【作者】 李丰；

【导师】 张粒子；

【作者基本信息】 华北电力大学 ， 电力系统及其自动化， 2014， 博士

【摘要】 风电等可再生能源的大力发展有利于实现我国能源结构优化调整,促进电力系统的节能减排,然而风电的波动性、不确定性和逆调峰性也给系统的安全稳定运行带来了诸多问题。本文考虑风电对于系统AGC备用、旋转备用和调峰的影响,对大规模风电并网条件下的发电调度优化模型及方法进行了系统和深入的研究。针对风电并网导致系统AGC备用需求增加的问题,提出了风电并网引发的AGC备用需求预测方法。采用极大熵谱分析法对风电的波动性进行了频谱分析,并运用滚动平均法对关键时间尺度的风功率波动分量进行分离,结合实际算例数据对风电的AGC容量需求进行了计算和分析。为充分发挥风电并网条件下各类电源提供AGC的自身优势,进一步促进网内AGC机组之间的充分竞争,本文结合AGC机组调节性能指标,提出了考虑风电接入系统的AGC备用容量优化调度方法；以电能量成本和AGC备用容量成本最小为目标函数,兼顾AGC调节速度、AGC调节容量和AGC调节范围约束,建立了考虑风电接入系统的AGC备用优化调度模型。算例分析表明,所提模型和方法能够量化分析风电并网引发的AGC备用容量需求,优先调用AGC调节性能优良的机组,充分发挥水电、火电等不同电源提供AGC服务的自身优势。针对风电出力不确定性和系统N-1故障对于系统旋转备用容量优化分配的影响,提出了考虑风电和负荷波动及系统N-1故障的旋转备用优化调度方法。基于负荷和间歇式电源功率的预测误差分布,确定合理的区间数,采用区间数优化方法建立表述风电出力与负荷不确定性的数学模型。在对不确定性场景进行筛选的基础上,综合考虑线路传输容量约束、故障场景下的网络拓扑结构变化、连续波动场景与瞬时离散故障场景下的爬坡约束,建立了考虑多场景的旋转备用容量和发电出力计划一体化优化调度模型；并针对模型的复杂性,采用Benders分解降低模型的求解规模。算例分析表明本文方法能够将机组组合方案与旋转备用容量计划进行协调同步优化,保障了系统在各预想场景下的自愈校正能力。采用概率性方法处理风电出力和系统N-1故障的不确定性问题；基于风电的预测出力与预测误差分布,运用拉丁超立方抽样法生成风电场景,并运用同步回代法对场景进行削减。采用马尔科夫链预测未来时段机组和线路的N-1故障状态概率。在此基础上,兼顾机组组合、机组经济功率分配以及AGC备用和旋转备用容量分层协调优化配置等问题,建立了考虑风电与储能系统的随机柔性优化调度模型。所提模型及方法综合考虑了系统N-1故障和风电出力的不确定性概率与严重性、系统AGC备用与旋转备用容量辅助服务成本、储能装置剩余可充放电能力以及网络约束的影响。针对该模型的复杂性,将其线性化后采用商用混合整数线性规划求解器CPLEX进行求解。算例分析表明,所提方法可将系统AGC备用容量与旋转备用容量分层协调的配置到各个机组,能够考虑间歇式电源不同时间尺度的波动性和不确定性对于系统的影响,充分发挥了储能柔性灵活的优势。在我国,大规模风电并网引发了系统调峰困难和弃风问题,本文提出了促进风电跨省消纳的解决方案和促进风电就地消纳的风储联合运行模式。针对本文所提调节市场的平衡调节相对偏差量最小调度模式,建立了相应的优化调度模型,实例分析验证了本文所提模型的适用性和有效性。针对如何调用储能装置促进风电消纳的问题,本文分析了不同储能容量场景和不同风电渗透率场景下的风电接纳能力及系统经济性,提出了风电经济接纳和风电最大接纳两种调度模式,建立了相应的混合整数线性规划模型。算例分析表明,风电经济接纳模式下储能系统能够更加柔性灵活地平衡调峰和弃风的矛盾,系统能耗更低。论文提出的考虑风电接入系统的发电调度优化模型和促进风电消纳的调度模式,不仅具有学术价值,而且对于提高电力系统的安全经济运行和促进可再生能源的健康发展具有一定的现实意义。更多还原

