Font Size: a A A

Hazard Assessment Of Ground Fissures In Xianyang Area Based On SVM And WOA-ELM Calculation Models

Posted on:2024-03-01Degree:MasterType:Thesis
Country:ChinaCandidate:X ChenFull Text:PDF
GTID:2530307157478134Subject:Geological engineering
Abstract/Summary:PDF Full Text Request
Frequent geological hazards pose a great threat to the safety of people’s lives and properties in China.Geological hazard evaluation can scientifically predict the level of regional geological hazard,and provide the necessary scientific basis for carrying out accurate geological hazard prevention and control as well as carrying out territorial spatial planning.In this paper,we take Xianyang area in Shaanxi Province as an example to carry out research on the hazard of ground fracture in Xianyang area,and evaluate the hazard of ground fracture in Xianyang area by high-dimensional support vector machine model(SVM)and extreme learning machine(WOA-ELM)model based on whale optimization algorithm,and compare and analyze the obtained model results,the results show that WOA-ELM model is more in line with the actual situation of the study area and The results show that the WOA-ELM model is more consistent with the actual situation in the study area,has higher accuracy,and is more consistent with the logical law of ground fracture development,and is more scientific and applicable in the regional ground fracture hazard evaluation.The main research contents and results of this paper are as follows:(1)It is found through the field survey that the ground fractures in Xianyang area are mainly distributed in Xianyang plain area,and the overall distribution characteristics are zonal with the characteristics of strong in the east and weak in the west.Most of the ground fractures in Xianyang area are structural ground fractures,which are mainly controlled by the fracture activities,and the directions are mainly NE,NEE and NW directions,which are basically consistent with the directions of fractures in Xianyang area.(2)The ground subsidence rate maps of the study area in 2020 and 2021 were obtained by using SBAS-In SAR technology,and through the analysis of the ground subsidence rate maps,it was obtained that four ground subsidence centers and two ground rebound centers mainly existed in the study area in 2020;six ground subsidence centers mainly existed in the study area in 2021,and the six ground subsidence centers existing in 2021 were Field investigation was conducted on the six ground settlement centers existing in 2021,and it was found that surface and house cracking existed in Yunyang town,Luqiao town and Xianyang airport settlement centers.(3)Combining with the analysis of typical ground crack causes in Xianyang area,based on the selection principles of comprehensiveness,scientificity,independence and representativeness,seven evaluation factors of topography and geomorphology,stratigraphic lithology,fracture activity and distance from fracture,hydrogeology,earthquake,ground subsidence and ground crack type and activity were selected,and the evaluation of ground crack hazard in Xianyang area was obtained by two calculation models of SVM and WOA-ELM The results show that the results obtained by the WOA-ELM model are more scientific and reasonable,more accurate,and more consistent with the actual situation in Xianyang area.(4)The results of the WOA-ELM model were used to evaluate the hazard,and the hazard of ground fracture in Xianyang area was classified into four levels:very low hazard area,low hazard area,medium hazard area and high hazard area,in which the area of high hazard area was 110.23km~2,accounting for 2.79%of the area of the study area,62 ground fractures were developed,the percentage of ground fractures was 44.30%,and the frequency ratio was 15.88,which is in line with the logical reasoning of high correlation of ground fractures in high-risk areas,and its evaluation results are relatively reasonable,the corresponding suggestions for ground fracture disaster prevention and control are given.
Keywords/Search Tags:Ground fracture, Ground settlement, Hazard evaluation, High-dimensional support vector machine, Whale optimization algorithm, Extreme learning machine
PDF Full Text Request
Related items