Performance Evaluation and Optimisation of Surface Grinding Process for Grinding of Aluminium Based Metal Matrix Composites using Response Surface Methodology and A Novel Genetic Algorithm Approach

Abstract

Aluminium-based metal matrix composites (MMCs) reinforced with ceramic particles are the advanced materials known for their good damping properties, high specific strength and high wear resistance. MMCs are increasingly used in aeronautical and automobile industries and in military applications. In addition, the sporting goods industry has also been in the forefront of MMCs development to capitalise on the materials high specific properties. Despite many advantages, full implementation of MMCs is cost-prohibitive. This is partially due to the poor machinability of the materials. Although near-netshape MMC components can be produced, finishing is still required for obtaining the desired dimensional accuracy and surface finish. Significant cost and fabrication problems, including machining, must be overcome for the successful application of these composites. Surface finish and surface integrity are important for surface sensitive parts subjected to fatigue. Unconventional processes produce better surface finish but they results in subsurface damage during the machining of MMCs. Hence, finishing processes such as grinding and allied abrasive machining are used to improve the surface integrity of machined MMCs. The grindability of aluminium-based MMCs reinforced with ceramic particles is investigated in this dissertation. By the analysis of variance, a complete realization of the grinding process and their effects was achieved. Mathematical model is established for specific energy, metal removal rate and surface roughness from Response surface methodology (RSM). The main objective of this research is to determine the favourable grinding conditions for aluminium-based MMCs reinforced with ceramic particles. Not many researchers have attempted the optimisation of the surface grinding process by considering the specific energy as a performance parameter during grinding of MMCs. A novel approach of multi-objective optimization based on Genetic Algorithm and Desirability function approach was conducted to achieve the desired objective. Very few research works have been attempted towards multi objective optimisation involving surface roughness, metal removal rate and specific energy as the performance parameters in total. The first part of the presented research concentrates on influence of process variables on specific energy, metal removal rate and surface roughness obtained ingrinding of Al6061-SiC35P composites using Taguchi’s design of experiments. From the above investigation, it is observed that feed is the dominant factor affecting the specific energy. Depth of cut is the dominant factor affecting the Metal removal rate and volume percentage of SiC is the dominant factor affecting the surface roughness. The second part of presented research concentrates on mathematical modelling RSM. From the study, it is revealed that the second order RSM model developed for the performance parameters indicates good fit with the experimental results. Desirability function approach for multi-objective optimisation is adopted to choose the process variables that are favourable to achieve the optimal values of specific energy, metal removal rate and surface roughness. The third part of the research involves the application of novel genetic algorithm on multi objective optimisation of specific energy (u), metal removal rate (Qw) and surface roughness (Ra). The results obtained from this novel genetic algorithm were compared with RSM and the results obtained were in fairly close agreement. Finally, the confirmatory experiments were carried out to validate the results obtained from RSM and novel genetic algorithm. From the experiments, it was observed that, deviation between the experimental and predicted responses were within 14%. However novel genetic algorithm compilation consumes less amount time in comparison to conventional non-dominated genetic algorithm (NSGA-II). Hence from the study, it can be concluded that the developed novel genetic algorithm model can be effectively used for the prediction of specific energy, metal removal rate and surface roughness. The understanding gained from Taguchi’s design of experiments, RSM, Desirability function approach and novel genetic algorithm in this research can be used to develop future guidelines for grinding of aluminium-based MMCs reinforced with ceramic particles.