| The non-standard weighing behavior of overloaded vehicles affects the weighing results of dynamic truck scales.Among them,the axle-loaded dynamic truck scale is more susceptible to the influence of non-standard weighing behavior.Aiming at this problem,this paper proposes a non-standard weighing behavior recognition algorithm based on the moving trajectory of the axle to reduce the influence of non-standard weighing behavior on its weighing results.The main research contents include:1、This paper introduces the composition and working process of axle load dynamic truck scale,analyzes the factors influencing the calculation results of axle weight and axle number,introduces the common non-standard weighing behavior,and analyzes the influence of non-standard weighing behavior on axle load dynamic truck scale.2、The calculation method of the equivalent contact point between the tire and weighing platform and axle moving track was analyzed,the identification characteristics and calculation method of irregular weighing based on axle moving track were studied,and the calculation model of identification characteristics of axle moving track and irregular weighing behavior was established by Labview.3、According to the collected original data and calculation model,the identification feature data set of non-standard weighing behavior is established.The recognition accuracy and speed of several machine learning classification algorithms are compared.Among them,the recognition accuracy of different machine learning classification algorithms to the recognition feature data sets can reach more than 90%.At the same time,the recognition accuracy of the KNN algorithm is the highest,reaching 97.5%,and the recognition speed is fast,requiring only 1.29 milliseconds.The results show that the identification features of irregular weighing behaviors based on axle moving trajectory can effectively distinguish irregular weighing behaviors.KNN algorithm has high recognition accuracy and fast recognition speed,which can meet the requirements of recognition accuracy and rapidity of axle load dynamic truck scale to non-standard weighing behavior. |
