Machine Learning Strategy for Solar Energy optimisation in Distributed systems

Abstract

Using renewable energies such as wind and solar energy, two types of renewable energy, we may adjust the structure of the energy system, addressing both energy and environmental challenges at the same time. As a result of its impact on the environment, wind and solar energy generation are inherently unreliable sources of energy. Lithium batteries, a technology that is becoming increasingly mature in terms of energy storage, are a critical component of the answer to the problem of instability. In order to avoid waste and expense increases, the capacity should not be too large or too small, respectively. Power consumption restricts the amount of energy that may be stored, but industrial power usage is unpredictable and non-periodic. This is a significant task that needs the development of a model that can dispatch while still providing a reasonable amount of storage. In this paper, we develop a KNN classification model that considers the test cyclic of photovoltaic (PV) generation that includes battery installation, data on electricity consumption and data on PV generation in India. These metrics are used to develop an energy management model. The model aims at the reduction of operation cost and optimal storage of energy that should satisfy the grid demands. The results of simulation and the comparison of the theoretical results shows that the proposed model has higher optimisation of energy in the storage devices in case of distributed systems.

Keywords

• Solar Energy
• Distributed Systems
• PhotoVoltaic Cells
• Machine Learning