Weld Quality Prediction Using Artificial Intelligence Technique
SAMRIDDHI A Journal of Physical Sciences Engineering and Technology
Decision-making process in manufacturing environment is increasingly difficult due to the rapid changes in design and demand of quality products. To make decision making process online, effective and efficient artificial intelligent tools like neural networks are being attempted. Usually, the desired welding parameters are determined using traditional methods like welder's experiences, charts and handbooks (preferred values) which are simple and inexpensive. But this does not ensure that the
... ensure that the selected welding parameters result in satisfactory welding and this method is not applicable to new welding process. To overcome this problem, various methods of obtaining the desired output variables through models to correlate input variables with output variables have been developed. Fractional factorial techniques, Mathematical modeling, curvilinear regression equations, linear regression equations, response surface methodology, finite element modeling, grey-based Taguchi method and sensitivity analysis were used to model SAW process. These methods are limited in application due to difficulties in modeling, time consuming and cumbersome. Due to the inadequacy and inefficiency of the mathematical models to explain the nonlinear properties existing between the input and output parameters of welding lead to the development of intelligent modeling techniques. Precise simulation and analysis of the process needs attention which helps to predict the wide variety of process parameters to set the factory floor in real time. The type of artificial intelligence capable of responding to changes in the automated manufacturing environment, and having the ability to capture vast manufacturing knowledge is Adaptive Neuro Fuzzy Inference System (ANFIS). It is becoming widely used in all aspects of manufacturing process to assist humans. Realizing that matter, ANFIS a state of the art artificial intelligent method, has the possibility to enhance the prediction of weld quality to find the best combination of independent variables which is welding current (I), speed (S) and welding voltage (V) as the input variables in order to achieve desired weld quality. Thus, the main objectives of this project is to develop ANFIS model to predict weld quality.