A spatiotemporal layout method for integrated monitoring of ecological environment in mining areas

Abstract

A method for the spatiotemporal layout of integrated monitoring of the ecological environment in mining areas, involving the technical field of ecological environment in mining areas, constructing a three-dimensional digital twin model of mining areas and an environmental characteristic database, storing ecological environment monitoring records of target mining areas and other mining areas, and obtaining vulnerability coefficients of various types of characteristic indexes in meteorological characteristic data; constructing a meteorological characteristic prediction model, obtaining fitting curves of estimated vulnerability coefficients of various types of characteristic indexes of various three-dimensional triangular grids in the current monitoring period, determining the number of ecological monitoring points, point distribution and monitoring time interval in each three-dimensional triangular grid; comparing the real-time meteorological characteristic data collected by each ecological monitoring point with the estimated meteorological characteristic data, performing an intraday adjustment mechanism according to the comparison results, performing real-time monitoring of environmental characteristic data, and updating the environmental characteristic database in real time according to the real-time monitoring results. The efficiency and accuracy of ecological environment monitoring in mining areas are significantly improved.

Description

Technical Field

[0001] The invention relates to the technical field of mining area ecological environment, and in particular to a mining area ecological environment integrated monitoring time and space layout method. Background Art

[0002] The prior art CN115170979A "A method for fine land classification in mining areas based on multi-source data fusion" first obtains multi-source data for preprocessing and multi-scale segmentation to obtain image objects, and then crops high-resolution image blocks of the image objects for deep feature extraction; then, the multi-branch convolutional neural network model automatically extracts feature column vectors and deep semantic features; uses a multi-source feature deep fusion module to perform feature fusion and uses a feature importance weight calculation module to calculate the importance weights of the deep features of multi-source data to achieve adaptive fusion of effective information; finally, a random forest classifier is used to use the deep fusion features to perform fine classification of mining area objects to obtain fine classification results of mining land.

[0003] The prior art CN118015494A "A regional enhanced multi-temporal stereoscopic monitoring method for erosion gullies in mining areas" has the following steps: step A, obtaining the mining area erosion gully data for preprocessing and constructing a sample data set; step B, constructing a mining area erosion gully identification model; step C, constructing a three-dimensional reconstruction model of the mining area erosion gullies; step D, using the data set to train the network model; step E, inputting the test data set into the network model trained with the training data set to obtain the identification results and stereoscopic model of the mining area erosion gullies.

[0004] The existing mining ecological environment methods have the problem of uneven distribution of monitoring points. The layout of monitoring points may not be reasonable, resulting in insufficient coverage of some important areas. At the same time, improper monitoring frequency, too high or too low monitoring frequency may affect the effectiveness of monitoring. Too high frequency may lead to waste of resources, while too low frequency may miss important environmental changes. At the same time, the data processing capacity is limited. As the amount of monitoring data increases, the ability to process and analyze data may become a bottleneck. At the same time, when the existing mining ecological environment monitoring system monitors the grassroots data of several areas, these data are scattered and stored in various monitoring systems, making the data isolated. Without effective data recording and analysis methods, it is difficult to obtain meaningful features from the monitoring, especially when it comes to complex data sets, it is impossible to conduct comprehensive data analysis of the data. How to conduct comprehensive and intelligent analysis based on multi-source data is one of the current problems that need to be solved. Summary of the invention

[0005] In order to solve the above technical problems, the purpose of the present invention is to provide a method for the spatial and temporal layout of integrated monitoring of the ecological environment in a mining area, comprising the following steps:

[0006] Step s1: Divide the target mining area into a number of three-dimensional triangular grids according to the terrain features, terrain elevation and SAR image data of the target mining area, and match the three-dimensional triangular grids with the environmental characteristic data and meteorological characteristic data to generate a three-dimensional digital twin model of the mining area;

[0007] Step s2: construct an environmental characteristic database to store the ecological environment monitoring records of the target mining area and other mining areas, and obtain the vulnerability coefficient of each type of characteristic index in the meteorological characteristic data based on the ecological environment monitoring records;

[0008] Step s3: construct a meteorological characteristic prediction model based on deep learning, obtain the fitting curve of the estimated vulnerability coefficient of each type of characteristic index of each three-dimensional triangular grid in the current monitoring period, and determine the number, point distribution and monitoring time interval of each three-dimensional triangular grid according to the fitting curve and the surface coverage area of ​​each three-dimensional triangular grid;

[0009] Step s4: Compare the real-time meteorological characteristic data collected by each ecological monitoring point with the estimated meteorological characteristic data, implement a daily adjustment mechanism based on the comparison results, monitor the environmental characteristic data in real time, and update the environmental characteristic database in real time based on the real-time monitoring results.

[0010] Furthermore, the target mining area is divided into a number of three-dimensional triangular grids according to the terrain features, terrain elevation and SAR image data of the target mining area, and the process of matching the three-dimensional triangular grids with the environmental characteristic data and the meteorological characteristic data to generate a three-dimensional digital twin model of the mining area includes:

[0011] The terrain features, terrain elevation, SAR image data and meteorological characteristic data of the target mining area are obtained through GIS means, and a plane triangulated network model of the target mining area is constructed according to the terrain features. The target mining area is divided into several plane triangular grids with different terrain features. The elevations on the vertices of each plane triangular grid in the plane triangulated network model are set according to the terrain elevation to generate a three-dimensional triangular grid. At the same time, the Internet method is used to determine the environmental characteristics that need to be collected when conducting ecological environment monitoring of the mining area and the band combination related to the environmental characteristics. The spectral reflection data of each band in the band combination reflected by the surface of the target mining area is obtained according to the SAR image data. The environmental characteristic data within the target mining area is obtained according to the spectral reflection data, and each three-dimensional triangular grid is matched with the corresponding environmental characteristic data and meteorological characteristic data to generate a three-dimensional digital twin model of the mining area.

