多维非线性特征重构与融合的复杂产品工期预测方法

doi:10.3901/JME.2023.06.294

机械工程学报 ›› 2023, Vol. 59 ›› Issue (6): 295-309.doi: 10.3901/JME.2023.06.294

• 交叉与前沿 • 上一篇

扫码分享

多维非线性特征重构与融合的复杂产品工期预测方法

常建涛¹, 乔子萱¹, 孔宪光¹, 杨胜康¹, 罗才文²

1. 西安电子科技大学机电工程学院西安 710071;
2. 成都中电锦江信息产业有限公司成都 610057

收稿日期:2022-04-20 修回日期:2022-09-10 出版日期:2023-03-20 发布日期:2023-06-03
通讯作者: 常建涛(通信作者),男,1980年出生,博士,副教授,博士研究生导师。主要研究方向为智能制造与数字化制造、工业大数据技术及系统、工业人工智能技术。E-mail:taocj@xidian.edu.cn
作者简介:乔子萱,女,1998年出生。主要研究方向为智能制造与工业大数据技术。E-mail:qiaozx98@stu.xidian.edu.cn;孔宪光,男,1974年出生,博士,教授,博士研究生导师。主要研究方向为面向智能制造的工业大数据与数字孪生技术。E-mail:kongxianguang@xidian.edu.cn;杨胜康,男,1994年出生,博士研究生。主要研究方向为智能制造与工业大数据技术、智能运维技术。E-mail:skyang1@stu.xidian.edu.cn;罗才文,男,1994年出生,硕士。主要研究方向为智能制造与工业大数据技术。E-mail:caiwenluo@qq.com
基金资助:
陕西省重点研发计划(2020ZDLGY07-08)、陕西省科技重大专项(2019zdzx01-01-02)和国家自然科学基金(51505357)资助项目。

Novel Product Duration Prediction Method of Complicated Product Based on Multi-Dimensional Nonlinear Feature Reconstruction and Fusion

CHANG Jiantao¹, QIAO Zixuan¹, KONG Xianguang¹, YANG Shengkang¹, LUO Caiwen²

1. School of Mechano-Electronic Engineering, Xidian University, Xi'an 710071;
2. Chengdu Zhongdian Jinjiang Information Industry Co., Ltd., Chengdu 610057

Received:2022-04-20 Revised:2022-09-10 Online:2023-03-20 Published:2023-06-03

摘要/Abstract

摘要： 制造企业复杂产品零部件种类众多、加工和装配工序复杂，质量、工艺、设备等各类数据呈现多维度、多尺度、多噪声等特点，工期的关键影响特征提取难度大，预测精度难以保证。针对上述问题，提出一种多维非线性特征重构与融合的复杂产品工期预测方法，首先提出基于集成堆栈式自编码器的多维非线性特征重构与融合方法并构建相应模型，建立特征间的复杂关联耦合关系，形成工期关键因素特征池；基于深度学习算法构建多维非线性重构与融合的复杂产品工期预测模型，实现复杂产品工期的准确预测。选取某企业断路器和3D打印产品为对象进行工期预测的应用验证和对比分析，本方法的均方根误差平均值为1.28，平均绝对百分比误差平均值为3.01%，与未进行特征重构融合的方法，以及支持向量机、神经网络等方法相比，在精度方面均有提升，方均根误差至少降低了约10.87%，平均绝对百分比误差至少降低了约7.74%，证明所提方法的有效性和实用性。

关键词: 特征重构, 特征融合, 工期预测, 深度神经网络, 集成堆栈式自编码器

Abstract: There are so many species of components for complicated products in manufacturing enterprise, which has complicated machining and assembly processes. And the quality, process, equipment and other types of data has multi-dimensional, multi-scale, multi-noise characteristics, which makes it difficult to accurately predict the duration. To address the issues mentioned above, one kind of complicated product duration prediction method based on multi-dimensional nonlinear feature reconstruction and fusion is proposed. First, multi-dimensional nonlinear feature reconstruction and fusion using integrated Stacked Auto Encoder is performed and the corresponding model is constructed, and a complicated correlation coupling connection between features is built to provide a feature pool comprising critical duration factors. Second, complicated product duration prediction model based on multi-dimensional nonlinear feature reconstruction is built to accomplish the complicated product duration prediction. Finally, the proposed model is applied in enterprises to predict the circuit breakers and 3D printed products’ duration. The mean root mean square error of this method reached 1.28, and the mean absolute percentage error reached 3.01%, which is higher than the accuracy of the method without feature reconstruction and fusion, support vector machine, neural network, and other methods. The root mean square error decreased by at least 10.87%, the mean absolute percentage error decreased by at least 7.74%, demonstrating the usefulness and applicability of this method.

