Application Research Based on Artificial Neural Network (ANN) to Predict No-Load Loss for Transformer's Design

Transformer is one of the vital components in electrical network which play important role in the power system. The continuous performance of transformers is necessary for retaining the network reliability, forecasting its costs for manufacturer and industrial companies. The major amounts of transformer costs are related to its no-load loss, so the cost estimation processes of transformers are based on reduction of no-load loss. This paper presents a new method for classification of transformer no-load losses. It is shown that ANNs are very suitable for this application since they present classification success rates between 78% and 96% for all the situations examined. The method is based on Multilayer Perceptron Neural Network (MPNN) with sigmoid transfer function. The Levenberg-Marquard (LM) algorithm is used to adjust the parameters of MPNN. The required training data are obtained from transformer company.

I. INTRODUCTION

Construction of transformers of high quality at minimum possible cost has been a crucial for any transformer manufacturing industry facing market competition. A critical measure of transformer quality is transformer noload loss. The less the transformer no-load loss, the higher the transformer quality and efficiency [2]. The transformer designer can reduce no-load loss by using lower loss core materials or reducing core flux density or flux path length. Electric utilities use more generating capacity to produce additional electrical energy to compensate for transformer energy losses. The production of this additional electrical energy increases electrical energy cost as well as greenhouse gas emissions. Although transformers inherently have high energy transfer efficiencies, the accumulated transformer energy losses in an electric utility distribution network are high since a large number of transformers are installed. In addition, transformer no-load loss appears 24 hours per day, every day, for a continuously energized transformer. Thus, it is in general preferable to design a transformer for minimum no-load loss. Transformer actual (measured) no-load loss deviates from designed no-load loss due to variability in the production process. Reduction of transformer actual no-load loss is a very important task for any manufacturing industry, since (1) it helps the manufacturer not to pay no-load loss penalties, and (2) it reduces the material cost (since a smaller no-load loss design margin is used) [11]. Artificial Neural Network is one of methods that mostly have been used in the recent years, in this field. Transformer insulation aging diagnoses, the time left from the life of transformers oil, transformers protection and selection of winding material in order to reduce the cost, are few topics that have been performed [4]–[8]. In this paper Artificial Neural Network based method have been used to estimate no load losses during design phase. ANN are used to predict no-load losses as a function of core design parameters. In following artificial neural networks with Levenberg-Marquard back propagation algorithm have been used to estimate no-load losses of transformers. The extracted data from transformer manufacturing company has been used to train the ANN and the best parameters for this network have been presented graphically. Finally result given by trained neural network have been compared with actual manufactured transformer prove the accuracy of presented method to estimate no-load losses as a function of core design parameters