ENERGY COOPERATION OPTIMIZATION IN MICROGRIDS WITH RENEWABLE ENERGY INTEGRATION

Abstract

Microgrids are key components of future smart grids, which integrate distributed renewable energy generators to efficiently serve the load locally. However, the intermittent nature of renewable energy generations hinders the reliable operation of microgrids. Besides the commonly adopted methods such as deploying energy storage system (ESS) and supplementary fuel generator to address the intermittency issue, energy cooperation among microgrids by enabling their energy exchange for sharing is an appealing new solution. In this project, we consider the energy management problem for two cooperative microgrids each with individual renewable energy generator and ESS. First, by assuming that the microgrids renewable energy generation/load amounts are perfectly known ahead of time, we solve the off-line energy management problem optimally. Based on the obtained solution, we study the impacts of microgrids energy cooperation and their ESSs on the total energy cost. Next, inspired by the off-line optimization solution, we propose online algorithms for the real-time energy management of the two cooperative microgrids. It is shown via simulations that the proposed online algorithms perform well in practice, have low complexity, and are also valid under arbitrary realizations of renewable energy generations/loads. Finally, we present one method to extend our proposed online algorithms to the general case of more than two microgrids based on a clustering approach.

Existing System

In the existing system, game-theoretic methodologies is used in the transition from legacy systems toward smart and intelligent grids.

Proposed System

In this project, the real-time energy management problem for a system with two cooperative microgrids that belong to the same entity or different entities with common interests are investigate. Assuming that the microgrids can exchange energy via the transmission line connecting them, and each comprises renewable energy generators, ESS, and an aggregate load. The main results of this project are, the off-line energy management problem by assuming that the microgrids’ net energy profiles, i.e., the renewable energy generation offset by the aggregate load of individual microgrids, are perfectly known ahead of time are formulated. The impacts of microgrids energy cooperation and their ESSs on the total energy cost saving via simulations based on the real wind generation data of Tuscon power system was studied. The results show that although both energy cooperation and ESSs can be used to save the energy cost, one can be more effective than the other depending on the system setup. For instance, energy cooperation reduces the total energy cost more considerably when the microgrids’ net energy profiles are highly uncorrelated. However, ESSs reduce the total energy cost more effectively when the net energy profiles are correlated and/or the energy loss in the transmission line is high. Based on the results obtained from the off-line optimization, we propose two online algorithms of low complexity for the real-time cooperative energy management of microgrids, namely store-then-cooperate and cooperate-then-store. The proposed algorithms can be applied under arbitrary realizations of microgrids’ net energy profiles. Simulation results reveal that our online algorithms perform very close to the optimal solution derived from the off-line optimization. The proposed online algorithms is extended to the general case of more than two microgrids based on a clustering approach. We show that the proposed clustering based approach performs fairly close to the optimal off-line solution, with performance losses of only 4:78% and 6:14% in the noisy environment with 15% and 30% renewable energy prediction errors, respectively.

Architecture

SYSTEM MODEL