether the geological event occurs is judged based on the prediction result and the similarity sequence.

[0023] As a preferred technical solution of the present invention, judging whether the geological event occurs based on the prediction result and the similarity sequence includes:

[0024] When the similarity in the similarity sequence increases over time, if the prediction result is that the target geological event occurs, it can be determined that the final prediction result of the target geological event will occur after the predetermined time period; when the similarity in the similarity sequence increases over time; if the prediction result is that the target geological event does not occur, when the first similarity data in the similarity sequence is greater than a set value, it can be determined that the final prediction result of the target geological event will occur after the predetermined time period, otherwise the target geological event does not occur.

[0025] As a preferred technical solution of the present invention, by inputting the element data obtained within the predetermined time into the geological event prediction model, when the output result is that the geological event will occur, corresponding measures are taken to prevent the occurrence of geological disasters.

[0026] As a preferred technical solution of the present invention, the geological environment data at least includes mining environment data, soil environment data and water environment data, and the basic geographic data at least includes the landform, vegetation, water system, soil, rock, geological structure, animal distribution, climate and natural resources of the target area.

[0027] The present invention also provides a geological environment situation system based on multi-scale observation and three-dimensional field model, which is used to implement the above method, and includes:

[0028] A data collection unit is used to collect geological data of the target area, wherein the geological data includes basic geographical data and geological environment data, and the geological environment data includes multiple environmental data;

[0029] a model creation unit, configured to construct a three-dimensional model of the target area based on the remote sensing image in the basic geographic data, fuse the basic geographic data, the geological environment data, and the geological environment dynamic image corresponding to the geological environment data, obtain a dynamic three-dimensional field model of the target area, and periodically update the geological data and the geological environment dynamic image;

[0030] an analysis unit, configured to construct a geological environment database based on the dynamic three-dimensional field model, obtain historical geological data within a predetermined time period before each geological event occurs based on the geological environment database, and obtain factor data affecting each geological event by analyzing the differences between the historical geological data and normal geological data;

[0031] The model creation unit is further used to create a geological event prediction model by using the historical element data corresponding to the geological event in the geological database, and to train the geological event prediction model by using the historical element data;

[0032] A prediction unit is used to predict the occurrence of the geological event after the predetermined time period based on the trained geological event prediction model and the element data before the predetermined time period.

[0033] The present invention also provides a computer storage medium, wherein the storage medium stores program instructions, wherein when the program instructions are executed, the device where the storage medium is located is controlled to execute the above method.

[0034] Compared with the prior art, the beneficial effects of the present invention are at least as follows:

[0035] The present invention constructs the above-mentioned three-dimensional model of the above-mentioned target area through remote sensing images in basic geographic data, obtains the geological environment dynamic image corresponding to the above-mentioned geological environment data by standardizing the above-mentioned basic geographic data and the above-mentioned geological environment data, and fuses the standardized above-mentioned basic geographic data, the above-mentioned geological environment data and the corresponding above-mentioned geological environment dynamic image based on the above-mentioned three-dimensional model to obtain the above-mentioned dynamic three-dimensional field model, thereby improving the observation effect, grasping the geological development dynamics in real time, obtaining historical geological data within a predetermined time period before the occurrence of N above-mentioned geological events, and obtaining factor data affecting each of the above-mentioned geological events through the difference between the historical geological data and the normal geological data, so as to facilitate the prediction of each of the above-mentioned geological events by establishing a geological event prediction model, obtaining the historical factor data corresponding to each of the above-mentioned geological events through the above-mentioned geological database, and dividing the above-mentioned historical factor data set into the above-mentioned training data set and test data set, This facilitates better training of the above-mentioned geological event prediction model, and also divides the above-mentioned test data into multiple test data subsets according to the similarity with the above-mentioned training data set, and tests the above-mentioned geological event prediction model, and obtains the accuracy corresponding to each test data subset. Through the above-mentioned technical solution, not only can the above-mentioned geological event prediction model after training be obtained, but also the accuracy of the above-mentioned geological event prediction model for each element data segment corresponding to the above-mentioned test data subset can be obtained more accurately. When the accuracy of the prediction of the element data segment corresponding to the above-mentioned test data subset is not high enough, the final prediction result is further determined according to the changing trend of the dynamic image of the geological environment corresponding to the above-mentioned geological event. Through the mutual coordination of the above-mentioned technical solutions, not only the dynamic three-dimensional model of multi-scale observation can be obtained, but also the dynamic image of the geological environment in the above-mentioned dynamic three-dimensional model can be coordinated with the above-mentioned geological event prediction model, so as to more accurately predict the above-mentioned geological event. BRIEF DESCRIPTION OF THE DRAWINGS

