Closed-Loop Battery Charging With Precomputed Current Constraints
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current battery charging protocols, such as CC-CV, are ad-hoc and conservatively tuned, leading to suboptimal charging times and lifetimes, and advanced methods like MPC-based approaches are computationally intensive, making them unsuitable for low-cost devices.
Innovation Solution
A closed-loop charging method using a dynamic electrochemical model with a constraint model and state observer algorithm to determine a maximum charging value, reducing computational power and ensuring safe, fast charging while limiting degradation.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Loss of time
If MPC-based charging protocol with electrochemical model is used, then charging time is reduced and degradation is limited, but computational complexity increases significantly
Solution Approach 1:
The patent pre-calculates and stores optimal charging current profiles for different battery states (SOC, temperature, age) before operation. During charging, the system simply retrieves and applies the pre-determined profile based on current battery conditions, avoiding real-time complex optimization calculations while maintaining optimal charging performance
Solution Approach 2:
The patent creates simplified lookup tables and pre-computed charging profiles that replicate the behavior of complex MPC algorithms. These tables store the essential charging characteristics without requiring the full computational machinery of real-time optimization, enabling low-cost microcontrollers to achieve MPC-like performance
2Reliability
If conservative parameter tuning in CC-CV protocol is used, then battery safety is ensured, but charging time increases and lifetime is not optimized
Solution Approach 1:
The patent implements dynamic charging current adjustment based on real-time battery state estimation (SOC, temperature, internal resistance). The charging protocol transitions from static conservative current limits to dynamic current profiles that adapt to actual battery conditions, enabling higher currents when safe and lower currents when degradation risks are present
Solution Approach 2:
The patent incorporates continuous feedback from battery measurements (voltage, current, temperature) to adjust charging parameters. The system monitors battery response during charging and modifies the charging profile accordingly, ensuring safety while maximizing charging speed and optimizing lifetime through data-driven decisions
3Productivity
If high computational power hardware is used, then MPC-based charging control can be implemented, but hardware cost increases significantly
Solution Approach 1:
The patent replaces expensive high-performance computational hardware with inexpensive microcontrollers and pre-computed lookup tables. The system accepts that the lookup tables occupy memory space but eliminates the need for expensive processors, making fast optimized charging accessible to low-cost battery management systems
Data Source
AI summary
A method for controlling a charging signal for charging a rechargeable battery cell, which is based on a dynamic electrochemical model using a functional or tabular relation for computing a maximum charging value for the charging signal. During charging, constraints of a constraint model are satisfied for a charging signal smaller than or equal to this maximum charging value. A battery cell not at an end-of-charge condition is charged with a closed-loop charging method, repetitively performing: i) measuring measurable signal(s) of the battery cell, ii) estimating state variable(s) of the model using a state observer algorithm, with the measured measurable signal(s) as input, iii) computing the maximum charging value using the functional or tabular relation with the estimates of the state variable(s) as input, and iv) applying to the battery cell a charging signal equal to or lower than the maximum charging value. Also, a charging device implementing the method.