Analysis For The Economic Loss Of Landslid And Debris Flow On Rough Set And BP Neural Network

Landslide refers to the slippage of the rock and soil along the shear failure surface. Debris flow is in the mountains or valleys and steep terrain of the region, since the Blizzard of the torrential rain or other natural disasters caused by landslides and carry a large amount of sand and rocks in flood, debris flow with the feature of sudden and fast flow rate, flowlarge material capacity and undermine the strong. Landslides and debris flows often destroy highway, rail and other transportation facilities even the towns and villages, are devastating sudden natural disasters, more common in the mountains, there are great similarities due to landslide and debris flow, and the two often occur simultaneously, it is often to discuss the two kinds of disasters. China’s vast territory and abundant resources are victims of the landslide and debris flow disaster-prone countries and affected more severely. Scholars have more danger and risk analysis, which is the affected area since the role of prevention and vigilance in the study of landslides and debris flows. At present, however, economic loss assessment for landslides and debris flows lack of systematic research, and disaster economic loss of value can not be accurate statistics and computing, because the statistics of the losses caused by disasters not partakers hazard refinement, often according to the broad categories such asgeological disasters and statistics by the general category of natural disasters; Second, because the disaster loss statistics that disaster quantitative assessment itself, there are a number of scientific and technical issues to be resolved. To consider the actual needs on the current level of disaster research and disaster prevention and mitigation, due to data acquisition objective, accurate, comprehensive and other reasons, making the statistical summary of disaster losses there is a big shortage and difficulty.In this paper, by rough sets and BP neural network theory, establish the rough BP neural network evaluation model of the landslide and debris flow disaster economic losses. First.the model is applied attribute reduction method to the easy collection of the index set:{number of people affected, death, missing, emergency transfer, area of affected crops, inundated area, crops area, the number of houses collapsed, damaged houses}, the reduced index set:{number of people affected, death and missing persons, emergency transfer, crop area affected, the number of houses collapsed}. And then use the BP neural network has strong applicability of landslide and debris flow disaster economic loss rate. Calculate the economic value of the loss rate.then according to the interval grading.landslide and debris flow losses are divided into high, medium and low disasterloss level. Then case studies,use28sets of data which the2009nationwide group of landslide and debris flow disasters, random the order, the25sets of data as training samples to enter the BP neural network training; and then enterthree groups of data:loss of data in Qinghai, Ningxia and Xinjiang provinces, then use of trained BP neural network, calculate the economic loss rate of landslides and debris flows; Finally, according to the grading standards, get the Qinghai, Ningxia,Xinjiang’s2009landslide and debris flow disaster loss level.In this paper.rough BP neural network evaluation model for landslide and debris flow disaster economic loss assessment provides a feasible and effective method, the results of the assessment can provide a theoretical basis for the government and relevant departments in the landslide and debris flow disaster reduction and prevention. many experts use a variety of methods to assess the landslide and debris flow disaster damage, but did not use rough BP neural network method to assess.therefore, this article is the first time in the field of landslide and debris flow disaster damage assessment, try to use rough BPneural network for landslide and debris flow disaster damage assessment, and case analysis shows that the method is effective, and may be extended to other disasters’s damage assessment.