One of the major challenges of the future power systems is managing the substantial unreliability and unpredictability induced by the large share of renewables. As one of the promising solutions, energy cooperatives have recently drawn attention. Unlike most of the existing proposed strategies (e.g., storage), there is not much known yet about the practical importance of energy cooperatives. Taking a data-driven approach, we quantify the power variability reduction (which we also call the reliability gain) for three neighborhood level types of cooperatives: energy generation, energy demand, and energy prosumer. We show how the reliability is scaled by the size of coalition and the time scale for these three energy cooperatives. We gain several interesting insights which are vital for designing and planning neighborhood level energy cooperatives. For example, we find that energy demand cooperatives have the largest reliability gain among all. Using the central limit theorem asymptotic results, we show that the demand energy variations up to 15 minutes of different homes are almost independent of each other. We also find that small coalitions of sizes 10 and 20, respectively, for energy demand and energy prosumer cooperatives are sufficient to capture most of the reliability gain of the grand coalition. For the special case of energy generation cooperatives, our results differ from an existing empirical study (in the context of geographical diversity) at a different location, suggesting that designing an efficient energy generation cooperative is jurisdiction-dependent.

doi.org/10.1109/SmartGridComm.2016.7778849, hdl.handle.net/1765/95617
7th IEEE International Conference on Smart Grid Communications, SmartGridComm 2016
Rotterdam School of Management (RSM), Erasmus University

Van Gelder, C. (Christian), & Ghiassi-Farrokhfal, Y. (2016). On the reliability gain of neighborhood coalitions: A data-driven study. In 2016 IEEE International Conference on Smart Grid Communications, SmartGridComm 2016 (pp. 736–741). doi:10.1109/SmartGridComm.2016.7778849