Battery Runtime Forecasting via Dynamic Charge Profile Segmentation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current information handling systems provide inaccurate battery runtime predictions due to linear estimation based on instantaneous discharge rates, leading to varying and confusing runtime estimates for users.
Innovation Solution
A method for battery runtime forecasting that identifies power usage, charge profiles, and power delivery capabilities to generate a predicted relative state of charge (RSOC) profile, allowing for a more accurate runtime estimate by considering the entirety of the charge profile and current state of charge.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Device complexity
If linear predication based on instantaneous discharge rate is used, then calculation simplicity is maintained, but battery runtime prediction accuracy deteriorates
Solution Approach 1:
The battery charge profile is segmented into multiple discrete charging levels (e.g., 0-20%, 20-40%, etc.), each with its own time period and charging rate. This segmentation allows the system to account for non-linear charging behavior at different state of charge ranges while maintaining a structured, manageable calculation framework.
Solution Approach 2:
The patent transitions from static linear predication to dynamic prediction by continuously updating the runtime estimate based on the current charging level and time period. The system adapts to changing battery characteristics as the charging progresses through different stages, improving accuracy without requiring overly complex calculations.
2Device complexity
If averaging function is applied to current rate of charge, then calculation complexity is reduced, but runtime prediction stability deteriorates
Solution Approach 1:
The patent pre-establishes a charge profile that contains time periods and charging levels for the entire charging process. By having this comprehensive profile available in advance, the system can directly reference appropriate values for each charging stage without needing to perform complex real-time averaging calculations, thereby maintaining both simplicity and stability.
3Speed
If instantaneous discharge rate is used for prediction, then response time is fast, but prediction accuracy deteriorates due to non-linear charge profiles
Solution Approach 1:
The system dynamically selects the appropriate charging level and time period from the pre-established charge profile based on the current state of charge. This dynamic approach allows the prediction to adapt to non-linear charging behavior at different stages while maintaining fast response times through efficient lookup and calculation.
Data Source
AI summary
System and method for battery runtime forecasting, including identifying a power usage of a IHS; identifying a charge profile of a battery of the IHS, the charge profile including charging levels each associated with a respective time period; determining a state of charge of the battery based on a current time; identifying power delivery capabilities of a power source providing power to the battery; generating a predicted relative state of charge (RSOC) profile of the battery based on i) the power usage of the IHS, ii) an entirety of the charge profile of the battery, iii) the state of charge of the battery, and iv) power delivery capabilities of the power source; and identifying a runtime estimate of the battery.