Battery SOC Estimation via Extended Kalman Filter

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

Existing methods for determining the state of charge (SOC) and state of health (SOH) of battery systems in electric and hybrid vehicles are inaccurate due to the inability to directly measure key parameters like battery resistance, equilibrium potential, and diffusion voltage, making it difficult to maintain optimal SOC and detect performance degradation.

Innovation Solution

A battery management system that includes sensors and a controller using an extended Kalman filter and equivalent circuit model to estimate SOC by calculating open-circuit voltage and accounting for solid-state diffusion voltage effects, allowing for accurate prediction of power capabilities.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Measurement precision

If equivalent circuit models are used to determine SOC, then a model for determining SOC is provided, but diffusion in the battery system is not accounted for

Engineering Contradiction:

ImproveSOC determination accuracyVSAvoiddiffusion voltage information

Core Design Contradiction:

Measurement precisionVSLoss of information

Solution Approach 1:

The battery voltage is segmented into multiple components: equilibrium potential, hysteresis voltage, Ohmic voltage drop, and diffusion voltage. Each component is modeled separately with appropriate circuit elements (voltage sources, resistors, capacitors), allowing the diffusion component to be captured independently while maintaining overall SOC determination accuracy.

Inventive Principle:

Principle #1Segmentation

Solution Approach 2:

A diffusion voltage source is introduced as an intermediary element in the equivalent circuit model. This diffusion voltage acts as a mediator that captures the solid-state diffusion effects in the battery, which cannot be directly measured but significantly influences the terminal voltage and SOC determination.

Inventive Principle:

Principle #24Intermediary (Mediator)

2Ease of operation

If battery parameters such as resistance, equilibrium potential, and diffusion voltage are not directly measurable, then measurement simplicity is maintained, but accurate determination of SOC and SOH becomes difficult

Engineering Contradiction:

Improvemeasurement simplicityVSAvoidSOC and SOH determination accuracy

Core Design Contradiction:

Ease of operationVSMeasurement precision

Solution Approach 1:

Direct electrical measurement methods are replaced with an electrochemical model-based estimation approach. The system uses measurable terminal voltage, current, and temperature to drive an equivalent circuit model that calculates unmeasurable parameters (equilibrium potential, diffusion voltage, internal resistance) through mathematical relationships rather than direct sensing.

Inventive Principle:

Principle #28Mechanics substitution (Replace mechanical system)

Solution Approach 2:

The model dynamically adjusts parameters such as equilibrium potential, hysteresis voltage, and diffusion resistance based on changing battery conditions (SOC, temperature, current). This allows the system to maintain measurement simplicity while achieving accurate SOC and SOH determination through adaptive parameter estimation.

Inventive Principle:

Principle #35Parameter changes

Data Source

PatentUS9575128B2Battery state-of-charge estimation for hybrid and electric vehicles using extended kalman filter techniques

Publication Date: 2017.02.21 GM GLOBAL TECHNOLOGY OPERATIONS LLC

AI summary

Adaptive estimation techniques to create a battery state estimator to estimate power capabilities of the battery pack in a vehicle. The estimator adaptively updates circuit model parameters used to calculate the voltage states of the ECM of a battery pack. The adaptive estimation techniques may also be used to calculate a solid-state diffusion voltage effects within the battery pack. The adaptive estimator is used to increase robustness of the calculation to sensor noise, modeling error, and battery pack degradation.