Research On Multi-source Information Fusion In Slope Monitoring

The coal industry is an important basic industry which impacts the country’s economy.And at the same time, the situation of the coal’s safety is very grim. In this thesis, we use theinformation fusion method to forecast the slope stability based on the kinds of observationdata.Information fusion researches how to use the multi-information to realize the target. Thisthesis uses so many advantages of information fusion and combines the practical situation todo the jobs below:After studying the domestic and international methods of slope monitoring, we firstlystudy the main factors which impact the slope stability. Then according to the main factors,the monitoring work is arranged reasonably. After the data is preprocessed, we proposeseveral methods that can predict the slope stability.In order to determine the slope stability, a recognition method based on support vectormachine and D-S evidence theory is proposed to get the state of the slope stability timely andaccurately. This method is more useful than traditional method. Because it can predict thechanging trend of the slope. The experiment results confirm that this method can greatlyenhance the classification accuracy of slope stability comparing with traditional methods(more than5%). A better understanding of the state of slope is obtained by the probabilityoutput and the stability trend of the slope also can be obtained. So it is helpful for us to do themanagement and protection of the slope timely. At last, this predicting model is used on themonitoring web to help forecast the state of the slope and do the management timely.