Data Preparation in Machine Learning for Condition-based Maintenance

Abstract

Using Machine Learning (ML) prediction to achieve a successful, cost-effective, Condition-Based Maintenance (CBM) strategy has become very attractive in the context of Industry 4.0. In other fields, it is well known that in order to benefit from the prediction capability of ML algorithms, the data preparation phase must be well conducted. Thus, the objective of this paper is to investigate the effect of data preparation on the ML prediction accuracy of Gas Turbines (GTs) performance decay. First a data cleaning technique for robust Linear Regression imputation is proposed based on the Mixed Integer Linear Programming. Then, experiments are conducted to compare the effect of commonly used data cleaning, normalization and reduction techniques on the ML prediction accuracy. Results revealed that the best prediction accuracy of GTs decay, found with the k-Nearest Neighbors ML algorithm, considerately deteriorate when changing the data preparation steps and/or techniques. This study has shown that, for effective CBM application in industry, there is a need to develop a systematic methodology for design and selection of adequate data preparation steps and techniques with the proposed ML algorithms.

Penilaian Kinerja Akurasi Metode Klasifikasi dalam Dataset Penerimaan Mahasiswa Baru Universitas XYZ

“…Transformasi yang dilakukan dengan mengubah data mentah menjadi format yang dipahami untuk dianalisis sehingga memberikan informasi. Proses data preparation dilakukan dengan pembersihan, transformasi dan konsolidasi data [16]. Didalam penelitian ini dilakukan data preparation karena beberapa faktor, antara lain: (1) Data perlu diformat sesuai dengan kebutuhan perangkat lunak Jupyter Notebook; (2) Data mentah cenderung 'kotor', seperti data tidak lengkap, terdapat nilai yang outlier, inkonsistensi nilai antar atribut terkait, tertukar antara kolom dan baris; (3) Dalam satu kolom yang sama terdapat banyak variabel.…”

Digital Twin Science of Metal Powder Bed Fusion Additive Manufacturing: A Selective Review of Simulations for Integrated Computational Materials Engineering and Science

“…CPSs and DTs are constructed on various scales (1,(3)(4)(5)(6)(7)(8)(9)(10)(11) . On the macroscopic scale, they are applied in areas such as weather forecasting (6) , optimization of logistics networks for supply chains (6) , and failure prognostic for parts replacement of machines (7)(8)(9)(10)(11) .…”

“…Particularly, we aim to establish the DT for the powder bed fusion (PBF) process (Fig. 1) (12) , which is a mainstream technique for metal AM (7,13) . In the DT of PBF, each process step is simulated by numerically solving various equations based on physical and chemical laws.…”