In recent years much research has been conducted to study the variations in welding parameters and consumables on the mechanical properties of steels to optimize weld integrity. The quality of weld is a very important working aspect for the manufacturing and construction industries. In the present work, an attempt has been made to apply an efficient technique, fuzzy based desirability method to solve correlated Multiple response optimization problems, in the field of flux cored arc welding. This approach converts the complex Multiple objectives into a single fuzzy reasoning grade. Based on fuzzy reasoning grade, optimum levels of parameters (Welding current, arc voltage and electrode stickout) were identified. Experiments were performed based on Taguchi method. Weld bead hardness and material deposition rate are selected as quality targets. Significant contributions of parameters are estimated using Analysis of Variance (ANOVA). Confirmation test is conducted and reported. It is found that the electrode stickout is the most significant controlled factor for the process according to the weighted fuzzy reasoning grade of the maximum weld bead hardness and material deposition rate. The proposed technique allows manufacturers to develop intelligent manufacturing system to achieve the highest level of automation.

Radial Basis function Neural Network (RBFNN) is a type of artificial neural networks used for supervised learning. They rely on radial basis functions (RBFs), nonlinear mathematical functions employed to approximate complex nonlinear data. Determining the architecture of the network is challenging, impacting the achievement of optimal learning and generalization capacities. This paper presents a Multi--objective model for optimizing and training RBFNN architecture. The model aims to fulfill three objectives: the first is the summation of distances between the input vector and the corresponding center for the neurons in the hidden layer. The second objective is the global error of the RBFNN, defined as the discrepancy between the calculated output and the desired output. The third objective is the complexity of the RBFNN, quantified by the number of neurons in the hidden layer. This innovative approach utilizes Multiple objective simulated annealing to identify optimal parameters and hyperparameters for neural networks. The numerical results provide accuracy and reliability of the theoretical results discussed in this paper, as well as advantages of the proposed approach.

Multi-hole orifices have better performance than single-hole orifices. In this paper, Multi-objective optimization of Multi-hole orifices is performed using a Fluid-Solid Interaction (FSI) analysis and Multi-objective genetic algorithm (NSGA II). In all numerical analysis, the governing equations of the solid and the governing equations of the fluid are carried out for orifice and fluid around orifice respectively. All calculations are made for a 16-hole orifice with circular holes. The design variable in the optimization process is the distance between the holes of the orifice and thus the amount of shrinkage or expansion of the orifice geometry. The objective functions are the pressure drop created on the sides of the orifice, the deformation and tension created in the orifice structure, which should be maximized, minimized and minimized respectively. In the results section, the Pareto front are presented which represent useful information for designing the Multi-hole orifices geometry, and five orifices are also introduced as final design options that have better performance. The results of the sensitivity analysis of the various parameters are also presented and discussed in detail in the Multi-hole orifices.