Abstract—Shot peening is a process of cold working a part that increase its resistance to metal fatigue and some forms of stress corrosion. Shot peening causes plastic deformation in the surface of the peened part and leads some changes in mechanical and metallurgical properties of it. Artificial intelligence (AI) systems such as artificial neural networks (ANNs) have found many applications to predict and optimize the engineering problems in the last few years. In present study effects of SP on mechanical and metallurgical properties of 18CrNiMo7-6 are investigated by ANN. Network has been developed based on back propagation error algorithm. In order to train the network data of experimental tests results were used. Experimental tests were concluding different SP types: single step SP and dual step SP with different SP intensities. Testing of the ANN is accomplished using experimental data not used during networks training. Distance from the surface and Almen intensity are considered as input parameters and residual stress, remnant austenite content, Cauchy breath, domain size and microhardness are regarded as output parameters of the network. The comparison of obtained results of ANN’s response and experimental values indicates that the networks are tuned well and the ANN can be used to predict the SP effects on mechanical and metallurgical properties of materials.
Index Terms—Step shot peening, mechanical properties, artificial neural network, back propagation algorithm.
E. Maleki is with the Sharif University of Technology-International Campus, Kish Island, Iran (e-mail: maleki_erfan@kish.sharif.edu, maleky.erfan@gmail.com).
K. Sherafatnia is with the Sharif University of Technology, Tehran, Iran (e-mail: sherafatnia@mech.sharif.edu).
[PDF]
Cite: E. Maleki and K. Sherafatnia, "Investigation of Single and Dual Step Shot Peening Effects on Mechanical and Metallurgical Properties of 18CrNiMo7-6 Steel Using Artificial Neural Network," International Journal of Materials, Mechanics and Manufacturing vol. 4, no. 2, pp. 100-105, 2016.