Battery Pack Lithium Plating Prediction From Fleet Sensor Data
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Lithium-ion batteries face premature degradation and reduced lifespan due to lithium plating, which can lead to short circuits and increased recycling demands, with existing detection methods being invasive and inefficient.
Innovation Solution
A computer-implemented method using machine learning models trained with sensor data from battery packs to predict lithium plating occurrences, allowing for early detection and preventative actions such as replacing faulty cells, thereby extending battery life and reducing recycling burdens.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Speed
If lithium-ion batteries are charged quickly or used in cold temperatures, then charging speed and power delivery are improved, but lithium plating occurs which reduces battery lifespan and causes degradation
Solution Approach 1:
The system performs preliminary detection of lithium plating conditions by analyzing voltage profiles and operational parameters before plating occurs. Machine learning models predict plating risk based on historical data, allowing preventive actions to be taken before the harmful effect manifests, thus enabling fast charging without compromising battery lifespan
Solution Approach 2:
The system continuously monitors battery voltage, current, and temperature, comparing real-time measurements against learned patterns of plating conditions. This feedback loop allows dynamic adjustment of charging parameters to avoid plating while maintaining fast charging capability, resolving the contradiction between speed and lifespan
2Measurement precision
If lithium plating is detected using traditional invasive methods, then detection accuracy is improved, but device complexity and operational disruption increase
Solution Approach 1:
The system uses voltage profiles as an intermediary indicator to detect lithium plating indirectly. Instead of directly measuring plating (which would require invasive methods), the system analyzes voltage characteristics that change in response to plating conditions. This intermediary approach maintains high detection accuracy while avoiding device complexity and operational disruption
Solution Approach 2:
The patent replaces physical/invasive detection methods with computational analysis of electrical signals. Machine learning models process voltage profile data to detect plating, substituting mechanical or chemical inspection methods with software-based detection that reduces device complexity while maintaining precision
3Reliability
If battery capacity drops below 70-80% threshold, then safety is improved by preventing failures, but productivity and resource utilization decrease due to premature replacement
Solution Approach 1:
The system performs preliminary detection of lithium plating conditions before they cause capacity degradation or safety failures. By identifying plating risk early through voltage profile analysis and machine learning prediction, the system enables preventive maintenance that extends battery life beyond traditional capacity thresholds, improving both safety and productivity
Solution Approach 2:
The system provides self-service monitoring and prediction capabilities that track battery health in real-time. This continuous self-assessment allows the battery management system to optimize utilization by detecting actual degradation mechanisms rather than relying on arbitrary capacity thresholds, thereby extending productive service life while maintaining safety
Data Source
AI summary
In one aspect, computer-implemented method may include receiving, from computing devices, fleet data pertaining to battery packs each including first cells. The fleet data includes false positive images of lithium plating affecting at least a first cell, true positive images of the lithium plating affecting at least a second cell, or both. The method may include training, using at least the fleet data, machine learning models to predict occurrences of the lithium plating, receiving, from sensors associated with second cells, measurements pertaining to current, voltage, temperature, or some combination thereof, and inputting the measurements into the machine learning models to predict the occurrences of the lithium plating for the second cells.