Adaptive compensation for eccentric error in truck scale

The conventional compensation for eccentric error in truck scale is realized by repeatedly regulating the potentiometer in junction box to adjust gain of each channel with load cell, which is fussy and labor-intensive. In this paper, eccentric error sources are analyzed, and an error model is established. A method of adaptive compensation for eccentric error is proposed, and its model of compensation based on radial basis function neural network (RBFNN) is established, which considers the output signals of multiple load cells as its input variables. A learning algorithm of RBFNN is also presented. Experiments and verifications in field show that with adaptive compensation the eccentric error from some nonlinear factors of truck scale is greatly reduced, and it is less than the maximum permissible error of scales with medium accuracy defined by international standard OIML R76 “Nonautomatic Weighing Instruments”.