A Study On The Prediction Of The Surface Roughness Of The Cutting Surface Using Elman Neural Network

As the ship's system becomes more complex, the type and quantity of pipes to be installed constantly increase. In particular, the offshore pipeline transports crude oil and gas developed in the sea and transports and transmits various fluids such as water and air. It is an essential element. Cutting in the manufacturing process of the pipeline is the first step in the process, and high-quality cutting work is indispensable. In this study, a prediction model was developed using an Elman neural network to secure and predict the surface roughness of the pipe cutting surface in the plasma cutting process. In the experiment, the arc current and cutting speed were selected as process variables, and the effect of the process variables on the surface roughness was analyzed using the experimental design method. Based on the result data of grasping the effect of process variables on the surface roughness, the quality of the cut surface was predicted using the Elman neural network, and the predictive ability was confirmed by comparing it with the experimental value.