变负荷条件下追踪电力系统安全最优运行轨迹的参数化方法

提出一种追踪线性约束下凸可分规划问题最优解轨迹的参数化方法。该参数优化算法可在对偶松弛凸可分规划算法的主循环之外,通过少量参数化扩展得到。参数分析表明最优解轨迹的性态是一条分段线性曲线,解轨迹上的破点和不可行现象存在密切关系。将这种方法应用到电力系统有功最优潮流问题中,得到一种统一经济调度和安全约束调度的参数化安全约束调度(security constrained economic dispatch,SCED)新算法,它可快速追踪变负荷条件下系统安全最优运行轨迹。算法在IEEE14-300节点测试系统及2个省级实际系统上通过测试,数值试验和几何分析表明了该方法的计算特性和物理内涵,同时清楚地显示了目前调度模式中存在的问题和改进方向。     

求解电力系统多目标环境经济调度的帕累托最优MFO算法

本文提出了一种利用MFO算法解决电力系统环境经济调度的新方法,该算法利用飞蛾扑火原理对设定目标进行螺旋式搜索,并在目标位置进行重复检索。MFO算法对于大规模非线性规划问题具有较强的适应性和有效性。在求解环境经济调度问题中,结合实际发电系统运行过程中应满足的功率平衡约束和容量约束等,以总燃料成本和污染排放最低为目标建立多目标规划数学模型。运用帕累托最优前沿求取帕累托非劣性最优解,得到帕累托最优配置方案,在可行域中搜索出全局最优解。在MATLAB仿真平台对含40台发电机组系统进行仿真计算,结果表明本文提出算法在求解电力系统环境经济调度中具有较高的收敛性和较强的适应性。     

水电站群优化调度非线性全局优化方法

水电站群优化调度是大规模、高维、非凸、非线性优化问题。传统解析式规划、动态规划(dynamic programming,DP)及系列方法、智能群体算法等很难保证在可接受时间内获得原问题的全局最优解。该文引入一种非线性全局优化方法,采用凸分析、区间分析、代数分析将原非凸、非线性问题转换为一系列凸、线性子问题,利用分支定界法遍历所有子问题,直至求得全局最优解。以澜沧江和金沙江水电站群长期调度为例,与DP等经典算法相比,该方法可以获得全局最优解,最大降低内存占用率99%以上,10座水电站的优化计算平均耗时仅5s,计算速度比DP逐次逼近法提高约50倍,为破解大规模水电优化调度维数灾难题提供新的技术途径。     

基于Q-learning算法的配电网储能装置控制策略研究

通过在配电网末端接入用于系统调压等辅助服务的储能装置，能有效解决可再生能源的高度间歇性和负荷需求波动导致的配变过载问题。基于强化学习的Q-learning算法，针对储能电池运行情况进行建模仿真，通过单时段优化内嵌的Q值得到各时段储能电池荷电状态的最优调度方案。实例试验分析表明，当迭代次数达到一定数量时，利用Q-learning算法能够达到理论上的最优解。最后，通过将Q-learning算法与动态规划算法生成的标准最优调度方案进行对比，证明了Q-learning算法能够与动态规划算法达成一致最优解。     

基于多智能体的混合发电机组调度问题研究

合理分配机组负荷,提高运行效率,降低生产成本是发电厂面临的重要任务,传统机组调度算法一般通过非线性规划或者群智能算法实现,但是随着技术的发展,新能源机组和传统机组混合调度问题的出现,传统的方法难以得到满意的调度规划结果。针对混合发电机组复杂调度问题,采用多智能体来联合调度优化。仿真实验表明,所提出的基于多智能体的优化模型能够得到复杂调度模型的最优解。     

基于改进灰狼算法的交直流混联电网经济调度方法

为提升交直流混联电网经济调度问题求解效率,提出了一种基于改进灰狼算法的日前经济调度方法。直流输电线路与交流输电线路在网损表达式、传输特性等方面存在较大差别,造成交直流混联电网经济调度模型较传统交流电网更为复杂。由于该问题涉及大量的混合整数规划变量,传统智能算法容易陷入局部最优解,难以获得满足要求的最优解。通过引入初始对立搜索、自适应局部搜索等改进措施,提出了一种改进灰狼算法,并将其应用于含网损的交直流混联电网日前经济调度问题中。基于IEEE RTS-96三区域节点系统的算例表明,相比于粒子群算法、飞蛾扑火算法,改进算法能保证相近的求解效率基础上,避免陷入局部最优解,具有更强的搜索能力。     