Key words: feature reconstruction, feature fusion, duration prediction, deep neural network, integrated stacked auto encoder

中图分类号:

TH166

常建涛, 乔子萱, 孔宪光, 杨胜康, 罗才文. 多维非线性特征重构与融合的复杂产品工期预测方法[J]. 机械工程学报, 2023, 59(6): 295-309.

CHANG Jiantao, QIAO Zixuan, KONG Xianguang, YANG Shengkang, LUO Caiwen. Novel Product Duration Prediction Method of Complicated Product Based on Multi-Dimensional Nonlinear Feature Reconstruction and Fusion[J]. Journal of Mechanical Engineering, 2023, 59(6): 295-309.

参考文献

[1] GUO Jianxia,HU Changming,BAO Rong. Predicting the duration of a general contracting industrial project based on the residual modified model[J]. KSCE Journal of Civil Engineering,2019,23(8):3275-3284.
[2] LIU Fang,ZHENG Lanlan,LI Mingyang,et al. Analysis and prediction of the interval duration between the first and second accidents considering the spatiotemporal threshold[J]. Journal of Advanced Transportation,2022,2022:14.
[3] YU Min,FENG Shuwei,YANG Yu. Combined prediction method of marine rainfall with different duration based on mathematical probability and statistics[J]. Journal of Coastal Research,2019,93(sp1):381-387.
[4] THOMAS N,THOMAS A. Regression modelling for prediction of construction cost and duration[J]. Applied Mechanics and Materials,2016,857(857):195-199.
[5] 裴洪,胡昌华,司小胜,等. 基于机器学习的设备剩余寿命预测方法综述[J]. 机械工程学报,2019,55(8):1-13. PEI Hong,HU Changhua,SI Xiaosheng,et al. Review of machine learning based remaining useful life prediction methods for equipment[J]. Journal of Mechanical Engineering,2019,55(8):1-13.
[6] ZHANG Qingchen,YANG L T,CHEN Zhikui,et al. A survey on deep learning for big data[J]. Information Fusion,2018,42:146-157.
[7] WANG Jinjiang,MA Yulin,ZHANG Laibin,et al. Deep learning for smart manufacturing:Methods and applications[J]. Journal of Manufacturing Systems,2018,48:144-156.
[8] 刘玉敏,张帅. 基于多主元特征与支持向量机的动态过程质量异常监控模型[J]. 计算机集成制造系统,2018,24(3):703-710. LIU Yumin,ZHANG Shuai. Dynamic process quality anomaly monitoring model based on multi principal component feature and support vector machine[J]. Computer Integrated Manufacturing Systems,2018,24(3):703-710.
[9] 鲍中新,文成林,马雪. 一种基于数据变化率的预处理及主元分析故障诊断方法[J]. 电子学报,2021,49(11):2234-2240. BAO Zhongxin,WEN Chenglin,MA Xue. Data preprocessing and PCA fault diagnosis method based on rate of change transformation[J]. Acta Electronica Sinica,2021,49(11):2234-2240.
[10] 李锵,武昱忻,关欣,等. 基于SRD感受野自调节的肺炎病灶检测[J]. 天津大学学报(自然科学与工程技术版),2021,54(5):508-516. LI Qiang,WU Yuxin,GUAN Xin,et al. Detection of pneumonia focus based on SRD receptive field self-regulation[J]. Journal of Tianjin University(Science and Technology),2021,54(5):508-516.
[11] HANSEN L K,SALAMON P. Neural network ensembles[J]. IEEE Transactions on Pattern Analysis & Machine Intelligence,2002,12(10):993-1001.
[12] 孟华,孙浩,裴迪,等. 基于聚类及优化集成神经网络的地铁车站空调负荷预测[J]. 同济大学学报(自然科学版),2021,49(11):1582-1589. MENG Hua,SUN Hao,PEI Di,et al. Air-conditioning load prediction of subway station based on clustering and optimization algorithm ensemble neural network[J]. Journal of Tongji University(Natural Science),2021,49(11):1582-1589.
[13] 张永峰,陆志强. 基于集成神经网络的剩余寿命预测[J]. 工程科学学报,2020,42(10):1372-1380. ZHANG Yongfeng,LU Zhiqiang. Remaining useful life prediction based on an integrated neural network[J]. Chinese Journal of Engineering,2020,42(10):1372-1380.
[14] 闫长斌,汪鹤健,周建军,等. 基于Bootstrap-SVR-ANN算法的TBM施工速度预测[J]. 岩土工程学报,2021,43(6):1078-1087. YAN Changbin,WANG Hejian,ZHOU Jianjun,et al. Prediction of TBM advance rate based on Bootstrap method and SVR-ANN algorithm[J]. Chinese Journal of Geotechnical Engineering,2021,43(6):1078-1087.
[15] 张洁,赵新明,张朋,等. 面向火箭总装过程的工期延误预警方法[J]. 上海交通大学学报,2020,54(3):322-330. ZHANG Jie,ZHAO Xinming,ZHANG Peng,et al. Early warning method of construction period delay for rocket assembly process[J]. Journal of Shanghai Jiaotong University,2020,54(3):322-330.