[0036] Figure 1 This is a flow chart of the geological environment situation analysis method based on multi-scale observation and three-dimensional field model of the present invention;

[0037] Figure 2 This is a structural diagram of the geological environment situation analysis system based on multi-scale observation and three-dimensional field model of the present invention. DETAILED DESCRIPTION

[0038] In order to make the purpose, technical solutions and advantages of the present invention more clearly understood, the present invention will be further described in detail below with reference to 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 intended to limit the present invention.

[0039] The present invention provides a geological environment situation analysis method based on multi-scale observation and three-dimensional field model, by constructing a dynamic three-dimensional field model of multi-scale observation of the target area, and by analyzing the correlation between each type of data in the geological data and the geological event, and obtaining the element data of the above geological event, and creating a geological event prediction model based on the above element data, based on the geological event prediction model and the geological environment dynamic image in the dynamic three-dimensional field model, accurately predicting the occurrence of each of the above geological events, such as Figure 1 As shown, the method includes:

[0040] Step S1: collecting geological data of a target area through a data collection unit, wherein the geological data includes basic geographical data and geological environment data, and the geological environment data includes a variety of environmental data;

[0041] Step S2: constructing a three-dimensional model of the target area based on the remote sensing image in the basic geographic data, fusing the basic geographic data, the geological environment data, and the geological environment dynamic image corresponding to the geological environment data to obtain a dynamic three-dimensional field model of the target area, and periodically updating the geological data and the geological environment dynamic image;

[0042] Specifically, the geological data of the target area is collected by the data collection unit, and the basic geographic data and the geological environment data in the geological data are obtained, wherein the geological environment data at least include mining environment data, soil environment data and water environment data, and the basic geographic data at least include the landform, vegetation, water system, soil, rock, geological structure, animal distribution, climate and natural resources of the target area, thereby providing data support for creating the dynamic three-dimensional field model. The basic geographic data also includes remote sensing images of the target area, and the three-dimensional model of the target area is constructed based on the remote sensing images. Since the three-dimensional model can only reflect the landform and terrain characteristics of the target area and cannot observe the geological environment of the target area at multiple scales, the basic geographic data and the geological environment data are standardized to obtain the geological environment dynamic image corresponding to the geological environment data, and the standardized basic geographic data, the geological environment data and the corresponding geological environment dynamic image are fused based on the three-dimensional model to obtain the dynamic three-dimensional field model. By periodically updating the geological data and the corresponding geological environment dynamic image, the observation effect is improved and the geological development dynamics are grasped in real time.

[0043] Step S3: constructing a geological environment database based on the dynamic three-dimensional field model, and obtaining historical geological data within a predetermined time period before each geological event occurs based on the geological environment database, and obtaining factor data affecting each geological event by analyzing the difference between the historical geological data and normal geological data;

