Sinusoidal Charging Current for Battery Parameter Estimation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing battery parameter estimation methods, such as the Extended Kalman Filter (EKF), require dynamic inputs to converge accurately and may return inaccurate values during periods of constant or slowly changing currents, which is common during battery charging, leading to inefficient battery power capability estimation.
Innovation Solution
A controller in the vehicle system changes the charging current from a normal constant current to a predetermined sinusoidal current that varies between a maximum and minimum, ensuring persistent excitation for the EKF, allowing accurate parameter estimation during charging by oscillating between these limits to achieve convergence of residual errors to zero.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If a constant current is used for battery charging, then the charging process is simple and stable, but the parameter estimation algorithm cannot converge accurately due to lack of dynamic input
Solution Approach 1:
The patent applies periodic action by superimposing a sinusoidal current signal on the constant charging current. This periodic variation provides the dynamic input needed for accurate parameter estimation while maintaining the overall charging function. The sinusoidal signal causes persistent excitation in the battery system, enabling the extended Kalman filter to converge to accurate parameter values.
Solution Approach 2:
The patent transforms the static constant current into a dynamic signal by adding a time-varying sinusoidal component. This dynamic current profile I(t) = I_constant + I_ac * sin(2πft) provides the necessary temporal variation for parameter estimation algorithms to accurately track battery parameters while still performing charging functionality.
2Measurement precision
If a sinusoidal current is applied for parameter estimation, then accurate battery parameters are obtained, but the charging process is disrupted and time is lost
Solution Approach 1:
The patent merges two functions - charging and parameter estimation - into a single integrated process. By superimposing the sinusoidal estimation signal on top of the constant charging current rather than using separate measurement phases, the system simultaneously charges the battery and performs parameter identification, eliminating the need to choose between charging speed and estimation accuracy.
Solution Approach 2:
The charging current serves dual purposes: it provides energy to charge the battery and simultaneously provides the dynamic excitation signal needed for parameter estimation. This multi-functional current waveform eliminates the need for dedicated measurement periods, maximizing charging efficiency while maintaining accurate parameter tracking.
3Measurement precision
If the sinusoidal current amplitude is increased for better estimation, then parameter convergence improves, but battery stress and potential damage increase
Solution Approach 1:
The patent carefully selects and adjusts the parameters of the sinusoidal signal, specifically the amplitude I_ac and frequency f, to achieve the minimum necessary excitation for parameter convergence while staying within safe operating limits. The controller is configured with specific parameter values that balance estimation accuracy with battery safety, preventing excessive stress while ensuring sufficient dynamic input for the estimation algorithm.
Data Source
AI summary
A vehicle includes a battery that is rechargeable using an external power source coupled to the vehicle. The vehicle includes a controller that is programmed to estimate parameters of the battery using a parameter estimation algorithm. The controller is programmed to change a charging current when connected to the external power source to provide an input to the parameter estimation algorithm that is sufficiently dynamic such that the parameter estimation algorithm converges to an accurate solution.