Predicting Operating Windows for High-Frequency Induction Aluminum Tube Welding

Abstract

High-frequency (HF) induction welding is a practical welding technique widely used in various industries. Although it is generally robust, HF induction welding of aluminum tubes is complicated by the very high line speed, which requires high and accurate power input, and, therefore, a small fluctuation or variation in power input could result in drastically different welds. This work is dedicated to analyzing the influence of welding parameters, line speed, power input, and other unknown random factors, such as those induced by weather or work shift, especially those induced by the change of aluminum stock and adjustment/maintenance of the induction welding coil. Through the machine learning process, statistical models defining the normal operating windows were developed based on experimental data. The operating windows, defined by the overheat-normal and normal-cold boundaries, are expressed in terms of probabilities of producing normal welds. These trained models can be used to make new predictions, i.e., new operating windows, by collecting a small sample (a very limited number of calibrating data points). This procedure was verified experimentally.