An efficient algorithm for optimal generation mix taking into account fuzziness of decision making and planning parameters

This paper presents an efficient computational algorithm for selecting the optimal generation mix under uncertain circumstances. Subjective, experiential or linguistic uncertainties are selected from among various uncertainties, i.e., we treat fuzziness in generation expansion planning. The fuzziness can be divided into: (1) the fuzziness of decision making; and (2) the fuzziness of some planning parameters, such as load growth, fuel price, and so on. Both classes of fuzziness are integrated into a fuzzy decision based on fuzzy sets theory, and then the optimal generation mix can be determined by the Fuzzy Dynamic Programming (FDP) technique. The proposed method, which is based on the dynamic programming technique, is extended by using the Bellman-Zadeh maximizing decision. In the method, each generation technology and generation capacity are selected as a stage and state, respectively. The proposed method can easily accommodate not only the fuzziness but also many constraints of generation expansion planning, such as integer solutions of unit capacities, condition of existing units, and so on. Furthermore, the arbitrary shape of membership function can be used. The effectiveness and feasibility of the proposed method are demonstrated on a typical power system model.     

基于并行动态规划的水库发电优化

动态规划是水库发电优化调度研究中的经典方法。为了提高动态规划算法计算性能以及适应多核计算技术发展趋势,本文在分析传统串行动态规划算法计算特点的基础上,建立基于主从模式的并行动态规划模型,并将其运用到水布娅水库的发电优化计算中。计算结果表明,在多核运算平台下,并行动态规划算法能够充分利用多核资源,有效提高算法计算效率,缩短计算时间。     

大渡河梯级水库群联合优化调度函数研究

采用＂优化-拟合-再优化-随机仿真＂技术框架研究梯级水库群联合优化调度函数：首先建立梯级水库群优化调度模型,通过改进的动态规划方法得到最优样本;然后对最优样本进行回归分析得到联合调度函数;并基于模拟调度结果采用非线性规划法直接修正调度函数;最后通过水文随机模拟技术生成径流系列,进一步检验、评价所得调度函数的效率。以长江上游大渡河梯级水库长期优化调度为对象进行实例研究,计算结果表明：对于1959-2000年实测径流系列,梯级水库群联合优化调度函数较常规调度年均发电量增幅达1.66%;对于Copula随机模型生成的水文径流系列,年均发电量提高了1.42%,经济效益显著。     

基于连续线性规划的梯级水电站优化调度

梯级水电站优化调度是一个多时段、多变量和多约束条件的大规模优化问题,其求解过程非常复杂。文章尝试采用连续线性规划的优化方法来解决梯级水电站长期优化调度问题。通过采用泰勒级数一阶描述形式,对优化调度目标函数和约束条件中的非线性约束进行线性化处理,建立了基于连续线性规划算法的优化调度数学模型,提出了用连续线性规划技术求解梯级水电站优化调度问题的算法,并采用迭代步长的动态比例缩减因子保证算法能快速准确地收敛到优化问题的最优解。利用Matlab7.0编制连续线性规划梯级水电站优化调度程序,一个两级梯级水电站群的仿真分析结果表明,该算法可用于求解梯级水电站优化调度问题,并可快速得到非线性问题的最优解。     

Proposal of a simplified method for integrating reduction of effective power of pumped storage into a generation system planning method

Pumped storage hydropower stations are extensively used as peaking and reserve power plants. The current trend in this area is to increase their power output relative to the capacity of reservoirs, which results in a shorter maximum duration of full‐power operation while availability of the energy required for the pumping becomes less evident. This increases the possibility of reducing the effective power of pumped storage units in the case of a tight supply–demand balance in the power system. Although many studies have examined a role of pumped storage in generation systems, few studies have explicitly dealt with the power reduction. This paper proposes a new linear programming model that incorporates the reduction of effective power of pumped storage. The proposed model makes it possible to easily incorporate the power reduction in computations by employing hypothetical daily load curves and dispatching loads to the curves. The model also includes LNG combined cycle power generation plants that are now being intensively implemented. The developed model is applied to a power system model. The sample study reveals that the reduction of effective power considerably varies from season to season and the reduction highly affects the optimal generation mix, that is, the optimal share of pumped storage. © 2001 Scripta Technica, Electr Eng Jpn 134(4): 50–61, 2001     

基于动态规划的分布式电源优化配置

电源规划是电力系统中非常重要的规划。本论文在分析平顺绿色能源资源条件、电力负荷需求的基础上,根据当地实际情况,应用动态规划的优化方法结合各种约束条件,得到分布式电源的规划方案。     