[16] 刘道元,郭宇,黄少华,等. 基于DBN-DNN的离散制造车间订单完工期预测方法[J]. 计算机集成制造系统,2020,26(9):2445-2452. LIU Daoyuan,GUO Yu,HUANG Shaohua,et al. Order completion time prediction method for discrete manufacturing workshop based on DBN-DNN[J]. Computer Integrated Manufacturing Systems,2020,26(9):2445-2452.
[17] 王燕青,袁逸萍,李晓娟,等. 应用RBF神经网络的订单完成时间(OCT)预测[J]. 机械设计与制造,2020(12):228-231. WANG Yanqing,YUAN Yiping,LI Xiaojuan,et al. Order completion time (OCT) prediction using RBF neural network[J]. Machinery Design & Manufacture,2020(12):228-231.
[18] CHANDA S,RAGHUCHARAN M C,REDDY K K,et al. Duration prediction of Chilean strong motion data using machine learning[J]. Journal of South American Earth Sciences,2021,109:103253.
[19] 倪维健,孙宇健,刘彤,等. 基于注意力双向循环神经网络的业务流程剩余时间预测方法[J]. 计算机集成制造系统,2020,26(6):1564-1572. NI Weijian,SUN Yujian,LIU Tong,et al. Business process remaining time prediction using bidirectional recurrent netural networks with attention[J],Computer Integrated Manufacturing Systems,2020,26(6):1564-1572.
[20] 汪俊亮,张洁. 大数据驱动的晶圆工期预测关键参数识别方法[J]. 机械工程学报,2018,54(23):185-191. WANG Junliang,ZHANG Jie. Big data driven key factor identification for cycle-time forecasting of wafer lots in semiconductor wafer fabrication system[J]. Journal of Mechanical Engineering,2018,54(23):185-191.
[21] CHEN T,WANG Yucheng. Enhancing the effectiveness of cycle time estimation in wafer fabrication-efficient methodology and managerial implications[J]. Sustainability,2014,6(8):5107-5128.
[22] WU H,CHEN T. CART-BPN approach for estimating cycle time in wafer fabrication[J]. Journal of Ambient Intelligence and Humanized Computing,2015,6(1):57-67.
[23] CHEN T. Combining statistical analysis and artificial neural network for classifying jobs and estimating the cycle times in wafer fabrication[J]. Neural Computing and Applications,2015,26(1):223-236.
[24] CHEN T. Estimating job cycle time in a wafer fabrication factory:A novel and effective approach based on post-classification[J]. Applied Soft Computing,2016,40:558-568.
[25] 王久健,杨绍普,刘永强,等. 一种基于空间卷积长短时记忆神经网络的轴承剩余寿命预测方法[J]. 机械工程学报,2021,57(21):88-95. WANG Jiujian,YANG Shaopu,LIU Yongqiang,et al. A method of bearing remaining useful life estimation based on convolutional long short-term memory neural network[J]. Journal of Mechanical Engineering,2021,57(21):88-95.
[26] GAB-ALLAH A A,IBRAHIM A H,HAGRAH O A. Predicting the construction duration of building projects using artificial neural networks[J]. International Journal of Applied Management Science,2015,7(2):123-141.
[27] 姜洪权,王岗,高建民,等. 一种适用于高维非线性特征数据的聚类算法及应用[J]. 西安交通大学学报,2017,51(12):49-55+90. JIANG Hongquan,WANG Gang,GAO Jianmin,et al. A clustering algorithm for high-dimensional nonlinear characteristic data and its application[J]. Journal of Xi'an Jiaotong University,2017,51(12):49-55+90.
[28] 袁非牛,章琳,史劲亭,等. 自编码神经网络理论及应用综述[J]. 计算机学报,2019,42(1):203-230. YUAN Feiniu,ZHANG Lin,SHI Jinting,et al. Summary of theory and application of self coding neural network[J]. Chinese Journal of Computers,2019,42(1):203-230.
[29] 孔祥强,刘晓东,尚燕平,等. 基于多元线性回归直膨式太阳能热泵性能预测[J]. 太阳能学报,2022,43(1):443-449. KONG Xiangqiang,LIU Xiaodong,SHANG Yanping,et al. Performance prediction of DX-SAHP based on multiple linear regression algorithm[J]. Acta Energiae Solaris Sinica,2022,43(1):443-449.
[30] 翁剑成,付宇,林鹏飞,等. 基于梯度推进决策树的日维度交通指数预测模型[J]. 交通运输系统工程与信息,2019,19(2):80-85+93. WENG Jiancheng,FU Yu,LIN Pengfei,et al. GBDT method based on prediction model of daily dimension traffic index[J]. Journal of Transportation Systems Engineering and Information Technology,2019,19(2):80-85+93.
[31] ZHANG Yali,LI Shan,ZHANG Honghai. Forecast of air traffic controller demand based on SVR and parameter optimization[J]. Transactions of Nanjing University of Aeronautics and Astronautics,2021,38(6):959-966.
[32] 何存富,蔡燕超,刘秀成,等. 基于磁巴克豪森噪声的S136钢表面硬度定量预测模型对比[J]. 机械工程学报,2019,55(18):15-21. HE Cunfu,CAI Yanchao,LIU Xiucheng,et al. Comparison of quantitative prediction models of surface hardness of S136 steel based on Magnetic Barkhausen noise[J]. Journal of Mechanical Engineering,2019,55(18):15-21.

