A method for urban geological safety risk assessment
By combining the adaptive learning rate of multiple geological data and LSTM network structures in urban geological safety risk assessment, a comprehensive evaluation model was established, which solved the problem of lack of comprehensiveness and comprehensiveness in the existing technology, and achieved more accurate and detailed .
Patent Information
- Application Number
- CN202411270812.8
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- Filing Date
- 2024-09-11
- Publication Date
- 2024-11-22
- Estimated Expiration
- 2044-09-11
AI Technical Summary
The existing urban geological disaster evaluation methods lack comprehensiveness and comprehensiveness, and fail to fully consider the comprehensive impact of various geological disasters, resulting in ineffective urban geological safety risk assessment.
A method of urban geological safety risk assessment is adopted. By obtaining data from various areas of the city and inputting geological safety risk assessment models, pre-processing, feature extraction and analysis are carried out, a comprehensive geological safety risk assessment coefficient is established, and risk type judgment is made based on the coefficients. It includes a first risk area, a second risk area and a third risk area. This method combines data on active fracture, collapsed landslide, goaf collapse, karst collapse and ground settlement, and uses the LSTM network structure for feature extraction and adaptive learning rate adjustment.
The effectiveness and comprehensiveness of urban geological safety risk assessment has been improved, the ability to comprehensively evaluate urban geological safety risks has been enhanced, and the accuracy and effectiveness of evaluation have been further improved by refining the risk type judgment.
Smart Images

Figure CN118780625B_ABST
Abstract
Description
Technical Field
[0001] The invention relates to the technical field of geological evaluation, and in particular to a method for evaluating urban geological safety risks. Background Art
[0002] With the continuous development of science and technology, geological disaster assessment technologies emerge in an endless stream, but the existing geological disaster assessment methods still have certain defects. The Chinese patent with the prior art application number CN202211159632.3 provides a geological disaster early warning method based on risk assessment, including setting a monitoring module, the monitoring module includes multiple data acquisition sensors, setting a data transmission storage module, the data transmission storage module is connected to multiple data sensors through a multi-channel acquisition card, obtains real-time data from the data sensor, and sets up a special database to store the real-time data of all data sensors, and sets up an early warning module. The Chinese patent with the prior art application number CN202310158226.3 provides a method for assessing the risk of urban geological disasters in the loess region based on groundwater level, including: the first step, data acquisition and collation analysis: obtain the required data and information from existing data and literature, and perform trend detection and analysis; the second step, establish a three-dimensional groundwater model of the working area: establish a three-dimensional groundwater model of the working area based on the relevant data obtained in the first step; the third step, calculate the loess collapse; the fourth step, calculate the sand liquefaction; the fifth step, geological disaster assessment. The above existing technologies for evaluating urban geology do not fully consider the comprehensive consideration of various geological hazards, which leads to the lack of comprehensiveness and integration of urban geological methods.
[0003] Therefore, a method for urban geological safety risk assessment is proposed. Summary of the invention
[0004] The purpose of the present invention is to provide a method for evaluating urban geological safety risks. The present invention relates to the field of geological evaluation technology, and specifically to a method for evaluating urban geological safety risks, comprising: obtaining first data of various regions of the city and inputting them into an urban geological safety risk evaluation model, firstly preprocessing the first data through a preprocessing layer to obtain first evaluation data; secondly, extracting and training the first evaluation data through a feature processing layer to obtain a first feature vector; then, analyzing the first feature vector in an output layer to obtain a risk evaluation coefficient of each region, and establishing a comprehensive geological safety risk evaluation coefficient based on the risk evaluation coefficient of each region; finally, establishing a piecewise function of an urban geological safety risk threshold, and judging the risk type of each region based on the piecewise function, wherein the risk type includes a first risk region, a second risk region, and a third risk region. The present invention improves the effectiveness of urban geological safety risk evaluation by establishing an urban geological safety risk evaluation model.
