Software Integration of Power System Measurement Devices with AI Capabilities

Abstract

The latest changes on the distribution network due to the presence of distributed energy resources (DERs) and electric vehicles make it necessary to monitor the grid using a real-time high-precision system. The present work centers on the development of an open-source software platform that allows for the joint management of, at least, power quality monitors (PQMs), phasor measurement units (PMUs), and smart meters (SMs), which are three of the most widespread devices on distribution networks. This framework could work remotely while allowing access to the measurements in a comfortable way for grid analysis, prediction, or control tasks. The platform must meet the requirements of synchronism and scalability needed when working with electrical monitoring devices while considering the large volumes of data that these devices generate. The framework has been experimentally validated in laboratory and field tests in two photovoltaic plants. Moreover, real-time Artificial Intelligence capabilities have been validated by implementing three Machine Learning classifiers (Neural Network, Decision Tree, and Random Forest) to distinguish between three different loads in real time.