Estimation of Remaining Useful Life Based on Time Series Analysis

Tekgoz H., Omurca S. I., Koc K. Y.

7th International Conference on Computer Science and Engineering, UBMK 2022, Diyarbakır, Türkiye, 14 - 16 Eylül 2022, ss.273-277 identifier

  • Yayın Türü: Bildiri / Tam Metin Bildiri
  • Doi Numarası: 10.1109/ubmk55850.2022.9919450
  • Basıldığı Şehir: Diyarbakır
  • Basıldığı Ülke: Türkiye
  • Sayfa Sayıları: ss.273-277
  • Anahtar Kelimeler: cnn-lstm, lstm, predictive maintance, remaining useful life, time series
  • Kocaeli Üniversitesi Adresli: Evet

Özet

© 2022 IEEE.With the introduction of Industry 4.0 into our lives and the creation of smart factories, predictive maintenance becomes even more important. The concept of predictive maintenance basically predicts when some physical data from machinery and equipment is taken and processed, with high accuracy, with various artificial intelligence models. In predictive maintenance systems, downtime differs according to the types of malfunctions. An important concern in predictive maintenance is the prediction of remaining useful life (RUL), which is a prediction of the time until failure for the machinery. Developing a prediction of RUL value increases the reliability and safety of the systems while reducing the cost of maintenance. Within the scope of this study, the effectiveness of the Machine Learning (ML) and Deep Learning (DL) models were investigated. The experiments were carried out with the NASA Turbofan Engine Corruption Simulation (C-MAPSS) dataset, which consists of 21 sensor data of the aircraft engine. In the results obtained, it has been observed that traditional ML models outperform LSTM, GRU, CNNLSTM algorithms.