[0005] To achieve the above object, the present invention provides the following technical solutions:
[0006] An urban geological safety risk assessment method includes an urban geological safety risk assessment model for assessing urban geological safety risks. The specific assessment process is as follows:
[0007] Obtaining first data of each area of the city, inputting the first data into the model, and sequentially passing through a preprocessing layer, a feature processing layer, and an output layer of the model;
[0008] The preprocessing layer is used to preprocess the first data to obtain first evaluation data;
[0009] The feature processing layer is used to extract and train features of the first evaluation data to obtain a first feature vector;
[0010] The output layer analyzes the first eigenvector to obtain the risk assessment coefficient of each region; a comprehensive geological safety risk assessment coefficient is established based on the risk assessment coefficient of each region to evaluate the comprehensive risk assessment level of each region in the city;
[0011] A piecewise function of the urban geological safety risk threshold is established, and the risk type of each area is judged according to the piecewise function, wherein the risk type includes a first risk area, a second risk area and a third risk area.
[0012] Preferably, the first data include active fault data, collapse and landslide data, mining subsidence data, karst subsidence data and ground subsidence data of each region.
[0013] Preferably, the active fault data are obtained through historical earthquake data and shallow seismic surveys in each region; the collapse and landslide data are obtained through the results of regional geological disaster surveys; the mining subsidence data are obtained through surface deformation monitoring and the boundaries of mining rights in each region; the karst collapse data are obtained based on the distribution of carbonate rock formations and historical collapses, and the ground subsidence data are obtained through InSAR monitoring data.
[0014] Preferably, the preprocessing includes outlier removal and missing value processing; the feature processing layer includes an LSTM network structure, and the network structure is used to extract and train features of the first evaluation data to obtain a first feature vector.
[0015] Preferably, the LSTM network structure includes a mean square error loss function, the mean square error loss function is minimized by a gradient descent method, and an adaptive learning rate based on regional features is established.
[0016] Preferably, the mean square error loss function is:
[0017] ;
[0018] in, Expressed as the mean square error loss function; is the number of samples; represents the predicted value; Represented as actual value.
[0019] Preferably, the learning rate calculation formula based on regional feature adaptation is:
[0020] ;
[0021] in, Expressed as Areas in The learning rate at each moment; Represents the learning rate to adjust the weight; Expressed as Areas in Activity fracture data at the moment; Expressed as Areas in Time-based collapse and landslide data; Expressed as Areas in Mining-out collapse data at all times; Expressed as Areas in Karst collapse data at the moment; Expressed as Areas in Ground subsidence data at the moment; It is expressed as the active fault impact factor; It is expressed as the impact factor of collapse and landslide; It is expressed as the impact factor of mining subsidence; It is expressed as the karst collapse influencing factor; Expressed as the ground subsidence impact factor.
[0022] Preferably, the output layer of the model is analyzed according to the first eigenvector to obtain the risk assessment coefficients of the regions, which include active fault risk assessment coefficients, collapse and landslide risk assessment coefficients, mining subsidence risk assessment coefficients, karst subsidence risk assessment coefficients and ground subsidence risk assessment coefficients, and the five risk assessment coefficients are weighted and superimposed to obtain a comprehensive geological safety risk assessment coefficient.
[0023] Preferably, the comprehensive geological safety risk assessment coefficient is:
[0024] ;
[0025] in, Expressed as Comprehensive geological safety risk assessment coefficient of each region; Expressed as Active fault risk assessment coefficient for each region; Expressed as The risk assessment coefficient of collapse and landslide in each region; Expressed as The risk assessment coefficient of mining subsidence in each region; Expressed as Karst collapse risk assessment coefficient for each region; Expressed as Land subsidence risk assessment coefficient for each region; It is expressed as the risk assessment weight of active faults; It is expressed as the risk assessment weight of collapse and landslide; It is expressed as the risk assessment weight of mining subsidence; It is expressed as the karst collapse risk assessment weight; Expressed as the ground subsidence risk assessment weight.
