A Review on The Application of Machine Learning To Predict The Battery State That Enables A Smart, Low-Cost, Self-Sufficient Drying And Storage System for Agricultural Purposes

Abstract

This study reviews studies on a more viable battery for the energy storage system, the development of battery technology is towards a high capacity, low cost, and long battery lifespan. An accurate prediction of battery state, such as the state of charge, is important to help control the battery charging and discharging and extend the battery lifespan. Several reviews have provided an insightful summary regarding the development of methods to predict battery state for energy storage. This study provides a review that explores the application of machine learning to predict the battery state, including state of charge, state of health, and remaining useful life. Recent studies within this review shown that 64.7% researcher used Neural Network to do prediction with few studies do method combination to further overcome battery dynamic condition in real world application with less computational time and cost to enable integration with IoT technology. Furthermore, the opportunity to implement the energy storage system techniques to enable a smart, low-cost, self-sufficient implementation of the smart solar dryer for agricultural purposes is also elaborated