[0044] Specifically, based on the above-mentioned geological database, N identical geological events are obtained, and historical geological data within a predetermined time period before each of the above-mentioned geological events occurs are obtained. By performing a difference operation between the numerical value corresponding to each data type in the N groups of the above-mentioned historical geological data and the corresponding normal numerical value in the normal geological data, N above-mentioned difference results corresponding to each data type in the above-mentioned historical geological data are obtained. All the above-mentioned difference results are clustered according to the difference size, and cluster clusters are obtained. Moreover, when the corresponding difference results in the above-mentioned clusters are all greater than a set threshold, and the number of difference results corresponding to the same data type in the above-mentioned clusters is greater than or equal to N, it is indicated that the above-mentioned data type is related to the above-mentioned geological event. Therefore, the data corresponding to the above-mentioned data type is used as the element data of the above-mentioned geological event, and the same method is used to obtain the element data corresponding to each other geological event. Through the above-mentioned technical solution, the element data corresponding to each geological event can be accurately obtained, thereby facilitating the prediction of each of the above-mentioned geological events by establishing a geological event prediction model.

[0045] Step S4: creating a geological event prediction model, and training the geological event prediction model using the historical element data corresponding to the geological event in the geological database and the historical element data and the corresponding geological event;

[0046] Specifically, the historical element data corresponding to each of the above-mentioned geological events is obtained through the above-mentioned geological database, and the above-mentioned historical element data set is divided into the above-mentioned training data set and the test data set, so as to facilitate better training of the above-mentioned geological event prediction model. The above-mentioned test data is also divided into multiple test data subsets according to the similarity with the above-mentioned training data set, and the above-mentioned geological event prediction model is tested, and the accuracy corresponding to each test data subset is obtained. Through the above-mentioned technical solution, not only the trained geological event prediction model can be obtained, but also the accuracy of the above-mentioned geological event prediction model for each element data segment corresponding to the above-mentioned test data subset can be obtained more accurately.

[0047] Step S5: predicting the occurrence of the geological event after the predetermined time period based on the trained geological event prediction model and the element data before the predetermined time period.

[0048] Specifically, by using the above-mentioned geological event prediction model to timely predict the occurrence of the above-mentioned geological events, corresponding measures can be taken in time to prevent the occurrence of geological disasters and avoid unnecessary casualties and property losses.

[0049] Furthermore, obtaining the dynamic three-dimensional model of the target area includes:

[0050] The target area is divided into grids, and the remote sensing images in the basic geographic data are mapped to construct a three-dimensional model of the target area according to the position correspondence between the remote sensing images and the grids;

[0051] Standardizing the other basic geographic data and the environmental data except the remote sensing image, and marking them in the three-dimensional model, and interpolating at locations where the geological data are discontinuous to obtain continuous geological data;

[0052] Specifically, the target area is divided into grids, and mapping is performed according to the positional relationship between each grid and the remote sensing image of the target area, thereby constructing the three-dimensional model of the target area. Since the three-dimensional model can only reflect the landform and terrain characteristics of the target area, and cannot observe the geological environment of the target area at multiple scales, the format and unit of the other basic geographic data and the environmental data other than the remote sensing image are standardized to facilitate data fusion. The other basic geographic data and the environmental data are also annotated in the three-dimensional model, that is, annotated at the corresponding position in the three-dimensional space of the three-dimensional model, and at the position where the geological data is discontinuous, the first position with the highest similarity and the closest distance to the discontinuous position is found, and the geological data at the first position is used to interpolate the geological data at the discontinuous position, thereby obtaining continuous geological data. Through the above technical solution, a foundation is laid for further data fusion and obtaining dynamic images of the geological environment corresponding to each geological environment.

[0053] The geological environment data corresponding to each grid is integrated with the basic geographic data marked in the three-dimensional model, and the geological environment data within a set time period is input into a corresponding existing neural network model to obtain dynamic images of the geological environment from multiple visual angles;

[0054] The dynamic geological environment image corresponding to each geological environment data is mapped and fused based on the three-dimensional model according to a visual angle to obtain a dynamic three-dimensional field model of the target area.

