Starter Battery Prediction Using Vehicle-Specific Engine-Start Data
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional methods for estimating the remaining capacity and predicting the engine starting performance of electrical energy storage systems, particularly start-stop batteries, are inaccurate due to reliance on direct measurements and assumptions of fully charged states, neglecting varying operating conditions and battery degradation, which leads to unreliable predictions.
Innovation Solution
A method that generates characteristic engine start data, including voltage behavior, temperature, and state of charge, using sensors and machine learning algorithms to create a prediction model that accounts for different vehicle brands, models, and variants, optimizing the estimation of aging state and engine starting performance under real conditions.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If conventional direct measurement methods are used to estimate battery condition, then the measurement process is simple, but the prediction accuracy deteriorates due to errors in measured variables directly affecting the output variable
Solution Approach 1:
The patent introduces characteristic curves as an intermediary between direct measurements and battery condition determination. Instead of directly using measured variables to determine battery state, the method maps measurements through characteristic curves that represent typical battery aging behavior, thereby filtering out measurement errors and providing more reliable predictions.
Solution Approach 2:
The patent creates characteristic curves that copy or represent typical battery aging patterns and behavior. These curves serve as reference models that can be compared against actual measurements, allowing the system to determine battery condition based on how well actual behavior matches the characteristic patterns rather than relying solely on direct measurements.
2Reliability
If battery testing devices measure internal resistance to predict capacity, then the testing process is straightforward, but the prediction reliability deteriorates because batteries are not operated in fully charged states and degradation varies
Solution Approach 1:
The patent performs preliminary characterization of battery behavior by recording voltage, current, and temperature data during actual engine starting operations. This preliminary data collection and analysis creates characteristic curves that capture the battery's true behavior under real operating conditions, including partial charge states and aging effects, before these patterns are used for prediction.
Solution Approach 2:
The system continuously monitors actual battery performance during engine starting and compares it against the characteristic curves. This feedback mechanism allows the system to detect deviations from expected behavior that indicate aging or degradation, enabling timely updates to the characteristic curves and maintaining prediction accuracy over time.
3Measurement precision
If conventional methods assume fully charged battery states for capacity specification, then the capacity specification is standardized, but the prediction accuracy deteriorates because start-stop batteries operate in partially charged states
Solution Approach 1:
The patent applies local quality by creating separate characteristic curves for different operating conditions, particularly different charge states. Instead of using a single universal capacity specification, the system adapts the characteristic curves to match the specific operating conditions of start-stop batteries, which frequently operate in partially charged states, thereby providing accurate predictions for each local operating condition.
Solution Approach 2:
The system dynamically adapts the characteristic curves based on actual operating conditions. Rather than relying on static capacity specifications, the characteristic curves are updated and adjusted to reflect the battery's actual behavior under varying charge states, temperatures, and aging conditions, making the prediction system versatile across different operating scenarios.
Data Source
Figure 1
AI summary
The invention relates to a method for predicting engine-start performance of an electrical energy storage system, in particular of a motor vehicle starter battery. The method comprises the following steps: generating engine-start data characteristic of the electrical energy storage system; evaluating the generated engine-start data; and outputting a result of the evaluation, which result relates to a prediction relating to the engine-start performance of the electrical energy storage system. According to the invention, in particular a vehicle make, a vehicle model and/or a vehicle variant of the vehicle to be started by the electrical energy storage system is taken into account in order to evaluate the generated engine-start data.