Break-Out Detection And Adaptive Control For Small Hole Fast EDM Drilling

Small hole fast electrical discharge machining(EDM)drilling is one of the EDM technologies,which is widely used to produce groups of film cooling holes on turbine blades of aeroengines and gas turbines.An entire drilling process of a through hole can be divided into three stages,i.e.the touch-down stage,the drill-in stage,and the breakout stage.At present,there are two problems in the small hole fast EDM drilling.First,since the characteristics of different stages have significant differences,the common control strategy that uses fixed machining parameters impedes the improvement of machining efficiency and stability.Second,due to the severe tool wear,it is difficult to determine when the hole is completed.Especially when drilling holes like film cooling holes on turbine blades that have inner channels inside,the hole completion has to be determined effectively to avoid any back-strikes.In order to improve the overall machining efficiency,it is necessary to distinguish different machining stages and adopt a stage-wise adaptive control strategy.And to determine the hole completion,a completion determination method should be proposed on the basis of the online distinguishing of machining stages.However,the research on the causes of distinct characteristics of stages,which should serve as the basis of the optimized control strategy,is still lacking.Besides,since the tool suffers from severe wear and the process has a strong stochastic nature,the information of breakout,which marks off the drillin stage and the breakout stage,and hole completion,which indicates the end of the processing,cannot be acquired from the amount of electrode feed and processing time.Detection methods for the two events should also be proposed according to the characteristics of the machining process.Existing methods,however,have poor reliability.To this end,this dissertation conducted systematic research on the small hole fast EDM drilling process to address these problems.This dissertation focuses on the following issues.(1)In order to find the causes of distinct characteristics of each stage,an experimental setup was designed and direct observations were conducted through cameras on different time scales.Physical phenomena happening in the discharge gap were analyzed to have a deeper comprehension of the process and guide the subsequent researches.In the touch-down stage,the forced vibration of the electrode was observed.Analysis shows that the vibration is the root cause of the instability in this stage.In the drill-in stage,it was observed that electrode vibration and discharges in the side gap were common.This stage features high efficiency of debris evacuation and general machining stability.After stepping into the breakout stage,the flushing effect remained for a while right after the hole outlet appeared,but side-gap discharges surged and the machining condition deteriorated rapidly.After the hole is completed,massive debris accumulated in the gap and side-gap discharges and short circuits were still common.(2)Mathematical and finite element models were built for the observed phenomena and semi-quantitative simulations and analyses were conducted to find the underlying governing laws,which lays the foundation of optimized control strategies.The equation of the forced vibration of the electrode in the touch-down stage was derived by using the fluid-structure interaction(FSI)theory.It was found that the free length of electrode is the key to minimize the vibration amplitude.A finite element FSI model was built to analyze the electrode vibration in the drill-in stage and the breakout stage as well.Simulation results demonstrated that the pressure of the working fluid in the side gap have little impact on the vibration while the discharge location is a more significant factor.Then a series of fluid-particle interaction two-phase flow finite element models were built to study the debris evacuation before and after the breakout.The simulation results inferred that the concentration of discharge location and the difference of fluid pressure lead to more side-gap discharges and short circuits,resulting in the deterioration of machining condition during the entire breakout stage.(3)Since reliable detection methods are still lacking,considering the results of observation and simulations,the variations of discharge signals were investigated and the methods of breakout detection and hole completion determination were proposed.First,the classification of machining state graph(CMSG)method was proposed for breakout detection.The CMSG method employs multiple feature signals,and the changing trend curves of them are obtained the combined to generate the machining state graph(MSG).The MSGs are classified online into two groups,i.e.before breakout and after breakout,by a machine learning algorithm.In contrast with the present methods the CMSG method is more reliable to avoid misjudgments caused by signal fluctuation and disturbances.Then,a hole completion determination method was proposed by analysis of the discharge signals before and after the hole completion when the electrode is not retracted.Experimental results proved the validity and effectiveness of the proposed methods.Namely the CMSG method can correctly detect the breakout event within 1 second and 1 mm,and the completion determination can be made before the electrode was fed 2 mm out of the hole outlet.(4)In order to improve the stability and efficiency in different stages of the drilling process,based on the research in previous sections,stage-wise adaptive control strategies were proposed and implemented.Firstly,in the touch-down stage,the feedrate gain factor of the discharge gap servo controller was dynamically adjusted according to the machining status.So that the negative influence of electrode vibration can be minimized.Experimental validations show that the machining time of this stage was reduced by more than 50% and the discharge affected zone around the hole inlet was also reduced by up to49.8%.Secondly,when drilling holes with high aspect ratio,the machining efficiency and stability in the drill-in stage decreases gradually as the hole becomes deeper.To cope with the problem,a double-input double-output adaptive controller was designed based on self-tuning regulation.The adaptive controller dynamically regulates the parameters of the original servo controller.Experimental results show that the adaptive controller reduced the machining time by 10.3%.Lastly,in view of the complex and harsh machining condition in the breakout stage,machining parameters are optimized to improve the efficiency.A full factorial experiment was designed and conducted to study the impacts of the servo reference voltage and the gain factor of the controller on machining time.Preferable parameters were screened out,and experimental results show that the machining time of this stage can be reduced by up to 56.2%.In the end,the adaptive control strategies and event detection methods are combined to automate the entire drilling process.