Battery Thermal Analysis Model Using GAN-Generated Learning Data

Solution Overview

Problem

Thermal analysis of large-capacity battery systems requires significant time and resources, particularly in energy storage systems (ESS), necessitating a more efficient method.

Innovation Solution

A thermal analysis system utilizing a learning data generation device and a model construction device that employs a generative adversarial network (GAN) to generate and transform thermal analysis data, constructing a thermal analysis model for faster and resource-efficient analysis.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Measurement precision

If numerical thermal analysis is performed on large-capacity battery systems using governing equations, then measurement precision and reliability are improved, but loss of time and productivity deteriorate significantly

Engineering Contradiction:

Improvethermal analysis accuracyVSAvoidanalysis time

Core Design Contradiction:

Measurement precisionVSLoss of time

Solution Approach 1:

The patent pre-generates comprehensive thermal analysis data covering various operating conditions, battery states, and environmental parameters before actual thermal analysis is needed. This pre-computed data serves as training data for machine learning models, enabling rapid prediction during actual operation without repeating time-consuming numerical simulations

Inventive Principle:

Principle #10Preliminary action

Solution Approach 2:

The patent creates simplified surrogate models (neural networks, Gaussian processes, polynomial chaos expansions) that copy the behavior of complex numerical thermal analysis models. These surrogate models are trained on pre-generated data and can rapidly predict thermal characteristics without performing full numerical simulations, thus reducing analysis time while maintaining accuracy

Inventive Principle:

Principle #26Copying

2Reliability

If numerical thermal analysis is performed on large-capacity battery systems using governing equations, then measurement precision and reliability are improved, but productivity deteriorates due to enormous resource requirements

Engineering Contradiction:

Improvethermal analysis reliabilityVSAvoidanalysis efficiency

Core Design Contradiction:

ReliabilityVSProductivity

Solution Approach 1:

The patent pre-generates comprehensive thermal analysis data covering various operating conditions, battery states, and environmental parameters before actual thermal analysis is needed. This pre-computed data serves as training data for machine learning models, enabling rapid prediction during actual operation without repeating time-consuming numerical simulations

Inventive Principle:

Principle #10Preliminary action

Solution Approach 2:

The patent creates simplified surrogate models (neural networks, Gaussian processes, polynomial chaos expansions) that copy the behavior of complex numerical thermal analysis models. These surrogate models are trained on pre-generated data and can rapidly predict thermal characteristics without performing full numerical simulations, thus reducing analysis time while maintaining accuracy

Inventive Principle:

Principle #26Copying

3Measurement precision

If traditional thermal analysis methods are used for battery system design and evaluation, then measurement precision is maintained, but loss of time in design and evaluation processes increases

Engineering Contradiction:

Improvethermal analysis accuracyVSAvoiddesign and evaluation time

Core Design Contradiction:

Measurement precisionVSLoss of time

Solution Approach 1:

The patent creates simplified surrogate models (neural networks, Gaussian processes, polynomial chaos expansions) that copy the behavior of complex numerical thermal analysis models. These surrogate models are trained on pre-generated data and can rapidly predict thermal characteristics without performing full numerical simulations, thus reducing analysis time while maintaining accuracy

Inventive Principle:

Principle #26Copying

Solution Approach 2:

The patent transforms the thermal analysis approach by changing from direct numerical simulation parameters to machine learning model parameters. The system uses trained models with pre-learned thermal patterns to predict temperature distributions, heat generation rates, and thermal characteristics based on input parameters like battery state of charge, ambient temperature, and load conditions, significantly accelerating the analysis process

Inventive Principle:

Principle #35Parameter changes

Data Source

PatentUS20260036470A1Thermal analysis system and method for battery system

Publication Date: 2026.02.05 SAMSUNG SDI CO LTD

AI summary

The present disclosure relates to a thermal analysis system and method for a battery system. The thermal analysis system includes a learning data generation device configured to generate second thermal analysis data using first thermal analysis data and a first artificial neural network model. The first thermal analysis data is obtained through numerical thermal analysis of a battery system. The thermal analysis system also includes a model construction device configured to construct a thermal analysis model by using a second artificial neural network model with the first thermal analysis data and the second thermal analysis data as learning data.