Short term load forecasting using with adaptive resonance theory neural network in Tehran area operation center

Expansion planning, load flow, economic operation,security analysis and electric energy systems control are some of studies that effectively depend on the behaviour of the load profile. With the growth of consumption and connected load of power systems with each other the amount of incoming and outgoing data in order to plan and optimize utilization has been Increased. So that those responsible for decisions related to these in order to support operations of power system we must use more science and software such as intelligent networks to predict Intelligent networks and new computational methods to predict the output of the power grids to provide answers .The neural networks based on the adaptive resonance theory architecture have two fundamental characteristics that are extremely important for the network performance (stability and plasticity), which allow he implementation of continuous training. The fuzzy adaptive resonance theory neural network aims to reduce the imprecision of the orecasting results by a mechanism that separate the analog and binary data, processing them separately. Therefore, this epresents a reduction on the processing time and improved quality of the results, when compared to the Back-Propagation neural network, and better to the classical forecasting techniques .Finished the training,he adaptive resonance theory neural network is capable to forecast electrical loads 24 h in advance. To validate the methodology,data from Tehran operation center(TAOC) is used.

کلیدواژه ها:

Tehran area operation center(TAOC) ، fuzzy adaptive resonance theory neural network ، load forecasting