Cloud Battery Life Estimation Using Fleet Temperature Data
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing battery capacity estimation methods, whether direct or model-based, incur additional costs or negatively impact battery life, and lack the ability to improve accuracy using information from other devices.
Innovation Solution
A cloud-based system that wirelessly receives battery temperature and initial state data to generate and refine battery life estimates, utilizing weather data and historical data from similar batteries to update predictions.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If direct battery capacity measurement using an integrated circuit is implemented, then measurement precision of remaining battery capacity is improved, but device cost increases and battery life decreases due to additional power consumption
Solution Approach 1:
The patent extracts the battery capacity measurement function from the device itself and relocates it to a remote server. The device only transmits minimal data (temperature, usage events) while the computationally intensive model execution and capacity calculation occur remotely, eliminating the need for power-hungry on-device sensors and processing.
Solution Approach 2:
The patent introduces a remote server as an intermediary between the device and the battery capacity measurement function. The server receives limited data from the device and performs the complex modeling and calculation remotely, acting as a mediator that enables accurate measurement without burdening the device's power resources.
2Use of energy by moving object
If a local battery model is created and used on-device, then the need for direct battery capacity measurement is removed, but the model cannot be improved using information from other devices
Solution Approach 1:
The patent merges data from multiple devices with the battery model. The server collects temperature, usage events, and actual capacity measurements from numerous devices, combining this aggregated data to continuously refine and improve the battery model, making it more accurate and adaptable over time.
Solution Approach 2:
The patent implements a feedback mechanism where actual battery capacity measurements and usage data from devices are sent back to the server. This feedback is used to update and refine the battery model, creating a continuous improvement loop that enhances prediction accuracy across the device fleet.
3Duration of action of moving object
If battery life is extended by reducing direct measurements, then power consumption decreases, but measurement precision and reliability of battery life prediction are reduced
Solution Approach 1:
The patent performs preliminary actions by collecting and storing battery temperature, usage events, and capacity measurements over time. This historical data is accumulated and used to train and validate the battery model, enabling accurate predictions without requiring continuous direct measurements during operation.
Solution Approach 2:
The patent replaces the physical/mechanical battery sensor system with a computational model. Instead of using hardware sensors to directly measure capacity, the system uses algorithms that calculate remaining capacity based on temperature, usage patterns, and historical data, substituting computational processing for physical measurement.
Data Source
Figure 1
Figure 2
Figure 3
AI summary
Predicting battery life including wirelessly receiving an initial state of a battery and a tracked event associated with the battery at a cloud-based server, and updating an estimate of the battery life using the tracked event.