EAGER: Renewables: Supply Rate Control and Grid Stability with Renewable Power Generation and Co-located Storage

Rensselaer Polytechnic Institute, Troy NY

Investigators

Abstract

Renewable power generators face the problem of variability of their power output, which makes it difficult for them to compete with conventional generators. Investing in energy storage devices and systems can alleviate the problems associated with variable generation rates to a good extent. Co-located storage (storage located at the site of renewable generation) also helps them to fill the transmission pipeline, and thereby maximize the utility of the existing transmission systems. Under high wind conditions, for example, the power that cannot be transmitted on the sub-transmission system can be used to charge batteries, which can be used to supply energy during peak hours, when the wind is typically low. Therefore, there is a strong case for the use of co-located storage with renewable power generation. The project aims at investigating some key theoretical modeling, analysis, optimization and control challenges that relate to making efficient use of such storage-equipped renewable power generation. The technical challenges will be explored both from the perspective of an individual renewable power generator, and from that of the overall grid in attempting to integrate a number of renewable energy sources in a decentralized manner. Education and outreach aspects of the project include integration of research insights into graduate courses in power engineering and control systems at RPI. Communication with power companies will be used to maximize the industrial outreach of the project, to make the project well grounded and practically impactful. The project seeks to investigate - from a theoretical modeling and analysis perspective - some fundamental questions that the smart grid faces in integrating the variable-rate renewable energy supplies into the grid in an economically efficient manner, while ensuring the stability of the grid. The questions are also fundamentally related to the value of using storage resources - both from the viewpoint of the renewable power generators themselves, as well as the grid as a whole. The first of these involves optimizing the day-ahead planning as well as real-time operations of the renewable power generators (equipped with co-located storage): given the day-ahead (real-time) electricity prices and power generation rate forecasts, how should a renewable power generator with limited storage optimize in advance (dynamically control, resp.) its power supply contracts (rates, resp.) under transmission constraints to maximize its profitability. The second of the two issues we investigate in this project involves the study of decentralized dynamic control mechanisms that can be used for frequency regulation using large numbers of storage-equipped renewable power generators. This involves novel exploration of stability issues of randomized frequency regulation schemes, and tuning them to ensure convergence as well as good transient behavior. From a theoretical perspective, addressing these issues requires exploring novel challenging questions in stochastic optimization, queuing theory, decentralized control and stochastic stability, and their applications to power networks.