HAZARD ASSESSMENT SYSTEM FOR GROUND FISSURE BASED ON COUPLING OF ANN AND GIS--A case study of ground fissures in Yuci City, Shanxi Province
Qiang Wu · Seismology and Geology · 2002
Modern ground fissures have became a serious geological hazard widely spread in many countries all over the world. In this paper, hazard causing factors of ground fissures in Yuci City have been analyzed in detail. It is found that the formation and development of ground fissures in Yuci City are controlled mainly by tectonic activity and influenced by some non tectonic factors such as exploitation of underground water, difference in properties of strata, character of landform etc. On the basis of the analysis of each factor of ground fissures in Yuci City, Shanxi Province, the thematic layer in geological sense is established by using Geographic Information System (GIS). This paper introduces how to establish the hazard assessment model and hazard assessment system for ground fissure based on Artificial Neural Network (ANN) and Geographic Information System (GIS). The hazard assessment system consists of five function modules including input module, output module, search and inquire module, assessment module and maintenance module. The function of each module is introduced in this paper. As an example, a nonlinear appraisal model for the hazard of ground fissures in Yuci City based on Artificial Neural Network is introduced. The model is a back propagation ANN model, which has an input layer with four points, a hidden layer with four points, and an output layer with one point. Using this model the ground fissure hazard in Yuci City has been evaluated, and a map of the division of ground fissures risk for Yuci City has been established. This result may provide scientific basis for land planning and construction of Yuci City.