[0055] Specifically, by fusing the geological environment data corresponding to each of the above-mentioned grids with the above-mentioned basic geographic data, the geological characteristics of the grid can be reflected at multiple scales. The geological environment data within a set time period is also input into the corresponding neural network model to obtain geological environment dynamic images of multiple visual angles corresponding to each of the above-mentioned geological environment data, wherein the above-mentioned neural network model is a trained model. For example, by inputting geological water environment data into the corresponding neural network model, a geological water environment dynamic image can be output. The geological water environment data at least includes: geological water flow velocity, geological water composition, geological water distribution position and geological water topographic distribution. The geological water environment dynamic image can reflect the changes in water flow characteristics, water potential distribution and water color within the above-mentioned set time period. The geological environment dynamic images corresponding to different geological environment data are also mapped and fused based on the above-mentioned three-dimensional model according to the same visual angle to obtain the above-mentioned dynamic three-dimensional field model. Through the above-mentioned technical solution, when the user selects a set visual angle to view the above-mentioned dynamic three-dimensional field model, it is convenient to observe the geological environment of the above-mentioned target area at multiple scales.

[0056] Furthermore, element data of each geological event is obtained, including:

[0057] Constructing a geological database within the target area based on the dynamic three-dimensional field model, wherein the geological database includes a basic geographic database and multiple geological environment databases, and the environmental database includes continuous geological environment data and dynamic images of the geological environment from multiple visual angles;

[0058] Obtaining N identical geological events based on the geological database, and obtaining historical geological data within a predetermined time period before each geological event occurs, performing a difference operation between a value corresponding to each data type in the N sets of historical geological data and a normal value corresponding to the data type in the corresponding normal geological data, and obtaining a difference result corresponding to each data type in the historical geological data;

[0059] Clustering all the difference results corresponding to N groups of geological data according to the size of the difference, and when the number of difference results corresponding to the same data type in the clusters where each difference result is greater than a set threshold is greater than or equal to N, taking the data corresponding to the data type in the clusters as the element data of the geological event;

[0060] Similarly, the element data of each other geological event is obtained, where the value range of N is a positive integer greater than or equal to 2.

[0061] Specifically, since the geological data in the above-mentioned dynamic three-dimensional field model are standardized and continuous, the geological database of the above-mentioned target area can be created through the above-mentioned dynamic three-dimensional field model. The geological environment database in the above-mentioned geological database at least includes a mining environment database, a groundwater environment database and a soil environment database. Since the above-mentioned dynamic three-dimensional field model corresponds to a large number of geological data in the same area, and since the same geological event may be formed by the joint action of multiple factors, the above-mentioned historical geological data before the occurrence of N same geological events are obtained based on the above-mentioned geological database, and the corresponding values ​​of each data type in the above-mentioned historical geological data are obtained by performing a difference operation on the values ​​corresponding to each data type in the N groups of the above-mentioned historical geological data and the corresponding normal values ​​in the above-mentioned normal geological data. The N above-mentioned difference results are clustered according to the difference size through all the above-mentioned difference results, and cluster clusters are obtained. The corresponding difference results in the above-mentioned cluster clusters are all greater than the set threshold value, and when the number of difference results corresponding to the same data type in the above-mentioned cluster cluster is greater than or equal to N, that is, the difference between the numerical value corresponding to the above-mentioned data type in each of the above-mentioned historical geological data and the normal numerical value is large, it means that the above-mentioned data type is related to the above-mentioned geological event. Therefore, the data corresponding to the above-mentioned data type is used as the element data of the above-mentioned geological event, and the same method is used to obtain the element data corresponding to each other geological event. Through the above-mentioned technical solution, the element data corresponding to each geological event can be accurately obtained, thereby facilitating the prediction of each of the above-mentioned geological events by establishing a geological event prediction model.

[0062] Furthermore, the training of the geological event prediction model includes:

[0063] Creating the geological event prediction model, obtaining a historical element dataset corresponding to each geological event through the geological database, dividing the historical element dataset into a test dataset and a training dataset, and training the geological event prediction model using the training dataset;

[0064] The test data set is divided into multiple test data subsets according to the similarity with the data in the training data set, and each test data subset is input into the geological event prediction model, and an output result is obtained. The accuracy rate corresponding to the test data subset is calculated based on the output result and the geological event corresponding to the test data subset, wherein the number of data in the test data set is less than the number of data in the training data set.