[0012] Furthermore, the process of constructing an environmental characteristic database, storing the ecological environment monitoring records of the target mining area and other mining areas, and obtaining the vulnerability coefficients of various types of characteristic indexes in the meteorological characteristic data according to the ecological environment monitoring records includes:

[0013] Construct an environmental characteristics database, preset a monitoring period, use Internet technology to collect ecological environment monitoring records of the target mining area and several other mining areas and store them in the environmental characteristics database, and construct a sample data set based on the ecological environment monitoring records. The sample data set includes historical topographic characteristics, historical environmental characteristics data, and historical meteorological characteristics data of the mining area within several monitoring periods;

[0014] Preset the threshold intervals corresponding to each type of characteristic index in the environmental characteristic data under different terrain characteristic conditions, extract the numerical time period sequences corresponding to each type of characteristic index in the environmental characteristic data of several historical monitoring periods in the sample data set, compare the numerical time period sequences corresponding to each type of characteristic index in the environmental characteristic data of several historical monitoring periods with the corresponding threshold intervals, obtain the cumulative time when the numerical time period sequence corresponding to each type of characteristic index is not located in the corresponding threshold interval, and when the cumulative time of the type characteristic index is greater than the preset time threshold, mark the type characteristic index as an abnormal type characteristic index;

[0015] Based on the historical terrain characteristics, historical environmental characteristic data, historical meteorological characteristic data and abnormal type characteristic index of the mining area within several monitoring periods, the vulnerability coefficient of each type of characteristic index in the meteorological characteristic data under different terrain characteristics and meteorological characteristic data conditions is obtained, and the vulnerability coefficient is stored in the environmental characteristic database.

[0016] Furthermore, the process of obtaining the fitting curve of the estimated vulnerability coefficient of each type of characteristic index of each three-dimensional triangular grid in the current monitoring period includes:

[0017] A meteorological feature prediction model is constructed based on deep learning, and meteorological feature data of several monitoring periods of each three-dimensional triangular grid in the three-dimensional digital twin model of the mining area are obtained as a training set and a test set. The training set is input into the regional feature prediction model for training until the loss function training is stable, and the model parameters are saved. The meteorological feature prediction model is tested by the test set until it meets the preset requirements, and the meteorological feature prediction model is output;

[0018] According to the meteorological characteristic prediction model, the estimated numerical time period sequence corresponding to each type of meteorological index in the meteorological characteristic data of each three-dimensional triangular grid in the current monitoring period is obtained; according to the terrain characteristics of each three-dimensional triangular grid and the estimated numerical time period sequence corresponding to each type of meteorological index in the meteorological characteristic data, the estimated vulnerability coefficient of each type of characteristic index of each three-dimensional triangular grid in each time period in the current monitoring period is obtained; for the estimated vulnerability coefficient of each type of characteristic index of each three-dimensional triangular mesh in each time period in the current monitoring period, a fitting curve with the estimated vulnerability coefficient as the ordinate and time as the abscissa is constructed.

[0019] Furthermore, the process of determining the number, point distribution and monitoring time interval of ecological monitoring points in each three-dimensional triangular grid according to the fitting curve of the estimated vulnerability coefficient of each type of characteristic index of each three-dimensional triangular grid and the surface coverage area includes:

[0020] The surface coverage area of ​​each three-dimensional triangular grid in the three-dimensional digital twin model of the mining area is obtained, and the estimated vulnerability coefficient of the fitting curve of each type of characteristic index in each three-dimensional triangular grid is compared with the preset vulnerability coefficient threshold. If the estimated vulnerability coefficient is greater than the vulnerability coefficient threshold, the corresponding curve segment is intercepted in the corresponding fitting curve and marked as a high vulnerability segment. The high vulnerability segment in the current monitoring period is integrated over time to obtain the high vulnerability amount;

[0021] According to the current surface coverage area of ​​each three-dimensional triangular grid and the high vulnerability amount corresponding to each type of characteristic index as the evaluation index, the indicator weight matrix of the evaluation index is set, and the membership matrix of each three-dimensional triangular grid for the preset importance level is obtained through fuzzy comprehensive evaluation;

[0022] Obtain the importance level of each three-dimensional triangular mesh according to the membership matrix and the indicator weight matrix, and set the ecological monitoring point for the three-dimensional triangular mesh whose importance level is greater than the importance level threshold;

[0023] The surface coverage area of ​​the three-dimensional triangular grid whose importance level is greater than the importance level threshold is obtained, and the number of ecological monitoring points, point distribution and monitoring time interval are determined in combination with the importance level.

[0024] Furthermore, the process of obtaining the surface coverage area of ​​the three-dimensional triangular grid whose importance level is greater than the importance level threshold and determining the number of ecological monitoring points, point distribution and monitoring time interval in combination with the importance level includes:

[0025] Determine the monitoring type of the ecological monitoring point according to the meteorological characteristic data and environmental characteristic data of the three-dimensional triangular grid, and determine the number of ecological monitoring points and the monitoring interval time period of the ecological monitoring points according to the importance level of the three-dimensional triangular grid;

[0026] The surface monitoring area of ​​the ecological monitoring points is obtained, and the layout coverage rate of the ecological monitoring points is determined according to the surface monitoring area of ​​the ecological monitoring points, the number of ecological monitoring points and the surface coverage area of ​​the three-dimensional triangular grid. The layout coverage rate is used as the objective function, and the optimal layout position of the ecological monitoring points in the three-dimensional triangular grid is generated by the particle swarm algorithm.