[0026] Preferably, the urban geological safety risk threshold piecewise function is:
[0027] ;
[0028] in, Expressed as Type of risk in each area; Expressed as the Comprehensive geological safety risk assessment coefficient of each region; It is represented as the first threshold; Indicated as the second threshold.
[0029] Compared with the prior art, the present invention has the following beneficial effects:
[0030] 1. The present invention establishes a learning rate adaptive to regional characteristics by comprehensively considering the active fault data, collapse and landslide data, mining subsidence data, karst subsidence data and ground subsidence data in various urban areas, and realizes dynamic adjustment of the model learning rate formula. This makes the convergence speed of the urban geological safety risk model have better regional characteristic adaptability, further improving the effectiveness of urban geological safety risk assessment.
[0031] 2. The present invention establishes an urban geological safety risk model to obtain risk assessment coefficients for various areas in the city, including active fault risk assessment coefficients, collapse and landslide risk assessment coefficients, mining subsidence risk assessment coefficients, karst subsidence risk assessment coefficients and ground subsidence risk assessment coefficients for each area, and establishes a comprehensive geological safety risk assessment coefficient based on these five coefficients, thereby improving the comprehensiveness and comprehensiveness of urban geological safety evaluation and further improving the effectiveness of urban geological safety risk evaluation.
[0032] 3. The present invention establishes a piecewise function of the urban geological safety risk threshold to judge the risk type of the comprehensive geological safety risk assessment coefficient, including the first risk area, the second risk area and the third risk area. By establishing the piecewise function of the urban geological safety risk threshold, the level of urban geological safety risk assessment is further refined, thereby further improving the effectiveness of urban geological safety risk assessment. BRIEF DESCRIPTION OF THE DRAWINGS
[0033] Figure 1 A flow chart of a method for evaluating urban geological safety risks provided by an embodiment of the present invention;
[0034] Figure 2 This is a diagram of an urban geological safety risk assessment model provided by an embodiment of the present invention. DETAILED DESCRIPTION
[0035] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.
[0036] The effective evaluation of urban geological safety risks is of great significance to urban geological safety and the living standards of residents. In the current field of urban geological safety risk evaluation technology, there is a lack of comprehensive geological disaster risk evaluation, which affects the comprehensiveness and comprehensiveness of urban geological safety to a certain extent. In order to achieve a comprehensive and integrated evaluation of urban geological safety risks, the present invention will be described in detail based on the following two embodiments to prove that the method proposed in the present invention has a certain effectiveness in urban geological safety risk evaluation.
[0037] Although embodiments of the present invention have been shown and described, it will be appreciated by those skilled in the art that various changes, modifications, substitutions and variations may be made to the embodiments without departing from the principles and spirit of the present invention, and that the scope of the present invention is defined by the appended claims and their equivalents.
[0038] Embodiment 1
[0039] This example analyzes the geological safety risk assessment process of City A;
[0040] In the specific implementation process of the embodiment of the present application, the method of the present invention is applied to practical applications to achieve a comprehensive evaluation of urban geological safety risks. Specifically, the urban geological safety risk evaluation method proposed in the present invention corresponds to Figure 1 The steps include: S10. acquiring data; S20. data preprocessing; S30. feature processing; S40. result output;
[0041] The specific process of the above steps includes: obtaining the first data of each area of the city and inputting it into the urban geological safety risk assessment model, firstly preprocessing the first data through the preprocessing layer to obtain the first evaluation data; secondly, extracting and training the first evaluation data through the feature processing layer to obtain the first feature vector; then, analyzing the first feature vector in the output layer to obtain the evaluation results of each area, and establishing a comprehensive geological safety risk assessment coefficient based on the evaluation results; finally, establishing a piecewise function of the urban geological safety risk threshold, and judging the risk type of each area based on the piecewise function, and the risk type includes the first risk area, the second risk area and the third risk area. The present invention improves the effectiveness of urban geological safety risk assessment by establishing an urban geological safety risk assessment model.
