ORIGINAL ARTICLE Artificial intelligent modeling to predict tensile strength of inertia friction-welded pipe joints Simranpreet Singh Gill & Jagdev Singh Received: 25 December 2009 / Accepted: 3 July 2013 # Springer-Verlag London 2013 Abstract Adaptive neural network-based fuzzy inference system (ANFIS) is an artificial intelligent neuro-fuzzy tech- nique used for modeling and control of ill-defined and un- certain systems. The present paper proposes this novel tech- nique of ANFIS to predict the tensile strength of inertia friction-welded tubular pipe joints with the aid of artificial neural network approach combined with the principle of fuzzy logic. The proposed model is multiple input–single output type of model which uses rotational speed and forge load as input signals. The set of rules has been generated directly from the experimental data using ANFIS. The per- formance of the proposed model is validated by comparing the predicted results with the actual practical results obtained by conducting the confirmation experiments. The applica- tion of χ 2 test confirms that the values of tensile strength predicted by proposed ANFIS model are well in agreement with the experimental values at 0.1 % level of significance. The proposed model can also be used as intelligent online adaptive control model for pipeline welding. Keywords ANFIS . Artificial intelligence . Tensile strength . Inertia friction welding . Artificial neural network . Fuzzy logic 1 Introduction Pipelines in one form or another have been used since the early days of civilization for transportation of liquid elements. This is mainly because of its economical mode of transporta- tion compared to any other modes of transportation, especially for liquid and gaseous elements. Today, pipes have been used for transportation of various liquid and gaseous elements like water/sewerage, oil, and gas over long distances. The need to construct pipelines over long distances has led to an increased demand to improve the productivity of pipeline girth welding [1]. Many novel techniques have been tried in the past to achieve productivity gains, including laser welding, flash butt welding, homopolar welding, and inertia friction welding. Inertia friction welding is pioneering pipe girth welding tech- nique and has been optimized in the past to produce the maximum productivity possible with this process. The advan- tages of the process are high reproducibility, short production time, and low energy input. Inertia friction welding is a complicated process, in which the heat for welding is produced by the direct conversion of mechanical energy stored in rotational flywheel to thermal energy at the welding interface of the welding parts. It does not need to apply electrical energy or heat from other sources to the parts. It is widely used in various industry fields espe- cially in pipe girth welding industries. In this welding process, the joining surface of the samples is heated to the desired temperature through frictional heat, and then a forging pres- sure is introduced to weld the parts. Inertia friction welding can be used to join metals of widely differing thermal and mechanical properties. Often, the combinations that can be inertia friction-welded cannot be joined by other welding techniques because of the formation of brittle phases, which make the joint poor in mechanical properties. The submelting temperatures and short weld times of inertia friction welding allow many combinations of work metals to be joined [2, 3]. The inertia friction welding process is a nonlinear process because of the interaction between the temperature field and the material properties as well as the friction force [4]. In practice, many aspects of the inertia friction welding process are difficult to detect experimentally. This is particularly true for tensile strength and the associated constitutive behavior of the material as it softens and deforms plastically. This situation means that the analysis of the phenomena occurring S. S. Gill (*) : J. Singh Department of Mechanical Engineering, Beant College of Engineering and Technology, Gurdaspur 143521, Punjab, India e-mail: Ritchie_223@yahoo.com Int J Adv Manuf Technol DOI 10.1007/s00170-013-5177-5