Battery State Determination via Feature Extraction and Server Diagnosis
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing battery state determination systems in vehicles have lower accuracy when relying on external server devices to assess battery deterioration, as the cycle for determining battery state based on transmitted physical quantities is longer than the cycle for detection at the vehicle, leading to reduced determination accuracy.
Innovation Solution
A battery state determination system comprising an in-vehicle device and a server device, where the in-vehicle device acquires and observes physical quantity data, generates feature data expressing state changes, and transmits it to the server device for diagnosis, allowing the server to accurately determine battery deterioration by referencing feature data from multiple observation times, including relationships between voltage, current, and temperature.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Use of energy by moving object
If battery state is determined by transmitting physical quantity data to an external server device, then operational load on the in-vehicle device is reduced, but determination accuracy decreases due to longer processing cycles
Solution Approach 1:
The determination process is segmented into two parts: the in-vehicle device performs feature extraction from physical quantity data, while the server device performs the actual deterioration state determination. This division allows the in-vehicle device to reduce operational load by not performing full determination processing, while the server device receives pre-processed feature data to maintain determination accuracy despite the external processing location.
2Device complexity
If battery state is determined by transmitting physical quantity data to an external server device, then device complexity is reduced, but communication bandwidth pressure increases
Solution Approach 1:
The in-vehicle device extracts essential features from raw physical quantity data and transmits only these feature data to the server device. This extraction process removes unnecessary data elements, significantly reducing the communication bandwidth required while maintaining the server's ability to perform accurate deterioration state determination.
3Measurement precision
If battery state is determined at the vehicle using detected physical quantities, then determination accuracy is maintained, but operational load increases
Solution Approach 1:
The system dynamically adjusts the processing distribution between in-vehicle device and server device. The in-vehicle device performs feature extraction (a lighter computational task), while the server device performs the more intensive deterioration state determination. This dynamic division of labor maintains determination accuracy while optimizing operational load distribution based on the capabilities and constraints of each component.
Data Source
AI summary
A battery state determination system is a battery state determination system including an in-vehicle device and a server device, the in-vehicle device includes an acquisition part acquires physical quantity data showing a physical quantity related to a state of a battery mounted in a vehicle, an observation part that observes a feature related to a state change of the battery on the basis of the physical quantity data at a plurality of different observation times, and a transmission part that transmits a plurality of pieces of feature data which expresses the feature observed each observation time to the server device, and the server device includes a reception part that receives the plurality of pieces of feature data transmitted from the in-vehicle device, and a diagnosis part that diagnoses a deterioration state of the battery on the basis of the plurality of pieces of feature data.