[0042] According to step S10, 100 different areas to be evaluated in city A are selected to obtain first data of each evaluation area, wherein the first data includes active fault data, collapse and landslide data, mining subsidence data, karst subsidence data and ground subsidence data of each area.
[0043] The active fault data are obtained through historical earthquake data and shallow seismic surveys in each region; the collapse and landslide data are obtained through the results of regional geological disaster surveys; the mining subsidence data are obtained through surface deformation monitoring and the boundaries of mining rights in each region; the karst collapse data are obtained based on the distribution of carbonate rock formations and historical collapses, and the ground subsidence data are obtained through InSAR monitoring data.
[0044] Furthermore, the first data is input into the geological safety risk assessment model, and sequentially passes through the preprocessing layer, feature processing layer and output layer of the geological safety risk assessment model; the preprocessing layer is used to preprocess the first data to obtain first evaluation data; the feature processing layer is used to extract and train features of the first evaluation data to obtain a first feature vector; the output layer analyzes the first feature vector to obtain risk assessment coefficients for each region, including active fault risk assessment coefficients, collapse and landslide risk assessment coefficients, mining subsidence risk assessment coefficients, karst subsidence risk assessment coefficients and ground subsidence risk assessment coefficients for each region, and weights are superimposed based on the five risk assessment coefficients to obtain a comprehensive geological safety risk assessment coefficient, which is used to evaluate the comprehensive risk assessment level of each region in the city; and an urban geological safety risk threshold piecewise function is established, and the risk type of each region is judged based on the urban geological safety risk threshold piecewise function, and the risk type includes a first risk area, a second risk area and a third risk area.
[0045] Each layer of the urban geological safety risk assessment model corresponds to steps S20 to S40 respectively. The urban geological safety risk assessment model is shown in FIG. Figure 2 As shown;
[0046] Further, the first data is preprocessed using the preprocessing layer in the urban geological safety risk assessment model, corresponding to the above step S20; the preprocessing includes outlier removal and missing value processing;
[0047] Furthermore, the feature processing layer of the urban geological safety risk assessment model is used to perform feature processing on the first assessment data, corresponding to the above-mentioned step S30; the feature processing layer includes an LSTM network structure, and the network structure is used to extract and train features of the first assessment data to obtain a first feature vector; wherein, the LSTM network structure includes a mean square error loss function, the mean square error loss function is minimized by the gradient descent method, and an adaptive learning rate based on regional features is established.
[0048] Furthermore, the mean square error loss function is:
[0049] ;
[0050] in, Expressed as the mean square error loss function; is the number of samples; represents the predicted value; Represented as actual value.
[0051] Furthermore, the learning rate calculation formula based on regional feature adaptation is:
[0052] ;
[0053] in, Expressed as Areas in The learning rate at each moment; Represents the learning rate to adjust the weight; Expressed as Areas in Activity fracture data at the moment; Expressed as Areas in Time-based collapse and landslide data; Expressed as Areas in Mining-out collapse data at all times; Expressed as Areas in Karst collapse data at the moment; Expressed as Areas in Ground subsidence data at the moment; It is expressed as the active fault impact factor; It is expressed as the impact factor of collapse and landslide; It is expressed as the impact factor of mining subsidence; It is expressed as the karst collapse influencing factor; Expressed as the ground subsidence impact factor.
[0054] This embodiment establishes a learning rate that is adaptive to regional characteristics by comprehensively considering the active fault data, collapse and landslide data, mining subsidence data, karst collapse data and ground subsidence data in various urban areas, and realizes dynamic adjustment of the model learning rate formula. This makes the convergence speed of the urban geological safety risk model have better adaptability to regional characteristics, further improving the effectiveness of urban geological safety risk assessment.
