English (United Kingdom)

https://curated-unify.zendy.io/wp-json/zendy-region/v1/featured_content/oa?rat=en

https://curated-unify.zendy.io/wp-json/zendy-region/v1/highlighted_journal/

Global: Zendy Plus annual

Presents the access of premium content as premium feature

Premium Content

Presents the keyphrase highlighting as premium feature

Keyphrase Highlighting

Presents the summarisation as premium feature

Summarisation

Insights

Presents the pdf analysis as premium feature

PDF Analysis

Presents the zaia usage as premium feature

ZAIA

Global: Zendy Plus monthly

Global: Zendy Tools annual

Global: Zendy Tools monthly

Global: Zendy Free Plan

As the use of Lithium-ion batteries continues to grow, it becomesincreasingly important to be able to predict their remaining useful life. Thiswork aims to compare the relative performance of different machine learningalgorithms, both traditional machine learning and deep learning, in order todetermine the best-performing algorithms for battery cycle life predictionbased on minimal data. We investigated 14 different machine learning modelsthat were fed handcrafted features based on statistical data and split into 3feature groups for testing. For deep learning models, we tested a variety ofneural network models including different configurations of standard RecurrentNeural Networks, Gated Recurrent Units, and Long Short Term Memory with andwithout attention mechanism. Deep learning models were fed multivariate timeseries signals based on the raw data for each battery across the first 100cycles. Our experiments revealed that the machine learning algorithms onhandcrafted features performed particularly well, resulting in 10-20% averagemean absolute percentage error. The best-performing algorithm was the RandomForest Regressor, which gave a minimum 9.8% mean absolute percentage error.Traditional machine learning models excelled due to their capability tocomprehend general data set trends. In comparison, deep learning models wereobserved to perform particularly poorly on raw, limited data. Algorithms likeGRU and RNNs that focused on capturing medium-range data dependencies were lessadept at recognizing the gradual, slow trends critical for this task. Ourinvestigation reveals that implementing machine learning models withhand-crafted features proves to be more effective than advanced deep learningmodels for predicting the remaining useful Lithium-ion battery life withlimited data availability.

Forecasting Lithium-Ion Battery Longevity with Limited Data Availability: Benchmarking Different Machine Learning Algorithms