多维非线性特征重构与融合的复杂产品工期预测方法

Novel Product Duration Prediction Method of Complicated Product Based on Multi-Dimensional Nonlinear Feature Reconstruction and Fusion

PDF

可视化

摘要/Abstract

引用本文

使用本文

参考文献

相关文章 11

编辑推荐

Metrics

本文评价

[1]	狄子钧, 袁东风, 李东阳, 梁道君, 周晓天, 信苗苗, 曹凤, 雷腾飞. 基于多尺度-高效通道注意力网络的刀具故障诊断方法[J]. 机械工程学报, 2024, 60(6): 82-90.
[2]	王润琼, 宋清华, 彭业振, 刘战强. 基于特征自适应融合和集成学习的高性能铣削刀具状态监测[J]. 机械工程学报, 2024, 60(1): 149-158.
[3]	陆晨浩, 陈耀, 何若琪, 范维莹, 冯健. 基于深度神经网络的四折痕锥形折纸结构设计[J]. 机械工程学报, 2023, 59(21): 167-176.
[4]	林京, 焦金阳. 可解释机械智能诊断技术的研究进展与挑战[J]. 机械工程学报, 2023, 59(20): 215-224.
[5]	江星星, 彭德民, 沈长青, 刘颉, 郭剑峰, 朱忠奎. 快速固有成分滤波特征融合的轴承故障诊断方法[J]. 机械工程学报, 2022, 58(22): 129-139.
[6]	张立, 陈江涛, 熊芬芬, 任成坤, 李超. 基于元学习的多可信度深度神经网络代理模型[J]. 机械工程学报, 2022, 58(1): 190-200.
[7]	喻曦, 赵欢, 李祥飞, 丁汉. 基于双深度神经网络的轮廓误差补偿策略研究[J]. 机械工程学报, 2019, 55(3): 130-137.
[8]	侯文擎, 叶鸣, 李巍华. 基于改进堆叠降噪自编码的滚动轴承故障分类[J]. 机械工程学报, 2018, 54(7): 87-96.
[9]	汪俊亮, 张洁. 大数据驱动的晶圆工期预测关键参数识别方法[J]. 机械工程学报, 2018, 54(23): 185-191.
[10]	孙文珺, 邵思羽, 严如强. 基于稀疏自动编码深度神经网络的感应电动机故障诊断[J]. 机械工程学报, 2016, 52(9): 65-71.
[11]	柯映林;刘云峰. 基于三角Bézier曲面的复杂特征模型重建及特征融合技术研究[J]. , 2004, 40(6): 85-90.