[0065] Specifically, the historical element data corresponding to each of the above-mentioned geological events is obtained through the above-mentioned geological database, and the above-mentioned historical element data set is divided into the above-mentioned training data set and the test data set, wherein the number of data in the above-mentioned test data set is smaller than the number of data in the above-mentioned training data set, so as to facilitate better training of the above-mentioned geological event prediction model, and the above-mentioned test data is also divided into multiple test data subsets according to the similarity with the above-mentioned training data set, and the above-mentioned geological event prediction model is tested, and the accuracy corresponding to each test data subset is obtained. Through the above-mentioned technical solution, not only the above-mentioned geological event prediction model after training can be obtained, but also the accuracy of the above-mentioned geological event prediction model for each element data segment corresponding to the above-mentioned test data subset can be obtained more accurately.

[0066] Furthermore, when the accuracy corresponding to the test data subset is less than the set accuracy, a first similarity between the test data set and the training data is obtained, and the training data is used as target training data, and the geological event corresponding to the target training data is used as the target geological event. When the data collection unit collects new geological data of the target area, target element data of the target geological event is extracted from the new geological data, and a second similarity between the target element data and the corresponding target training data is calculated. When the difference between the second similarity and the first similarity is less than or equal to the set difference, the target element data is input into the geological event prediction model, and a prediction result is obtained. By analyzing the first geological environment dynamic image of the geological environment corresponding to the target geological event in the new geological data and the K second geological environment dynamic images before the first geological environment dynamic image, they are compared in chronological order with the historical geological environment dynamic images corresponding to the occurrence of the target geological event to obtain a similarity sequence, and judging whether the geological event has occurred based on the prediction result and the similarity sequence.

[0067] Specifically, when acquiring new geological data and predicting the target geological event within a predetermined time period in the future, first determine whether the target element data in the geological data falls within the range where the accuracy is less than the set accuracy, for example: 80%, that is, first calculate the second similarity between the target element data in the new geological data and the target training data, and compare the second similarity with the first similarity. When the difference between the two is greater than or equal to the set difference, it is considered that the prediction is accurate. When the difference between the two is less than the set difference, it is considered that the target element data cannot obtain an accurate prediction result through the geological event prediction model. Since when the target geological event occurs, it can be predicted by the target geological event belonging to the target geological event. The geological environment dynamic image is intuitively reflected, for example: mine rock collapse, etc., can be displayed through the mine environment dynamic image. Therefore, the newly acquired first geological environment dynamic image and the K second geological environment dynamic images before the first geological environment dynamic image are compared with the historical geological environment dynamic image, and a similarity sequence is obtained. Based on the similarity change of the similarity sequence and the first output result, it is judged whether the target geological event occurs. Through the above technical solution, in the element data segment that the geological event prediction model cannot accurately predict, the similarity sequence and the first output result are further used to make a comprehensive judgment, thereby improving the accuracy of identifying the target geological event.

[0068] Furthermore, judging whether the geological event occurs based on the prediction result and the similarity sequence includes:

[0069] When the similarity in the similarity sequence increases over time, if the prediction result is that the target geological event occurs, it can be determined that the final prediction result of the target geological event will occur after the predetermined time period; when the similarity in the similarity sequence increases over time; if the prediction result is that the target geological event does not occur, when the first similarity data in the similarity sequence is greater than a set value, it can be determined that the final prediction result of the target geological event will occur after the predetermined time period, otherwise the target geological event does not occur.

