A Three-Stage Strategy With Settlement for an Energy Community Management Under Grid Constraints

Abstract

This paper focuses on a local energy market where a group of households in a low voltage (LV) grid is organized as an energy community. A three-stage management strategy is proposed under the French collective self-consumption framework. In the first day-ahead stage, households coordinate with the community manager to minimize the overall energy bills for the next day. Then, real-time operation (second stage) focuses on the mitigation of forecast uncertainties and voltage violation, by utilizing the reserve of the local production/storage assets. Finally, the third stage contractually allocates the community energy among the households on 30 min basis (based on the French regulation), in order to ensure fair individual cost reduction and possibly create economic surplus in the community. The proposed community management strategy is evaluated on a LV grid with 55 households, and the results show the proposed strategy achieve an average of 30 % individual cost reduction among the users compared to the base scenarios. Furthermore, different pricing scenarios are compared through sensitivity studies and scalability tests are run on larger systems.

Methods for mitigating uncertainty in real-time operations of a connected microgrid

“…Furthermore, optimization problems are often split into multiple stages to mitigate uncertainty. The initial stage addresses the optimization problem at a lower granularity (on the order of days), while the subsequent stages handle it at a finer granularity (on the order of seconds) [19][20][21][22][23][24]. Nevertheless, these strategies require the expected values or potential scenarios for solar generation and load demand, which is often challenging to obtain accurately.…”

Mitigation of grid parameter uncertainties for the steady-state operation of a model-based voltage controller in distribution systems

“…In this paper, two different voltage management architectures (i.e., centralized and distributed) are considered, adapted from our previous works [16,12]. In the centralized control architecture , there exists a dedicated centralized entity (e.g., DSO or aggregator) who is responsible for voltage management.…”

“…The centralized voltage control is formulized as a multi-objective optimal power flow (OPF) problem with second-order conic programming (SOCP) relaxation [17]. The objective of the OPF problem is to minimize PV curtailments over a predefined time horizon T , subjected to DistFlow (i.e., grid) model, PV and voltage constraints that are not represented here for the sake of clarity, but described in [16,12]. The objective function is illustrated in (1).…”

A two-stage coordination strategy for the control of distributed storage at the household level—Arbitrage between users preferences and distribution grid objectives

“…This is especially of interest for power system operators with a reduced need for carbon-intensive energy reserves to fulfill the demand/load balance in realtime thanks to more predictable power profiles at the end-user level. Both [9] and [10] adopt ADMM for the real-time coordination of users paying attention to the fulfillment of distribution grid constraints (i.e., current/voltage limitation). However, the number of iterations before convergence cannot be controlled with regular ADMM implementations.…”