Fuzzy Clustering and Nonlinear Regression Imputation for Incomplete Data of Tunnel Boring Machine

被引：0
|
作者：
Wang Y. [1 ]
Pang Y. [1 ]
Zhang L. [2 ]
Shi Y. [1 ]
Sun W. [1 ]
Song X. [1 ]
机构：
[1] School of Mechanical Engineering, Dalian University of Technology, Dalian
[2] School of Control Science and Engineering, Dalian University of Technology, Dalian
来源：
Jixie Gongcheng Xuebao/Journal of Mechanical Engineering | 2023年 / 59卷 / 12期
关键词：
data modeling; fuzzy clustering; in-situ data; incomplete data imputation; tunnel boring machine;
D O I：
10.3901/JME.2023.12.028
中图分类号：
学科分类号：
摘要：
The integrity and reliability of the in-situ data are the necessary prerequisites and key factors to promote the development of various industries in the era of industrial big data. As a key equipment of tunnel engineering, tunnel boring machine (TBM) has the characteristics of complex system structure and high relationship between the attributes, and it is a national tool for urban underground construction in national strategies such as “One Belt and One Road”. However, in the process of TBM operation, missing values frequently occurs in the acquisition of measured data of TBM due to various reasons, such as environmental interference, acquisition interruption, equipment failure, which seriously reduces the quality and reliability of data and affects the progress of the project. According to the characteristics of measured data of TBM, a high-precision missing value imputation algorithm based on fuzzy clustering and nonlinear regression is proposed. Firstly, the measured data under different working conditions are divided into several linear subsets by fuzzy clustering method. Then, a linear regression model is established for each subset, and an alternating learning strategy is used to solve the model parameters, which effectively mines the correlation between attributes. Experimental results show that the proposed method performs well both in clustering incomplete data and imputation missing data. The proposed data imputation algorithm can effectively solve the problem of data division and recovery, and provides a reliable foundation for the actual big data mining of shield machine. © 2023 Editorial Office of Chinese Journal of Mechanical Engineering. All rights reserved.
引用
收藏
页码：28 / 37
页数：9
相关论文
共 23 条
  • [1] LAKSHMINARAYAN K, HARP S A, SAMAD T., Imputation of missing data in industrial databases[J], Applied Intelligence, 11, pp. 259-275, (1999)
  • [2] CHEN Shuai, ZHAO Ming, GUO Dong, Et al., Missing data imputation using SVD-KDR algorithm in industrial monitoring data[J], Journal of Mechanical Engineering, 57, 2, pp. 30-38, (2021)
  • [3] ZHANG S., Nearest neighbor selection for iteratively KNN imputation[J], Journal of Systems and Software, 85, 11, pp. 2541-2552, (2012)
  • [4] TIAN Y, ZHANG K, LI J Y, Et al., LSTM-based traffic flow prediction with missing data[J], Neurocomputing, 318, pp. 297-305, (2018)
  • [5] MEEYAI S., Logistic regression with missing data：A comparison of handing methods，and effects of percent missing values[J], Journal of Traffic and Logistics Engineering, 4, 2, pp. 128-134, (2016)
  • [6] YANG K, LI J, WANG C., Missing values estimation in microarray data with partial least squares regression[C], Computational Science-ICCS, 3992, pp. 662-669, (2006)
  • [7] DEMPSTER A P, LAIRD N M, RUBIN D B., Maximum likelihood from incomplete data via the EM algorithm[J], Journal of Royal Statistical Society. Series B, 39, 1, pp. 1-22, (1977)
  • [8] GAJAWADA S, TOSHNIWAL D., Missing value imputation method based on clustering and nearest neighbors[J], International Journal of Future Computer and Communication, 1, 2, pp. 206-208, (2012)
  • [9] HAO Shengxuan, SONG Hong, ZHOU Xiaofeng, Novel approach for missing data imitation based on biclustering[J], Application Research of Computers, 32, 3, pp. 674-678, (2015)
  • [10] MA Yongjun, WANG Rui, LI Yajun, Et al., Data filling using cluster analysis and outlier detection[J], Computer Engineering and Design, 48, 15, pp. 90-95, (2019)
123