The abnormality detection of lithium-ion battery pack is crucial to ensure the safety of electric vehicles (EVs). However, the dynamic and complex operating conditions of EVs making it challenging for algorithms. ••The proposed method is based on unsupervised learning, avoiding the. EVs Electric vehiclesANN Artificial neural networkAE. Transportation electrification has been considered as a promising solution to environmental problems and has experienced rapid growth in recent years, leading to a glob. In practice, data acquisition during a thermal runaway is almost impossible, meaning that only few samples can be collected for algorithm design. Consequently, tr. 3.1. Data acquisitionTo incorporate real-world EV charging profiles, in this work, datasets from the National Bigdata Alliance Open Laboratory of NEVs (NBAOL.
Why is a single encoder a problem?
A model with a single encoder (temporal-wise) may struggle with the complexity and the multi-dimensional nature of the data in large-scale EV applications. Such a model tends to be insufficient in isolating channel-specific information and temporal dependencies simultaneously, leading to less accurate fault predictions.
Can aging data be used to identify battery lifetime abnormalities?
Here, we proposed to solve this issue by “creating” more abnormal data. The aim of this work was to use the data collected from the first cycle of the aging test to identify the lifetime abnormality. However, as shown in Figure 1 and many other battery aging datasets, [22, 35, 36] the battery's behaviors in the first few cycles were highly similar.
Can a multi-fault coding method detect abnormal changes in Battery Data?
The use of improved Lyapunov exponent method to detect abnormal changes in battery data has a stable effect on different levels of abnormal conditions. The proposed multi-fault coding method can use unified coding to characterize different fault scenarios, and has good robustness.
Can a battery detection method detect abnormal batteries?
Verified with the largest known dataset with 215 commercial lithium-ion batteries, the method can identify all abnormal batteries, with a false alarm rate of only 3.8%. It is also found that any capacity and resistance-based approach can easily fail to screen out a large proportion of the abnormal batteries, which should be given enough attention.
How to develop a reliable and efficient early warning model for battery failures?
Therefore, developing a reliable and efficient early warning model for battery failures is not just about selecting an optimal embedding time. It also necessitates understanding the nature and severity of potential faults and the anticipated prediction tasks. This knowledge is as crucial as the selection of embedding time.
Is there a lifetime abnormality detection method for lithium-ion batteries?
This work proposes a lifetime abnormality detection method for batteries based on few-shot learning and using only the first-cycle aging data. Verified with the largest known dataset with 215 commercial lithium-ion batteries, the method can identify all abnormal batteries, with a false alarm rate of only 3.8%.