改进BP神经网络的车载称重测量方法

打开文本图片集

关键词：车载称重系统；鹦鹉优化算法；BP神经网络；胎压传感器；货车载重；测量精度中图分类号：TN912.2-34；TP274 文献标识码：A 文章编号：1004-373X（2025）20-0069-05

Method of on-board weighing measurement based onimproved BP neural network

WANGHaoqi1，YINJihui1，ZHENGXuguang²，CHENDongdong'，ZHANGPenglei² （1.SchoolofElectricalandMechanicalEngineering，NortheastForestryUniversity，Harbin15oo40，China; 2.Qingdao JunfengHuachuangElectronic TechnologyCo.，Ltd.，Qingdao 266199，China）

Abstract：In alusion to the low measurement accuracyof on-board weighing systemadynamic weighing model based on BP neuralnetworkoptimizedbyParrotoptimization（PO）algorithmisproposedtocopewiththeoverloadingoftrucksandrealizethe dynamicandacuratemeasurementof truck’sloadcapacity.Theon-board weighing systemisintroduced briefly，andthesignal collectedbythesensorispreprocessdbymeansofthecompositefiltering method.Theweightsandthresholdsof theneural networkareoptimizediterativelybymeansofthePO-BPalgorithmtoconstructaweighing modelwith thepreprocessedsignals oftirepressure，，whelrotationalacelerationand temperatureasinputs，soastostimatetheloadcapacityofthetruck.The experimental resultsshow thatthePOoptimization basedBP neural network weighing algorithmcan decrease theroot-meansquare error to 2.3% and theaverage absolute error to1.9% within a smaller numberof iterations.Incomparison with the traditionalBPneuralnetwork，thealgorithmhashighermeasurementaccuracy，themeasuredvalueof truckloadcapacityis closer to the real value，and the relative error is within 5% ，which can meet the accuracyrequirement of on-board weighing.

Keywords：on-board weighing system;Parrot optimization algorithm;BP neural network;tire pressure sensor;truck load; measurement accuracy

0 引言

随着交通运输业的快速发展，大型货车的数量也迅速增加，车辆超载问题日益严重，这一现象缩短了公路桥梁的寿命，并引发了交通安全隐患。（剩余6368字）