Current-Pulse Battery Diagnosis for Accurate SOH and SOC Estimation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current battery diagnosis methods for lithium-ion batteries, such as direct-current pulse and alternating-current impedance methods, fail to accurately diagnose degradation due to errors in observation data and incorrect evaluation of pseudo equivalent circuit parameters, leading to inaccurate determination of State of Health (SOH) and State of Charge (SOC).
Innovation Solution
A battery diagnosis apparatus and method using a current pulse technique that applies a constant-current pulse, with noise filtering and data resampling, and incorporates a diagnostic algorithm based on machine learning to correlate voltage responses with battery state factors, minimizing the effects of hysteresis and thermal changes.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If direct-current pulse method or alternating-current impedance method is used for battery diagnosis, then the diagnosis can be performed, but the measurement precision of SOH and SOC is inaccurate due to errors in observation data and incorrect evaluation of pseudo equivalent circuit parameters
Solution Approach 1:
The patent changes the measurement parameters by applying a constant-current pulse with optimized amplitude and width ratios, and by measuring voltage at specific time points (t1 and t2) after pulse application. This parameter optimization resolves the contradiction by achieving accurate SOH and SOC estimation without relying on error-prone pseudo equivalent circuit parameters, thereby improving both measurement precision and diagnosis reliability
Solution Approach 2:
The patent replaces the traditional electrical measurement approach (direct-current pulse and alternating-current impedance methods that rely on pseudo equivalent circuit models) with a simplified voltage response measurement method. By measuring voltage at specific time points after constant-current pulse application and using predetermined coefficients to calculate SOH and SOC, the patent eliminates the need for complex circuit parameter evaluation, thereby improving diagnosis accuracy
2Productivity
If traditional battery diagnosis methods are used, then the diagnosis process can be completed, but the diagnosis time is lengthy and the process is complex
Solution Approach 1:
The patent applies preliminary action by pre-calculating and storing predetermined coefficients (K1, K2, K3, K4) that relate voltage responses to SOH and SOC values. During actual diagnosis, these pre-prepared coefficients are directly applied to measured voltage values, eliminating the need for complex real-time calculations and significantly reducing diagnosis time while maintaining high productivity
Solution Approach 2:
The patent segments the diagnosis process into distinct measurement points (t1 and t2) after pulse application, where each time point provides specific information. By dividing the continuous voltage response into discrete measurement segments, the patent simplifies the diagnosis process, reduces the time needed for data collection and analysis, and enables rapid determination of battery state
Applied Scientific Principles
This section explains which scientific principles are used to turn an abstract innovation direction into a practical engineering solution.
Function Achieved in This Case
Enables accurate and rapid battery degradation diagnosis, improving the precision of SOH and SOC estimation, and extending the battery's life by providing a high-accuracy, short-time diagnosis of battery health and charge state.
Implementation Method 1
the hysteresis phenomenon occurs in the output potential of lithium-ion batteries, that is, a potential difference of several tens of millivolts (mV) arises per battery depending on the current operation direction after a charge process or a discharge process
Data Source
Figure 1~2
Figure 3
Figure 4(a)~4(g)
AI summary
A battery diagnosis apparatus and a battery diagnosis method for accurately diagnosing a secondary battery are proposed. A pulse current generator, a voltage measuring instrument that measures a voltage response to application of a current pulse, a first data processing device that obtains a chronopotentiogram (CP) indicating a change in the voltage response over time and normalizes the CP, a database that saves normalized data, and a second data processing device that uses a correlation between the saved data and a battery state expressing factor prepared in advance to make a battery diagnosis are used. Desirably, the current pulse is a current in the same direction at the time of data obtainment and at the time of a diagnosis. Further, a noise filter for an input signal of the CP and resampling means for reducing the number of pieces of data input to the first data processing device are provided.