The work in this thesis is focused on the development and validation of an automotive battery monitoring system that estimates the health of a lead-acid battery during engine cranking and provides a low state of health (SOH) warning of potential battery failure. A reliable SOH estimation should assist users in preventing a sudden battery failure and planning for battery replacement in a timely manner.

Most commercial battery health estimation systems use the impedance of a battery to estimate the SOH with battery voltage and current; however, using a current sensor increases the installation cost of a system due to parts and labor. The battery SOH estimation method with the battery terminal voltage during engine cranking was previously proposed. The proposed SOH estimation system intends to improve existing methods. The proposed method requires battery voltages and temperature for a reliable SOH estimation. Without the need for a costly current sensor, the proposed SOH monitoring system is cost-effective and useful for automotive applications.

Measurement results presented in this thesis show that the proposed SOH monitoring system is more effective in evaluating the health of a lead-acid battery than existing methods. A low power microcontroller equipped prototype implements the proposed SOH algorithm on a high performance ARM Cortex-M4F based MCU, TM4C123GH6PM. The power dissipation of the final prototype is approximately 144 mW during an active state and 36 mW during a sleep state. With the reliability of the proposed method and low power dissipation of the prototype, the proposed system is suitable for an on-board battery monitoring as there is no on-board warning that estimates the health of a battery in modern cars.