Ensure Quality of Battery Cells with the help of an AI Solution

duration | 2 hours

Description

The course will deliver a full boiler plate for ai-based quality assurance in battery welding. It starts with consideration of the laser-based welding process that joins cell connectors. Selection of suitable sensors and a discussion of data acquisition will follow before code is being created to analyse the data. 

Learning Outcomes

By the end of this course, participants will be able to:

  • Consider suitable sensors to monitor process properties in laser-based welding
  • Plan the integration of such sensors into manufacturing equipment
  • Prepare data storage to acquire actual data from the sensor system
  • Set up a programming environment for the AI solution
  • Implement a first set of algorithms for training and evaluation of acquired data

    Introduction to laser-based welding

    duration | 0.5 hours

    Learning Outcomes

    • Understand parameters of the laser-based welding process
    • Identify quality indicators for cell welding
    • Consider properties to monitor for determination of product quality

    Activities

    • Interactive lecture
    • Group discussion: “How do deviations look like?”
    • Case example of laser-based battery welding

    Sensor selection and application

    duration | 0.5 hours

    Learning Outcomes

    • Understand sensor properties
    • Map sensors to process properties
    • Select and apply sensors for specific process features

    Activities

    • Lecture on sensor properties
    • Group exercise: Mapping of sensor capabilities to process features
    • Quiz

    Create an AI data processing pipeline

    duration | 1 hours


    Learning Outcomes

    • Set up a software development environment
    • Select promising AI algorithms 
    • Decide about pre-processing of acquired data
    • Run model training on a split data set
    • Validate the trained model on a test data set

    Activities

    • Excercise: installation and set up of programming environment
    • Lecture with mapping of suitable algorithms
    • Excercise: execute training and validation

    Reserve a seat