Version 1
: Received: 17 November 2023 / Approved: 20 November 2023 / Online: 20 November 2023 (14:02:30 CET)
Version 2
: Received: 2 January 2024 / Approved: 3 January 2024 / Online: 3 January 2024 (09:43:49 CET)
Al-Meer, M.H. A Deep Learning Method for the Health State Prediction of Lithium-Ion Batteries Based on LUT-Memory and Quantization. World Electr. Veh. J.2024, 15, 38.
Al-Meer, M.H. A Deep Learning Method for the Health State Prediction of Lithium-Ion Batteries Based on LUT-Memory and Quantization. World Electr. Veh. J. 2024, 15, 38.
Al-Meer, M.H. A Deep Learning Method for the Health State Prediction of Lithium-Ion Batteries Based on LUT-Memory and Quantization. World Electr. Veh. J.2024, 15, 38.
Al-Meer, M.H. A Deep Learning Method for the Health State Prediction of Lithium-Ion Batteries Based on LUT-Memory and Quantization. World Electr. Veh. J. 2024, 15, 38.
Abstract
Accurately estimating the State-of-Health (SOH) of lithium-ion batteries is of great importance within the field of battery management systems. Some new technique is now being developed to ensure the secure operation of lithium-ion batteries. The model being suggested in this study employs a deep learning framework that integrates discretized input data. This study utilized two distinct neural network architectures, including a standard fully connected neural network (FCNN) and a bidirectional long short-term memory (LSTM) architecture. The process of converting digitized feature values into binary bits enables the storing of inferred values within a Lookup Table (LUT-Memory). The efficiency and speed of the inference process are expected to improve when inferring a pre-trained deep neural network architecture directly. The primary objective of this study is to accurately build a lookup table that efficiently correlates the state of health (SOH) of lithium-ion batteries, while ensuring a tolerable degree of imprecision. The findings derived from the lithium-ion battery dataset provided by NASA PCoE provide evidence to support the claim that the suggested methodology exhibits similar performance to complete models that require inference during testing. The error assessment metrics, namely Root Mean Square Error (RMSE), Mean Absolute Error (MAE), and Mean Absolute Percentage Error (MAPE), have been employed for quantitative analysis of the accuracy of status of health (SOH) prediction. The aforementioned indicators exhibit a notable level of precision in forecasting the State of Health (SOH).
Keywords
Lithium-Ion Batteries; SoH; SoC; RUL; Batteries; Deep Learning; LUT
Subject
Computer Science and Mathematics, Artificial Intelligence and Machine Learning
Copyright:
This is an open access article distributed under the Creative Commons Attribution License which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.