[0027] Furthermore, the real-time meteorological characteristic data collected at each ecological monitoring point is compared with the estimated meteorological characteristic data, and the process of performing the intraday adjustment mechanism according to the comparison results includes:

[0028] Obtain the numerical time period sequence corresponding to each type of meteorological indicator in the meteorological characteristic data collected by each ecological monitoring point in the current monitoring period, perform similarity comparison on the numerical time period sequence corresponding to each type of meteorological indicator and the estimated numerical time period sequence corresponding to each type of meteorological indicator, obtain the similarity corresponding to each type of meteorological indicator, judge whether each type of meteorological indicator is less than a preset similarity threshold, if so, obtain the absolute value of the difference between the similarity and the similarity threshold, construct an intra-day adjustment comparison table, the intra-day adjustment comparison table includes monitoring interval time periods and predicted time period lengths corresponding to different absolute values ​​of the difference, and obtain new monitoring interval time periods and predicted time period lengths according to the absolute value of the difference and the intra-day adjustment comparison table;

[0029] The numerical time period sequences corresponding to various types of meteorological indicators in the meteorological characteristic data collected during the current monitoring period are used as supplementary training data for the meteorological characteristic prediction model. The meteorological characteristic prediction model is trained, and the trained meteorological characteristic prediction model is used to output the estimated numerical time period sequences corresponding to various types of meteorological indicators within the prediction time period length, and the monitoring interval time period of each ecological monitoring point within the prediction time period length is updated to the new monitoring interval time period.

[0030] Furthermore, the process of real-time monitoring of the environmental characteristic data and real-time updating of the environmental characteristic database according to the real-time monitoring results includes:

[0031] Obtain the numerical time period sequence corresponding to each type of characteristic index in the environmental characteristic data collected by each ecological monitoring point within the current monitoring period, compare the numerical time period sequence corresponding to each type of characteristic index with the corresponding threshold interval, obtain the cumulative time that the numerical time period sequence corresponding to each type of characteristic index is not located in the corresponding threshold interval, and when the cumulative time of the type characteristic index is greater than the preset time threshold, mark the type characteristic index as an abnormal type characteristic index, and generate security alarm information of the abnormal type characteristic index; at the same time, when the type characteristic index is marked as an abnormal type characteristic index, the vulnerability coefficient of the type characteristic index corresponding to the abnormal type characteristic index under the current terrain characteristics and meteorological characteristic data conditions in the environmental characteristic database is updated in real time.

[0032] Compared with the prior art, the beneficial effects of the present invention are as follows: a series of three-dimensional triangular grids are divided by using the terrain features, terrain elevation data and SAR image data within the target mining area, and these grids are combined with environmental characteristic data and meteorological characteristic data to generate a three-dimensional digital twin model that can reflect the actual conditions of the mining area; a database containing several mining area ecological environment monitoring records is established, different types of meteorological characteristic data are extracted from these records, and their respective vulnerability coefficients are calculated, which reflect the sensitivity of the ecosystem under specific environmental conditions; a meteorological characteristic prediction model is established using deep learning technology, which can predict a period of time in the future based on the data in the current monitoring period. The meteorological characteristics of each grid in the mine are analyzed. The vulnerability coefficient fitting curve output by the model is analyzed to determine the number, location and monitoring frequency of ecological monitoring points required for each grid. The actual meteorological characteristic data collected are compared with the data predicted by the model. The monitoring strategy is adjusted according to the differences, and the environmental characteristic database is updated to reflect the latest monitoring results. The location and frequency of monitoring points are reasonably planned, and limited monitoring resources are effectively utilized. Real-time monitoring and timely adjustment of strategies are carried out, which is conducive to rapid response to sudden environmental events. At the same time, based on a large amount of historical data and real-time monitoring data, a scientific basis is provided for decision-making, which significantly improves the efficiency and accuracy of ecological environmental monitoring in mining areas, thereby helping to better protect the environment and promote sustainable development. BRIEF DESCRIPTION OF THE DRAWINGS

[0033] Figure 1 This is a schematic diagram of a method for integrated spatiotemporal layout of ecological environment monitoring in a mining area according to an embodiment of the present application. DETAILED DESCRIPTION

[0034] The following is a clear and complete description of the technical solutions in the embodiments of the present application in conjunction with the drawings in the embodiments of the present application. Obviously, the described embodiments are part of the embodiments of the present application, not all of the embodiments. Based on the embodiments in the present application, all other embodiments obtained by those skilled in the art without creative work are within the scope of protection of the present application.

