Battery Charging Level Prediction Using Fleet Machine Learning
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing battery management systems lack the granularity to optimize battery pack performance by not considering individual cell characteristics and operating conditions, leading to suboptimal charging and potential heat-related issues, which can reduce battery life and efficiency.
Innovation Solution
A system utilizing machine learning to predict battery pack usage based on real-world data from a fleet of vehicles, generating a predictive model that determines a target charging level by considering cell arrangements, travel conditions, and temperature, allowing for optimized charging decisions.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Duration of action of moving object
If the battery pack is charged to a higher level, then the vehicle can travel longer distances, but the battery pack life is reduced
Solution Approach 1:
The system dynamically adjusts the target charging level based on real-time conditions including route characteristics, driving patterns, weather forecasts, and actual battery performance data. Rather than using a fixed charging strategy, the system adapts the charging level to match specific travel requirements, allowing higher charges only when necessary and lower charges when sufficient, thereby extending battery life while maintaining adequate travel range.
Solution Approach 2:
The system changes the charging parameter (target charging level) based on multiple variables including route distance, elevation changes, temperature conditions, and historical battery performance. By adjusting this parameter dynamically rather than maintaining a constant high charging level, the system reduces unnecessary stress on the battery while ensuring sufficient energy for actual travel needs.
2Use of energy by moving object
If the battery pack is charged to a higher level, then the vehicle has more energy available, but the charging time increases
Solution Approach 1:
The system applies partial charging action by determining the minimum necessary charging level required for the specific route and driving conditions. Instead of consistently charging to maximum capacity, the system charges only to the extent needed, which reduces charging time while maintaining sufficient energy availability for the planned journey.
Solution Approach 2:
The target charging level is dynamically adjusted based on route characteristics, current battery state, and predicted driving behavior. This dynamic approach allows the system to optimize the balance between energy availability and charging time by charging to higher levels only when the route demands it, rather than always charging to maximum capacity.
3Power
If the battery pack is charged to a higher level, then the vehicle can handle demanding routes, but the charging cost increases
Solution Approach 1:
The system changes the charging parameter (target charging level) based on route demands, vehicle usage patterns, and energy costs. By adjusting this parameter to match actual needs rather than maintaining maximum capacity, the system reduces unnecessary charging costs while ensuring sufficient power capacity for demanding routes when required.
Solution Approach 2:
The system applies partial charging by determining the precise amount of energy needed for specific routes and charging only to that level. This avoids the excessive action of consistently charging to maximum capacity, thereby reducing charging costs while maintaining adequate power capacity for when it is truly needed.
4Productivity
If the battery management system considers more operating conditions, then the charging optimization improves, but the system complexity increases
Solution Approach 1:
The system achieves multi-functionality by using a single machine learning model that processes multiple types of input data (route information, driving patterns, weather conditions, battery performance) and generates optimized charging recommendations. This universal approach consolidates what could be multiple separate systems into one integrated solution, improving charging optimization without proportionally increasing system complexity.
Solution Approach 2:
The machine learning model automatically processes and integrates multiple operating conditions without requiring manual intervention or complex rule-based systems. The system self-learns from historical data and autonomously determines optimal charging levels based on current conditions, reducing the complexity that would otherwise be required to manually manage all these variables.
Data Source
AI summary
Systems, methods, and storage media for determining a target charging level of a battery pack for a drive route are disclosed. A method includes receiving data pertaining to cells within a battery pack installed in each vehicle of a fleet of vehicles, the data received from at least one of each vehicle in the fleet of vehicles, providing the data to a machine learning server, directing the machine learning server to generate a predictive model, the predictive model based on machine learning of the data, receiving a vehicle route request from the vehicle, the vehicle route request corresponding to the drive route, estimating travel conditions of the vehicle based on the route request, determining a temperature of the battery pack in the vehicle, determining a target battery charging level based on the predictive model, the travel conditions, and the temperature, and providing the target battery charging level to the vehicle.