Eddy Current Loss Based Non-Intrusive State-of-Charge Estimation System for Lithium Based Batteries

Lithium-ion batteries are regarded as the batteries that could potentially change the world. From consumer electronics and electric vehicles to energy storage systems and aerospace applications, Li-ion based batteries have become the norm. Although these batteries show a lot of promise to rid the world of several challenges in the future, there are still a few shortcomings of this battery that need to be addressed. It is also important to note the recent accidents caused due to the explosion of these lithium-ion batteries, to name a few: Samsung Galaxy Note 7 explosion, e-cigarettes battery explosion, overheating of lithium batteries present in Apple MacBook Pro laptop. This calls for a more reliable and accurate Battery Management System (BMS). One of the major shortcomings in today’s available battery management systems is the accuracy of the measurement of charge present in lithium-ion batteries, termed as State-of-Charge (SoC) and State-of-Energy (SoE) of the battery.

To address this problem, a highly sensitive and a high-resolution system is developed to estimate the State-of-Charge based on the changes in impedance of a sensor coil which is caused due to the effect of Eddy Current Power Loss in the battery. The redox reaction taking place inside a battery suggest that lithium ions are exchanged back-and-forth between anode and cathode during an event of charging and/or discharging of the battery. This gives rise to change in electrical resistivity of the battery electrode materials. A sensor coil which is excited with an AC magnetic field induces Eddy currents on the internal components of the battery. Based on the change in resistivity of the electrode materials, eddy current and hence the power loss due to Eddy currents change. This in turn changes the complex impedance of the sensor coil, which is mapped to estimate the SoC of the battery. The results confirm the superiority of the proposed technique in terms of sensitivity, resolution, computational complexity and cost of the measured SoC in comparison with other existing methods of estimating SoC. This can be a potential method to estimate SoE of the battery as well.