Weigh-in-motion based on multi-sensor and RBF neural network

The work presented in this paper focuses on several aspects of weigh-in-motion. According to the mathematical model of weigh-in-motion, this paper proposed the methods of multi-sensor data acquisition and axle weight detection by using quadratic mean. The Radial Basis Function (RBF) neural network was used to construct the weighing system. In the modeling and training of RBF neural network, three different types of test were given: dead load, normal load and overweight. The results have indicated that using RBF neural network in weigh-in-motion has a significant effect on weighing test precision.