Cellular Battery Energy Estimation Using SOH and SOC States
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing methods for estimating available energy in cellular batteries, such as those used in vehicles, are either too pessimistic, inaccurate, especially as cells approach end-of-life, or require extensive databases and are not responsive in real-time.
Innovation Solution
A method that estimates the total energy available in a cellular battery by considering the current resistance states of health, capacity states of health, and states of charge of each cell, along with a selected discharge current and reference temperature, providing a more accurate and reliable estimation.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If the first solution (using maps from new cell characterizations and considering the most limiting cell) is used to estimate available energy, then the estimation is conservative and safe, but the estimation accuracy deteriorates and becomes too pessimistic
Solution Approach 1:
The patent changes the parameters used for estimation from static maps based on new cell characterizations to dynamic parameters including resistance state of health (SOHR), capacity state of health (SOHC), and state of charge (SOC) that evolve with cell aging. This allows the estimation to adapt to actual cell conditions rather than relying on conservative initial characterizations.
Solution Approach 2:
The patent implements feedback by continuously monitoring the actual state of each cell (resistance, capacity, charge) and using this information to update the available energy estimation. This closed-loop approach allows the system to learn from actual cell behavior and improve estimation accuracy over time while maintaining safety.
2Device complexity
If the second solution (considering only current estimated capacities and SOHCs) is used to estimate available energy, then the estimation is simpler, but the accuracy deteriorates especially as cells approach end-of-life
Solution Approach 1:
The patent adds the resistance state of health (SOHR) parameter to the estimation model, working in conjunction with capacity state of health (SOHC) and state of charge (SOC). This additional parameter becomes increasingly important as cells age, allowing the system to accurately capture the growing impact of resistance on available energy without significantly increasing overall system complexity.
3Ease of manufacture
If the third solution (using off-line calibrated durability models) is used to estimate available energy, then the model can be pre-calibrated, but it cannot account for sudden loss of energy from a cell at end-of-life due to usage outside experimental limits
Solution Approach 1:
The patent enables the estimation system to self-adapt to actual cell behavior through continuous monitoring of resistance, capacity, and charge states. Rather than relying solely on pre-calibrated models, the system learns from real-world cell performance and adjusts estimates accordingly, allowing it to handle unexpected usage conditions and sudden changes at end-of-life.
4Measurement precision
If the fourth solution (using black box artificial intelligence learning methods with off-line databases) is used to estimate available energy, then the system can learn from large datasets, but it monopolizes wireless network bandwidth and does not offer true real-time responsiveness
Solution Approach 1:
The patent extracts only the essential parameters (resistance, capacity, charge states) needed for accurate estimation and processes them locally in real-time. This eliminates the need to transmit large datasets over wireless networks while maintaining estimation accuracy, as the system works with compact, critical information available at the cell level.
Data Source
AI summary
A method is provided for estimating at least one item of information in relation to a cellular battery of a system comprising N cells able to store electrical energy, where N>1, and each having a current state of charge, a current resistance state of health and a current capacity state of health. This method comprises a step (10-40) of estimating first information representative of a total energy available in the cellular battery as a function of the current resistance states of health, current capacity states of health and current states of charge, and of a time interval for which the cellular battery is allowed to discharge with a chosen discharge current and at a reference temperature.