Research On Risk Assessment Model Of Meteorological Disaster In Water Traffic

In recent years,due to frequent meteorological disasters such as strong winds,heavy rainfall and low visibility,the number of water traffic accidents has been increasing.If the meteorological factors of the ship's navigation route can be analyzed in time and the risk assessment of the channel can be adopted,the occurrence of accidents can be reduced to a certain extent.Therefore,based on the fuzzy comprehensive evaluation method and BP neural network,this paper constructs a risk assessment model and an improved BP neural network prediction model based on genetic algorithm.The research and application of this model can not only effectively evaluate the risk of ship navigation,but also assist ship safety prevention.This paper takes the main waterway of hunan province as the research object,and the main research contents are as follows:(1)The risk assessment index system of water traffic meteorological disaster in hunan province is put forward.Firstly,the data of water traffic accidents in hunan province are statistically calculated,and the influencing factors of water traffic safety navigation are analyzed,and the main influencing indexes such as visibility,wind speed,precipitation,channel width,channel length,channel depth,channel bending radius and guarantee rate of minimum navigable water level are obtained.Then,the expert evaluation method is used to determine the impact index value,and the risk evaluation index system is proposed.The risk assessment model based on the system can effectively reduce the risk of water traffic.(2)The risk assessment model of water traffic meteorological disaster in hunan province is established by fuzzy comprehensive evaluation method.Firstly,channel data and meteorological data are collected from the databaseof hunan meteorological bureau.Then,factor set,evaluation set and membership function based on risk assessment index are constructed,and AHP analytic hierarchy process is adopted to calculate the weight.Finally,the risk assessment model of water traffic meteorological disaster in hunan province is established to evaluate the channel risk.Although some historical data are missing,the accuracy of the model is still 91%.The experimental results show that the evaluation result of this model is in line with the actual risk situation of navigation channel,which provides a new direction for the study of meteorological disaster risk.(3)An improved BP neural network based on genetic algorithm is used to construct the risk prediction model of water traffic meteorological disaster in hunan province.Based on genetic algorithm,this improved algorithm optimizes the weight and threshold of the algorithm,and verifies the feasibility and stability of the new algorithm.Firstly,sample data were collected from hunan meteorological bureau database and processed.Then,take the risk assessment results as sample output,and use the improved BP neural network based on genetic algorithm to predict the risk of the main waterway in hunan province,and the highest accuracy of the model prediction reaches 96.875%.The research shows that the improved BP neural network model based on genetic algorithm can effectively predict the risk of waterway and provide technical support for the prediction of meteorological disaster risk of water traffic in hunan province.