摘要：Non- conventional machining processes play a critical role in the manufacturing of advanced components for high-added value sectors such as aerospace and bioengineering. Zero defect manufacturing is a key objective in these sectors, which requires more efficient monitoring techniques than those classically used in other sectors. In the classical approach, the engineer him/herself has to decide the statistical variables from which relevant information about the process will be obtained. Scientific literature shows that threshold levels were manually set for the voltage signal for monitoring and control of the Wire Electrical Machining (WEDM) process applied to aerospace components. However, now that the amount of data available is extremely large (Big Data), the decision on the statistics is not always straightforward. In this context, unsupervised Artificial Intelligence (AI) techniques provide a very interesting approach to the problem. In this paper, unsupervised machine learning techniques are used to extract relevant information from the voltage signal in the wire electrical discharge machining process.
