In this paper, an arbitrage strategy is proposed for renewable-based microgrids (MGs) to overcome the volatile behavior of renewable energy sources (RESs) such as photovoltaic and wind in a newly emerged business space in which peer-to-peer (P2P) energy-trading in transactive energy markets (TEMs) set up between a day-ahead market (DAM) and real-time markets (RTMs). To identify arbitrage opportunities created from the price difference between the P2P and real-time trades, a bi-level risk-constrained stochastic programming with interval coefficients (BRSPIC) is presented. In the first stage of the decision-making, scenarios are employed to deal with the DAM prices uncertainties. In the second stage, P2P energy-trading competition is modelled by a bi-level programming based on non-cooperative leader-follower games. While the social welfare of peers is maximized at the lower level, the MG maximizes its profit at the upper level. By getting closer to real-time, interval coefficients are considered in the third stage to cope with the uncertainties of RESs and loads, as well as RTM prices. The conditional value-at-risk (CVaR) is enforced the model to control the risk of profit variability. By using Karush-Kuhn-Tucker (KKT), the BRSPIC is transformed into a single level optimization. Then, it is linearized and solved by a mixed-integer linear programming (MILP) solver. By evaluating the proposed model on a test system, it is evident that the MG increases more than 3.1% of its profit by the arbitrage strategy. By considering CVaR, a fully risk-averse decision decreases the profit of MG by 27%, although it would be so conservative decision.

中文翻译：

---

对等能源交易的基于可再生能源的微电网套利策略

本文提出了一种针对可再生能源的微电网（MG）的套利策略，以克服可再生能源（RESs）的波动行为，例如光伏和风能在对等（P2P）新兴业务空间中的波动在日前市场（DAM）和实时市场（RTM）之间建立了无源能源市场（TEM）中的能源交易。为了识别由P2P和实时交易之间的价格差异产生的套利机会，提出了一种具有区间系数（BRSPIC）的双层风险约束随机规划。在决策的第一阶段，采用情景来处理DAM价格的不确定性。在第二阶段，通过基于非合作的领导者跟随者游戏的双层编程对P2P能源交易竞争进行建模。在较低级别上最大化同行的社会福利，而在较高级别上，MG最大化其利润。通过接近实时，在第三阶段将考虑间隔系数，以应对RES和负载以及RTM价格的不确定性。条件风险价值（CVaR）被强制执行模型以控制利润变动的风险。通过使用Karush-Kuhn-Tucker（KKT），BRSPIC被转换为单级优化。然后，通过混合整数线性规划（MILP）求解器对其进行线性化和求解。通过在测试系统上评估提出的模型，很明显，MG通过套利策略增加了超过3.1％的利润。通过考虑CVaR，完全避免风险的决策会使MG的利润降低了27％，尽管这是一个保守的决策。