[0035] like Figure 1 As shown, a method for spatial and temporal layout of integrated monitoring of ecological environment in mining areas includes the following steps:

[0036] Step s1: Divide the target mining area into a number of three-dimensional triangular grids according to the terrain features, terrain elevation and SAR image data of the target mining area, and match the three-dimensional triangular grids with the environmental characteristic data and meteorological characteristic data to generate a three-dimensional digital twin model of the mining area;

[0037] Step s2: construct an environmental characteristic database to store the ecological environment monitoring records of the target mining area and other mining areas, and obtain the vulnerability coefficient of each type of characteristic index in the meteorological characteristic data based on the ecological environment monitoring records;

[0038] Step s3: construct a meteorological characteristic prediction model based on deep learning, obtain the fitting curve of the estimated vulnerability coefficient of each type of characteristic index of each three-dimensional triangular grid in the current monitoring period, and determine the number, point distribution and monitoring time interval of each three-dimensional triangular grid according to the fitting curve and the surface coverage area of ​​each three-dimensional triangular grid;

[0039] Step s4: Compare the real-time meteorological characteristic data collected by each ecological monitoring point with the estimated meteorological characteristic data, implement a daily adjustment mechanism based on the comparison results, monitor the environmental characteristic data in real time, and update the environmental characteristic database in real time based on the real-time monitoring results.

[0040] It should be further explained that, in the specific implementation process, the target mining area is divided into a number of three-dimensional triangular grids according to the terrain characteristics, terrain elevation and SAR image data of the target mining area. The three-dimensional triangular grids are matched with the environmental characteristic data and meteorological characteristic data to generate a three-dimensional digital twin model of the mining area. The process includes:

[0041] The terrain features, terrain elevation, SAR image data and meteorological characteristic data of the target mining area are obtained through GIS means, and a plane triangulated network model of the target mining area is constructed according to the terrain features. The target mining area is divided into several plane triangular grids with different terrain features. The elevations on the vertices of each plane triangular grid in the plane triangulated network model are set according to the terrain elevation to generate a three-dimensional triangular grid. At the same time, the Internet method is used to determine the environmental characteristics that need to be collected when monitoring the ecological environment of the mining area and the band combination related to the environmental characteristics. The environmental characteristics include vegetation distribution, bare land distribution and air quality. The spectral reflection data of each band in the band combination reflected by the surface of the target mining area is obtained according to the SAR image data. The environmental characteristic data within the target mining area is obtained according to the spectral reflection data, and each three-dimensional triangular grid is matched with the corresponding environmental characteristic data and meteorological characteristic data to generate a three-dimensional digital twin model of the mining area.

[0042] It should be further explained that, in the specific implementation process, the process of obtaining environmental characteristic data within the target mining area according to the spectral reflectance data includes:

[0043]

[0044] TK=ρ red -ρ NIR ;

[0045]

[0046] It should be noted that the various types of characteristic indexes in the environmental characteristic data include the normalized difference vegetation index, the particle difference index and the bare soil index. Among them, the normalized difference vegetation index (NDVI) can fully reflect the distribution information of vegetation such as vegetation biomass, leaf area index and vegetation coverage. The particle difference index (TK) is used to reflect the PM2.5 concentration that is more sensitive to the red light band and the near-infrared band. BI represents the bare soil index, and ρ NIR , red , Blue , SWIR They are respectively expressed as the band reflectance of near-infrared band, red light band, blue light band and short-wave infrared band. At the same time, meteorological characteristic data are a data set used to describe atmospheric conditions and climate change. Various types of meteorological indicators in meteorological characteristic data include: temperature, humidity, air pressure data, rainfall, snowfall, wind speed and direction, solar radiation, ultraviolet index, etc.

[0047] It should be further explained that, in the specific implementation process, the process of constructing an environmental characteristic database, storing the ecological environment monitoring records of the target mining area and other mining areas, and obtaining the vulnerability coefficients of various types of characteristic indexes in the meteorological characteristic data based on the ecological environment monitoring records includes:

[0048] An environmental feature database is constructed, and a monitoring cycle is preset. The monitoring cycle is generally one year long, and the monitoring period in the monitoring cycle is generally one day. The ecological environment monitoring records of the target mining area and several other mining areas are collected using Internet technology and stored in the environmental feature database. A sample data set is constructed based on the ecological environment monitoring records. The sample data set includes historical terrain features, historical environmental feature data, and historical meteorological feature data of the mining area within several monitoring cycles. The historical terrain features include valleys, grasslands, forests, gentle hills, plains, and altitudes;

[0049] Preset the threshold intervals corresponding to each type of characteristic index in the environmental characteristic data under different terrain characteristic conditions, extract the numerical time period sequences corresponding to each type of characteristic index in the environmental characteristic data of several historical monitoring periods in the sample data set, compare the numerical time period sequences corresponding to each type of characteristic index in the environmental characteristic data of several historical monitoring periods with the corresponding threshold intervals, obtain the cumulative time when the numerical time period sequence corresponding to each type of characteristic index is not located in the corresponding threshold interval, and when the cumulative time of the type characteristic index is greater than the preset time threshold, mark the type characteristic index as an abnormal type characteristic index;

[0050] According to the historical terrain characteristics, historical environmental characteristic data, historical meteorological characteristic data and abnormal type characteristic index of the mining area in several monitoring periods, the vulnerability coefficient of each type of characteristic index in the meteorological characteristic data under different terrain characteristics and meteorological characteristic data conditions is obtained, and the vulnerability coefficient is stored in the environmental characteristic database;

[0051] Among them, the calculation formula for obtaining the vulnerability coefficient of each type of characteristic index under different terrain characteristics and meteorological characteristic data conditions is:

[0052]

[0053] Among them, bq (r，type，E) It represents the vulnerability coefficient of characteristic index type r under the conditions of terrain characteristic type and meteorological characteristic data E, failure (r) Indicates the total number of times the feature index type r marks the abnormal type feature index, failure (r，type，E) It represents the number of times the characteristic index of the abnormal type is marked under the conditions of terrain feature type and meteorological feature data E. Type represents the terrain feature, and E represents the numerical interval time period sequence corresponding to each type of meteorological index in the meteorological feature data.

