基于APSO_LSSVM的模块化产品设计时间预估研究  

Research on Prediction of Modular Product Design Time Based on APSO_LSSVM

作　　者：王兆华[1] 黄丽[2] WANG Zhaohua;HUANG Li(School of Management, Northwestern Polytechnical University, Xi’an 710072;School of Electrical and Information Engineering, Jiangsu University, Zhenjiang 212013)

机构地区：[1]西北工业大学管理学院,西安710072 [2]江苏大学电气信息工程学院,镇江212013

出　　处：《微型电脑应用》2019年第4期4-7,共4页Microcomputer Applications

基　　金：国家自然科学基金项目(71572147);江苏省高校自然科学基金(12KJB210001)

摘　　要：为了实现模块化产品设计时间精确预估,提出了采用改进的自适应粒子群和最小二乘支持向量机(APSO_LSSVM)相结合的预估方法。首先构建模块化产品设计最小二乘支持向量机(LSSVM)时间预估模型;其次应用改进后的自适应粒子群(APSO)对LSSVM模型参数进行优化,以加快模型收敛速度并寻找全局最优;最后将APSO_LSSVM时间预估模型用于打印机模块化设计中。结果表明,基于APSO_LSSVM时间预估模型对时间的预估与实际情况基本一致,与基于tPSO_FNN时间预估模型相比,在输入变量较多的情况下,收敛速度更快、预估精度更高。In order to predict modular product design time accurately, a design time prediction method is proposed. It is based on the combination of the adaption particle swarm optimization (APSO) and the least squares support vector machine (LSSVM). Firstly, the LSSVM time prediction model of modular design is established. And then, the improved method of APSO is presented to optimize the LSSVM model in order to quicken convergence rate of the model and search global optimum. At last, this model is verified by the time prediction of printer combination machine modular design. The results show that the predicted value of the APSO_LSSVM model is basically identical to the actual situation. And compared with the tPSO_FNN model, the APSO_LSSVM model has faster convergence speed and higher prediction accuracy for the case of multiple input variables.

关 键 词：产品开发 模块化设计 时间预估 粒子群算法 最小二乘支持向量机 

分 类 号：TP183[自动化与计算机技术—控制理论与控制工程]