Artificial neural network-based bank collapse prediction model;

Mahu dyke,which is the place of bank collapse occurring frequently(e.

The authors studied the relationship between bank collapse and seepage failure of embankment foundation and the impact of flood soaking and water level variation on bank stability during flooding periods.

On part dike collapse cause in Zhongyang and corresponding measure;

The paper introduced Jiangxingzhou embankment collapse and its project harnessing measures and construcction of concrete hinge mattress were discussed.

The result shows that the safety degree of Beimenkou reach is smaller than that of others,which is verified by its frequent bank-failures,meanwhile that fuzzy statistic clustering theory can be applied for the study on bank-failures.

As an example, the sections of Shishou reaches whose parameters were measured in September, 1998, were quantitatively evaluated using the above method and the result showed that the bank-failure dangerous extent of Beimenkou section is higher than .

Using the Evaluation Model based on Analytic Hierarchy Process for the influential factors of bank failure, this paper carried out a quantitative analysis of multiple influential factors of bank failure, which laid a solid foundation for in-depth research on the mechanism and prevention of bank failure.

But local section will be made regulation to some extent,as a result,the scour of nearbank river bed and the collapse and recession of bank line will happen,the bank failure may take place for the section without prote.

Reservoir bank failure is a typical form of collapse.