梯级水电站群优化调度控制研究及解决方案

研制了分层的乌江梯级水电站群优化调度控制系统,由3个子系统组成:水情测报系统、水库调度系统和梯级远程集中监控系统。将调度控制的优化分解为3个子系统的优化控制。运用了梯级最大发电量、梯级最大蓄能量和梯级各电站最小库水位越限程度3个优化准则,并针对各个优化准则建立了相应的数学模型。在对适用于梯级水电站群优化调度计算的各类优化算法进行分析并结合实际运用情况的基础上,提出应优先考虑采用改进动态规划算法求解梯级水电站群优化调度问题。实际运行情况表明,优化调度控制大幅提高了乌江梯级水电站群的经济效益和运行管理水平。     

基于动态规划法的电力市场下最优发电计划的实现

负荷分配是编制短期发电计划所面临的主要问题。为了有效地解决各种类型报价曲线的负荷经济分配 ,并考虑到计算速度、精度及收敛性等问题 ,采用了一种改进的动态规划法———局部加密动态规划法 ,并采取了限制机组组合状态的措施。详细论述了应用这种改进算法解决实际问题的方案设计思想及软件实现过程 ,并用C语言编制了程序     

Optimal planning of rural medium voltage distribution networks

A model is suggested for selection of main initial parameters and timing of the reconstructions of rural distribution networks in long-term planning to meet the increasing load demands with minimum total present worth cost. The model incorporates capital and exploitation costs as well as the costs due to undelivered energy and load curtailments. The optimal investment policy is determined using a constrained dynamic programming technique which indicates the best choice among possible options while taking into account all relevant technical criteria and common-sense rules.     

An evaluation of the optimal generation mix for controlling carbon dioxide emissions

This paper presents an efficient computational algorithm for selecting the optimal generation mix considering CO₂ emissions. To demonstrate the effectiveness and feasibility of the proposed method, a fundamental study of the evaluation of the optimal generation mix for controlling CO₂ emissions is indicated. Furthermore, by using a parametric analysis which considers load characteristics as parameters, a general trend for the optimal generation mix which is affected by controlling CO₂ can be derived. The proposed method is based on an optimization method known as simulated annealing. In the method, solutions in a generation mix problem are equivalent to state of a physical system, and the cost of a solution is equivalent to the energy of a state. The proposed method can easily accommodate not only CO₂ emissions but also many practical constraints of generation expansion planning, such as integer solutions of unit capacities, condition of existing units, and so on. Case studies with various annual load patterns (combinations of annual load factors and the shapes of annual load duration curve) are presented and discussed. Consequently, a general trend for selecting generation technologies that should be added to a power system is derived, i.e., a useful guideline for studying generation expansion planning under controlling CO₂ emissions can be provided.     

Application of parallel genetic algorithms to generation expansion planning using parallel processors

This paper presents an application of parallel genetic algorithms (PGA) to the optimal long-range generation expansion planning. The problem can be formulated as a combinatorial optimization problem that determines the order of introduced generation units at each interval of the year. The proposed method considers introduced power limits of each technology, maximum loads at each interval, and load duration curves at each interval. Appropriate string representation for the problem is presented. Binary and decimal coding and three selection methods are compared. The method is developed on a transputer that is one of the parallel processors. The feasibility of the proposed method is demonstrated using a typical expansion problem with four technologies and five intervals. The method is then compared with conventional dynamic programming and a simple genetic algorithm with promising results.     

基于决策支持系统的配电网优化规划

开发了基于决策支持系统的配网优化规划软件包, 具有系统规划、方案分析、方案决策、方案绘图和数据库管理系统等功能。静态规划建立了混合整数规划模型, 利用基于线性规划换基运算的支路交换法有效地求解该模型。动态规划, 采用临界可行匹配算法和伪动态匹配法在全临界匹配空间找到最优化动态决策。采用支路自动分解、最优乘子的快速分解法潮流计算, 收敛性好、速度快、占内存少。算例说明该软件包适用于大规模配电网规划, 使用方便灵活     

基于运行数据的水电站MILP模型最优代表水头选择取方法

针对采用设计水头的水电站混合整数线性规划（mixed integer linear programming, MILP）调度模型计算的出库流量与实际出库流量偏差较大的问题,提出了基于运行数据的水电站MILP模型最优代表水头选取方法。首先基于运行数据,采用MILP模型拟合出使模型计算出库流量过程与水电站实际出库流量过程偏差最小的代表水头,然后在实际调度中,以日平均入库流量和日平均出力作为该代表水头特征向量,根据预测入库流量和日计划电量即可选取最优代表水头。计算实例表明,相比于传统固定水头,该方法能够更好的反应水电站实际的出库过程,有利于提高电网制定调度计划中梯级水电站上下游水量匹配精度。     

