带有风能的智能电网多周期买电策略研究

本文研究带有风能随机供给的智能电网中传统能源的多周期买电问题,假设存在一个能源运营商集中负责智能电网传统能源的购买和消费。通过构建并求解动态规划模型,找到能源运营商在风能供给不确定性下的传统能源最优多周期买电策略。在最优买电策略下,能源运营商只有在当期电价足够小时才购买传统能源,其买电量与风能分布、价格信息和时间信息有关。在实际数据的基础之上,提供详实的数值实验对比研究了本文的最优买电策略和其他两种策略(实践中只考虑风能估计的策略和放弃利用风能的策略)在最小化总成本方面的效果,并验证了本文的最优买电策略在真实风能数据中的鲁棒性。

Smart Grid Energy Procurement Strategies under the Supply of Wind Energy

The changes of electricity market from the introduction of smart grid are envisioned to be revolutionary on both power generation side and power consumption side. Wind energy reduces greenhouse gas emissions and other pollution but its supply is uncertain. How to deal with the uncertainty of wind energy for the stability of smart grid is still indispensable, especially with the integration of flexible demand side management. In this paper, we analyze a multi-period energy procurement problem in a smart grid community with wind energy integration, under the pre-announced information of day-ahead real time prices. There is an energy aggregator who is responsible for centralized controlling of energy procurement and consumption. It is assumed that the aggregator can delay time-adjustable demand if necessary during all operating periods. By using finite-horizon dynamic programming, we show the optimal procurement policy for the aggregator is to procure traditional energy only when the price is less than a certain threshold that depends on the information of the price statistics, the wind energy distribution and the time left to the deadline. To decide the optimal procurement amount, the aggregator makes trade-offs between two supplies, traditional energy with cost or free wind energy with uncertainty, to choose a more cost-effective energy. Numerical studies for comparing our optimal policy with other two policies, procuring energy on estimation of wind energy and procuring energy on arrived demand, are conducted. The cost-savings of our optimal policy are remarkable if given great price fluctuations. We also examine the robustness of our optimal policy in the scenarios with historical and seasonal wind energy data. The results show that our optimal policy can achieve remarkable cost-savings even with small wind scenarios and/or high percentage of adjustable demand. Moreover, the robustness of our optimal policy in different seasons is well observed.