[0054] It should be further explained that, in the specific implementation process, the process of constructing a meteorological characteristic prediction model based on deep learning and obtaining the fitting curve of the estimated vulnerability coefficient of each type of characteristic index of each three-dimensional triangular grid in the current monitoring period includes:

[0055] A meteorological feature prediction model is constructed based on deep learning, and meteorological feature data of several monitoring periods of each three-dimensional triangular grid in the three-dimensional digital twin model of the mining area are obtained as a training set and a test set. The training set is input into the regional feature prediction model for training until the loss function training is stable, and the model parameters are saved. The meteorological feature prediction model is tested by the test set until it meets the preset requirements, and the meteorological feature prediction model is output;

[0056] According to the meteorological characteristic prediction model, the estimated numerical time period sequence corresponding to each type of meteorological index in the meteorological characteristic data of each three-dimensional triangular grid in the current monitoring period is obtained; according to the terrain characteristics of each three-dimensional triangular grid and the estimated numerical time period sequence corresponding to each type of meteorological index in the meteorological characteristic data, the estimated vulnerability coefficient of each type of characteristic index of each three-dimensional triangular grid in each time period in the current monitoring period is obtained; for the estimated vulnerability coefficient of each type of characteristic index of each three-dimensional triangular mesh in each time period in the current monitoring period, a fitting curve with the estimated vulnerability coefficient as the ordinate and time as the abscissa is constructed.

[0057] It should be further explained that, in the specific implementation process, the process of determining the number of ecological monitoring points, point distribution and monitoring time interval in each three-dimensional triangular grid according to the fitting curve of the estimated vulnerability coefficient of each type of characteristic index of each three-dimensional triangular grid and the surface coverage area includes:

[0058] The surface coverage area of ​​each three-dimensional triangular grid in the three-dimensional digital twin model of the mining area is obtained, and the estimated vulnerability coefficient of the fitting curve of each type of characteristic index in each three-dimensional triangular grid is compared with the preset vulnerability coefficient threshold. If the estimated vulnerability coefficient is greater than the vulnerability coefficient threshold, the corresponding curve segment is intercepted in the corresponding fitting curve and marked as a high vulnerability segment. The high vulnerability segment in the current monitoring period is integrated over time to obtain the high vulnerability amount;

[0059] According to the current surface coverage area of ​​each three-dimensional triangular grid and the high vulnerability amount corresponding to each type of characteristic index as the evaluation index, the indicator weight matrix of the evaluation index is set, and the membership matrix of each three-dimensional triangular grid for the preset importance level is obtained through fuzzy comprehensive evaluation;

[0060] Obtain the importance level of each three-dimensional triangular mesh according to the membership matrix and the indicator weight matrix, and set the ecological monitoring point for the three-dimensional triangular mesh whose importance level is greater than the importance level threshold;

[0061] The surface coverage area of ​​the three-dimensional triangular grid whose importance level is greater than the importance level threshold is obtained, and the number of ecological monitoring points, point distribution and monitoring time interval are determined in combination with the importance level.

[0062] It should be further explained that, in the specific implementation process, the process of obtaining the importance level of each three-dimensional triangular mesh according to the membership matrix and the index weight matrix includes:

[0063] The indicator weight matrix and the membership matrix of the evaluation indicator are integrated through a formula to obtain a fuzzy comprehensive evaluation matrix of the evaluation indicator, the membership of each three-dimensional triangular mesh for different importance levels is obtained according to the fuzzy comprehensive evaluation matrix, the importance level with the highest membership corresponding to each three-dimensional triangular mesh is screened out, and the importance level with the highest membership corresponding to each three-dimensional triangular mesh is used as the importance level of each three-dimensional triangular mesh;

[0064] Wherein, the formula is:

[0065] M = αM1 × βM2;

[0066] Among them, M is the fuzzy comprehensive evaluation matrix of the evaluation index, M1 is the indicator weight matrix of the evaluation index, M2 is the membership matrix, "×" represents the multiplication of the elements at the corresponding positions of the weight matrix of the evaluation index and the membership matrix, and α and β are weighting parameters for controlling the balance between the weight matrix and the membership matrix in the fuzzy comprehensive evaluation matrix of the evaluation index.

[0067] It should be further explained that, in the specific implementation process, the process of obtaining the surface coverage area of ​​the three-dimensional triangular grid whose importance level is greater than the importance level threshold and determining the number of ecological monitoring points, point distribution and monitoring time interval in combination with the importance level includes:

[0068] Determine the monitoring type of the ecological monitoring point according to the meteorological characteristic data and environmental characteristic data of the three-dimensional triangular grid, and determine the number of ecological monitoring points and the monitoring interval time period of the ecological monitoring points according to the importance level of the three-dimensional triangular grid;

[0069] Obtaining the surface monitoring area of ​​the ecological monitoring points, determining the layout coverage rate of the ecological monitoring points according to the surface monitoring area of ​​the ecological monitoring points, the number of ecological monitoring points and the surface coverage area of ​​the three-dimensional triangular grid, taking the layout coverage rate as the objective function, and generating the optimal layout position of the ecological monitoring points in the three-dimensional triangular grid by using a particle swarm algorithm;

