Modeling and Control of Nd:YAG Laser Percussion Drilling of Nickel-Based Super Alloys

Abstract

Non-Conventional methods have recently been employed in the drilling of materials such as super alloys which are hard to machine using mechanical conventional methods. These materials are mainly used in turbines in the power generation and aerospace industries. One of the recently developed non-conventional methods for drilling is using pulsed lasers whereby a laser beam is focused to a spot equal in diameter to the hole to be drilled. Pulsed lasers such as Nd:YAG are mainly employed for this process. Laser percussion drilling is commonly employed to produce holes with small diameters, usually less than 1mm. With percussion drilling, the control of hole parameters such as taper, entrance and exit hole variation and roundness is di cult and these parameters are of utmost importance for small holes. Selection of machining parameter combinations for obtaining optimum circularity at entry and exit and minimum hole taper is a challenging task owing to the presence of a large number of process variables. There is therefore need to develop a control system that is able to adjust the various process parameters to the optimum values and hence control the drilling process. The main aim of this study was to model the laser percussion drilling process and to develop a control system for optimizing the process by controlling the laser power and pulse duration based on the hole diameters and taper. A neuro-fuzzy strategy for control of machining parameter settings for the generation of the maximum circularity at entry and exit and minimum hole taper was adopted. A controller based on adaptive neural fuzzy inference system (ANFIS) was developed for the laser percussion drilling process. In this study, the e ects of laser machining parameters, namely laser power and pulseduration, were successfully investigated. These have been proven to bear a great in uence on the laser drilling process. A model for predicting the optimal laser parameters for quality drilling of Nickel-based super alloy-Inconel 718 was successfully developed. Inconel 718 was chosen since it is one of the commonly employed nickel alloys in the turbines. The model satisfactorily predicted the hole tapers and diameters, given inputs of laser power and pulse duration. The results showed that both laser power and pulse duration have great in uence on the hole geometric parameters. Hole taper reduces with increase in both laser pulse duration and laser power. Hole diameters increase with increase in both laser power and pulse duration. A neuro-fuzzy controller was then developed based on MATLAB R and LabV IEW R platforms. The controller was used to control the laser percussion drilling by simulating a drilling environment. It was demonstrated that a neuro-fuzzy based controller e ectively controls the hole diameters and taper through in-process adjustments of laser power and pulse duration. While using the controller, the diameters increase with increase in peak power and pulse duration upto an optimum level beyond which the peak power and pulse duration remain constant. The hole taper decreases with increase in peak power and pulse duration upto an optimum level beyond which the peak power and pulse duration remain constant. The ANFIS based neuro-fuzzy contoller read the input values from a spreadsheet le. Using the learning capabilty of arti cial neural networks, the controller was able to compute the optimum values for laser peak power and pulse duration. Arti cial neural networks have excellent capabality for approximation of process performance. Thus, the controller helps maintain the peak power and pulse duration at optimum levels.