基于遗传算法的电力超负荷平衡调度研究

在对电力超载负荷平衡调度的研究中,由于传统的基于时空双尺度方法采取的是单路径搜索最优解,会存在只有初值在最优值附近时才可能找到最优解,否则得到的只能是局部最优解的问题。针对这样的问题,提出了基于遗传算法优化的电力超载负荷平衡调度方法,在获得电力负荷多种状态数据的基础上,建立电力超载负荷平衡调度的精确数学模型,针对电力超载负荷平衡调度问题的求解,设计了一种基于遗传算法的超负荷调度方法,能够将同时间内电力超载负荷约束在一定的差异范畴内,有效地搜索了全局最优解,并通过数值计算验证了该方法的有效性。实验结果表明,利用遗传算法的电力超载负荷平衡调度稳定、准确度高。     

宣威市调度、通信与自动化规划研究

针对城市电网规划的通讯和调度优化问题,重点研究了城市光缆规划原则和光传输网规划原则,进而提出了宣威中压配网自动化规划方案和宣威市配网通信规划方案研究,最后提出了通信介质和接口的选择,同时做出了投资的估算。算例优化表明,该文方法实用性,适用性均较强,可以快速有效地得到网架结构规划全局最优解。     

发电机组启停机的智能优化经济调度研究

遗传算法用于解决启停机优化调度问题中,常出现群体早熟和有时收敛于局部最优解等问题.而模拟退火算法在接受新解时却显示出较好的特性.在遗传算法的评价函数中引入模拟退火算法以及在选择操作中采用模拟退火算法的接受准则,将两者进行混合可有效地缓解其选择压力,增强算法的全局收敛性.采用十进制编码,无需解码,可减少计算误差和时间.算例分析表明该算法可以在满足安全可靠的多种约束条件下,较好地改善机组启停计划的经济性.     

变尺度混沌优化算法在梯级水电站水库优化调度中的应用

利用变尺度混沌优化算法(Mutative Scale Chaos Optimization Algorithm,MSCOA)对梯级水电站水库调度问题进行优化调度。主要思想是利用混沌运动的随机性,由Logistic方程随机生成混沌序列;将其载波到包含水电站目标函数可行域S的一个区域;利用随机性、遍历性和规律性,不断缩小优化变量的搜索空间和提高搜索精度进行全局寻优,从中搜索属于可行域S的解;同时在搜索中引入解向量优选,将解向量中那些接近全局最优解的分量找出,构成一个新的向量,代入目标函数中进行计算,从而找出全局最优解,最终求出水电站水库发电调度的最优调度线。实例计算结果表明,算法可以求解具有复杂约束条件的非线性梯级水电站水库优化调度问题。算法求解精度高,具有较大的实用价值,为求解梯级水电站水库优化调度问题提供了一种有效算法。     

A solution method using neural networks for the generator commitment problem

This paper studies the feasibility of applying the Hopfield-type neural network to unit commitment problems in a large power system. The unit commitment problem is to determine an optimal schedule of what thermal generation units must be started or shut off to meet the anticipated demand; it can be formulated as a complicated mixed integer programming problem with a number of equality and inequality constraints. In our approach, the neural network gives the on/off states of thermal units at each period and then the output power of each unit is adjusted to meet the total demand. Another feature of our approach is that an ad hoc neural network is installed to satisfy inequality constraints which take into account standby reserve constraints and minimum up/down time constraints. The proposed neural network approach has been applied to solve a generator scheduling problem involving 30 units and 24 time periods; results obtained were close to those obtained using the Lagrange relaxation method.     

机组组合问题的仿射可调整鲁棒优化模型与算法

为了有效应对电力系统调度决策中的不确定因素,尤其是大规模间歇式能源并网所带来的不确定性,提出基于仿射可调整鲁棒优化理论的不确定机组组合求解方法。建立了不确定机组组合问题的仿射可调整鲁棒优化模型,利用线性决策规则建立决策变量与不确定参数之间的仿射关系,从而将两阶段问题转化为单个阶段优化问题,在此基础上,采用对偶理论将模型转化为可以直接求解的标准混合整数规划模型。通过标准算例测试,验证了该方法的有效性。     

A new method for unit commitment with ramping constraints

Unit commitment with ramping constraints is a very difficult problem with significant economic impact. A new method is developed in this paper for scheduling units with ramping constraints within Lagrangian relaxation framework based on a novel formulation of the discrete states and the integrated applications of standard dynamic programming for determining the optimal discrete states across hours, and constructive dynamic programming for determining optimal generation levels. A section of consecutive running or idle hours is considered as a commitment state. A constructive dynamic programming (CDP) method is modified to determine the optimal generation levels of a commitment state without discretizing generation levels. The cost-to-go functions, required only for a few corner points with a few continuous state transitions at a particular hour, are constructed in the backward sweep. The optimal generation levels can be obtained in the forward sweep. The optimal commitment states across the scheduling horizon can then be obtained by standard dynamic programming. Numerical testing results show that this method is efficient and the optimal commitment and generation levels are obtained in a systematic way without discretizing or relaxing generation levels.     