[0055] Furthermore, the output layer of the urban geological safety risk assessment model analyzes the first eigenvector to obtain the risk assessment coefficients of each region, including the active fault risk assessment coefficient, collapse and landslide risk assessment coefficient, mining subsidence risk assessment coefficient, karst subsidence risk assessment coefficient and ground subsidence risk assessment coefficient of each region; a comprehensive geological safety risk assessment coefficient is established based on this coefficient to evaluate the comprehensive risk assessment level of each region in the city; a piecewise function of the urban geological safety risk threshold is established, and the risk type of each region is judged based on the piecewise function, and the risk type includes the first risk region, the second risk region and the third risk region. Corresponding to the above step S40;
[0056] Furthermore, the comprehensive geological safety risk assessment coefficient is:
[0057] ;
[0058] in, Expressed as Comprehensive geological safety risk assessment coefficient of each region; Expressed as Active fault risk assessment coefficient for each region; Expressed as The risk assessment coefficient of collapse and landslide in each region; Expressed as The risk assessment coefficient of mining subsidence in each region; Expressed as Karst collapse risk assessment coefficient for each region; Expressed as Land subsidence risk assessment coefficient for each region; It is expressed as the risk assessment weight of active faults; It is expressed as the risk assessment weight of collapse and landslide; It is expressed as the risk assessment weight of mining subsidence; It is expressed as the karst collapse risk assessment weight; Expressed as the ground subsidence risk assessment weight.
[0059] This embodiment establishes an urban geological safety risk assessment model to obtain risk assessment coefficients for various areas in the city, including active fault risk assessment coefficients, collapse and landslide risk assessment coefficients, mining subsidence risk assessment coefficients, karst subsidence risk assessment coefficients and ground subsidence risk assessment coefficients for each area, and establishes a comprehensive geological safety risk assessment coefficient based on these five coefficients, thereby improving the comprehensiveness and comprehensiveness of urban geological safety assessment and further improving the effectiveness of urban geological safety risk assessment.
[0060] Furthermore, the urban geological safety risk threshold piecewise function is:
[0061] ;
[0062] in, Expressed as Type of risk in each area; Represented as the Comprehensive geological safety risk assessment coefficient of each region; It is represented as the first threshold; Indicated as the second threshold.
[0063] This embodiment establishes a piecewise function of the urban geological safety risk threshold to judge the risk type of the comprehensive geological safety risk assessment coefficient, including the first risk area, the second risk area and the third risk area. By establishing the piecewise function of the urban geological safety risk threshold, the urban geological safety risk assessment level is further refined, thereby further improving the effectiveness of the urban geological safety risk assessment.
[0064] The first risk area, the second risk area and the third risk area represent three areas with risk levels from high to low.
[0065] In order to verify the effectiveness of the urban geological safety risk assessment model proposed in this embodiment, this embodiment conducted 5 comparative experiments, mainly by comparing the evaluation levels of urban geological safety risks of different models, the evaluation level is mainly represented by the average judgment accuracy of the risk types of the 100 detected areas, and the evaluation levels of urban geological safety risks of different models A are shown in Table 1;
[0066] Referring to Table 1, the first model is the urban geological safety risk assessment model proposed by the present invention, the second model is based on the first model without considering the first data, and the third model is based on the first model and replaces the LSTM model with the RNN model. The results show that the urban geological safety risk assessment model proposed in this embodiment has the highest evaluation level in the urban geological safety risk assessment task.
