The Research On The Highway Pavement System With Neural Network And Combination Forecast

With the gradual improwement of Highway Nerwork in China, The Road management department work focus will from large-scale construction road network to conserve.rebuild,maintain and improve the existing road network.,whick highly show the importance of pavement management system.Therefore,this paper does some researches on the highway pavement pavement management system, including:study the damaged area highway pavement types and reasons, road conditions evaluation and pavement performance prediction, key research neural network theory in the highway pavement management system application.Pavement performance of the evaluation and prediction is an important part of pavement management system, it is reasonable formulate pavement maintenance maintenance plan and capital allocation scheme foundationo This paper first analyze highway pavement structure and division of classification, respectively for flexible pavement and rigid pavement of conventional diseases and causes and then give the maintenance plan.Secondly, the paper analyze the highway pavement performance single-item indicators, discusses the current pavement performance comprehensive evaluation method, in view of the current road test data is fuzzy and nonlinear characteristics, established based on fuzzy neural network of comprehensive evaluation method.Finally, in the highway pavement performance prediction, the research in China and abroad are summarized application performance prediction model, analyses the influence factors of pavement performance and changing trends.The paper use combination forecast method increase the prediction accuracy of thought, firstly by using grey prediction model and the BP neural network model, then based on entropy method of combination forecast model,use the combination-weighted coefficient。Through examples, this paper combined forecasting model on the method and theory is feasible, for highway pavement managers in pavement maintenance and management of decisions can provide a new thought and help.