CN-122020062-A - Railway along-line photovoltaic site landslide hazard evaluation method, system, medium, equipment and product

CN122020062ACN 122020062 ACN122020062 ACN 122020062ACN-122020062-A

Abstract

The invention discloses a railway along-line photovoltaic site landslide hazard evaluation method, a system, a medium, equipment and a product, which belong to the technical field of electrified railway new energy development. The method comprises the steps of carrying out pearson correlation analysis on a railway along-line photovoltaic field landslide risk preliminary evaluation factor of a target site to obtain a railway along-line photovoltaic field landslide risk evaluation factor, carrying out landslide prediction by utilizing a BP neural network according to the railway along-line photovoltaic field landslide risk evaluation factor and a sample data set of a project place to obtain a prediction result, and obtaining a contribution value of each railway along-line photovoltaic field landslide risk evaluation factor in the prediction process to the prediction result by utilizing a SHAP model to obtain a railway along-line photovoltaic field landslide risk evaluation result. Aiming at the railway along-line photovoltaic scene, the invention solves the problems of difficult site selection and the like, and can effectively improve the construction efficiency and quality of the railway along-line photovoltaic.

Inventors

ZHANG YUHONG
JIANG KE
ZHU DONGSHENG
CHEN ZHEFENG
ZHANG SIQI
WANG ZHANGFAN
Shi Houfu
YANG JINGXU
Fang Lele

Assignees

中国能源建设集团投资有限公司
中国能源建设集团江苏省电力设计院有限公司

Dates

Publication Date: 20260512
Application Date: 20260213

Claims (10)

1. The railway along-line photovoltaic site landslide risk evaluation method is characterized by comprising the following steps of: Carrying out pearson correlation analysis on the railway along-line photovoltaic site landslide risk preliminary evaluation factors of the target site to obtain railway along-line photovoltaic site landslide risk evaluation factors; According to the railway along-line photovoltaic site landslide risk evaluation factors and the sample data set of the project location, performing landslide prediction by using a BP neural network to obtain a prediction result; And acquiring a contribution value of each railway along-line photovoltaic field landslide risk evaluation factor to a prediction result in the prediction process by using the SHAP model to obtain a railway along-line photovoltaic field landslide risk evaluation result.
2. The railway on-line photovoltaic site landslide hazard assessment method according to claim 1, wherein the railway on-line photovoltaic site landslide hazard preliminary assessment factors include a distance from a slope, a road distance, lithology, curvature, elevation, a distance from a water system, temperature, precipitation, gradient, distance from a fault, land utilization and topography relief; the lithology and land are discrete data, and the preliminary evaluation factors of landslide dangers of the rest railway along-line photovoltaic sites are continuous data.
3. The railway along-line photovoltaic site landslide risk evaluation method of claim 2, wherein the pearson correlation analysis is performed on the railway along-line photovoltaic site landslide risk preliminary evaluation factor of the target site to obtain the railway along-line photovoltaic site landslide risk evaluation factor, and the method comprises the following steps: Carrying out correlation analysis on the railway along-line photovoltaic field landslide risk preliminary evaluation factors of the target field by using the pearson correlation coefficient, and obtaining a pearson correlation coefficient matrix by using significance test; According to the pearson correlation coefficient matrix, if the absolute value of the pearson correlation coefficient between the two railway along-line photovoltaic field landslide risk preliminary evaluation factors is larger than a preset value, eliminating one of the railway along-line photovoltaic field landslide risk preliminary evaluation factors; and taking the preliminary evaluation factors of the landslide dangers of all the rest railway along-line photovoltaic sites as the evaluation factors of the landslide dangers of the railway along-line photovoltaic sites.
4. A railway on-line photovoltaic site landslide risk assessment method according to claim 3, wherein the additive interpretation expression of the SHAP model is as follows: , Wherein, the Representing SHAP model for sample Is used to determine the predicted value of (c), A baseline is represented for the SHAP model, Representing a sample Is a set of all railway along-line photovoltaic site landslide hazard assessment factors, Representing a sample Railway along-line photovoltaic site landslide risk evaluation factor Is used for the contribution value of (a), As an indication function.
5. The method for evaluating the landslide hazard of a railway on-line photovoltaic site of claim 4, wherein the sample Railway along-line photovoltaic site landslide risk evaluation factor Contribution value of (2) Is obtained by the following formula: , Wherein, the For the sample Is not included in the railway along-line photovoltaic site landslide hazard evaluation factor A railway along-line photovoltaic site landslide risk assessment factor subset, Representing BP neural network based on samples Railway along-line photovoltaic site landslide risk evaluation factor subset The obtained prediction result; representing BP neural network based on samples Railway along-line photovoltaic site landslide risk evaluation factor subset Adding railway along-line photovoltaic site landslide hazard evaluation factors And (5) obtaining a prediction result.
6. The method for evaluating the landslide risk of a railway along-line photovoltaic site of claim 5, wherein the indication function The expression of (2) is as follows: , Wherein, the Representing BP neural network for use with samples And (5) predicting a combined set of all railway along-line photovoltaic site landslide risk evaluation factors.
7. A railway along-line photovoltaic site landslide hazard assessment system, comprising: The correlation analysis module is used for carrying out pearson correlation analysis on the railway along-line photovoltaic site landslide risk preliminary evaluation factors of the target site to obtain the railway along-line photovoltaic site landslide risk evaluation factors; the landslide prediction module is used for performing landslide prediction by utilizing a BP neural network according to the railway along-line photovoltaic field landslide risk evaluation factors and a sample data set of the project location to obtain a prediction result; And the influence contribution analysis module is used for acquiring contribution values of the railway along-line photovoltaic field landslide risk evaluation factors to the prediction result in the prediction process by utilizing the SHAP model to obtain the railway along-line photovoltaic field landslide risk evaluation result.
8. A computer readable storage medium having stored thereon computer instructions which, when executed by a processor, implement the steps of the railway line photovoltaic site landslide risk assessment method of any one of claims 1-6.
9. A computer device, comprising: a memory for storing computer instructions; A processor for executing the computer instructions to implement the steps of the railway along-line photovoltaic site landslide risk assessment method of any one of claims 1-6.
10. A computer program product comprising computer instructions which, when executed by a processor, carry out the steps of the railway line photovoltaic site landslide risk assessment method of any one of claims 1 to 6.

