ANN based On-Load Tap Changer for Distribution Network with Distributed Generation

Abstract

Number of distributed generation (DG) plants integrated to utility network increases due to rising of electricity consumption. DG plants have a big impact on voltage level of network in accordance with this increasing. Especially DG plants based on renewable resources may cause voltage rise and drop because of their intermittently energy generation. Generally, the voltage level is regulated by reactive power control technique and On-load tap changer (OLTC). Generally, HV/MV transformers use OLTC to regulate the voltage level under load. The conventional voltage control techniques depend on current of branch at the main substation end. If the rising of DG installation is taken into consideration, the conventional methods can be inadequate. Artificial intelligence techniques can be used to mitigate this problem. In this paper, an Artificial Neural Networks (ANN) model is designed for determination of OLTC best tap position. ANN model is trained by using different values of network voltage level, DG generation values and amount of load demand. Designed model is applied on electrical network of Selcuk University. The electrical network of Selcuk University is modeled by using PSCAD/EMTDC. 1MW Photovoltaic (PV) power plant is determined as DG plant.