火电厂厂级负荷优化分配系统研制

提出一种新颖的火电厂厂级负荷优化分配系统的设计与解决方案.采用OPC服务器与IEC608705-104远动通信方式替换了传统负荷分配系统方案中的硬接线,简化了系统硬件结构.同时软件上增加了机组负荷指令偏置调整和手动功能,分配方式更加多样.通过分析基于动态规划的多目标负荷优化分配算法,同时要求兼顾负荷响应快速性和电厂经济...     

动态规划和粒子群算法在水电站厂内经济运行中的应用比较研究

随着电站装机容量和机组台数的不断增加,利用动态规划求解水电站厂内经济运行问题,将面临"维数灾"和实效性问题.近些年,粒子群算法作为一种新型的群体智能优化方法,由于能够弥补动态规划计算时间长、内存占用量大等诸多不足,在水电站厂内经济运行等方面得到了广泛重视.现有文献,大多数从方法的应用角度探讨较多,但从替代动态规划的必然性和潜力方面探讨较少,鲜有实例分析.本文以百万级装机千瓦的乌江渡水电站为实例,深入分析与比较了粒子群算法与动态规划的优劣,认为粒子群算法是代替动态规划、求解装机规模庞大的巨型水电站厂内经济运行的有效方法.     

动态规划在电力建设项目投资决策中的应用

电力建设项目的投资决策是一个复杂的决策系统，同时也是一个多阶段，多状态的动态决策过程，作者利用动态规划的多阶段寻优的特点，建立了动态规划的数学模型，对电力建设项目进行了分阶段的评价和选择，通过求解动态规划模型，获得电网建设的最优投资规模和投资时。文章将方法应用在汕头的220KV电网建设上，得出了汕头电网建设的动态决策结果，得到规划期内各年的最佳电网建设项目安排，使电网建设投资具有科学性和可行性。     

梯级水电站长期优化调度的细粒度并行离散微分动态规划方法

伴随水电规模的扩大,水电站群优化调度的计算量不断增加,需要探求新的方法.在分析离散微分动态规划(discrete differentiation and dynamic programming,DDDP)算法的基础上,提出了基十分治模式的梯级水电站长期优化调度的细粒度并行离散微分动态规划(parallel discrete differentiation and dynamic programming,PDDDP)方法,并以澜沧江梯级的6个电站系统长期优化调度问题为应用实例,在多核计算环境下进行验证.结果表明,多核环境下的PDDDP方法简便易行,能充分利用闲置计算资源、大幅度提高优化调度的计算效率,是解决大规模复杂水电系统调度的高效和实用方法.     

水电厂参与市场竞争下的市场模拟及其市场势力削弱

为评估水电厂参与市场竞价时的市场势力问题,建立了多时段市场模拟模型,模型由动态规划法求解,各时段各状态下的决策为多厂商博弈的均衡点,各均衡点由迭代算法求解。通过比较分散调度与统一调度下计算结果的方法评估市场势力。算例表明,从中长期看,水电厂分散调度比统一调度电价更高且波动性大,采取双边合同可明显降低实时电价、减弱水电厂的市场势力。     

基于动态规划法的电力市场下最优发电计划的实现

负荷分配是编制短期发电计划所面临的主要问题。为有效解决各种类型报价曲线的负荷经济分配,并考虑到计算速度、精度及收敛性等问题,采用了一种改进的动态规划法——局部加密动态规划法,并采取了限制机组组合状态的措施。详细论述了应用这种改进算法解决实际问题的方案设计思想及软件实现过程,并用C语言编制程序。     

熵分析在水电能源系统管理中的应用

水电能源系统是一个典型的开放系统,与外界环境存在着频繁的物质和能量交换。根据热力学熵的基本概念及熵概念泛化的基本特性,提出了水电能源系统熵的定义,并进行了系统电站熵变分析;结合水电能源系统电站的特点,提出了径流调节过程中基于决策出流量的水头、水量潜在熵增计算表达式;建立了系统电站发电量优化规划即熵增最小目标方程,应用凸规划原理及凸函数性质论证了目标方程存在最优解;最后提出了水电能源系统优化规划模型,实例验证该模型符合实际规划需求。