Dealing with missing data as it pertains of e-maintenance

Abstract

Purpose Centrifugal compressors are integral components in oil industry, thus effective maintenance is required. Condition-based maintenance and prognostics and health management (CBM/PHM) have been gaining popularity. CBM/PHM can also be performed remotely leading to e-maintenance. Its success depends on the quality of the data used for analysis and decision making. A major issue associated with it is the missing data. Their presence may compromise the information within a set, causing bias or misleading results. Addressing this matter is crucial. The purpose of this paper is to review and compare the most widely used imputation techniques in a case study using condition monitoring measurements from an operational industrial centrifugal compressor. Design/methodology/approach Brief overview and comparison of most widely used imputation techniques using a complete set with artificial missing values. They were tested regarding the effects of the amount, the location within the set and the variable containing the missing values. Findings Univariate and multivariate imputation techniques were compared, with the latter offering the smallest error levels. They seemed unaffected by the amount or location of the missing data although they were affected by the variable containing them. Research limitations/implications During the analysis, it was assumed that at any time only one variable contained missing data. Further research is still required to address this point. Originality/value This study can serve as a guide for selecting the appropriate imputation method for missing values in centrifugal compressor condition monitoring data.

Data Preparation in Machine Learning for Condition-based Maintenance

“…Although extensive work has been carried out under ML for CBM, yet little attention has been paid to the data preparation phase. According to (Bennane and Yacout, 2010;Loukopoulos et al, 2017;Diez-Olivan et al, 2019), the relevance of the data preparation phase has been widely recognized in the literature but still few research efforts have been carried out to address this issue in CBM context.…”

“…Data were cleaned using the Logical Analysis of Data (LAD) model; then a supervised learning algorithm was used to predict the health state of an oil transformer system. Loukopoulos et al (2017) have also presented different imputation techniques to handle the missing data, for the CBM application on centrifugal compressors. Among these techniques, autoregressive model, k-NN imputation, Self Organizing Map (SOM) and Bayesian Principal Components Analysis (BPCA) were used to fill the missing data.…”

“…The neighborhood size k selection plays an important role in resulting in a good performance of k-NN. However, as pointed out by (Loukopoulos et al, 2017), no global rule is set for determining this optimal k. In the present paper, preliminary experiments as the one done by (Thanh Noi and Kappas, 2018), with different values of k between 1 and 20, are conducted. Then the k value which gave the lowest value of MAPE is selected Data normalization consists of scaling the features so they can fall within a smaller range, improving the efficiency and the accuracy of ML algorithms, (Han et al, 2011).…”

“…However, as pointed out in (Bennane and Yacout, 2010;Loukopoulos et al, 2017;Diez-Olivan et al, 2019), the majority of these works have centered in comparing performances of different ML algorithms in degradation prediction; however, they did not give enough insight on the data preparation phase. Only a few works, such as (Bukhsh et al, 2020) have exhibited the data preparation technique used before applying predictive ML models for bridges intelligent maintenance.…”

An improved k-nearest neighbours method for traffic time series imputation

“…It has been shown that different PCA based methods perform similar with regards to accuracy [14]. Another work introduces an ad hoc category which includes mean, median and last observation carried forward (LOF) [13]. Among those methods, kNN is accurate and efficient [8], [11].…”