Machine Learning Prediction for Battery Electrode Suspension Quality
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing methods for predicting the quality of battery suspensions during the production of lithium-ion battery electrodes are imprecise due to the complexity of the extrusion process and reliance on subjective human predictions, leading to costly rejects and loss of process knowledge when skilled personnel leave.
Innovation Solution
A method using a machine-learned prediction algorithm that estimates properties of the suspension based on actual and target process parameters, ambient conditions, and historical data, allowing for inline, mechanical, and reliable quality prediction with minimal time and cost expenditure.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Ease of operation
If physical models are used for prediction, then the prediction can be made mechanically and inline, but the precision is insufficient due to process complexity and disturbance variables
Solution Approach 1:
The patent replaces traditional physical/mechanical models with a machine learning-based prediction system. The machine learning model is trained on historical process data and suspension property data to learn complex non-linear relationships, substituting the insufficient physical models while maintaining inline prediction capability through automated data processing and algorithmic analysis.
2Reliability
If human expert prediction is used, then process knowledge can be utilized, but the prediction is unreliable due to subjective feelings and loss of knowledge when experts leave
Solution Approach 1:
The patent creates a digital copy of expert knowledge by training a machine learning model on historical data that encapsulates years of process expertise. This digital replica captures and preserves process knowledge in a structured format, eliminating dependence on individual experts and preventing knowledge loss when personnel leave the organization.
Solution Approach 2:
The machine learning system enables the organization to serve itself by automatically learning and applying process knowledge without requiring continuous human expert intervention. The system independently analyzes data, makes predictions, and can be continuously improved through automated retraining, reducing reliance on external or internal human expertise.
3Ease of manufacture
If traditional prediction methods are used, then the process is simple, but costly rejects occur due to insufficient quality prediction
Solution Approach 1:
The patent implements a feedback mechanism where prediction results are continuously monitored and used to adjust process parameters in real-time. This closed-loop system allows for immediate corrective actions when quality deviations are predicted, preventing the production of defective products and reducing material rejects while maintaining process simplicity through automated control.
Data Source
AI summary
The invention relates to a method for producing at least one electrode for a battery cell, the method comprising at least the following steps: a) providing a suspension for creating the at least one electrode, at least one target process parameter (1) being specifiable for providing the suspension; b) capturing at least one actual process parameter (2) while providing the suspension in step a); c) performing a prediction of at least one property (4) of the suspension by means of a machine-learned prediction algorithm (5), which estimates the at least one property (4) of the suspension depending on the at least one actual process parameter (2) and taking into account information on previously provided suspensions; d) defining at least one target process parameter (1) for providing the suspension in step a) depending on the prediction results from step c).