Online Battery Resistance Estimation Algorithm With Down-Sampling and Preprocessing Filter in Embedded DC-DC Power Converters

Abstract

The safety and reliability of lithium-ion batteries have become increasingly critical due to the rising risk of fire incidents in electric vehicles and energy storage systems. Among various diagnostic methods, AC internal resistance (AC-IR) has emerged as a promising metric for fault detection. While electrochemical impedance spectroscopy (EIS) offers high accuracy across a wide frequency range, it is unsuitable for real-time applications due to bulky hardware and limited cost-effectiveness. To overcome these limitations, this paper proposes an online AC-IR measurement algorithm integrated into a DC–DC converter operating under constrained digital signal processing resources. The proposed method employs a frequency-domain analysis framework with tunable down-sampling and FIR-based anti-aliasing filtering, enabling low-frequency impedance estimation with enhanced resolution. This approach addresses the inherent limitations in frequency resolution imposed by fixed sampling rates and restricted FFT size in embedded systems. The algorithm is experimentally validated using a 50-W bidirectional DC–DC converter and a lithium-ion battery, achieving less than 1.34% error compared to a commercial EIS device over the frequency range of 6.5 Hz to 990.5 Hz.