Published Paper Details:

Load forecasts are very important for the smooth working in the field of electricity. It has various uses such as purchasing of energy and production, switching the load, base development, and for evaluating contacts. Energy planning and generation of power play an important role in this scenario. Their knowledge can be used in the development of smart grids. It is of three types: long-term, short-term, and medium-term. A study concerning the already existing approaches of short-term electricity forecasting shows that an example of an improved technique is still in need. In this article, the use of a neural network called long-short term memory has been explored considering tackling of load projecting accuracy issues. The result is analyzed and the steps that can be taken to get better results along with the future scope of the project are discussed. Factors affecting load forecasting accuracy are also discussed. Forecasting permits using energy storage systems to decrease the cost of energy for the consumer. Knowing future electricity consumption along with future electricity prices makes it possible to decide when to engage a battery storage system as opposed to drawing power from the grid. It creates a path for future research and development of high-efficiency algorithms for load forecasting. A comparison of the merits and demerits of various load prediction methods is also performed.