梯级水库群发电优化调度的大系统分解协调模型北大核心CSCD

针对标准动态规划算法应用于梯级水库群发电优化调度求解时的"维数灾"问题,基于大系统分解协调原理建立了二级(分解、协调)递阶结构的梯级水库群发电优化调度模型。在分解级,应用多线程技术实现了标准动态规划算法的并行化,以对分解后的单库子系统进行优化求解;单库子系统间的耦合关联作为协调级,反复与分解级交换信息,经迭代计算求得模型的最优解。通过应用于李仙江三库梯级系统的实例验证分析,结果表明该方法可在一定程度上提高求解效率,为复杂工程系统优化提供了理论参考。     

基于分解协调方法的光储系统配置—运行协同优化

以光伏-储能电池系统为研究对象,构建光储系统的配置-运行优化分解协调模型.首先,将考虑储能投资成本与光储系统运行成本在内的综合费用作为目标函数,利用Benders分解方法将模型分解为规划配置主问题和运行调度子问题,然后,通过分解协调优化得到储能电池的最优配置方案与系统的优化运行策略,通过设定光伏消纳率、自给率、内部收益率、动态投资回收期等一系列指标,对模型进行评价.此外,实现了 Benders分解法收敛过程和Bender割集累积过程的可视化,直观展现出容量配置与优化运行的交互作用.基于此,进一步利用K-means聚类分析算法建立了年光伏出力模型,求解长时间尺度上储能电池的最优配置,探讨了分解协调优化模型在中长期尺度下求解的可行性.     

梯级水库群发电优化调度的大系统分解协调模型

针对标准动态规划算法应用于梯级水库群发电优化调度求解时的"维数灾"问题,基于大系统分解协调原理建立了二级(分解、协调)递阶结构的梯级水库群发电优化调度模型。在分解级,应用多线程技术实现了标准动态规划算法的并行化,以对分解后的单库子系统进行优化求解;单库子系统间的耦合关联作为协调级,反复与分解级交换信息,经迭代计算求得模型的最优解。通过应用于李仙江三库梯级系统的实例验证分析,结果表明该方法可在一定程度上提高求解效率,为复杂工程系统优化提供了理论参考。     

计及阀点效应的电力系统经济运行方法

动态经济调度在电力系统运行中占据重要作用,这是一个在时间和空间上相互耦合的复杂优化决策问题.当计及机组阀点效应后,该问题更是呈现为不光滑、非凸特性.本文提出一种沿连续两次迭代计算次梯度之差方向的空间扩张算法来解决动态经济调度问题,并采用自适应更新罚因子的不光滑精确罚函数处理约束条件.通过仿真计算,结果表明该方法求解精度高、收敛速度快,验证了此方法的可行性和有效性,从而为电力系统动态经济调度问题的求解提供了一条新途径.     

最小化运行费用的时变重构全局优化算法

该文提出了一种以运行费用最小为目标的时变重构全局优化算法。将时变重构问题分解为多个静态重构子问题，使用核心模式遗传最短路算法(CSGSA)分别求解每个子问题，得出候选放射状网络结构，并使用动态规划，根据各子问题的结果求解全局最优的时变重构策略。算法还对开关操作次数约束进行了处理，使之不影响动态规划的全局最优性。所提算法是配电网络实时调度，规划和离线分析的有效工具。     

含风电场的电力系统多目标优化调度

针对火力发电对环境造成的严重污染问题,建立了含风电场的电力系统多目标优化调度模型。以火电机组燃料费用和污染气体排放量最低为目标,用虚拟解理论将多目标优化问题转化为单目标优化问题,以降低问题的复杂性;用动态搜索步长及动态搜索概率对布谷鸟算法改进,用改进后的算法求解所建立调度模型。仿真结果表明,改进算法在符合有功平衡约束、机组出力约束及机组爬坡约束等同样的前提条件下,具有节约发电成本,减少污染物排放,加快系统运行速度的优点。     

基于免疫克隆选择算法的水电站水库优化调度

在研究了人工免疫系统中的克隆选择学说和克隆选择算法的基础上,研究了1种新的人工免疫算法——免疫克隆选择算法,并将其应用到水库优化调度中,提出了1种基于免疫克隆选择算法的水库优化调度方法.该算法通过在克隆选择算法中引入免疫基因操作,提高了算法的求解精度和求解效率,避免了“维数灾”和早熟问题.实例研究结果表明,相对于动态规划,免疫克隆选择算法计算速度快、收敛性好,提高了计算效率,较好地解决了传统的动态规划方法求解水库(群)优化调度问题存在“维数灾”问题.     

