Vehicle Battery SOH Estimation Using Current Distribution Patterns

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Estimating the state of health (SOH) of a battery during driving is difficult due to frequent changes in charging and discharging currents, making it challenging to grasp the battery deterioration degree.

Innovation Solution

A vehicle system that includes a current detector, processor, and memory to acquire, derive, and store current-value frequency and percentage distributions, and estimate traveling patterns and battery deterioration by combining predetermined distributions with actual driving data.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Measurement precision

If current values are frequently acquired during driving to capture battery state changes, then measurement precision of battery deterioration is improved, but device complexity and data processing burden increase

Engineering Contradiction:
Improvebattery deterioration measurement precisionVSAvoiddata processing complexity
Core Design Contradiction:
Measurement precisionVSDevice complexity

Solution Approach 1:

The patent segments the continuous current data into discrete frequency distributions by dividing the current value range into multiple bins or intervals. Each bin accumulates frequency counts, transforming continuous measurement data into discrete, manageable segments that are easier to process and analyze for battery deterioration assessment.

Inventive Principle:
Principle #1Segmentation

Solution Approach 2:

The patent creates a simplified representation (copy) of the actual current data by generating frequency distribution histograms. Instead of processing raw current values directly, the system works with these derived frequency distributions that capture the essential characteristics of battery usage patterns while reducing data complexity.

Inventive Principle:
Principle #26Copying

2Reliability

If multiple current value parameters are measured and processed to derive frequency distributions, then reliability of battery SOH estimation is improved, but loss of time for data processing increases

Engineering Contradiction:
Improvebattery SOH estimation reliabilityVSAvoiddata processing time
Core Design Contradiction:
ReliabilityVSLoss of time

Solution Approach 1:

The patent performs preliminary processing by accumulating frequency counts in binned intervals during normal operation. This preliminary organization of data into frequency distributions prepares the information in advance for rapid SOH estimation, avoiding the need for complex real-time calculations when deterioration assessment is needed.

Inventive Principle:
Principle #10Preliminary action

Solution Approach 2:

The patent transforms the raw current value parameter into a frequency distribution parameter representation. By changing the parameter form from individual current measurements to aggregated frequency counts across intervals, the system enables more reliable SOH estimation through statistical analysis while reducing processing time through simplified mathematical operations.

Inventive Principle:
Principle #35Parameter changes

Data Source

PatentUS12351066B2Vehicle
Publication Date: 2025.07.08 SUBARU CORP
  • US12351066B2 patent drawing
  • US12351066B2 patent drawing
  • US12351066B2 patent drawing

AI summary

A vehicle includes at least one processor, at least one memory, and a storage. The processor functions as a current-value acquirer that acquires a current value of a battery mounted on the vehicle, a current-value frequency distribution deriver that increments an acquisition count for one of classes, to which the current value belongs, and derives a current-value frequency distribution representing acquisition counts for the respective classes, and a current-value rate distribution deriver that transforms the acquisition counts to acquisition rates each indicating a rate of the each of the acquisition counts to a total of the acquisition counts and derives a current-value rate distribution representing the acquisition rates. The processor further functions as a traveling pattern rate deriver that derives on the basis of an actual current-value rate distribution and predetermined current-value rate distributions stored in the storage, traveling pattern rates indicating rates of the respective predetermined current-value rate distributions.