Research On The Maintenance Management System Of The Steel Bridge Deck Pavement

The design and construction technology of the steel bridge deck pavement becomes more perfect at home and abroad, and the maintenance management of the steel bridge deck pavement being concerned gradually. The steel bridge deck pavement performance results can be got by the evaluation model with measured bridge deck diseases data, to find suitable forecasting model and explore the development trend of pavement performance index, avoid unfavorable overhaul period and formulate current conservation measures with maintenance decision tree, which is beneficial to delay the development of disease and prolong the pavement service life.Firstly, investigate the existing commonly used steel bridge deck pavement material and structure; analyze common diseases of different pavement structure and their causes. On the basis of comparison of similarities and differences between the traditional asphalt pavement and steel deck pavement, figure out evaluation model suitable for steel bridge deck pavement.Secondly, according to the investigation of the disease a real bridge pavement, analyze the factors of the decline performance of steel deck pavement monthly, according to the variation characteristics of each factor under short interval, choose gray model and Markov model to predict separately.Thirdly, demarcate the peer lane regionally, calculate the pavement indexes with evaluation model, this paper selects ring slow lane crack condition index of Wuhan Baishazhou Bridge to predict, than based on the selection current prediction model, using BP neural network to learn and predict the model parameters value.Finally, with reference to the relevant norms, define the maintenance engineering of steel bridge deck pavement, establish maintenance decision tree based on evaluation model suitable for steel bridge deck pavement, according to the characteristics of deck diseases pictures selected content feature method to process the images; after that, treated pictures was similar image recognized, ues the best match picture to identify specific information of the disease, so that bridge management department can make the maintenance plan and analysis the development of the disease.