This project is done in cooperation with Charmilles Technologies in Meyrin, Geneva, which is one of the world leaders in the field of Electro-Discharge Machining. The process is quite complex, the behaviour is stochastic and no reliable models are available. Also, the a large set of parameters need to be chosen for different regimes and their optimal values vary considerably. The main goal of the project is to use a new optimization scheme based on measurements rather than on the knowledge of a model. The main challenges are i) to getting reliable measurements out of machining sequences and ii) use these noisy measurements in an optimization algorithm. So, the role of statistics and stochastic modelling plays a key role in this project.
Funded by the Commission pour la technologie et l’innovation (CTI)
Keywords : ;