SOC-estimation-of-lithium-ion-batteries

Various machine learning algorithms have been used to estimate state-of-charge (SOC) of calendar-aged lithium-ion pouch cells. Calendar life data was generated by applying galvanostatic charge/discharge cycle loads at different storage temperature (35°C and 60°C) and conditions (fully-discharged and fully-charged). The data was obtained at various C-rates for duration of 10 months at one-month intervals. The wininng model, Random Forest (RF), has achieved a R2 score of 99.98% and a mean absolute error (MAE) of 0.14% over test data, confirming the ability of RF to capture input-output dependency. The model will be employed to estimate the SOC of calendar-aged lithium-ion batteries which is essential for the reliable operation of electic vehicles (EVs).

Prediction-of-lithium-ion-batteries-SOH

Artificial Neural Network (ANN) has been used to estimate state-of-health (SOH) of lithium-ion batteriess. The batteries were stored at different storage temperature (35°C and 60°C) and conditions (fully-discharged and fully-charged) and their capacity was recorded for the duration of 10 months at one-month intervals.

Credit-Risk-Management

A binary classification model is developed to predict the probability of paying back a loan by an applicant. Customer previous loan journey was used to extract useful features using different strategies such as manual and automated feature engineering, and deep learning (CNN, RNN). Various machine learning algorithms such as Boosted algorithms (XGBoost, LightGBM, CatBoost) and Deep Neural Network are used to develop a binary classifier and their performances were compared.