系统分解协调算法的抽水蓄能电站优化调度

建立了包含抽水蓄能电站的电网统一调度优化模型,即以调度周期内火电燃料成本为最小目标函数,满足系统及各机组约束条件。利用系统分解协调思想,开发了一个结合拉格朗日松弛方法和粒子群优化算法的混合算法,将原优化问题分解为两层优化问题。上层拉格朗日算子优化利用次梯度算法求解,下层各子问题利用粒子群优化算法求解,经过迭代寻优得到最优对偶解后,利用一个启发式算法求得满足系统约束及各机组运行约束的原问题的可行解。最后通过算例验证了模型的合理性及算法的有效性。     

水电站水库优化调度几种求解方法的比较研究

针对水电站水库优化调度模型的求解,比较了常用的动态规划法、逐欢优化法(POA)和粒子群优化算法(PSO)3种求解方法.给出了3种算法的实现过程及约束处理方法,尤其针对粒子群优化算法,纠正了常见的约束处理和算法不结合问题.结论认为PSO水库优化调度模型和解法融合了模拟调度的思想,更容易为调度人员所接受,且能有效避免动态规划"维数灾"问题.但存在计算时间长,实时性差等问题,在实效要求较低情况下可以应用.     

基于混合粒子群算法并计及概率的梯级水电站短期优化调度

针对梯级水电站短期优化调度的不确定性问题,研究了不确定性因素的概率分布规律,并根据实际系统的运行要求,给出了概率分布密度函数的假设检验方法。探索发电用水量与各种随机因素的互动关系及影响机理,构建了一种新的计及概率的梯级水电站短期优化调度策略。把灾变理论、混沌优化思想和基本粒子群算法结合起来,形成一种混合粒子群算法。该算法扩大了种群的搜索空间,增加了种群的多样性,改善了基本粒子群算法摆脱局部极值点的能力,并能从理论上证明其依概率收敛至全局最优解。将混合粒子群算法嵌入蒙特卡罗随机模拟中对本文提出的模型进行求解,求解方法简单有效。仿真结果表明,该策略能较好地处理不确定性条件下梯级水电站的短期优化调度问题。     

含风电场电力系统经济调度的模糊建模及优化算法

随着人们对风电的日益关注,新型风电场的容量在并网系统中所占比例不断增加,这对传统电力系统的经济调度问题提出了新的要求。特别是风电场输出功率的随机变化给系统的经济调度带来了更多的不确定性因素。文中应用模糊理论建立了含风电场电力系统动态经济调度的模糊模型,使调度结果能够表达决策者的意愿,从而更好地适应风机输出功率的随机性。在优化算法上,利用下降搜索思想对传统粒子群算法进行了改进,并将改进后的算法用于求解提出的动态经济调度问题。在算例中进行了测试,结果验证了所提出的方法的可行性。     

A practical approach to real time economic dispatch consideringunit's prohibited operating zones

This paper develops an efficient and practical approach for determining the feasible optimal solution of the economic dispatch problem when some of the online units have prohibited operating zones. When a unit has prohibited operating zones, its operating region will be broken into isolated sub-regions, which results in multiple decision spaces for the economic dispatch problem. For each of the decision spaces, it can be either feasible or unfeasible with respect to the system demand. The feasible optimal solution for the dispatch problem can only reside in one of the feasible, spaces, and can be found by performing the conventional λ-δ iterative search based on the equal-incremental-cost criterion in each of the feasible spaces. However, for a system with a large number of decision spaces, such an exhaustive search method would not be acceptable in real time operation due to its high computational time requirement. In this paper, a novel approach is proposed. It defines a small and advantageous set of decision spaces, uses an efficient algorithm to determine the most advantageous space and then utilizes the conventional λ-δ iterative method to find the feasible optimal dispatch solution     

A Framework for Economic Load Frequency Control Design Using Modified Multi-objective Genetic Algorithm

This article deals with the coordination of security-constrained economic dispatch and load frequency control in an interconnected power system. The realistic and performance optimization inherent of the load frequency control (LFC) and security-constrained economic dispatch are fully considered without simplifying assumptions. For this purpose, modeling security-constrained economic dispatch as a discontinuous control action in the continuous frequency response model of a power system is well addressed. Considering conflict behavior of LFC and security-constrained economic dispatch beside the powerfulness of the multi-objective genetic algorithm (GA) to solve high-dimensional problems with conflicted objective functions makes it attractive for the automatic generation control coordination problem. The employed security-constrained economic dispatch utilizes the advantages of dynamic economic dispatch to achieve more realistic results. The GA is used to compute the decentralized control parameters and centralized generation levels of the on-line units to achieve an acceptable operating point. A significant modification in convergence speed has been performed by using LFC model properties in corporation with the genetic algorithm, so the proposed method gives considerable promise for implementation in multi-area power systems. The efficiency of the proposed algorithm and modification is demonstrated on a three control area power system.     

