Parameter Estimation for Lithium-Ion Battery State Monitoring
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
The existing parameter estimation methods, such as the least-squares method, often result in local solutions rather than the global optimum, especially when dealing with systems having many independent parameters, leading to decreased calculation speed and reduced accuracy in real-time detection of device parameters like residual capacity rate and life in lithium-ion batteries.
Innovation Solution
A parameter estimating device and method that utilize a state quantity measurement unit, a state quantity calculation unit, and an optimum estimated value identification unit to calculate error evaluation functions based on absolute errors and their variation, adjusting coefficients to prioritize minimizing the degree of variation first, followed by reducing absolute errors, to identify optimal parameter values.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Ease of manufacture
If the least-squares method is used for parameter estimation, then the calculation process is simple, but the estimation accuracy deteriorates due to falling into local solutions
Solution Approach 1:
The patent segments the parameter estimation process into multiple stages: initial estimation using least-squares method, followed by refinement using a two-stage optimization approach. This segmentation allows the system to benefit from the simplicity of least-squares for initial values while achieving higher accuracy through subsequent refinement stages that escape local minima.
Solution Approach 2:
The patent performs preliminary action by using the least-squares method to obtain initial parameter estimates before applying the more sophisticated two-stage optimization. This preliminary estimation provides a starting point that accelerates convergence and improves final accuracy without sacrificing the overall simplicity of the approach.
2Measurement precision
If the number of independent parameters is increased to improve estimation comprehensiveness, then the measurement accuracy improves, but the calculation speed deteriorates significantly
Solution Approach 1:
The patent uses preliminary action by pre-calculating optimal weighting coefficients and organizing measurement data structures before the actual estimation process. This preparation work enables the system to handle multiple parameters efficiently during real-time operation, maintaining high calculation speed while processing comprehensive parameter sets.
Solution Approach 2:
The patent applies dynamics by implementing an adaptive two-stage optimization process that dynamically adjusts the estimation strategy. The system transitions from a fast initial least-squares stage to a more accurate refinement stage only when necessary, allowing flexible adjustment between speed and accuracy based on real-time requirements and measurement quality.
3Productivity
If real-time successive detection of device parameters is implemented, then the operational monitoring capability is improved, but the calculation complexity increases making the algorithm unrealistic
Solution Approach 1:
The patent segments the complex real-time estimation problem into manageable discrete time-step iterations. Each time step performs a simplified two-stage optimization using only current and recent measurement data, avoiding the need to process all historical data simultaneously. This segmentation makes the algorithm computationally feasible for real-time implementation while maintaining accuracy.
Solution Approach 2:
The patent applies partial action by focusing the optimization process on the most critical parameters and using a simplified two-stage approach rather than exhaustive optimization. This selective focus achieves sufficient estimation accuracy for practical applications while keeping computational complexity manageable for real-time operation.
Data Source
Figure 1
Figure 2~3
Figure 4
AI summary
A parameter estimating device includes a state quantity measurement unit 11 which acquires measured values for state quantities having a correlation with a plurality of parameters, a state quantity calculation unit 12 which calculates and acquires calculated values corresponding to the measured values based on a correlation expression of the estimated values of the plurality of parameters and the state quantities, and an optimum estimated value identification unit 13 which finds an error evaluation function value calculated based on a value obtained by multiplying the total sum of the absolute quantities of the errors of the calculated values corresponding to the measured values by a first coefficient and a value obtained by multiplying the degree of variation of the errors by a second coefficient, and identifies an optimum estimated value having a minimum error evaluation function value while changing the estimated values.