Abstract
The applications of Wire Arc Spraying (WAS) include large-area corrosion protection coatings e.g. the protection of off-shore wind power plants. While WAS is cost efficient and well-known, the inherent instabilities can lead to coating defects and subsequent vulnerabilities in the corrosion protection coating. The occurrence of these process-related fluctuations cannot be predicted by deterministic models. However, these fluctuations can be monitored in situ, analyzed and finally minimized. A sensor unit is set up on the free jet of a WAS process using ZnAl15 wire. Voltage, amperage, noise and wire feed rate are measured in situ at a sampling rate of 80 MHz. Following a design of experiments approach, 64 different parameter settings are run and measured. For that purpose, voltage, atomizing gas and wire feed rate of the free gas jet have been varied. A generalized linear model (GLM) is trained on the dataset. A Fast Fourier Transformation (FFT) in conjunction with smoothing filters is conducted. Adopting the GLM enabled the calculation of parameters that minimize process fluctuations. Plots in the form of response surfaces depict the influence of the varied parameters on the process stability. A signal analysis using FFT revealed major periodic changes of the voltage in the range of 0.5-1 kHz next to process control-related frequencies at 20 kHz. The mounting and structuring of the data as well as the calculation of key figures is fully automated. Due to the high degree of automation, large quantities of data can be processed. In the future, a simplified version of the adopted sensor unit may be adopted to optimize parameters in an autonomous way. This can ensure not only the minimization of process fluctuations for any chosen feed wire, but also indicate irregularities in the process. The high-resolution recording and automated analysis of the data allows the determination of optimized parameters as well as major underlying frequencies.
