Font Size: a A A

Geological Hazard Warning Issues Based On Intelligent Genetic Algorithm

Posted on:2012-08-12Degree:MasterType:Thesis
Country:ChinaCandidate:T WangFull Text:PDF
GTID:2210330374953549Subject:Applied Mathematics
Abstract/Summary:PDF Full Text Request
Geological disasters have frequently occurred all over the world in the recent years, including earthquakes, landslides and debris flows. We strongly fear recently happened geological disasters, especially "Wenchuan Earthquake", "Yushu Earthquake" and "Qinghai Debris Flow". The advent of these disasters is attributed to activity of poor geological function and increasing construction activities by human beings. Therefore, the geological disasters have become one of important factors to restrict the economic and social development and improvement of people's living standard. Since the geological disasters are attributed to many reasons, they are uncertain and the classification of regional geological hazard risk becomes the key problem in the research of geological hazards.In the thesis, I mainly make a study on geological disasters of Jilin Province in the flood seasons and conduct an analysis of important disturbing factors, including the rainfall, so as to divide the grade of disaster regions from high to low and thus build a feasible early warning and forecast model.This thesis mainly focuses on the research of early warning of geological disasters in flood seasons in Jilin Province. Though the topography of Jilin Province is relatively stable, geological hazards are likely to happen in the flood seasons. This thesis does a research of geological disasters in flood seasons in Jilin Province by means of the concept of intelligent genetic algorithm. On the basis of advantages of BP (Backpropugation Neural Network) algorithm in the solution of nonlinear problems, the genetic algorithm is introduced to compute the initial weights and thresholds so as to avoid the local optimization of the results. Meanwhile, to reduce the number of iterations and to improve the solution speed, LM (Levenberg-Marquardt) algorithm is used to substitute the alternative gradient descent method, obtaining good results. This method could be used to set up a simulation model of geological hazard risk evaluation, provide objective evaluation of geological hazard risk in Jilin Province, and finally obtain the pragmatic geological hazard risk zoning.
Keywords/Search Tags:Intelligent Genetic Algorithm (IGA), iteration of L-M, weight, hazard evaluation
PDF Full Text Request
Related items