Modeling and Predicting Abrasive Wear Behaviour of Poly Oxy Methylenes Using Response Surface Methodolgy and Neural Networks

Abrasive wear is defined wear due to hard particles or hard protuberances forced againts and moving along a solid surface. This defination encompasses several different mechanisms by which material removal occoures. The five types of wear are abrasive, adhesive, erosion, fatigue and fretting. The abrasive wear of polymeric materials is the interest and the subject of quite number of literature. Sahin 1 carried out experiments to analyze the influence of applied load, abrasive grain size and sliding distance on weight loss of metal matrix and its composite using Taguchi method. Sahin and Ozdin 2 investigated the abrasive wear behaviour of aluminium based composites using pin on disc type of machine and developed in terms of the applied load, sliding distance and particle size using factorial design. Sahin 3 studied wear behaviour of aluminuim alloy and its composites reinforced by SiC particles using statistical analysis and he expressed in terms of applied load, sliding distance and particle size using a linear factorial design approach. Franklin 4 focused on the wear performances of several engineering polimer based materials under dry reciprocating sliding conditions to estimate the wear. Lin and Chou 5 used to response surface method to express the wear rate parameter and reciprocal of the contact temperature as a function of sliding speed and applied load. The solution predicted by the polynomials were compared with the experimental results. Farias et al. 6 studied the sliding wear of austenitic stainless steels. They adopted to obtain an empirical model of wear rate as a function of applied load and sliding velocity using RSM. Shipway and Ngao 7 investigated the abrasive behaviour of polimeric materials in microscale level. They concluded that the behaviour rates of polymers dependent critically on the polymer type. Durmus et al. 8 used neural networks for the prediction of wear loss and surface roughness of AA 6351 aluminium alloy. Zhang et al. 9 applied an ANN model to predict the erosive wear of three polymers. Unal et al. 10 studied abrasive wear behaviour of aliphatic polyketone (APK), polyoxymethylene, ultrahigh molecular weight (UHMWPE) polyethylene, polyamide 66 (PA 66), and 30 % glass fibre reinforced polyphenylene sulfide (PPS+30 %GFR) engineering polymers at room temperature using pin on disc. Tests were at 1 m/s test speed, load value of 10 N and different sliding distances. Liu et al. 11 studied the influence of the parametres; sliding distance, contact pressure and sliding speed on the wear performance of polyamide and UHMWPE using statistical wear analysis.