Study On Detection Of WEDM Discharge Status Of Spray Based On LVQ Optimization

Wire electrical discharge machining(WEDM) in gas and water spray has been developed in recent years. During processing, compared to electrical discharge machining in liquid,the discharge of energy is less and inter-electrode gap is small and the machining process has no pollution to the environment,etc.In WEDM, the gap discharge status analysis is helpful to reveal the machining mechanism of the gas and water spray in the medium. Therefore, it is significant to study the high performance gap discharge status detection system to optimize the machining parameters, to ensure the machining quality and to improve the machining efficiency. On the basis of consulting a large amount of documents and materials, The detection method of discharge state in traditional WEDM is analyzed, Combined with a variety of intelligent detection technology at home and abroad, and based on the learning vector quantization neural network, the discharge status detection system is proposed to identify the gap discharge status in real time. In the process of analyzing the state of gap discharge, the discharge status usually includes open status, normal discharge and short status. In this paper, the effects of partial open status and short circuit on the machining process are also studied, The experiments show that a longer period of partial open pulses or partial short pulses may cause the subsequent pulses become open or short pulses, which can affect the machining precision and machining efficiency. So detecting the partial open and partial short discharge status in WEDM will help to achieve the best control of electrical discharge machining.In this paper, gap voltage and current are selected as the input signals of the detection system. The data acquisition card is used to complete the collection of the WEDM gap signals. Based on the Lab VIEW virtual platform theory, the learning vector quantization neural network is used to set up the logic processing program and the appropriate hardware is chosed to build the corresponding discharge state detection system. In the end, the real-time detection of five kinds of discharge status, which isopen, partial open circuit, normal discharge, partial short circuit and short circuit, is realized.During the processing, the digital storage oscilloscope is used to collect the signal of discharge gap, the waveform obtained by the digital storage and the signal waveform of the detection system are compared to verify the reliability of the detection system.Then the experiment of rough machining in liquid is carried out. The detection system is used to study the discharge gap characteristics of liquid, gas and water spray. The discharge waveform and the characteristics of discharge status and the distribution range of spark rate are obtained; Finally, with the pulse width, the pulse interval ratio,the peak current, the offset and the feed rate as variable factor respectively, the single factor experiments has been carried out.