Automating the Battery Cell Microstructure Evaluation
Our client, KeraCel, is focused on the transformation of additive manufacturing with multi-material 3D printing solutions that allow production of active electro-mechanical devices. We were tasked with helping them with their battery cell microstructure evaluation.
Simple Lattice Structure
Material properties have a large impact on machine performance. As part of the production process of a battery cell, a lattice structure of particles is created.
The structure requires an even distribution of particles in order to be acceptable for the following phase of the production process. Only through visual inspection of the microstructure, is it possible to verify that the distribution of particles is acceptable or not.
The quality assurance step of visually inspecting the battery cell microstructure is
We trained our application to recognize and measure different features in the microstructure.
Where a human was previously required to review the individual lattice pictures, now through deep learning, engineers are notified in real time of abnormal data sets.
*not actual customer data
OUR PROCESS & RESULTS
Training a Deep Learning Model
Through deep learning, the Machinery Analytics model is trained to recognize and measure different features in the microstructure. Certain features are immediately indicative of problems, while other features form part of a multi-variable, non-linear quality function which indicates product quality.
An Improved Production Process
Where a human was previously required to review the individual lattice pictures, now through deep learning, Engineers are notified in real time of abnormal data sets. This increases Quality Assurance throughput significantly, which in turn increases efficiency and ultimately decreases issues found post production.
Provide Actionable Data
Are you facing similar obstacles in your organization?
At Machinery Analytics, we believe that skilled professionals should always have the best tools for the job. That’s why we are developing a cloud based, deep learning platform which can easily distinguish changes in actual performance from changes in the environment or operating conditions. Our customers are innovators who want to develop quality products faster.