Implementation of iot based battery state of health and charge estimation system for electric and hybrid vehicles

In this master's thesis, the focus was on IoT-based lifetime prediction for lithium-ion batteries used in electric and hybrid vehicles, whose usage has been rapidly increasing. Knowing the battery lifespan is crucial for solidifying the position of electric and hybrid vehicles in the market. In this study, a cross-sectional lithium-ion battery model's characteristic data was determined, and based on these characteristics, lifetime estimation of equivalent battery models was performed. During the characteristic determination process, current, voltage and surface temperature data of the battery pack were collected for each charge-discharge cycle for 500 cycles. The collected data revealed the expected decrease in charge-discharge profiles. The maintainable capacity of the battery model dropped below 70% after the 450th cycle, indicating the end of usable lifetime. A program was developed using the Arduino IoT cloud system based on the charge-discharge profile and maintainable capacity change data obtained during the battery characteristic inference process. Charge-discharge cycles of the lithium-ion battery packs to be used for lifetime prediction and the states of the cells were monitored and the data was collected through an IoT tracking system. Based on the collected data, estimations of battery health and charge status were made. Current, voltage, battery surface temperature, ambient temperature, and humidity values were obtained through sensors from the equivalent battery pack used for lifetime estimation and transferred to the interface in an IoT-based manner. By processing the instantaneous health status, charge status, and surface temperature data of the battery through the program created based on the acquired sensor data, the information was transferred to the interface in a way that the end-user could visualize. Consequently, a successful module capable of IoT communication on the lithium-ion battery pack was achieved. With the developed module, lifetime prediction can be made even if the lithium-ion battery pack is located remotely, as long as it is connected to the internet.