Research on discharge detection system of electric discharge machining based on neural network

Electric discharge machining (EDM) uses electrical energy to etch and remove materials instead of traditional machining methods. It is well-suited for processing geometrically complex surfaces and difficult-to-process hard materials, and has broad applications in the fields of mould-making, aerospace, and the military. In order to obtain satisfactory processing quality, it is essential to maintain an appropriate gap discharge state for EDM. Accurately detecting the gap discharge state in EDM plays a crucial role in ensuring high quality and efficiency during processing. Based on a review of both national and international literature, this paper proposes a method to identify the gap discharge state during EDM machining using a trained neural network. This approach aims to improve the accuracy and recognition rate of the discharge state due to its complexity and unpredictability. Furthermore, it will offer valuable insight into enhancing the servo system. After undergoing experimental verification, the recognition of the discharge state has shown a higher level of consistency with the given value.