【作者】 王静；

【导师】 段建东；

【作者基本信息】 西安理工大学 ， 电力系统及其自动化， 2020， 硕士

【摘要】 为提升分布式发电单元在电网的渗透率,微网技术应运而生。分布式可再生能源单元出力具较强随机性,加之需求响应工作的进一步推广,其源荷侧的不确定性更加突出。当前基于人工智能的预测技术在数据处理和模型表达等方面具有很大优势,但预测精度尚不能满足可再生能源接入下微网的调度需求。如何在源荷侧功率波动情况下,既能保证调度方案的最优性,又能提升可再生能源的利用率,是微网优化调度的关键。鲁棒优化方法提前考虑了不确定性参数的波动,所得最优解具有较强抗干扰性,受到广泛关注。当前微网鲁棒优化调度研究中,仍存在解的保守性强及不能提升可再生能源消纳能力等方面的问题,而开展多能互补微网两阶段鲁棒优化调度是解决以上问题的可行方法,对其深入研究将具有重要的理论价值与实际意义。本文首先分析了微网中分布式风电、光伏以及负荷的出力特性,指出了微网优化调度中存在的多种不确定性因素。为解决上述不确定性因素下的优化调度问题,引出了鲁棒优化方法。其次,研究计及需求响应的微网两阶段鲁棒优化调度方法,并构建了min-max-min 形式的两阶段优化模型,其预调度阶段根据源荷侧的预测值来制定各分布式电源出力、需求响应负荷运行及备用容量设置的最优方案;再调度阶段旨在辨识不确定变量的“恶劣”场景,并检验预调度决策方案在该场景下的可行性。接着,以综合能源系统为发展背景,考虑了多种能源协同优化作用,研究了电-气-热联供型微网两阶段鲁棒优化调度方法,其预调度阶段根据源荷侧的预测值来制定系统购能及多能耦合设备出力的最优方案;再调度阶段以期通过热电联供等机组快速调节出力来应对不确定性变量的波动。针对大多已有求解算法在解的精确度及迭代次数等方面的不足,本文将所建模型分解为主问题和子问题并应用列约束生成算法进行交互迭代求解。最后,在典型微网系统中设置多种方案对比验证了本文方法的有效性,结果表明本文所得的调度方案具有良好的经济性和鲁棒性,并能降低弃风弃光和切负荷风险。更多还原

【Abstract】 To increase the penetration of distributed generation units in the power grid,microgrid technology came into being.The distributed renewable energy output has strong randomness,coupled with the further promotion of demand response strategy,the uncertainty on the source and load sides is more prominent.The current prediction technology based on artificial intelligence has great advantages in data processing and model expression,but the prediction accuracy cannot yet meet the scheduling needs of microgrids with renewable energy access.How to ensure the optimality of the scheduling scheme,and improve the utilization rate of renewable energy is the key to the optimal scheduling of the microgrid.The robust optimization method considers the fluctuations of uncertain parameters in advance,and the obtained optimal solution has strong anti-interference and is widely concerned.In the current research on microgrid robust optimization schedulingr there are still problems with strong conservative and the inability to improve the capacity of renewable energy consumption.The two-stage robust optimal scheduling is a feasible method to solve the above problems,and the in-depth study of it will have important theoretical value and practical significance.Firstly,this paper analyzes the output characteristics of distributed wind power,photovoltaic,and load in the microgrid,and proposes a variety of uncertain factors in the optimal scheduling of the microgrid.In order to solve the optimal scheduling problem under the above uncertain factors,a robust optimization method is introduced.Secondly,the two-stage robust optimization scheduling method of microgrid considering demand response is studied,and a two-stage min-max-min optimization model is constructed.In the pre-scheduling stage formulates the optimal scheme of each distributed generation output,the demand response load operation,and the reserve capacity settings based on the predicted values of the source and load sides.The re-scheduling stage aims to identify the "worst" scenario of uncertain variables and test the feasibility of the pre-scheduling decision scheme in this scenario.Then,taking the test the feasibility of the pre-scheduling decision scheme in this scenario.Then,taking the integrated energy system as the development background,considering the coordinated optimization of multiple energy sources,the two-stage robust optimization scheduling method of the electricity-gas-heat co-generation microgrid is studied.The pre-dispatching stage formulates the optimal plan for system energy purchase and multi-energy coupling equipment output based on the predicted values of the source and load sides.The re-scheduling stage is to cope with fluctuations in uncertain variables through quickly adjust the output of the combined heating and power and other units.In view of the deficiencies of most existing two-stage solution algorithms in terms of solution accuracy and number of iterations,this paper applies the column-and-constraint generation method to the solution of the above two-stage optimization model,and decomposes the model into a main problem and a subproblem,which to be solved by interactively and iteratively.Finally,a variety of schemes are set in a typical microgrid system to verify the effectiveness of the proposed method.The results show that the scheduling scheme obtained in this paper has a good economy and robustness,and can reduce the risk of wind/photovoltaic abandonment and load shedding.更多还原