| Title |
Comparative Study of Data Mining-Based State Determination Method for Smart EOCR |
| Authors |
이경민(Kyung-Min Lee) ; 박철원(Chul-Won Park) |
| DOI |
https://doi.org/10.5370/KIEE.2025.74.3.411 |
| Keywords |
Data mining; LSTM; MCC; Motor; Smart EOCR; State determination method; SVM |
| Abstract |
When a motor fault happens, the life and productivity of motor are reduced, and enormous recovery time and cost occur, so a protection plan must be established. Recently, the concept of predictive maintenance based on the prognosis according to the facility status has been attracting attention. In this paper, as part of the project to develop AI-based predictive maintenance technology for MCC's smart EOCR, state determination method using data mining is proposed. First, the data is collected from an electric motor system using an actual pump system, and then the training data and test data sets that can determine various states are configured.
Among data mining technique, the state determination method is designed using the SVM model and the LSTM model, and implemented using the Python language. Finally, the performance of the two proposed data mining models are compared through evaluation metrics such as Precision, Recall, and F1_Score, etc. |