基于条件风险价值的智能区域代理商报价策略及EV充电管理

智能区域代理商整合区域内电动汽车(electric vehicle,EV)参与电力市场交易,然而在联营市场(包括日前和实时市场)交易机制下,联营市场电价的随机性增加了代理商的市场交易风险.基于此,提出了以联营市场条件期望购电成本最小为目标的报价决策模型,该模型考虑了联营市场电价的随机特性,并将模型转换成带条件风险价值约束的随机优化问题.由于含随机性变量概率密度函数的积分计算困难,因此利用蒙特卡罗模拟,预测电价分布特性,采用抽样平均近似法对模型进行离散化处理,并通过构造函数将原问题化为凸优化问题进行求解.数值实验表明代理商通过控制EVs参与联营市场竞价交易,对EVs进行充电管理,可有效降低峰谷差,有助于电网的稳定性.

Optimal bidding strategy and EV charging management of intelligent regional aggregators based on CVaR method

Intelligent regional aggregator (IRA) integrates regional electric vehicles (EVs) to participate in electric power market transaction. However, in the pool market (include the day-ahead and real-time markets), the transaction risk of IRA will be increased by the randomness of pool market price. To this end, the bidding strategy model of IRA aims to minimize the conditional expectation of electricity purchase cost in pool market with considering the pool market price uncertainty. And the model is reformulated to be a stochastic optimization problem with the conditional value-at-risk (CVaR) constraints. The integration function with probability density of random variable is difficult to solve, so the Monte-Carlo simulation method is adopted for prediction of the distribution characteristics of the pool market price. And the model is transformed and discretized by sample average approximation as a solvable convex programming problem. Numerical experiments show that the IRA controls EVs bidding in the pool electricity market, and manages EVs charging profile to achieve the purposes of low peak valley difference and enhance stability of power grid.