Designation And Empolder Of Highway Car-Loading Weigh-In-Motion System

The Highway Car-loading Weigh-in-Motion system(WIMS)has undoubtedly played an antive electronic role and social value in the protection of highway trancport. Along with the development of the Highway Transportation,Industrial production and Business Trade, it is required that the Highway WIMS needs to saticfy more qualification of modernization and scientization for the rapid automatically and the enforcement of the overloading rule.As to the Highway Car-loading WIMS, the weighing precision of the vehicle moving is the most important standard specification. It indicates the technical level of the Highway Car-loading WIMS. While the weighing signal processing of the actual Weigh-in-Motion is simple digital filter and the model is not fit well with non-leneared object. Even further signal processing technique of kinds of disturbing factors. So the WIM system's precision is hardly improved.Because there are so many factors affect the WIMS'accuracy, and there are not exact function relations among these factors, this paper introduces some new arithmetic models to settle these factors compared with other models.In theory of improving the WIM system's precision, final chance is choosing BP neural network. It mostly introduces the basic idea, calculated process, executes steps and existed problems of BP Neural Network, and Finally puts forward the improved method to satisfy weigh-in-motion requirement.Through analyzing Highway WIMS'cars,this paper bases BP Neural Nerwork model. The weigh-in-motion datas, which gathered in scene and trained by the BP Neural Network model, indicate that the precision accords with practical requirement and international standard. In hardware, the system use single-chip microcomputer and introduced the choosing condition of the single-ship microcomputer. The WIM system of the paper described can implement not only the basic function of gathering data, processing data, display data, and query data,but preserved the gathering data into the database and print in form. The system mainly use C language to write in the single-chip microcomputer, and use Visual C++ 6.0 to write in the computer for processing data. The RS-232 is used to communicate the serial.Otherwise, the paper use MATLAB R2007 language to model weighing system, simulink and train the weighing datas.