[0070] Among them, the calculation formula for determining the layout coverage rate of ecological monitoring points based on the surface monitoring area of ​​ecological monitoring points, the number of ecological monitoring points, and the surface coverage area of ​​the three-dimensional triangular grid is:

[0071]

[0072] Among them, fp v represents the layout coverage rate, S represents the surface coverage area of ​​the three-dimensional triangular grid, and S v represents the surface monitoring area of ​​ecological monitoring point v, d vq represents the distance from the ecological monitoring point v to the surface geometric center q of the triangular grid, δ represents the conversion coefficient, and n1 represents the number of ecological monitoring points;

[0073] Then the particle swarm size is determined, and the initial position and speed are randomly generated for each particle (i.e., the position of each ecological monitoring point in the three-dimensional triangular grid), and the parameters are set. Then, the optimization process is performed, and the optimal solution is determined after multiple iterations, i.e., the optimal position of the swarm gbest, which represents the optimal layout position of the ecological monitoring point in the three-dimensional triangular grid. The calculation formula of the optimization process is:

[0074] v i (n+1)=ω·vi (n)+c1·r1·(pbest i -x i (n))+c2·r2·

[0075] (gbest-x i (n));

[0076] x i (n+1)=x i (n+1)+v i (n+1);

[0077] Among them, ω represents the inertia weight; c1 and c2 represent the acceleration coefficients, which are used to weight the optimal position of the particle itself and the optimal position of the group respectively; v i (n) represents the velocity of particle i at the current iteration n; x i (n) represents the position of particle i at the current iteration n; r1 and r2 are random numbers between 0 and 1; pbest i represents the optimal position of particle i.

[0078] It should be further explained that, in the specific implementation process, the real-time meteorological characteristic data collected by each ecological monitoring point is compared with the estimated meteorological characteristic data, and the process of performing the intraday adjustment mechanism based on the comparison results includes:

[0079] Obtain the numerical time period sequence corresponding to each type of meteorological indicator in the meteorological characteristic data collected by each ecological monitoring point in the current monitoring period, perform similarity comparison on the numerical time period sequence corresponding to each type of meteorological indicator and the estimated numerical time period sequence corresponding to each type of meteorological indicator, obtain the similarity corresponding to each type of meteorological indicator, judge whether each type of meteorological indicator is less than a preset similarity threshold, if so, obtain the absolute value of the difference between the similarity and the similarity threshold, construct an intra-day adjustment comparison table, the intra-day adjustment comparison table includes monitoring interval time periods and predicted time period lengths corresponding to different absolute values ​​of the difference, and obtain new monitoring interval time periods and predicted time period lengths according to the absolute value of the difference and the intra-day adjustment comparison table;

[0080] The numerical time period sequences corresponding to various types of meteorological indicators in the meteorological characteristic data collected during the current monitoring period are used as supplementary training data for the meteorological characteristic prediction model. The meteorological characteristic prediction model is trained, and the trained meteorological characteristic prediction model is used to output the estimated numerical time period sequences corresponding to various types of meteorological indicators within the prediction time period length, and the monitoring interval time period of each ecological monitoring point within the prediction time period length is updated to the new monitoring interval time period.

[0081] It should be further explained that, in the specific implementation process, the numerical time period sequence corresponding to each type of meteorological indicator is compared with the estimated numerical time period sequence corresponding to each type of meteorological indicator for similarity, and the calculation formula for obtaining the similarity corresponding to each type of meteorological indicator is:

[0082]

[0083] Among them, dc a Indicates the similarity corresponding to meteorological index a; D at represents the average value of the numerical time period sequence of meteorological index a in the tth time period; DaA represents the average value of the numerical time period sequence of meteorological index a; DW at It represents the average value of the estimated numerical time period sequence of meteorological indicator a in the tth time period; DWaA represents the average value of the estimated numerical time period sequence of meteorological indicator a; n2 represents the number of time periods in the numerical time period sequence.

[0084] Since the number of ecological monitoring points, point distribution and monitoring time interval in each three-dimensional triangular grid proposed in the present invention are based on the estimated vulnerability coefficient of each type of characteristic index in the meteorological characteristic data as the basis for the layout strategy, when the prediction of the meteorological characteristic data deviates from the actual situation, step s3 may select a non-optimal layout strategy. The meteorological characteristic data estimation has the characteristic that the accuracy increases as the time scale decreases. Therefore, in the intraday scheduling link of step s4, while combining the ultra-short-term meteorological characteristic data estimation, a dynamic monitoring interval time period adjustment mechanism is introduced to continuously revise the layout strategy formulated in step s3 to ensure the real-time, accuracy and stability of the layout strategy, and significantly improve the accuracy and comprehensiveness of mining area ecological environment data monitoring.

[0085] It should be further explained that, in the specific implementation process, the process of real-time monitoring of environmental characteristic data and real-time updating of the environmental characteristic database according to the real-time monitoring results includes:

[0086] The numerical time period sequence corresponding to each type of characteristic index in the environmental characteristic data collected by each ecological monitoring point within the current monitoring period is obtained, the numerical time period sequence corresponding to each type of characteristic index is compared with the corresponding threshold interval, and the cumulative time that the numerical time period sequence corresponding to each type of characteristic index is not located in the corresponding threshold interval is obtained. When the cumulative time of the type characteristic index is greater than the preset time threshold, the type characteristic index is marked as an abnormal type characteristic index, and a safety alarm information of the abnormal type characteristic index is generated. According to the safety alarm information, relevant personnel are arranged to perform manual monitoring of the area covered by the three-dimensional triangular grid. At the same time, when the type characteristic index is marked as an abnormal type characteristic index, the vulnerability coefficient of the type characteristic index corresponding to the abnormal type characteristic index under the current terrain characteristics and meteorological characteristic data conditions in the environmental characteristic database is updated in real time.

