Smart grid power demand and supply scheduling to preserve energy wastage using machine learning technique

At present the demand for electrical energy is growing continuously because of the increasing population with their need for electricity. Additionally, to fulfil this need more and more power is being generated by using different energy generation resources. However, it is currently sufficient for our needs but shortly, we need to advance our ways of power generation and management. By the influence of this complexity, this paper is motivated to design and develop a deep learning model-assisted intelligent power demand and supply management system. In this context, two main types of power consumers are assumed and then the subcategories of consumers have been defined. Here, for each power consumption behaviour, a neural network model has been trained and future power demand has been predicted. Further, the predicted power demand is converted into the required amount of power supply. Based on power demand and supply consequences of 33 hours, the experimental results have been measured in terms of hit and miss ratio. Based on the experimental results the proposed model has provided an 87.5% hit ratio for best case and 62.5% hit ratio for worst case scenarios.