ABSTRACT:
This paper presents a novel weighing method for truck scale based on neural network with weight-smoothing constraint (WSCNN). In this method, the truck scales' prior knowledge, i.e., the correlation of the load cells' output signals and the independence of the truck scale's weighing results with the loading positions, is used to construct the constraint condition for optimizing a neural network (NN) in case of lacking samples, then the NN's performance index is created and the detail algorithm of WSCNN is given. The experimental results show that the weighing errors of the truck scale with WSCNN are far less than those of DIMNN (DIMNN is a training algorithm of NN by using only the data samples, not the prior knowledge), and the errors reduce by one order of magnitude against those of DIMNN. In addition, the WSCNN's generalization ability is more excellent than that of DIMNN especially the lack of training samples.
ABSTRACT:This paper presents a novel weighing method for truck scale based on neural network with weight-smoothing constraint (WSCNN). In this method, the truck scales' prior knowledge, i.e., the correlation of the load cells' output signals and the independence of the truck scale's weighing results with the loading positions, is used to construct the constraint condition for optimizing a neural network (NN) in case of lacking samples, then the NN's performance index is created and the detail algorithm of WSCNN is given. The experimental results show that the weighing errors of the truck scale with WSCNN are far less than those of DIMNN (DIMNN is a training algorithm of NN by using only the data samples, not the prior knowledge), and the errors reduce by one order of magnitude against those of DIMNN. In addition, the WSCNN's generalization ability is more excellent than that of DIMNN especially the lack of training samples.
การแปล กรุณารอสักครู่..