Comparison of data Preprocessing Methods for Support Vector Regression based Remaining Useful Life Prediction
Authors:
- 22 July 2022
- conference paper
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- p. 1-5
Abstract
The prediction of the remaining useful life (RUL)of Lithium-ion batteries (LIBs) is more challenging due to the strong interference of internal characteristics and external factors. Therefore, the two different preprocessing methods are proposed in this paper to alleviate this problem, and the reliability and applicability of different preprocessing techniques for RUL prediction need to be concerned and studied. The aim of this study is to compare the capabilities of two preprocessing technology approaches and to evaluate the performance of lithium-ion battery RUL and capacity prediction. Firstly, the historical degraded data are decomposed using empirical modal decomposition and wavelet transform, respectively. Secondly, the real data are reconstructed by correlation coefficients and alignment entropy association coefficients (AC). Finally, support vector regression (SVR) model is built and the RUL of LIBs is predicted. This experimental data is obtained from the actual test Lithium-ion batteries 18650 provided by NASA PCoE, and additionally, some quantitative metrics are used to analyze and evaluate the battery RUL estimation results.Keywords
This publication has 16 references indexed in Scilit:
- Prediction of the Remaining Useful Life of Lithium-Ion Batteries Based on Empirical Mode Decomposition and Deep Neural NetworksIEEE Access, 2020
- Remaining Useful Life Prediction of Lithium-Ion Batteries Using Support Vector Regression Optimized by Artificial Bee ColonyIEEE Transactions on Vehicular Technology, 2019
- Analysis and Comparison of Permutation Entropy, Approximate Entropy and Sample Entropy2018
- Online State of Health Estimation for Lithium-Ion Batteries Based on Support Vector MachineApplied Sciences, 2018
- A novel Gaussian process regression model for state-of-health estimation of lithium-ion battery using charging curveJournal of Power Sources, 2018
- Long Short-Term Memory Recurrent Neural Network for Remaining Useful Life Prediction of Lithium-Ion BatteriesIEEE Transactions on Vehicular Technology, 2018
- A new hybrid method for the prediction of the remaining useful life of a lithium-ion batteryApplied Energy, 2017
- Capacity Prognostics of Lithium-Ion Batteries using EMD Denoising and Multiple Kernel RVMIEEE Access, 2017
- Intelligent Prognostics for Battery Health Monitoring Using the Mean Entropy and Relevance Vector MachineIEEE Transactions on Systems, Man, and Cybernetics: Systems, 2014
- The empirical mode decomposition and the Hilbert spectrum for nonlinear and non-stationary time series analysisProceedings of the Royal Society A: Mathematical, Physical and Engineering Sciences, 1998