EV Battery Capacity Prediction Using Historical Charging Patterns
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing battery capacity prediction systems for electric vehicles require real-time communication and complex data processing, leading to high costs and complexity.
Innovation Solution
A prediction apparatus that utilizes past usage records and a trained machine learning model to predict battery capacity based on travel distance and charge information, with correction values applied to improve accuracy.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If real-time communication devices and servers are installed to acquire charging information, traffic information, and position information, then the remaining capacity prediction accuracy is improved, but the system cost and complexity increase
Solution Approach 1:
The patent extracts only the essential historical usage data (charge amount, travel distance, charging time) needed for prediction, removing the need for complex real-time communication infrastructure. The prediction is performed using only acquired data without requiring continuous real-time data streams from multiple sources.
Solution Approach 2:
The patent performs prediction using historically acquired data before real-time conditions are fully known. By using past usage patterns and pre-acquired information, the system provides advance prediction without needing complex real-time monitoring infrastructure during the prediction moment.
2Measurement precision
If a server acquires large amounts of real-time information from the EV side, then the prediction accuracy is improved, but the processing complexity and difficulty of realization increase
Solution Approach 1:
The patent extracts only the necessary historical features (charge amount, travel distance, charging time) from usage data, eliminating the need to process large volumes of real-time information. This selective extraction simplifies the prediction process while maintaining accuracy based on critical parameters.
Solution Approach 2:
The patent uses historical usage records as copies of past conditions to predict future states. By analyzing patterns in historical data rather than processing real-time information streams, the system achieves prediction without complex real-time processing infrastructure.
3Measurement precision
If correction values are calculated and applied to the predicted remaining capacity, then the prediction accuracy is improved, but the computational process becomes more complex
Solution Approach 1:
The system uses its own historical prediction data and actual usage outcomes to automatically calculate correction values. This self-correction mechanism improves accuracy without requiring external intervention or complex manual adjustment processes.
Solution Approach 2:
The patent implements a feedback mechanism where prediction errors from historical data are used to calculate correction values that are applied to future predictions. This closed-loop approach continuously improves accuracy using simple computational feedback rather than complex control systems.
Data Source
AI summary
A prediction apparatus for predicting a remaining capacity value of a battery of a vehicle includes a processor; and a memory storing instructions for executing a process including acquiring usage record data including information regarding a travel distance of the vehicle and information indicating a charge amount of the battery or information indicating a charging time of the battery as a past use record of the vehicle, and acquiring usage schedule data including information regarding a travel distance of the vehicle as a future usage schedule of the vehicle; determining the battery being charged when a predetermined travel distance is higher than a threshold; calculating a correction value as a predetermined charge amount; predicting a remaining capacity value of the battery at return; using the correction value to execute correction on the predicted remaining capacity value; and outputting a remaining capacity value of the battery after executing the correction.