【Abstract】 The great development of wind power and other renewable energy is conducive to China’s energy structure adjustment, the energy conservation and emission reduction of power system is also promoted. However, the volatility, uncertainty and inverse peaking character of wind power bring a lot of problems to the operation of the power system. The influence of AGC reserve, spinning reserve and peak regulation are considered in this dissertation, and the optimal scheduling model and algorithm of large-scale wind power integrated into power system are studied.A forecast method of AGC reserve demand caused by wind power is proposed. The volatility of wind power is analyzed by maximum entropy spectral analysis method, and the wind power fluctuation components of critical time-scale are separated by the rolling average method. AGC reserve capacity demand caused by wind power is calculated and analyzed with practical case data.In order to take full advantages of various types of AGC unit and promote the full competition of AGC units, an optimal AGC reserve scheduling approach of wind power integrated system is proposed. The objective function is to minimize both the cost of electrical energy and AGC reserve capacity. AGC regulation speed constraint, AGC regulation capacity constraint and AGC regulation range constraint are taken into account. The unit commitment and AGC reserve scheduling model of wind power integrated system is established. Case study demonstrats that the proposed approach and model can quantify AGC reserve capacity demand which is caused by wind power, and distinguish AGC regulation performance of different units effectively.In order to analyze the impact of N-1contingency and wind power on spinning reserve capacity scheduling, an optimal spinning reserve scheduling approach considering wind power and load fluctuations and N-1contingency is proposed. A reasonable range for each time interval is determined based on the wind power forecast error distribution and load forecast error distribution. Firstly, the uncertainty scenarios are determined. Secondly, the line transmission capacity constraints, structural changes of network topology in single contingency scenarios, and the ramping constraints in continuous fluctuation scenarios and instantaneous discrete fault scenarios are considered. At last, an optimal spinning reserve scheduling model with the consideration of multi-scene is established. Meanwhile, Benders decomposition is used to reduce the scale of the model. Numerical example results show that this approach can co-optimize the unit commitment and spinning reserve scheduling, and its self-healing correction capability in each scenario is assured.This dissertation adopted the probabilistic approach to deal with the uncertainty of wind power and N-1contingency. The wind power scenarios are generated by Latin hypercube sampling method based on the prediction of wind power and its prediction error distribution, and these scenarios are reduced by simultaneous backward and fast forward reduction method. The probability of N-1contingency is forecasted by markov chain. On this basis, a flexible stochastic optimal scheduling model considering wind power and energy storage system is developed with the consideration of the unit commitment and hierarchical coordinated optimal dispatch of AGC and reserve capacity. The proposed approach and model consider the uncertainty and severity of N-1contingency and intermittent power resources, the cost of AGC and reserve capacity ancillary services, the remaining charge and discharge ability of energy storage devices as well as the network constraints. This complicated model is linearized and solved by commercial mixed-integer linear programming solver CPLEX. Numerical example results demonstrats that the proposed approach can hierarchical optimize the AGC reserve and spinning reserve of each unit. The impact of different time-scale’s fluctuations and uncertainties of intermittent power are considered.In order to solve the consumptive problem caused by large-scale wind power in China, solutions to accommodate transprovincial wind power are proposed, and this dissertation studied the co-operation mode of wind power and energy storage system which is used to accommodate wind power locally. The wind power integration capacity under different peak regulation modes are analyzed. The minimum relative regulate deviation scheduling model of the regulation market is established. The applicability and effectiveness of the proposed model are verified. As for the scheduling mode of energy storage system which is used to accommodate wind power locally, the maximum wind power integration and economical wind power scheduling mode are proposed. The wind power integration capacity and system economy under different scenarios of wind power penetration and storage capacity are analyzed. Example shows that Energy Storage System(ESS) can balance the contradiction between peak regulation and abandoned wind power under economical wind power integrate mode.The optimal scheduling model of large-scale wind power integrated system and the scheduling model to accommodate wind power are proposed in this dissertation. It not only has academic value, but also has a practical value to promote the dispatch level of grid and the fast development of renewable energy.更多还原