Voltage security enhancement via coordinated control

The problem of analytically coordinating dissimilar voltage control actions to prevent voltage collapse in a large power system is addressed. A framework for hybrid voltage control based on coordination of controls with different response time and dynamic characteristics is presented. The proposed method is based on a security constrained steady-state approach. The minimum distance from the operating point to the bifurcation boundary is used to evaluate system voltage security. The optimal control direction toward adequate security is then obtained by calculating the sensitivity of the minimum distance with respect to control parameters. The actual dispatch of controls along the optimal direction, which takes into account impacts of economic cost and control availability, is determined as the solution of a multiple-stage optimization problem using differential dynamic programming. The algorithm is demonstrated for a system with dynamic load models representing the main grid of New South Wales, Australia     

Improved Interval Optimization Method Based on Differential Evolution for Microgrid Economic Dispatch

This article proposes an improved interval optimization method based on differential evolution for dynamic economic dispatch of a microgrid. First, the dynamic economic dispatch model applied is a microgrid model considering various distributed generations, energy storage systems, the transaction between the microgrid and power grid, as well as multiple kinds of loads. Both economic and environmental costs are taken into account. Second, an improved interval optimization method based on differential evolution is proposed to solve this non-linear optimization model. The proposed method is improved in the aspects of the branch-and-bound strategy, acceleration tool, and interval estimation method. Finally, a practical dynamic economic dispatch case of a microgrid considering the valve-point effect is studied. A comparison among the proposed method, traditional interval method, differential evolution method, and interior point method is given to verify the efficiency and practicality of the improved interval optimization method.     

考虑电网静态安全的输电线路动态增容系统

针对电网调控的特点和需求,将动态增容技术从设备分析层面提升到电网分析层面,提出了动态增容在电网调控系统中应用的总体实施方案和功能模块设计。首先分析了调控系统中动态增容的应用原则,在此基础上提出了动态增容系统的总体架构、功能模块、接口设计和硬件结构;然后对系统的数据接入与处理、线路和输电断面动态限值计算、增容过程安全评估、电网增容预测分析等关键技术进行了阐述;最后介绍了该系统在南京地区的应用,基于电网实际运行数据对系统应用效果进行了分析。     

基于深度强化学习的综合能源系统动态经济调度

综合能源系统的优化调度对于实现系统的多能互补和经济运行具有重要意义.然而,系统中可再生能源的间歇性以及用户用能需求的不确定性造成了系统中供需双方的随机波动,传统的调度方法难以准确地适应实际环境的动态变化.针对这一问题,提出了一种考虑可再生能源和负荷时变特性的综合能源系统动态经济调度方法.首先对综合能源系统动态经济调度问...     

基于风速预测和随机规划的含风电场电力系统动态经济调度

随着风力发电在电力系统中比重的持续增加,在电力系统经济调度中需要考虑风电场的影响;而并网风电场具有动态、随机、容量大等特点,传统的经济调度方法已不再适用。针对风电场出力的随机性,在风速预测的基础上,应用随机规划理论建立了考虑机组组合的含风电场电力系统动元网络和遗传算法的混合智能算法,提高了算法的收敛速度和搜索性能。以含风电场的IEEE30节点系统为算例验证了所提方法的可行性和有效性。     

Power system operating state forecasting for security analysis applications

A methodology for power system operating state forecasting with applications in the study mode security analysis is described. Additional application areas are advance or dynamic dispatch and measurement replacement. The operating state is represented by the active and reactive nodal injections which are forecast using a four-step procedure: bus active and reactive load forecast, simulation of the economic dispatch and on-line load flow. The multivariate Box and Jenkins approach and the adaptive Kalman filter are used for the active and reactive bus load forecast, respectively. The results of numerical experiments with the proposed methodology indicate a performance with potential for practical applications.     

集群空调负荷双层动态优化调度决策方法

为了解决常规空调调度中切除负荷带来的电网运行稳定问题,在集群空调的调度中考虑空间上的优化组合,提出一种考虑网损和调度偏差最小的自上而下双层优化动态调度决策方法。在空调的聚类建模中,研究空调参数和运行状态的差异性,分析并验证聚类中的误差约束值和分组数对调度准确性的影响;在双层优化调度中,通过最优潮流计算合理地分配空间上调度的容量,有效减小网损与电压越限的可能性;结合动态调度,提高调度的准确性,保证用户的舒适性。IEEE 33节点配电网算例结果验证了所提算法的可行性和准确性。