Font Size: a A A

Research On CNC System And Workpiece Height Estimation For Reciprocated Traveling WEDM

Posted on:2018-01-25Degree:MasterType:Thesis
Country:ChinaCandidate:G W HuangFull Text:PDF
GTID:2381330590977527Subject:Mechanical engineering
Abstract/Summary:PDF Full Text Request
Wire electrical discharge machining(WEDM)is widely used in manufacture of molds,and some processing fields with special requirements on materials and precise.With the development of technologies,people have a higher requirement for the quality and precision of WEDM.In the machining process of WEDM,when the thickness of workpiece becomes thinner,the heat density on the wire will increase,and the temperature of wire will also increase,which may leads to wire breakage;on the contrary,the processing speed and the machining efficiency will be reduced.When machining workpiece with a variable height,wire breakage and low machining efficiency are the two main problems that more likely to happen.So this thesis has done a serious research on CNC system,online workpiece height estimation and parameter adaptive control for WEDM processes.In this thesis,a network distribution reciprocating traveling WEDMCNC system based on Digitizer/Player architecture is developed.The Firefly board and Linux system are used as the hardware and software environment,and the CNC system is built all by Qt 5.The digitizer is responsible for non-real-time tasks such as NC code interpretation,motion trajectory encoding and file management;While the STM32 board is used as the running environment of the player,and the player mainly focuses on playing back the motion bit-stream,motion control,I/O control and state detection.The digitizer and player achieve stable communication by using the TCP/IP protocol.And a discharge detection system was also developed to distinguish various discharge states.In order to solve problems when machining workpiece with a variable height,the workpiece height online estimation function is realized by using machine learning.Due to the advantages of fast learning speed and good generalization performance,a workpiece height estimation model is built up based on the Extreme Learning Machine algorithm.The inputs of the model are discharge frequency,processing speed,pulse interval and feed rate,while the output of the model is the height of the workpiece.Then the model is introduced to the WEDM-CNC system by using HP-ELM package.The experiment shows that the processing efficiency is improved by 37% and 35% when machining a stair-shaped workpiece and a trapezoidal workpiece.
Keywords/Search Tags:wire electrical discharge machining, numerical control system, workpiece height estimation, extreme learning machine, HP-ELM
PDF Full Text Request
Related items