Implementing Neural Network Price Predictor for Residential Load Control System in Real-Time Electricity Pricing Environments

Smart grids are one of the newest world's technologies in transmitting electricity. The main goal in these grids is to provide customers with the best quality and the most efficient services, while imposing the least number of threats for the environment. The major topic in smart grid discussions is modelling a dynamic pricing system that can be applied in these grids. Though most of the researches done, confirm the positive effects of implementing these techniques in all aspects from customer to producer and protecting environment, it must be noted that lack of knowledge to answer price changes effectively and also lack of proper utilities for taking advantage of this form of pricing, in order to manage electrical costs, are the most significant barriers on the way of dynamic pricing tactics to work efficiently. In this research, designing a smart system for managing electrical costs for a residential customer in a time varying environment will be discussed with focusing on proposing an efficient price prediction unit by using neural networks. Results show that in comparison with other prediction tactics neural network systems in spite of their simplicity have a high ability of predicting the upcoming prices

Implementing Neural Network Price Predictor for Residential Load Control System in Real-Time Electricity Pricing Environments Keywords:

Smart Grids , Price Prediction , Dynamic Pricing , Optimizing Electricity Prices , Neural Networks