Document Abstract
Welding machines are widely used in each and every construction works, in various industries, automobiles, and in many more applications. They are large sources for harmonics,
interharmonics and subharmonics. This means, harmonics of different magnitude originate with fundamental frequency. Due to the origination of harmonics of different magnitude, the quality of
power system becomes poor which means power system gets polluted. When the load gets polluted power system, there occurs various kinds of losses and also lays an adverse effect on the working of equipment and also hampers generation, transmission, distribution and utilization. In order to maintain the quality of power distribution, it is important to find out the components of harmonics in the power system. An efficient measurement instrument should be used to find out the harmonic component in every power system. Different methods are used from time to time for this. In recent years, neural network has got special attention by the researchers because of its simplicity, learning and generalization ability and it has been applied in the field of engineering. The theory of neural network is becoming more and more mature and is also making certain breakthrough progress in various fields. It has the advantages of parallel information processing, learning, distribution patterns and memory which can be used in the measurement of the harmonic to construct an appropriate network. In this paper we used radial basis function neural network to find out the components of harmonics in power system generated by welding machine.