Ushasta Aich

Application of teaching learning based optimization procedure for the development of SVM learned EDM process and its pseudo Pareto optimization

Two-stage soft computing ((SVM-TLBO)-(PLM-TLBO-pseudo PARETO)) based virtual system of manufacturing process - EDM is developed.Virtual data generator of EDM process learned by support vector machine (SVM) with internal parameters (C, ? and ?) tuned by teaching learning based optimization (TLBO) is reported.Modifications namely population based termination criteria, initialize population with high dispersion and way of choosing teacher in case of multiple best learners performing same score, over standard TLBO are suggested. Further, a comparison between performances of TLBO and PSO in model development is studied.A simple procedure for pseudo Pareto front development by modified TLBO is proposed.Inverse solution procedure for selection of optimum available machine parameter setting corresponding to specific output combination is elaborated. Manufacturing processes could be well characterized by both the quantitative and the qualitative measurements of their performances. In case of conflicting type performance measures, it is necessary to get possible optimum values of all performances simultaneously, like higher material removal rate (MRR) with lower average surface roughness (ASR) in electric discharge machining (EDM) process. EDM itself is a stochastic process and predictions of responses - MRR and ASR are still difficult. Advanced structural risk minimization based learning system - support vector machine (SVM) is, therefore, applied to capture the random variations in EDM responses in a robust way. Internal parameters of SVM - C, ? and ? are tuned by modified teaching learning based optimization (TLBO) procedure. Subsequently, using the developed SVM model as a virtual data generator of EDM process, responses are generated at the different points in the experimental space and power law models are fitted to the estimated data. Varying the weight factors, different weighted combinations of the inverse of MRR and the ASR are minimized by modified TLBO. Pseudo Pareto front passing through the optimum results, thus obtained, gives a guideline for selection of optimum achievable value of ASR for a specific demand of MRR. Further, inverse solution procedure is elaborated to find the near-optimum setting of process parameters in EDM machine to obtain the specific need based MRR-ASR combination.

A simple procedure for searching pareto optimal front in machining process: electric discharge machining

Optimum control parameter setting in complex and stochastic type processes is one of the most challenging problems to the process engineers. As such, effective model development and determination of optimal operating conditions of electric discharge machining process (EDM) are reasonably difficult. In this apper, an easy to handle optimization procedure, weight-varying multiobjective simulated annealing, is proposed and is applied to optimize two conflicting type response parameters in EDM--material removal rate (MRR) and average surface roughness (Ra) simultaneously. A solution set is generated. The Pareto optimal front thus developed is further modeled. An inverse solution procedure is devised so that near-optimum process parameter settings can be determined for specific need based requirements of process engineers. The results are validated.

Multi-objective optimisation of abrasive water jet machining responses by simulated annealing and particle swarm

Optimum control parameter settings in complex and stochastic type processes are one of the most challenging problems to the process engineers. As such, effective model development and determination of optimal operating conditions of abrasive water jet machining process (AWJM) are reasonably difficult. In this article new modifications are proposed on two single-objective optimisation techniques, simulated annealing and particle swarm optimisation and applied to optimise two response parameters in AWJM – material removal rate (MRR) and depth of cut (DOC) simultaneously. For the optimisation purpose statistical models are developed from experimental data obtained from AWJ machining of amorphous material – borosilicate glass. Optimum results are validated and it is suggested to employ particle swarm-based proposed multi-objective optimisation technique in present case for newly modelled system due to its fast convergence and low memory space requirement.

Evaluation for Chaos in EDM Generated Surface Topography

Machined surface carries the inherent features of machining process. Investigation of surface topography generated by machining process is helpful to extract the features of surface development process. In the present study, roughness profiles measured on machined surface generated by EDM are considered as time series and used for extraction of inherent features of surface topography through phase space reconstruction. Presence of self-similarity in surface topography is assessed by estimating a second order fractal dimension, called as correlation dimension. Saturation of correlation exponents with the increase of embedding dimension indicates the presence of chaos in surface topography.

Support vector machine-based unified learning system for prediction of multiple responses in AWJM of borosilicate glass and SEM study

Modelling of responses in any manufacturing process is helpful for working in virtual world. As such, effective model development of stochastic processes working on heterogeneous materials is reasonably difficult. Hence, a robust unified learning system, multi-objective modelling with SVM, is proposed in this work to study the gross erosion behaviour of borosilicate glass in abrasive water jet machining. In this study, experiments are conducted on borosilicate glass with variation of the control parameters - water pressure, abrasive flow rate, traverse speed and standoff distance. Two process responses - material removal rate (MRR) and depth of cut (DOC) are trained through support vector machine (SVM)-based learning system for regression. An optimised single set of internal parameters of SVM, that would predict both MRR and DOC with their respective Lagrange multipliers, is estimated by minimising the training errors with the help of particle swarm optimisation (PSO) procedure. A modification of PSO is also proposed in this article. Further, scanning electron micrographs of cut wall are qualitatively examined to reveal the possible erosion behaviour of the amorphous material - borosilicate glass.

Searching for a Pareto Optimal Solution Set of EDM Responses Applying Multi-Objective Simulated Annealing on RSM Model

Simultaneous optimization of conflicting type responses like material removal rate (MRR) and average surface roughness (Ra) in stochastic type electrical discharge machining (EDM) process is a matter of concern to the process engineers. In this paper, EDM is first modeled by response surface methodology (RSM). Current setting, pulse on time and pulse off time were taken as the input parameters while material removal rate and average surface roughness as the responses. Multi-objective simulated annealing (MOSA) is then applied on these models. Pareto optimal solution set is thus developed. It would assist a process engineer to take decision regarding the optimal setting of the process parameters for a specific need-based requirement.