Search

Researchers are focused toward the development of electric vehicles and hybrid electric vehicles for the preservation of the ecosystem. Energy storage, conversely, became essential with these changes in vehicular technology. This opens new research fronts and introduces complex problems that are to be investigated regarding the primary storage element. Battery modeling offers improved efficiency, safety, and reliability. Lithium-ion batteries are a popular source of energy in many applications, especially for electric vehicles because of their high performance and high energy density. However, there exist some inherent issues like complex electrochemical reactions that make accurate monitoring a challenging task. This article performs a comparative study of the existing battery state of health (SOH) monitoring and capacity estimation techniques using the NASA dataset for lithium-ion batteries. Charging cycle datasets for voltage, current, and temperature are used for feature identification and extraction. After the modeling of extracted features, the SOH and the capacity of four batteries are estimated using three data-driven techniques. Including convolutional neural network, feed-forward neural network, and long short-term memory (LSTM), and their results are compared. The comparative analysis highlights an outstanding performance and high accuracy of the LSTM-based machine learning technique because of the inherited long-term memory of the LSTM. The study, therefore, recommends the use of LSTM to researchers for battery health monitoring and capacity estimation with the highest possible accuracy. [ABSTRACT FROM AUTHOR]

Batteries are combinations of electrochemical cells that generate electricity to power electrical devices. Batteries are continuously converting chemical energy to electrical energy, and require appropriate maintenance to provide maximum efficiency. Management systems having specialized monitoring features; such as charge controlling mechanisms and temperature regulation are used to prevent health, safety, and property hazards that complement the use of batteries. These systems utilize measures of merit to regulate battery performances. Figures such as the state-of-health (SOH) and state-of-charge (SOC) are used to estimate the performance and state of the battery. In this paper, we propose an intelligent method to investigate the aforementioned parameters using a data-driven approach. We use a machine learning algorithm that extracts significant features from the discharge curves to estimate these parameters. Extensive simulations have been carried out to evaluate the performance of the proposed method under different currents and temperatures. [ABSTRACT FROM AUTHOR]