Determination of source voltage from audible corona noise by neural networks

In this study, a different application of the signal recognition is presented for classification of source voltage level, which leads to produce corona noise in an experimental set-up, using recorded sound data of corona (electrical discharge) and utilizing probabilistic neural network (PNN). By applying different levels of 50 Hz ac high-voltage, the corona sound data are acquired experimentally from a test set-up intentionally producing corona sound. After successfully recording the sound data, they have been used in training and test sets of the probabilistic neural network. In this context, we can indicate the main objective for our study; to develop a model to determine exact source voltage level by only analyzing the recorded corona sound data. During the application of algorithmic method, linear prediction coefficients are used. It is shown that reasonable results can be obtained by following the proposed method.