Big Data Analytics Assisted Power Converter Performance Monitoring for Renewable Energy Integrated Smart Grid Applications

Abstract

In renewable energy-integrated smart grids, intermittent sources and varying loads introduce several concerns, such as managing the intermittency from sources to connected loads and power quality disturbances from the grid formation level up to a scale where these disturbances can be compensated for by an intelligent power conversion with sufficient performance monitoring. This work presents the framework for a Big Data Analytics (BDA)-aided monitoring system for power converters in Photovoltaic (PV)- and wind energy-based smart grid applications. Most of the existing state-of-the-art converter monitoring techniques are only isolated voltage regulation or power quality analysis, and few efforts have been devoted to the unified assessment of efficiency, DC-link stability, Total Harmonic Distortion (THD), temperature rise, and healthy operating status. The primary issue this work addresses is that there is no monitoring index driven by data that can swiftly reveal performance degradation in the presence of dynamic renewable input and random loads. The system consists of a renewable energy source interface, a power converter, a Direct Current (DC) link, an IoT-based data acquisition layer, a BDA processing layer, and a decision-support feedback system. The time-domain simulation environment assesses converter output power, DC-link voltage regulation, efficiency, THD, thermal behaviour, and the Performance Monitoring Index (PMI). The results show renewable input between 3.7 kW and 11.3 kW, converter efficiency between 94.2% and 96.6%, grid current THD between between $3.2\%$ and $4.8\%$, and converter temperature from 38°C to 58°C with PMI in the range of 0.76-0.93. This framework is applicable in use cases such as smart grids, microgrids, renewable energy plants, Electric Vehicle (EV) charging stations and industrial power quality monitoring.