[0067] Table 1 Evaluation level of urban geological safety risk in different models A
[0068] Number of experiments First Model Second Model The third model 1 99 86 87 2 98 84 86 3 98 85 85 4 97 87 86 5 97 86 84
[0069] Embodiment 2
[0070] This example analyzes the geological safety risk assessment process of City B;
[0071] In the specific implementation process of the embodiment of the present application, the method of the present invention is applied to practical applications to achieve a comprehensive evaluation of urban geological safety risks. Specifically, the urban geological safety risk evaluation method proposed in the present invention corresponds to Figure 1 The steps include: S10. acquiring data; S20. data preprocessing; S30. feature processing; S40. result output;
[0072] The specific process of the above steps includes: obtaining the first data of each area of the city and inputting it into the urban geological safety risk assessment model, firstly preprocessing the first data through the preprocessing layer to obtain the first evaluation data; secondly, extracting and training the first evaluation data through the feature processing layer to obtain the first feature vector; then, analyzing the first feature vector in the output layer to obtain the evaluation results of each area, and establishing a comprehensive geological safety risk assessment coefficient based on the evaluation data; finally, establishing a piecewise function of the urban geological safety risk threshold, and judging the risk type of each area according to the piecewise function, and the risk type includes the first risk area, the second risk area and the third risk area. The present invention improves the effectiveness of urban geological safety risk assessment by establishing an urban geological safety risk assessment model.
[0073] According to step S10, 100 different areas to be evaluated in city B are selected to obtain first data of each evaluation area, wherein the first data includes active fault data, collapse and landslide data, mining subsidence data, karst subsidence data and ground subsidence data of each area.
[0074] The active fault data are obtained through historical earthquake data and shallow seismic surveys in each region; the collapse and landslide data are obtained through the results of regional geological disaster surveys; the mining subsidence data are obtained through surface deformation monitoring and the boundaries of mining rights in each region; the karst collapse data are obtained based on the distribution of carbonate rock formations and historical collapses, and the ground subsidence data are obtained through InSAR monitoring data.
[0075] Furthermore, the first data is input into the geological safety risk assessment model, and sequentially passes through the preprocessing layer, feature processing layer and output layer of the geological safety risk assessment model; the preprocessing layer is used to preprocess the first data to obtain first evaluation data; the feature processing layer is used to extract and train features of the first evaluation data to obtain a first feature vector; the output layer analyzes the first feature vector to obtain risk assessment coefficients for each region, including active fault risk assessment coefficients, collapse and landslide risk assessment coefficients, mining subsidence risk assessment coefficients, karst subsidence risk assessment coefficients and ground subsidence risk assessment coefficients for each region, and weights are superimposed based on the five risk assessment coefficients to obtain a comprehensive geological safety risk assessment coefficient, which is used to evaluate the comprehensive risk assessment level of each region in the city; and an urban geological safety risk threshold piecewise function is established, and the risk type of each region is judged based on the urban geological safety risk threshold piecewise function, and the risk type includes a first risk area, a second risk area and a third risk area.
[0076] Each layer of the urban geological safety risk assessment model corresponds to steps S20 to S40 respectively. The urban geological safety risk assessment model is shown in FIG. Figure 2 As shown;
[0077] Further, the first data is preprocessed using the preprocessing layer in the urban geological safety risk assessment model, corresponding to the above step S20; the preprocessing includes outlier removal and missing value processing;
[0078] Furthermore, the feature processing layer of the urban geological safety risk assessment model is used to perform feature processing on the first assessment data, corresponding to the above-mentioned step S30; the feature processing layer includes an LSTM network structure, and the network structure is used to extract and train features of the first assessment data to obtain a first feature vector; wherein, the LSTM network structure includes a mean square error loss function, the mean square error loss function is minimized by the gradient descent method, and an adaptive learning rate based on regional features is established.
[0079] Furthermore, the mean square error loss function is:
[0080] ;
[0081] in, Expressed as the mean square error loss function; is the number of samples; represents the predicted value; Represented as actual value.