Generation scheduling with thermal stress constraints

This paper describes a scheduling method for representing the thermal stress of turbine shafts as ramp rate constraints in the thermal commitment and dispatch of generating units. The paper uses Lagrangian relaxation for optimal generation scheduling. In applying the unit commitment, thermal stress over the elastic limit is used for calculating the ramping cost. The thermal stress contribution to generation cost requires the calculation of a set that includes thermal stress at the end of each time step; this requirement presents a complicated problem which cannot be solved by an ordinary optimization method such as dynamic programming. The paper uses an improved simulated annealing method to determine the optimal trajectory of each generating unit. Furthermore, the paper uses linear programming for economic dispatch in which thermal stress limits are incorporated in place of fixed ramp rate limits. The paper illustrates the economics of frequently ramping up/down of low cost generating units versus the cost of replacement of their turbine rotors with a shorter life span. The experimental results for a practical system demonstrate the effectiveness of the proposed method in optimizing the power system generation scheduling.     

基于SCUC的可入网混合电动汽车优化调度方法

在可入网混合电动汽车(PHEV)有望规模化应用的背景下,以传统的计及安全约束的机组最优组合(SCUC)问题为基础,发展了能够容纳PHEV的电力系统优化调度数学模型。所发展的模型以保证系统安全运行为前提,兼顾了PHEV车主的经济效益与发电的碳排放成本。利用PHEV作为可移动电量储存单元的特性,将模型解耦为机组最优组合与计及交流潮流约束的充/放电计划优化2个子模型。应用混合整数规划方法和牛顿—拉夫逊潮流算法迭代求解优化问题,可以同时获取日前机组调度计划和各时段的PHEV最优接纳容量及充/放电计划等结果。最后,以6节点和IEEE 118节点2个系统为例,验证了所构建模型的正确性和有效性。     

计及可入网电动汽车最优时空分布的双层经济调度模型

为研究大规模电动汽车入网(V2G)对电网的影响,文中以传统计及网络安全约束的机组组合问题为基础,构建了计及V2G的经济调度双层优化数学模型,并充分考虑电动汽车的时空分布特性,将模型解耦为机组最优组合和电动汽车最优充/放电计划两个子模型。分别采用基于牛顿—拉夫逊潮流计算思路的非线性规划方法和粒子群优化算法求解该模型。算例表明,该模型可以同时获取次日机组调度计划和各时段电动汽车最优充放电计划。时间上,实现24时段实时优化控制;空间上,将V2G调度计划细分到各接入节点,对充放电站的规划选址具有指导意义。模型在实现降低发电成本的同时使得配电网网损最小,且削峰填谷作用明显。     

电力市场环境下解决机组组合问题的新方法

机组组合问题是电力市场环境下编制短期发电计划所面临的主要问题,在满足各种约束条件的情况下,如何合理地开、停机组、以及负荷如何在运行的发电机组之间经济地分配是一个比较困难的问题,特别是由于发电机组出力上升、下降速度的限制,使这个问题一直没有很好的解决方法。提出一种组合优化方法解决这一问题,即用启发式方法确定机组组合,用分段线性规划算法分配功率,并满足各种约束条件,特别是可以处理发电机组出力上升、下降速度约束、经实际系统检验是一种非常有效的算法。     

Power system operations scheduling using separable programming

The problem of electric power operations scheduling comprises maintenance scheduling, intermediate- and short-term hydro-thermal scheduling, unit commitment and production scheduling. Recognizing the inter-relationship among these subproblems, the effect on one another is analyzed. Computationally efficient formulations and algorithms for maintenance scheduling and intermediate- and short-term hydro-thermal scheduling are presented.A variation of the separable programming technique is presented. This results in a linear programming formulation of the nonlinear scheduling problem. This technique is applied to a hypothetical system containing nuclear, fossil, hydro-electric and pumped-storage units. The large problem of hydro-thermal scheduling due to the inclusion of a nuclear unit is decomposed into two stages. In the first stage, the relatively stable nuclear generation is optimized with respect to the generation from large fossil-steam units. Hourly generation levels for all units in the system are then determined in the second stage.     

Uncertainty in operational planning of power system generation

Short-term scheduling of a power generation system is discussed with regard to the probabilistic character of the load demand and the effects of generation capacity outages. The probabilistic method and a practical algorithm of optimal unit commitment and load distribution among power stations is presented. An example of the computations performed for a large power system is given.     

A new unit commitment method

This paper introduces a new unit commitment method based on a decommitment procedure for solving the power system resource scheduling problem. From an initial schedule of all available units committed over the study period, a `one-at-a-time' unit decommitment is accomplished by dynamic programming according to some specified economic criteria. The decommitment process continues until no further reduction in total cost is possible, or the unit schedules of two consecutive iterations over the time period remain unchanged without any violation of the spinning reserve constraint. Two criteria for decommiting a unit are introduced and described in detail. Comparisons of the proposed unit commitment method with the Lagrangian relaxation (LR) approach and Fred Lee's sequential unit commitment method (SUC) demonstrate the potential benefits of the proposed approach for power system operations planning