[0070] Specifically, since the similarity sequence is obtained by comparing the second geological environment dynamic image and the first geological environment dynamic image arranged in chronological order with the historical geological environment dynamic image, the similarity sequence is also a time series. When the similarity in the similarity sequence increases over time, it indicates that the possibility of the target geological event occurring is increasing. Although the prediction accuracy of the geological event prediction model is not high at the newly acquired element data position, it still has a certain reference value. When the similarity sequence and the prediction result both indicate that the target geological event may occur, it is considered that the final result of the target geological event will occur after the predetermined time period. However, when the prediction result is that the target geological event will not occur, the two are contradictory. Since the prediction result is not accurate and the minimum similarity in the similarity sequence is greater than the set value, that is, when each geological environment dynamic image in the similarity sequence is relatively similar to the historical geological dynamic image, it is also considered that the final prediction result of the target geological event will occur after the predetermined time period. Otherwise, it will not occur. Through the above technical solution, the target geological event can be further accurately predicted.

[0071] Furthermore, by inputting the element data obtained within the predetermined time into the geological event prediction model, when the output result is that the geological event will occur, corresponding measures are taken to prevent the occurrence of geological disasters.

[0072] Specifically, by using the above-mentioned geological event prediction model to timely predict the occurrence of the above-mentioned geological events, corresponding measures can be taken in time to prevent the occurrence of geological disasters and avoid unnecessary casualties and property losses.

[0073] Furthermore, the geological environment data includes at least mining environment data, soil environment data and water environment data, and the basic geographic data includes at least the landform, vegetation, water system, soil, rock, geological structure, animal distribution, climate and natural resources of the target area.

[0074] The present invention also provides a geological environment situation system based on multi-scale observation and three-dimensional field model, which is used to implement the above method, such as Figure 2 As shown, the system includes:

[0075] A data collection unit is used to collect geological data of the target area, wherein the geological data includes basic geographical data and geological environment data, and the geological environment data includes multiple environmental data;

[0076] a model creation unit, configured to construct a three-dimensional model of the target area based on the remote sensing image in the basic geographic data, fuse the basic geographic data, the geological environment data, and the geological environment dynamic image corresponding to the geological environment data, obtain a dynamic three-dimensional field model of the target area, and periodically update the geological data and the geological environment dynamic image;

[0077] an analysis unit, configured to construct a geological environment database based on the dynamic three-dimensional field model, obtain historical geological data within a predetermined time period before each geological event occurs based on the geological environment database, and obtain factor data affecting each geological event by analyzing the differences between the historical geological data and normal geological data;

[0078] The model creation unit is further used to create a geological event prediction model, and train the geological event prediction model through the historical element data corresponding to the geological event in the geological database and the historical element data and the corresponding geological event;

[0079] A prediction unit is used to predict the occurrence of the geological event after the predetermined time period based on the trained geological event prediction model and the element data before the predetermined time period.

[0080] In summary, the present invention constructs the above-mentioned three-dimensional model of the above-mentioned target area through the remote sensing image in the basic geographic data, obtains the geological environment dynamic image corresponding to the above-mentioned geological environment data by standardizing the above-mentioned basic geographic data and the above-mentioned geological environment data, and fuses the standardized above-mentioned basic geographic data, the above-mentioned geological environment data and the corresponding above-mentioned geological environment dynamic image based on the above-mentioned three-dimensional model to obtain the above-mentioned dynamic three-dimensional field model, thereby improving the observation effect, grasping the geological development dynamics in real time, obtaining the historical geological data within a predetermined time period before the occurrence of N above-mentioned geological events, and obtaining the element data affecting each of the above-mentioned geological events through the difference between the historical geological data and the normal geological data, thereby facilitating the prediction of each of the above-mentioned geological events by establishing a geological event prediction model, obtaining the historical element data corresponding to each of the above-mentioned geological events through the above-mentioned geological database, and dividing the above-mentioned historical element data set into the above-mentioned training data set and the test data set. A test data set is provided to facilitate better training of the above-mentioned geological event prediction model. The above-mentioned test data is also divided into multiple test data subsets according to the similarity with the above-mentioned training data set, and the above-mentioned geological event prediction model is tested, and the accuracy corresponding to each test data subset is obtained. Through the above-mentioned technical solution, not only the trained geological event prediction model can be obtained, but also the accuracy of the above-mentioned geological event prediction model for each element data segment corresponding to the above-mentioned test data subset can be obtained more accurately. When the accuracy of the prediction of the element data segment corresponding to the above-mentioned test data subset is not high enough, the final prediction result is further determined according to the change trend of the dynamic image of the geological environment corresponding to the above-mentioned geological event. Through the above-mentioned technical solution, not only the dynamic three-dimensional model of multi-scale observation can be obtained, but also the dynamic image of the geological environment in the above-mentioned dynamic three-dimensional model can be coordinated with the above-mentioned geological event prediction model, so as to more accurately predict the above-mentioned geological event.