[0087] The above embodiments are only used to illustrate the technical method of the present invention rather than to limit it. Although the present invention has been described in detail with reference to the preferred embodiments, those skilled in the art should understand that the technical method of the present invention may be modified or replaced by equivalents without departing from the spirit and scope of the technical method of the present invention.

Claims

1. A method for integrated monitoring of the spatial and temporal layout of the ecological environment in a mining area, characterized in that: The following steps are involved: Step s1: Divide the target mining area into a number of three-dimensional triangular grids according to the terrain features, terrain elevation and SAR image data of the target mining area, and match the three-dimensional triangular grids with the environmental characteristic data and meteorological characteristic data to generate a three-dimensional digital twin model of the mining area; Step s2: Construct an environmental characteristic database to store the ecological environment monitoring records of the target mining area and other mining areas, and obtain the vulnerability coefficients of various types of characteristic indexes in the meteorological characteristic data based on the ecological environment monitoring records, including: Construct an environmental characteristics database, preset a monitoring period, use Internet technology to collect ecological environment monitoring records of the target mining area and several other mining areas and store them in the environmental characteristics database, and construct a sample data set based on the ecological environment monitoring records. The sample data set includes historical topographic characteristics, historical environmental characteristics data, and historical meteorological characteristics data of the mining area within several monitoring periods; Preset the threshold intervals corresponding to each type of characteristic index in the environmental characteristic data under different terrain characteristic conditions, extract the numerical time period sequences corresponding to each type of characteristic index in the environmental characteristic data of several historical monitoring periods in the sample data set, compare the numerical time period sequences corresponding to each type of characteristic index in the environmental characteristic data of several historical monitoring periods with the corresponding threshold intervals, obtain the cumulative time when the numerical time period sequence corresponding to each type of characteristic index is not located in the corresponding threshold interval, and when the cumulative time of the type characteristic index is greater than the preset time threshold, mark the type characteristic index as an abnormal type characteristic index; According to the historical terrain characteristics, historical environmental characteristic data, historical meteorological characteristic data and abnormal type characteristic index of the mining area in several monitoring periods, the vulnerability coefficient of each type of characteristic index in the meteorological characteristic data under different terrain characteristics and meteorological characteristic data conditions is obtained, and the vulnerability coefficient is stored in the environmental characteristic database; Step s3: construct a meteorological characteristic prediction model based on deep learning, obtain the fitting curve of the estimated vulnerability coefficient of each type of characteristic index of each three-dimensional triangular grid in the current monitoring period, and determine the number, point distribution and monitoring time interval of each three-dimensional triangular grid according to the fitting curve and the surface coverage area of ​​each three-dimensional triangular grid; The process of building a meteorological feature prediction model based on deep learning includes: A meteorological feature prediction model is constructed based on deep learning, and meteorological feature data within several monitoring periods of each three-dimensional triangular grid in the three-dimensional digital twin model of the mining area is obtained as a training set and a test set. The training set is input into the meteorological feature prediction model for training until the loss function training is stable, and the model parameters are saved. The meteorological feature prediction model is tested by the test set until it meets the preset requirements, and the meteorological feature prediction model is output; Step s4: compare the real-time meteorological characteristic data collected at each ecological monitoring point with the estimated meteorological characteristic data, perform a daily adjustment mechanism based on the comparison results, monitor the environmental characteristic data in real time, and update the environmental characteristic database in real time based on the real-time monitoring results; The process of implementing the intraday adjustment mechanism includes: Obtain the numerical time period sequence corresponding to each type of meteorological indicator in the meteorological characteristic data collected by each ecological monitoring point in the current monitoring period, perform similarity comparison on the numerical time period sequence corresponding to each type of meteorological indicator and the estimated numerical time period sequence corresponding to each type of meteorological indicator, obtain the similarity corresponding to each type of meteorological indicator, judge whether each type of meteorological indicator is less than a preset similarity threshold, if so, obtain the absolute value of the difference between the similarity and the similarity threshold, construct an intra-day adjustment comparison table, the intra-day adjustment comparison table includes monitoring interval time periods and predicted time period lengths corresponding to different absolute values ​​of the difference, and obtain new monitoring interval time periods and predicted time period lengths according to the absolute value of the difference and the intra-day adjustment comparison table; The numerical time period sequences corresponding to various types of meteorological indicators in the meteorological characteristic data collected during the current monitoring period are used as supplementary training data for the meteorological characteristic prediction model. The meteorological characteristic prediction model is trained, and the trained meteorological characteristic prediction model is used to output the estimated numerical time period sequences corresponding to various types of meteorological indicators within the prediction time period length, and the monitoring interval time period of each ecological monitoring point within the prediction time period length is updated to the new monitoring interval time period.