[0082] Furthermore, the learning rate calculation formula based on regional feature adaptation is:
[0083] ;
[0084] in, Expressed as Areas in The learning rate at each moment; Represents the learning rate to adjust the weight; Expressed as Areas in Activity fracture data at the moment; Expressed as Areas in Time-based collapse and landslide data; Expressed as Areas in Mining-out collapse data at all times; Expressed as Areas in Karst collapse data at the moment; Expressed as Areas in Ground subsidence data at the moment; It is expressed as the active fault impact factor; It is expressed as the impact factor of collapse and landslide; It is expressed as the impact factor of mining subsidence; It is expressed as the karst collapse influencing factor; Expressed as the ground subsidence impact factor.
[0085] Furthermore, the output layer of the urban geological safety risk assessment model analyzes the first eigenvector to obtain the risk assessment coefficients of each region, including the active fault risk assessment coefficient, collapse and landslide risk assessment coefficient, mining subsidence risk assessment coefficient, karst subsidence risk assessment coefficient and ground subsidence risk assessment coefficient of each region; a comprehensive geological safety risk assessment coefficient is established based on this coefficient to evaluate the comprehensive risk assessment level of each region in the city; a piecewise function of the urban geological safety risk threshold is established, and the risk type of each region is judged based on the piecewise function, and the risk type includes the first risk region, the second risk region and the third risk region. Corresponding to the above step S40;
[0086] Furthermore, the comprehensive geological safety risk assessment coefficient is:
[0087] ;
[0088] in, Expressed as Comprehensive geological safety risk assessment coefficient of each region; Expressed as Active fault risk assessment coefficient for each region; Expressed as The risk assessment coefficient of collapse and landslide in each region; Expressed as The risk assessment coefficient of mining subsidence in each region; Expressed as Karst collapse risk assessment coefficient for each region; Expressed as Land subsidence risk assessment coefficient for each region; It is expressed as the risk assessment weight of active faults; It is expressed as the risk assessment weight of collapse and landslide; It is expressed as the risk assessment weight of mining subsidence; It is expressed as the karst collapse risk assessment weight; Expressed as the ground subsidence risk assessment weight.
[0089] Furthermore, the urban geological safety risk threshold piecewise function is:
[0090] ;
[0091] in, Expressed as Type of risk in each area; Represented as the Comprehensive geological safety risk assessment coefficient of each region; It is represented as the first threshold; Indicated as the second threshold.
[0092] The first risk area, the second risk area and the third risk area represent three areas with risk levels from high to low.
[0093] In order to verify the effectiveness of the urban geological safety risk assessment model proposed in this embodiment, this embodiment conducted 5 comparative experiments, mainly by comparing the evaluation levels of urban geological safety risks of different models, the evaluation level is mainly represented by the average judgment accuracy of the risk types of the 100 detected areas, and the evaluation levels of urban geological safety risks of different models B are shown in Table 2;
[0094] Referring to Table 2, the first model is the urban geological safety risk assessment model proposed by the present invention, the second model is based on the first model without considering the first data, and the third model is based on the first model and replaces the LSTM model with the RNN model. The results show that the urban geological safety risk assessment model proposed in this embodiment has the highest evaluation level in the urban geological safety risk assessment task.
[0095] Table 2 Evaluation level of urban geological safety risk in different models B
[0096] Number of experiments First Model Second Model The third model 1 98 85 84 2 98 84 86 3 98 86 85 4 98 86 86 5 97 87 87
[0097] Although embodiments of the present invention have been shown and described, it will be appreciated by those skilled in the art that various changes, modifications, substitutions and variations may be made to the embodiments without departing from the principles and spirit of the present invention, and that the scope of the present invention is defined by the appended claims and their equivalents.
