Study On Engineering Insurance Rate Determination Of Expressway Based On Particle Swarm Optimization Bp Neural Network

With the rapid development of Chinese economy,the expressway has developed rapidly as an important infrastructure for social development and economic construction.On account of the highway project has the characteristics of wide construction surface,long line and long construction time.The risk of its construction is large,and the loss caused by natural disaster or accident is huge.Engineering insurance can effectively transfer engineering risks to contractors in an economical form.However,the development of engineering insurance in China is relatively backward because the competition in the insurance market is not standardized,besides there are also lack of statistical data.As a result,the rate is determined to be confusing.The process of determining it depends on the experience on engineering,which it's difficult to match the actual risk level of the project and reduce the risk of protection.In order to scientifically and reasonably measure the insurance project insurance premium rate and effectively transfer the risk through insurance,this paper uses particle swarm optimization(PSO)algorithm to optimize the initial threshold and weight of BP neural network based on the consideration of engineering risk and actual insurance claims.A highway engineering insurance rate determination model was established in this study.The model was applied to 34 engineering insurance cases,and the relationship between risk indicators and rates in insurance samples was fitted by PSO-BP neural network to realize rate prediction.This article includes the following research work:(1)To introduce the research background and significance.Extensive reading of relevant,domestic and international literatures,expounding the status quo of highway development in China,the importance of engineering insurance for project development and the problems of insurance premium rate,on this basis,combing and summarizing domestic and international engineering risk of management and engineering insurance research.The current status and results provide a reference for the research of this paper;(2)The related theoretical foundations of risk management,engineering insurance,BP neural network and particle swarm optimization(PSO)are expounded,which provides theoretical basis for research development.(3)Based on relevant literatures and regulations,the Risk DecompositionStructure Method(RBS)was used to analyze the risk indicators of expressway engineering that affect the insurance premium rate,and the index system was used to optimize the index system to construct natural disasters and accidents.Risk assessment index system of seven indicators such as risk and project environmental risk;use risk matrix method to evaluate risk indicators and establish evaluation criteria;The analytic hierarchy process is used to determine the weight of risk indicators as a parameter of the neural network model.(4)Based on PSO-BP neural algorithm,the highway engineering insurance rate determination model is constructed.PSO is used to optimize the BP neural network to obtain the initial weight and threshold of the network.34 engineering insurance samples were selected,and the risk index weights were calculated for each sample,and the contract rate was revised according to the insurance loss rate,so that the sample revision rate was closer to the actual risk situation,thus constructing the network learning sample data;through PSO-BP The neural network fits the relationship between the risk index weight and the revised rate in the sample,constructs the rate determination model,and realizes the rate prediction.The comparative analysis of the PSO-BP neural network and the BP neural network rate simulation results,and concludes: The PSO-BP neural network model can better reflect the actual risk level of highway engineering,make the forecast rate close to the revised rate,the prediction accuracy is high,and the network convergence speed is fast,which is suitable for the insurance rate determination.(5)Case analysis.Taking a highway as an example,the model is used to determine the rate.The overall forecasting effect of 13 samples is analyzed.The results show that the rate obtained by using the PSO-BP neural network highway engineering insurance rate determination model is closer to the real risk situation of the project,and the rate is determined reasonably.