Research on the algorithm of dynamic weighing system has become a difficulty in the research of dynamic weighing system.In order to improve the measurement accuracy of dynamic weighing system, the shortrange signal of the sensor is analyzed in the condition of vehicle non stopping, and the high frequency components in the noise signal are filtered out by the Butterworth lowpass filter.The LevenbergMarquardt algorithm is used to fit the least square method, which can effectively suppress the lowfrequency periodic random interference and realize the separation of the load.Further,the weight of the vehicle is calculated by load separation.Experiments show that the algorithm can improve the measurement accuracy of dynamic weighing system.