CI-based Analytics for Photovoltaic Power Predictions and Tie-line Bias Control in Smart Grid

Abstract

The smart grid enables two-way flows of electricity and information by introducing vast quantities of data. Advanced metering technologies, such as weather sensors, phasor measurement units and smart meters generate variety of data at very high velocities. Proper analysis of this data allows intelligently monitored and controlled power systems having self-healing, fault-tolerant and secured functioning features. Increasing the integration of renewable energy sources, such as photovoltaic (PV) power has a significant impact on the power system operation and control. The variability and uncertainty of renewable energy poses the challenges of power and frequency fluctuations. Predictive analytics in power generation can provide a smarter grid with better control. Computational intelligence (CI) paradigms based on adaptive learning play a pivotal role in predictive analytics in smart grid. A comparison of CI approaches for predicting PV power and tie-line bias control is presented in this paper. Typical results indicate that PV power predictions improve tie-line bias control performance and better utilization of available PV power.