2. The method for integrated monitoring of the mining area ecological environment in time and space according to claim 1 is characterized in that: The target mining area is divided into several three-dimensional triangular grids according to the terrain features, terrain elevation and SAR image data of the target mining area. The process of matching the three-dimensional triangular grids with the environmental characteristic data and meteorological characteristic data to generate a three-dimensional digital twin model of the mining area includes: The terrain features, terrain elevation, SAR image data and meteorological characteristic data of the target mining area are obtained through GIS means, and a plane triangulated network model of the target mining area is constructed according to the terrain features. The target mining area is divided into several plane triangular grids with different terrain features. The elevations on the vertices of each plane triangular grid in the plane triangulated network model are set according to the terrain elevation to generate a three-dimensional triangular grid. At the same time, the Internet method is used to determine the environmental characteristics that need to be collected when conducting ecological environment monitoring of the mining area and the band combination related to the environmental characteristics. The spectral reflection data of each band in the band combination reflected by the surface of the target mining area is obtained according to the SAR image data. The environmental characteristic data within the target mining area is obtained according to the spectral reflection data, and each three-dimensional triangular grid is matched with the corresponding environmental characteristic data and meteorological characteristic data to generate a three-dimensional digital twin model of the mining area.

3. The method for integrated monitoring of the time and space layout of the mining area ecological environment according to claim 2 is characterized in that: The process of obtaining the fitting curve of the estimated vulnerability coefficient of each type of characteristic index of each three-dimensional triangular grid in the current monitoring period includes: According to the meteorological characteristic prediction model, the estimated numerical time period sequence corresponding to each type of meteorological index in the meteorological characteristic data of each three-dimensional triangular grid in the current monitoring period is obtained; according to the terrain characteristics of each three-dimensional triangular grid and the estimated numerical time period sequence corresponding to each type of meteorological index in the meteorological characteristic data, the estimated vulnerability coefficient of each type of characteristic index of each three-dimensional triangular grid in each time period in the current monitoring period is obtained; for the estimated vulnerability coefficient of each type of characteristic index of each three-dimensional triangular mesh in each time period in the current monitoring period, a fitting curve with the estimated vulnerability coefficient as the ordinate and time as the abscissa is constructed.

4. The method for integrated monitoring of the time and space layout of the mining area ecological environment according to claim 3 is characterized in that: The process of determining the number, point distribution and monitoring time interval of ecological monitoring points in each three-dimensional triangular grid according to the fitting curve of the estimated vulnerability coefficient of each type of characteristic index of each three-dimensional triangular grid and the surface coverage area includes: The surface coverage area of ​​each three-dimensional triangular grid in the three-dimensional digital twin model of the mining area is obtained, and the estimated vulnerability coefficient of the fitting curve of each type of characteristic index in each three-dimensional triangular grid is compared with the preset vulnerability coefficient threshold. If the estimated vulnerability coefficient is greater than the vulnerability coefficient threshold, the corresponding curve segment is intercepted in the corresponding fitting curve and marked as a high vulnerability segment. The high vulnerability segment in the current monitoring period is integrated over time to obtain the high vulnerability amount; According to the current surface coverage area of ​​each three-dimensional triangular grid and the high vulnerability amount corresponding to each type of characteristic index as the evaluation index, the indicator weight matrix of the evaluation index is set, and the membership matrix of each three-dimensional triangular grid for the preset importance level is obtained through fuzzy comprehensive evaluation; Obtain the importance level of each three-dimensional triangular mesh according to the membership matrix and the indicator weight matrix, and set the ecological monitoring point for the three-dimensional triangular mesh whose importance level is greater than the importance level threshold; The surface coverage area of ​​the three-dimensional triangular grid whose importance level is greater than the importance level threshold is obtained, and the number of ecological monitoring points, point distribution and monitoring time interval are determined in combination with the importance level.

5. The method for integrated monitoring of the time and space layout of the mining area ecological environment according to claim 4 is characterized in that: The process of obtaining the surface coverage area of ​​the three-dimensional triangular grid whose importance level is greater than the importance level threshold and determining the number of ecological monitoring points, point distribution and monitoring time interval in combination with the importance level includes: Determine the monitoring type of the ecological monitoring point according to the meteorological characteristic data and environmental characteristic data of the three-dimensional triangular grid, and determine the number of ecological monitoring points and the monitoring interval time period of the ecological monitoring points according to the importance level of the three-dimensional triangular grid; The surface monitoring area of ​​the ecological monitoring points is obtained, and the layout coverage rate of the ecological monitoring points is determined according to the surface monitoring area of ​​the ecological monitoring points, the number of ecological monitoring points and the surface coverage area of ​​the three-dimensional triangular grid. The layout coverage rate is used as the objective function, and the optimal layout position of the ecological monitoring points in the three-dimensional triangular grid is generated by the particle swarm algorithm.

6. A method for integrated monitoring of the time and space layout of the mining area ecological environment according to claim 5, characterized in that: The process of real-time monitoring of environmental characteristic data and real-time updating of the environmental characteristic database according to the real-time monitoring results includes: Obtain the numerical time period sequence corresponding to each type of characteristic index in the environmental characteristic data collected by each ecological monitoring point within the current monitoring period, compare the numerical time period sequence corresponding to each type of characteristic index with the corresponding threshold interval, obtain the cumulative time that the numerical time period sequence corresponding to each type of characteristic index is not located in the corresponding threshold interval, and when the cumulative time of the type characteristic index is greater than the preset time threshold, mark the type characteristic index as an abnormal type characteristic index, and generate security alarm information of the abnormal type characteristic index; at the same time, when the type characteristic index is marked as an abnormal type characteristic index, the vulnerability coefficient of the type characteristic index corresponding to the abnormal type characteristic index under the current terrain characteristics and meteorological characteristic data conditions in the environmental characteristic database is updated in real time.

Images