基于BP神经网络的电动堆高车货叉偏载检测  

Partial Load Detection of Electric Stacker Truck Forks Based on BP Neural Network

作　　者：贠海涛[1] 王成振 曹爱霞[2] 谢建新[2] YUN Haitao WANG Chengzhen CAO Aixia XIE Jianxin(School of Automobile and Traffic, Qingdao Technological University, Qingdao 266520, China School of Traffic and Marine Engineering, Qingdao Huanghai University, Qingdao 266427, China)

机构地区：[1]青岛理工大学汽车与交通学院,山东青岛266520 [2]青岛黄海学院交通与船舶工程学院,山东青岛266427

出　　处：《济南大学学报（自然科学版）》2017年第2期143-149,共7页Journal of University of Jinan(Science and Technology)

基　　金：国家自然科学基金项目(51205215);青岛开发区科技计划项目(2014-1-63);青岛理工大学名校工程建设项目(MX4-042)

摘　　要：为了避免电动堆高车货叉在装卸货物时发生偏载安全事故,通过在试验车上安装偏载传感器来采集相关数据,利用神经网络工具箱训练数据来获得网络权值以及阈值并生成Simulink模型,再利用权值和阈值和Simulink模型建模的方法,构建基于BP神经网络的电动堆高车货叉偏载检测人工算法。对误差的分析验证表明:该算法可进一步提高检测货叉偏载距离的精度;该偏载检测算法应用到工程实践,能够满足电动堆高车货叉偏载检测的设计要求。For avoiding the partial load security accident in the process of loading and unloading cargo with electric stacker truck forks, related data are collected through partial load sensors installed on the tested vehicles, and the weights and thresholds of network as well as the Simulink model were acquired from the training data of neural network toolbox. The artificial algorithm of partial load detection based on BP neural network was established by using of weights and thresholds and Simulink modeling. Calculation error analysis and verification demonstrate that the algorithm further improves the detection precision of partial load distance of forks, and engineering practice shows that this algorithm meets the design requirements of partial load detection.

关 键 词：电动堆高车 货叉偏载 BP神经网络 SIMULINK模型 

分 类 号：U273.99[机械工程—车辆工程]