Study of the effect of the cooling and lubricant fluid on the cutting performance of dimension stone through artificial intelligence models

• Main Authors: Seyed Mehdi Hosseini, Mohammad Ataei, Reza Khalokakaei, Reza Mikaeil, Sina Shaffiee Haghshenas
• Format: Article
• Language: English
• Published: Elsevier 2020-02-01
• Series: Engineering Science and Technology, an International Journal
• Online Access: http://www.sciencedirect.com/science/article/pii/S2215098618321761

In this study aims to provide an accurate practical anticipation model for investigating the effect of cooling and lubricant fluid on the cutting performance with a special emphasis on the consumed electrical current as one of the most important factors influencing the performance of cutting disks. In this study, 10 dimension stone samples were tested as hard rocks using three types of lubricant fluids, and totally the modelling was conducted based on 160 collected laboratory test samples for each fluid. The modelling was done using two methods of artificial neural networks (ANN) and the hybrid genetic algorithm – artificial neural network algorithm (hybrid GA-ANN algorithm). Furthermore, four influential physical and mechanical parameters of rocks, including the uniaxial compressive strength, Mohs hardness, Schimazek’s F-abrasiveness factor and Young modulus, and two operational parameters, including feed rate (Fr) and depth of cut (Dc) were used as input data, and also the maximum electrical current was used as the output data in these simulations. Finally, the results obtained by the algorithm performance indices such as the value account for (VAF), root mean square error (RMSE), and coefficient of determination (R2) were evaluated. According to the obtained results, totally 18 models were built which the Multi-Layer Perceptron (MLP) Neural Network had the highest efficiency for providing an accurate and stable anticipation model compared to the hybrid ANN-GA algorithm. In addition, the proposed MLP models prove that Soap Water with ratios of 1–40 can provide higher performance capacity in cutting process for reducing the consumed electrical current. This network is used for anticipating the maximum electrical current consumed by the cutting machine in order to evaluate the performance of sawing blade in hard rocks and the possibility of a good planning to increase the efficiency of machine using different lubricant fluids. Keywords: Sawing blade, Hard rocks, Lubricant fluids, ANN-GA, MLP, Intelligent techniques