Investigating the impact of building local photovoltaic power plants on the national grid, an artificial intelligence approach

Abstract

Today, simple electrical energy networks are becoming obsolete and are being replaced by smart networks. One of the advantages of smart grids is to facilitate energy management, so that one of the ideas of this method is to provide local energy sources (photovoltaics, wind turbines, generators, etc.) to consumers and predict the amount of energy production. In this way, not only the consumed electricity is supplied, but also it is determined the time that electricity can be exported to the national grid. For this purpose, it is necessary to predict the production and consumption of each unit in the network. In this article, an urban building complex in the Canadian state of Alberta is investigated. Its consumption is calculated using simulation in the calculation software, and then the amount of electricity produced by photovoltaic panels is also calculated and the result of production and consumption is estimated, then using a machine learning algorithm of energy balance for the following years are expected. The results of this study show that the use of photovoltaic systems and the production of electrical energy from it can have a significant impact on energy management, and not only the building's electricity is supplied in most months of the year, but also additional electricity can be used. Sending this excess energy to the main grid, can be an important step towards peak management of the power grid.

Keywords

• Photovoltaic panels
• Machine learning
• Power grid management
• Local power plant
• Demand for electrical energy
• Electrical energy balance