System Identification And Adaptive Control Of Small Hole Edm Drilling Processes

Small hole electrical discharge machining(EDM)drilling machines are specially designed for fast hole drilling,high aspect ratio drilling and,complicated cavity drilling.Small hole EDM drilling is extensively applied to machine cooling film holes of turbine blades in aeroengine and spray holes in diesel injectors.When machining deep holes,process stability can be negatively influenced by uncertain retractions caused by difficult debris evacuation or adhesive conditions between an electrode and a workpiece caused by a fast feedrate,which makes it hard to guarantee a high machining efficiency and a good surface finish.This dissertation aims to address this problem by developing adaptive controllers designed for small hole EDM processes based on an embedded platform.Firstly,a computer numerical control(CNC)system including upper and lower machine is constructed for the need of software transplant.Upper machine includes user interface and logical proceeding core whilst lower machine is composed of servo control,movement control,data communication and,motor driver modules.A pulse classification system for small hole EDM drilling is designed to provide EDM process monitoring indices,whose accuracy is evaluated up to 99.75%.Secondly,to solve the problems of process instability and hole inconsistency in small hole EDM drilling,adaptive controllers are designed based on ARMAX polynomial models which is able to cope with the timevariant and stochastic characteristics of EDM processes.Non-parameter identification through correlation analysis and spectrum analysis is conducted to obtain the a priori system knowledge on causality and stability.Then the model set is ascertained by parameter estimation and further evaluated by a residual test and a verification test.At last,online model is deduced by a recursive least squares which is solved by the UD factorization numerical method.Based on the ARMAX model,a sliding mode controller,a self-tuning regulator and a minimum-variance controller are designed for process stability.As a comparison,a fuzzy logic controller is also brought in based on discharging status.Finally,these control methods are testified by simulation.And high aspect ratio drilling experiments are then carried out for electrodes with diameter of 0.3mm,1.0mm,3.0mm.Machining tests show that within the adaptive controllers,the machining efficiency is improved by 39.25% and the electrode wear is reduced by 17.33% at most.Furthermore,standard deviation of machining time and electrode wear are improved mostly by 78.85% and 87.73% respectively,which prove the availability of adaptive controllers in improving process stability of small EDM drilling.