基于CBR-HJaya-BP的液压缸加工工时预测研究  

Research on Hydraulic Cylinder Machining Man-Hour Forecasting Based on Case-Based Reasoning and HJaya-BP Neural Network

作　　者：唐红涛[1,2] 杨思琴 张伟 黄浪[1,2] 官思佳 TANG Hongtao;YANG Siqin;ZHANG Wei;HUANG Lang;GUAN Sijia(School of Mechanical and Electronic Engineering,Wuhan University of Technology,Wuhan Hubei 430070,China;Hubei Provincial Engineering Research Center of Robotics and Intelligent Manufacturing,Wuhan Hubei 430070,China;General Department of Space Engineering,China Aerospace Science and Industry Co.,Ltd.,Beijing 100039,China)

机构地区：[1]武汉理工大学机电工程学院,湖北武汉430070 [2]机器人与智能制造湖北省工程研究中心,湖北武汉430070 [3]中国航天科工集团空间工程总体部,北京100039

出　　处：《机床与液压》2023年第8期124-129,共6页Machine Tool & Hydraulics

基　　金：国家自然科学面上项目(52075401)。

摘　　要：为了提高冶金行业液压缸加工工时预测精度,提出一种结合基于案例的推理(CBR)和混合Jaya算法优化BP神经网络的液压缸加工工时预测方法(CBR-HJaya-BP)。使用混合Jaya算法优化BP神经网络的初始权值和阈值,采取基于Sin混沌反向学习的种群初始化策略提高初始解的质量,引入阿基米德优化算法中的转移算子,在探索阶段采用均匀交叉产生中间种群,在开发阶段使用Jaya公式产生中间种群,在解的保留策略中引入了模拟退火算法中的Metropolis准则,以跳出局部最优。以某冶金液压缸制造企业的历史加工数据库为样本,采用CBR方法提取与待预测液压缸的特征参数相似度最高的历史数据,使用提出的HJaya-BP模型进行预测实验,并与改进前的Jaya-BP模型以及原始BP神经网络模型进行对比。实验结果表明,HJaya-BP模型的预测准确度和稳定性均为最优。In order to improve the accuracy of hydraulic cylinder machining man-hour forecasting in metallurgical industry,a new man-hour forecasting method based on case-based reasoning and BP neural network optimized by hybrid Jaya optimization algorithm(CBR-HJaya-BP)was proposed.The hybrid Jaya optimization algorithm was used to optimize the initial weights and thresholds of BP neural network,a population initialization strategy based on Sin chaos and backward learning was adopted to improve the quality of initial solutions.The transfer factor in the Archimedes optimization algorithm was used to balance the process of exploration and development,the uniform crossover was used to generate the intermediate population in the exploration phase and the Jaya formula was used to generate the intermediate population in the development phase.Finally,the Metropolis criterion of simulated annealing algorithm was used to jump out of the local optimum.Taking the historical processing database of a hydraulic manufacturing enterprise as the sample,the case-based reasoning method was used to extract the historical data which was similar to the hydraulic cylinder needed predicting.Then the prediction experiments were carried out by using the proposed HJaya-BP model,the Jaya-BP model and the initial BP neural network model.The results show that the forecasting accuracy and stability of HJaya-BP are all the best.

关 键 词：液压缸 工时预测 基于案例的推理 改进Jaya算法 BP神经网络 

分 类 号：TH16[机械工程—机械制造及自动化] TH166