Design and Arduino Implementation of Artificial Neural Network Based Intelligent Power Management System of Bangladesh

The research contrivance is to maintain and control load detachment or shedding in local distribution area and also utilize different types of power generation units such as conventional and non-conventional energy sources. The artificial neural network will anticipate, predict and exploit idea when generation is insufficient to meet the load demand. If the demand for the load is more than a generation, the artificial neural network will acknowledge the specific area where load detachment would be appropriately needed. Basically, an intelligent power management system of Bangladesh is developed by using artificial neural network (ANN) and implemented using Arduino for validation this work. We have designed the artificial neural network by utilizing and manipulating different types of seasonal and occasional load data. For further utilization, we actually divided an area into different sub-areas such as residential-load, industrial-load, commercial-load and VIP-load. However, it can suggest the area where load-shedding would be preferable on the basis of priority. Hence, artificial neural network suggests the area where maximum load-shedding is desired which is demonstrated by Arduino. Keywords: Power system, artificial neural network, supervised learning, solar radiation, regression plot, error histogram Cite this Article Mahfuzur Rahman, Joy Shaha, Navid Hossain et al. Design and Arduino Implementation of Artificial Neural Network Based Intelligent Power Management System of Bangladesh. Journal of Power Electronics & Power Systems . 2017; 7(3): 17–29p.