Research On Prediction Of Economic Loss Of Earthquake Disaster

China has been suffering from the disturbance of devastating earthquake for many years due to its extremely special geographical position,that is located in the middle of world's two active seismic zones.Taking 2018 as an example,according to the statistics of the relevant departments,there were 542 earthquakes above magnitude 3 in China in the whole year,which greatly affected the normal production and life of our people and seriously threatened the security of our people's lives and properties and caused huge economic losses.Therefore,how to accurately predict the direct economic losses of earthquake disasters has become a key issue that must be studied in the field of earthquake prevention and disaster reduction.It is possible to actively adopt comprehensive measures for earthquake prevention and disaster reduction through forecasting results,thereby minimizing the economic losses caused by earthquake disasters.First of all,based on the theory of grey relational analysis,some historical statistics of some earthquake disasters in China from 1996 to 2017 are supported by data.On the basis of analyzing the direct economic losses caused by different factors,calculate the grey correlation degree between different factors and direct economic losses of earthquake disaster,then determine the important and small factors that affect the direct economic losses of earthquake disaster according to the result of grey correlation degree.Among them,the total influencing factors are earthquake disaster occurrence time,earthquake magnitude,focal depth,the affected population,the area of the intensity area and casualties.Secondly,based on the grey correlation theory to analyze the direct economic losses of earthquake disasters,two direct earthquake economic losses prediction models based on Support Vector Machine(SVM)and BP neural network are established based on the earthquake disaster data.By using the average relative error,the maximum relative error and the minimum relative error as the evaluation indexes of the prediction model,we can see that the prediction accuracy of SVM model is higher than that of BP neural network model.The prediction accuracy of the SVM model is higher than that of the BP neural network prediction model.The SVM direct economic losses model established by historical earthquake disaster statistics shows that the SVM theory can be applied to the field of direct economic loss prediction of earthquake disasters,and compared with the BP neural network model,the prediction model has high prediction accuracy,feasibility and practicability,and can provide a theoretical reference for the prediction of direct economic losses of earthquake disasters.Figure 23;Table 9;Reference 48...