The Research Of Weigh-in-motiom System Based On DSP

WIM (weigh-in-motion) becomes an essential and emerging technology forhighway management nearly. This technology can effectively slow down the breakingof road surface, saving the cost of road maintenance, reduce traffic accidents causeddue to overloading. In order to improve the accuracy of WIM system, the design ofintelligent WIM system based on DSP was introduced in this paper.On the basis of referring a great deal of relevant literature, a number of factorsthat affect the accuracy of it were analyzed, including vehicle speed, vehicle vibrations,the ground is not flat, etc. Using three-dimensional simplified model of moving vehicle,the vehicle vibrations were classified and the reasons were found. In the last the idea ofimproving the precision of WIM system was suggested from both hardware andsoftware.Then, the hardware based on DSP of the WIM system had been designed. Thissystem used resistance strain load cell, and designed the signal amplifier circuit，Gain-limited low-pass filter circuit, A/D converter circuit, RS-232serialcommunication circuit,auxiliary circuit, and gave details of the developmentmethodology and development process of the software. Using the integrateddevelopment environment-CCS, the software of the system was designed. It includesfour parts-the main program module, A/D converter sub-module, data processingsub-modules and RS-232serial communications sub-module.Data processing has a very big impact on the accuracy of the system. Severalkinds of data processing algorithms on the dynamic weighing were introduced: theaverage algorithm, neural network algorithm, parameter estimation algorithm, thedisplacement integral algorithm and the EMD algorithm. In the last the averagealgorithm was applied in this system. On the one hand FIR low-pass filter was used toimprove the accuracy by reducing the affect of vehicle vibration. On the other hand theadaptive compensation algorithm was used to improve the dynamic characteristics ofthe system. To simulate the weight signal in MATLAB by using the above-mentionedmethod, good result was achieved.