Wireless Battery Management System with Data-Driven Modelling

In today's world, because of the universal use of batteries for different applications like electric vehicles, tools/appliances, and industrial equipment, battery monitoring and management have become a crucial element of research. For a large battery pack consisting of hundreds or thousands of battery cells, the battery parameters like Current, Voltage, and Temperature must be monitored carefully. Due to the diverging characteristics of the cell, every cell has to be monitored accurately and periodically. The parameters to be considered essentially are: 1) sensing the health condition of the battery pack timely to capture the turning points where the battery condition changes abruptly, 2) the frequency of monitoring the battery parameters (Voltage, Current, and Temperature) based on the load condition. A Battery Management System (BMS) monitors the state of the battery through Voltage, Temperature, Current, SOC, DOD, and SOH monitoring, and it uses serial, DC BUS, or CAN bus to communicate with the hardware at the cell level. This type of communication requires a large amount of wiring and circuit components, which increases system complexity and maintenance cost. In this project, a Wireless Battery Management System (WBMS) has been designed and implemented considering battery management principles. The communication between the battery pack (Battery Sensing unit) and the monitoring device (Battery Control unit) is based on Bluetooth and ZigBee. This approach aids in reducing the complexity and maintenance cost involved in the existing Battery Management System and helps in modularizing a battery pack. A subsequent analysis is performed to determine the type of communication best suited for a Wireless BMS and provide a conclusion. The implemented system is capable of monitoring the battery parameters without significant loss of information and latency. The necessary cell-level information from the battery pack is available at the Controller device for logging, monitoring, and analysis.