[0081] The technical features of the above embodiments can be combined arbitrarily. In order to make the description concise, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0082] The above embodiments merely represent several implementation methods of the present invention, and their descriptions are relatively specific and detailed, but they should not be construed as limiting the scope of the present invention. It should be noted that a person skilled in the art may make various modifications and improvements without departing from the scope of the present invention, and these modifications and improvements fall within the scope of protection of the present invention. Therefore, the scope of protection of the present invention shall be based on the appended claims.

[0083] The above are only preferred embodiments of the present invention and are not intended to limit the present invention. Any modifications, equivalent substitutions and improvements made within the spirit and principles of the present invention should be included in the scope of protection of the present invention.

Claims

1. A geological environment situation analysis method based on multi-scale observation and three-dimensional field model, characterized in that: include: Collecting geological data of the target area through a data collection unit, the geological data includes basic geographical data and geological environment data, and the geological environment data includes various environmental data; A three-dimensional model of the target area is constructed based on the remote sensing images in the basic geographic data, and the basic geographic data, geological environment data and the dynamic images of the geological environment corresponding to the geological environment data are fused to obtain a dynamic three-dimensional field model of the target area, including dividing the target area into a grid, and constructing the three-dimensional model of the target area by mapping the remote sensing images in the basic geographic data according to the position correspondence with the grid; other basic geographic data and environmental data other than the remote sensing images are standardized and annotated into the three-dimensional model, and interpolation is performed at the discontinuous position of the geological data to obtain continuous geological data, including finding a first position with the highest similarity and the closest distance to the discontinuous position, and using the geological data at the first position to interpolate the geological data at the discontinuous position, thereby obtaining continuous geological data; The geological environment data corresponding to each grid is integrated with the basic geographic data marked in the three-dimensional model. The geological environment data within a set time period is input into the corresponding existing neural network model to obtain corresponding geological environment dynamic images from multiple visual angles. The geological environment dynamic images corresponding to each geological environment data are mapped and integrated based on the three-dimensional model according to the visual angle to obtain a dynamic three-dimensional field model of the target area. The geological data and geological environment dynamic images are periodically updated. Constructing a geological environment database based on a dynamic three-dimensional field model, and obtaining historical geological data within a predetermined time period before each geological event based on the geological environment database, and obtaining factor data affecting each geological event by analyzing the differences between the historical geological data and normal geological data; Create a geological event prediction model using historical element data corresponding to geological events in the geological database and train the geological event prediction model using the historical element data; The occurrence of geological events after a predetermined time period is predicted based on the trained geological event prediction model and the element data before the predetermined time period.

2. The method according to claim 1, characterized in that Obtaining element data for each geological event includes: constructing a geological database within the target area based on a dynamic three-dimensional field model, the geological database including a basic geographic database and multiple geological environment databases, the environmental database including continuous geological environment data and dynamic images of geological environments from multiple visual angles; obtaining N identical geological events based on the geological database, and obtaining historical geological data within a predetermined time period before each geological event occurs, performing a difference operation between the numerical value corresponding to each data type in the N groups of historical geological data and the normal numerical value of the corresponding data type in the corresponding normal geological data, and obtaining the difference result corresponding to each data type in the historical geological data; clustering all the difference results corresponding to the N groups of geological data according to the size of the difference, and when the number of difference results corresponding to the same data type in a cluster whose each difference result is greater than a set threshold is greater than or equal to N, using the data corresponding to the data type in the cluster as element data for the geological event; similarly, obtaining element data for each other geological event, wherein the value range of N is a positive integer greater than or equal to 2.

