Research On Prediction And Application Of Overtopping Of Compound Breakwater Based On Random Forest Algorithm

Breakwater is an important hydraulic structure in coastal areas.It can effectively prevent waves from crossing the top of the dike and protect the workers and buildings behind the dike.A large number of breakwater collapse accidents show that overtopping is an important cause of breakwater failure.When a large number of waves over the top of the embankment,the embankment and the back slope will form a large impact force,so that the breakwater is damaged.Compound breakwater is a common type of breakwater,and its structure is more complex,so it is more difficult to calculate the amount of breakwater.In order to achieve the due strength when designing the compound breakwater,the prediction model of the overtopping volume is studied,in order to build a prediction model of the overtopping volume with better prediction accuracy and greater application value.The main research of this paper is as follows:This paper is based on the European CLASH overtopping dataset.Firstly,some important parameters and meanings of CALSH dataset are explained.Secondly,the dataset is cleaned to select the data which is consistent with the prediction of the overtopping volume of compound breakwater.According to the characteristics of the compound breakwater,the features which have great influence on the overtopping volume of the compound breakwater are selected and processed by dimensionless method.Then,a random forest model was established for the overtopping volume of the compound breakwater by using the Sklearn module provided by Python,and the parameters of the model were optimized by gridsearch(GridSearchCV)method.The accuracy of the model was evaluated by two evaluation indexes:root mean square error(MSE)and determination coefficient(R~2).In order to further verify the accuracy and efficiency of the prediction model,the model is compared with the ensemble neural network(ENN)model.Finally,based on the random forest model established in this paper,combined with PyQT5 and Qt Designer,a set of prediction system software for the overtopping volume of compound breakwater is developed,and the software is applied in actual engineering tests to verify the performance of the software.This paper proposes a compound breakwater slope overtopping volume prediction model based on the random forest algorithm.The results show that the random forest model established in this paper has high prediction accuracy,and its prediction results are basically within 5 times the tolerance zone.And the coefficient of determination(R~2)and root mean square error(MSE)reached 92.5%and 0.0035,respectively.However,the prediction results of the ensemble neural network(ENN)model are more than 5 times tolerance range,the coefficient of determination(R~2)and root mean square error(MSE)are 86.8%and 0.0046,respectively.In contrast,the random forest model has better prediction accuracy.Through actual engineering test,it is verified that the prediction system software has good prediction effect,is quick and convenient,and runs stably.Therefore,the research in this paper provides a reference for the calculation of overtopping of the compound breakwater.