Description

Railway along-line photovoltaic site landslide hazard evaluation method, system, medium, equipment and product Technical Field The invention relates to the technical field of new energy development of electrified railways, in particular to a railway along-line photovoltaic site landslide hazard evaluation method, a system, a medium, equipment and a product. Background The rapid development of electrified railways brings convenience to the people to travel and simultaneously leads to the rapid increase of the energy consumption requirements. As far as the full life cycle energy usage pattern of the current electrified railway is concerned, it is still one of the important industries of carbon emission. The photovoltaic power generation system can be built by utilizing the side slope along the railway, so that the carbon emission level of the railway industry can be effectively reduced. And the engineering disturbance of building the photovoltaic station can cause the damage of the rock-soil mass to form landslide. Landslide is used as one of common disaster types of geological disasters, and has the characteristics of high destructive power, high risk, strong burst property and the like. The area along the railway is large, the terrain is changeable, the manual survey difficulty is large, and the efficiency is low. At present, the application of the neural network in the aspect of terrain identification is mature, but the photovoltaic site selection needs to comprehensively consider various factors, and the mainstream prediction neural network is difficult to meet. Disclosure of Invention The invention aims to overcome the problems in the prior art and provides a railway along-line photovoltaic field landslide risk evaluation method, a system, a medium, equipment and a product. In order to solve the technical problems, the invention is realized by adopting the following technical scheme: In a first aspect, the invention provides a railway along-line photovoltaic site landslide hazard evaluation method, which comprises the following steps: Carrying out pearson correlation analysis on the railway along-line photovoltaic site landslide risk preliminary evaluation factors of the target site to obtain railway along-line photovoltaic site landslide risk evaluation factors; According to the railway along-line photovoltaic site landslide risk evaluation factors and the sample data set of the project location, performing landslide prediction by using a BP neural network to obtain a prediction result; And acquiring a contribution value of each railway along-line photovoltaic field landslide risk evaluation factor to a prediction result in the prediction process by using the SHAP model to obtain a railway along-line photovoltaic field landslide risk evaluation result. Optionally, the railway along-line photovoltaic site landslide hazard preliminary evaluation factors comprise a slope direction, a road distance, lithology, curvature, elevation, a water system distance, temperature, precipitation, gradient, a fault distance, land utilization and topography fluctuation; the lithology and land are discrete data, and the preliminary evaluation factors of landslide dangers of the rest railway along-line photovoltaic sites are continuous data. Optionally, the pearson correlation analysis is performed on the railway along-line photovoltaic field landslide risk preliminary evaluation factor of the target site to obtain the railway along-line photovoltaic field landslide risk evaluation factor, which comprises the following steps: Carrying out correlation analysis on the railway along-line photovoltaic field landslide risk preliminary evaluation factors of the target field by using the pearson correlation coefficient, and obtaining a pearson correlation coefficient matrix by using significance test; According to the pearson correlation coefficient matrix, if the absolute value of the pearson correlation coefficient between the two railway along-line photovoltaic field landslide risk preliminary evaluation factors is larger than a preset value, eliminating one of the railway along-line photovoltaic field landslide risk preliminary evaluation factors; and taking the preliminary evaluation factors of the landslide dangers of all the rest railway along-line photovoltaic sites as the evaluation factors of the landslide dangers of the railway along-line photovoltaic sites. Optionally, the additive interpretation expression of the SHAP model is as follows: , Wherein, the Representing SHAP model for sampleIs used to determine the predicted value of (c),A baseline is represented for the SHAP model,Representing a sampleIs a set of all railway along-line photovoltaic site landslide hazard assessment factors,Representing a sampleRailway along-line photovoltaic site landslide risk evaluation factorIs used for the contribution value of (a),As an indication function. Optionally, the sampleRailway along-line photovoltaic site landslide risk ev