3. The method according to claim 1, characterized in that The training of the geological event prediction model includes: creating a geological event prediction model, obtaining historical element data corresponding to each geological event through the geological database, dividing the historical element data into a test data set and a training data set, and training the geological event prediction model through the training data set; dividing the test data set into multiple test data subsets according to the similarity with the data in the training data set, and inputting each test data subset into the geological event prediction model, and obtaining the output result, and calculating the accuracy corresponding to the test data subset based on the output result and the geological event corresponding to the test data subset, wherein the number of data in the test data set is less than the number of data in the training data set.

4. The method according to claim 3, characterized in that When the accuracy rate corresponding to the test data subset is less than the set accuracy rate, the first similarity between the test data set and the training data is obtained, and the training data is used as the target training data, and the geological event corresponding to the target training data is used as the target geological event. When the data collection unit collects new geological data of the target area, the target element data of the target geological event is extracted from the new geological data, and the second similarity between the target element data and the corresponding target training data is calculated. When the difference between the second similarity and the first similarity is less than or equal to the set difference, the target element data is input into the geological event prediction model, and the prediction result is obtained. By analyzing the first geological environment dynamic image of the geological environment corresponding to the target geological event in the new geological data and the K second geological environment dynamic images before the first geological environment dynamic image, they are compared with the historical geological environment dynamic images corresponding to the occurrence of the target geological event in chronological order to obtain a similarity sequence, and judging whether the geological event occurs based on the prediction result and the similarity sequence.

5. The method according to claim 4, characterized in that Whether a geological event occurs is judged based on the prediction results and the similarity sequence, including: when the similarity in the similarity sequence becomes larger over time, if the prediction result is that the target geological event occurs, determining that the final prediction result of the target geological event will occur after a predetermined time period; when the similarity in the similarity sequence becomes larger over time; if the prediction result is that the target geological event does not occur, if the first similarity data in the similarity sequence is greater than a set value, determining that the final prediction result of the target geological event will occur after a predetermined time period, otherwise the target geological event does not occur.

6. The method according to claim 1, wherein By inputting the element data obtained within the predetermined time into the geological event prediction model, when the output result is that a geological event will occur, corresponding measures are taken to prevent the occurrence of geological disasters.

7. The method according to claim 1, characterized in that Geological environment data include at least mining environment data, soil environment data and water environment data, and basic geographic data include at least the landform, vegetation, water system, soil, rock, geological structure, animal distribution, climate and natural resources of the target area.

8. A geological environment situation system based on multi-scale observation and three-dimensional field model, the system is used to implement the method according to any one of claims 1 to 7, characterized in that: include: A data collection unit is used to collect geological data of the target area, wherein the geological data includes basic geographical data and geological environment data, and the geological environment data includes a variety of environmental data; A model creation unit is used to construct a three-dimensional model of the target area based on the remote sensing image in the basic geographic data, and to fuse the basic geographic data, geological environment data and the geological environment dynamic image corresponding to the geological environment data to obtain a dynamic three-dimensional field model of the target area, and to periodically update the geological data and the geological environment dynamic image; an analysis unit for constructing a geological environment database based on the dynamic three-dimensional field model, obtaining historical geological data within a predetermined time period before each geological event occurs based on the geological environment database, and obtaining factor data affecting each geological event by analyzing the differences between the historical geological data and normal geological data; The model creation unit is further used to create a geological event prediction model by using historical element data corresponding to geological events in the geological database, and to train the geological event prediction model by using the historical element data; The prediction unit is used to predict the occurrence of geological events after a predetermined time period based on the trained geological event prediction model and the element data before the predetermined time period.

9. A computer storage medium, characterized in that The storage medium stores program instructions, wherein when the program instructions are executed, the device where the storage medium is located is controlled to execute the method according to any one of claims 1 to 7.

Images