Claims
1. A method for evaluating urban geological safety risk, characterized in that: It includes an urban geological safety risk assessment model for evaluating urban geological safety risks. The specific assessment process is as follows: Acquire first data of each area of the city, input the first data into the model, and sequentially pass through the preprocessing layer, feature processing layer and output layer of the model; the first data includes active fault data, collapse and landslide data, mining subsidence data, karst subsidence data and ground subsidence data of each area; The preprocessing layer is used to preprocess the first data to obtain first evaluation data; the preprocessing includes outlier removal and missing value processing; The feature processing layer includes an LSTM network structure, which is used to extract and train features of the first evaluation data to obtain a first feature vector; the network structure includes a mean square error loss function, which is minimized by a gradient descent method, and an adaptive learning rate based on regional features is established; The output layer performs analysis based on the first feature vector to obtain a risk assessment coefficient for each region; The risk assessment coefficients for each region include active fault risk assessment coefficient, collapse and landslide risk assessment coefficient, mining subsidence risk assessment coefficient, karst subsidence risk assessment coefficient and ground subsidence risk assessment coefficient, and these five risk assessment coefficients are weighted and superimposed to obtain a comprehensive geological safety risk assessment coefficient; The comprehensive geological safety risk assessment coefficient is used to evaluate the comprehensive risk assessment level of each area in the city; the comprehensive geological safety risk assessment coefficient is: ; in, Expressed as Comprehensive geological safety risk assessment coefficient of each region; Expressed as Active fault risk assessment coefficient for each region; Expressed as The risk assessment coefficient of collapse and landslide in each region; Expressed as The risk assessment coefficient of mining subsidence in each region; Expressed as Karst collapse risk assessment coefficient for each region; Expressed as Land subsidence risk assessment coefficient for each region; It is expressed as the risk assessment weight of active faults; It is expressed as the risk assessment weight of collapse and landslide; It is expressed as the risk assessment weight of mining subsidence; It is expressed as the karst collapse risk assessment weight; It is expressed as the land subsidence risk assessment weight; An urban geological safety risk threshold value piecewise function is established, and the risk type of each area is judged according to the urban geological safety risk threshold value piecewise function, wherein the risk type includes a first risk area, a second risk area and a third risk area.
2. The urban geological safety risk assessment method according to claim 1 is characterized in that: The active fault data are obtained through historical earthquake data and shallow seismic surveys in each region; the collapse and landslide data are obtained through the results of regional geological disaster surveys; the mining subsidence data are obtained through surface deformation monitoring and the boundaries of mining rights in each region; the karst collapse data are obtained based on the distribution of carbonate rock formations and historical collapses, and the ground subsidence data are obtained through InSAR monitoring data.
3. The urban geological safety risk assessment method according to claim 1 is characterized in that: The mean square error loss function is: ; in, Expressed as the mean square error loss function; is the number of samples; represents the predicted value; Represented as actual value.
4. The urban geological safety risk assessment method according to claim 1 is characterized in that: The calculation formula of the adaptive learning rate based on regional features is: ; in, Expressed as Areas in The learning rate at each moment; Represents the learning rate to adjust the weight; Expressed as Areas in Activity fracture data at the moment; Expressed as Areas in Time-based collapse and landslide data; Expressed as Areas in Mining-out collapse data at all times; Expressed as Areas in Karst collapse data at the moment; Expressed as Areas in Ground subsidence data at the moment; It is expressed as the active fault impact factor; It is expressed as the impact factor of collapse and landslide; It is expressed as the impact factor of mining subsidence; It is expressed as the karst collapse influencing factor; Expressed as the ground subsidence impact factor.
5. The urban geological safety risk assessment method according to claim 1 is characterized in that: The urban geological safety risk threshold piecewise function is: ; in, Expressed as Type of risk in each area; Expressed as the Comprehensive geological safety risk assessment coefficient of each region; It is represented as the first threshold; Indicated as the second threshold.
Citation Information
Patent Citations
Geological disaster early warning method based on risk evaluation
CN115691051A
Loess area urban geological disaster risk evaluation method based on underground water level
CN116663881A
Fire early-warning method of fixed-point area unit
CN108053594A
Enterprise information security management level evaluation method and system based on big data analysis
CN114611928A
Intelligent monitoring and early warning system for urban geological safety risk
CN116663882A
