基于搜索+调整的两阶段萤火虫算法求解机组组合问题

提出了一种新颖的基于搜索+调整的两阶段萤火虫算法求解机组组合问题。算法将机组组合求解流程分解为具有离散变量和连续变量的两个优化问题,通过二进制编码的萤火虫算法求解含离散变量的机组启停主问题,利用改进的实数编码萤火虫算法解决连续变量的负荷经济分配子问题,采用调整策略校核和修复约束,实现主子问题的交替迭代求解。算法通过启发式的约束调整策略,以及两种编码方式实现了离散变量和连续变量的分解优化,提高了机组组合问题求解的效率和精度。通过对6个不同规模算例的计算及与其他经典算法的对比,验证了所提算法的有效性和优越性。     

用于机组组合优化的蚁群粒子群混合算法

提出了一种用于求解机组组合优化问题的蚁群粒子群混合优化算法。通过将机组组合解编码为机组操作序列,降低了蚁群算法搜索的难度,使其空间复杂度由指数型降为线性型,使采用蚁群算法求解更大规模的机组组合问题成为可能。采用协同粒子群算法求解多时段负荷的经济分配问题时,用一个粒子群处理一个时段的优化问题,通过共享粒子群间的惩罚项解决了机组爬升率的约束问题。10机和20机系统的仿真实验和分析结果验证了该方法正确性、有效性和优越性。     

求解机组组合问题的嵌入贪婪搜索机制的改进粒子群优化算法

提出了一种求解机组组合问题的嵌入贪婪搜索机制的改进粒子群优化算法。其特点包括:采用固定阈值处理表示机组运行状态的0、1整型变量,从而可直接应用粒子群算法求解机组组合问题,避免求解各时段中的经济负荷分配子问题;在粒子群算法迭代过程中应用变异操作更新进化速度缓慢的粒子,增强了算法的搜索能力;算法收敛后,采用基于优先列表的贪婪搜索机制做进一步寻优,既加快了算法收敛速度,又提高了解的质量。算例结果表明所提出的方法在求解机组组合问题时具有很强的搜索能力和适应性。     

基于改进布谷鸟搜索算法的机组组合优化模型

为了获得更加理想的机组组合方案,提出一种改进布谷鸟算法的机组组合优化模型。首先建立机组组合的数学优化模型;然后采用布谷鸟算法对该数学模型进行求解,得到最优机组组合方案,同时为了解决标准布谷鸟搜索算法存在的不足,引入选择性淘汰策略和决策域策略,加快搜索速度和寻优能力;最后采用3机组和10机组组合问题对其性能进行仿真测试。仿真实验结果表明,改进布谷鸟算法具有更快的求解速度和求解准确度,获得了更优的机组组合方案。     

一种求解大规模机组组合问题的混合智能遗传算法

针对传统的采用二进制编码的遗传算法在求解大规模机组组合问题时收敛速度慢、易早熟等问题,作者结合机组组合问题的特点,提出了一种混合智能遗传算法.该算法以机组状态作为个体编码,结合启发式方法的自适应智能变异算子求解目标函数,显著缩小了求解问题的规模,保证了群体多样性,提高了算法的搜索效率,改善了算法的收敛性.仿真计算结果表明了该算法的有效性和实用性.     

基于自身计划的发电公司报价方案求解方法

提出了一种在考虑自身发电计划前提下,以矩阵实数编码遗传算法为基础的制定发电公司报价方案的新方法。该方法以报价方案中的主能量电量为优化变量,采用矩阵实数编码方式进行编码,然后进行基本报价方案的优化求解。在基本报价方案基础上,通过最小二乘法拟合得出最终报价方案。文中通过算例仿真及与其他算法的对比分析,验证了所提出的方法在发电公司报价方案制定问题的优化求解上是切实可行的。     

电力系统机组启停优化问题的改进DPSO算法

该文从微粒群优化算法的原理和机组组合问题的特点出发，提出了一种适合机组启停优化问题求解的改进的离散二进制微粒群优化算法(DPSO)：文中结合机组启停优化问题的特点，采用改进的DPSO算法对机组的开停机状态进行优化组合，利用随机的顺序投入法初始化原始种群，将无希望／重希望准则引入搜索过程，通过重新初始化机制与变异操作克服DPSO易于陷入局部最优的缺点，并保证机组的开停状态组合满足单机约束和系统约束。保证搜索在问题的可行域进行。对2个算例系统的仿真计算及与其它方法的比较表明，该算法在搜索精度和搜索速度方面均具有很大的优越性。此算法兼顾了收敛速度和收敛精度2个方面，具有很好的适应性。这种寻优的方式不仅为机组肩停优化问题带来了新的解决思路，对于求解更广泛的组合优化问题亦具有普遍的意义。     

低碳调度下机组组合问题的混沌遗传混合优化方法

提出了一种求解低碳调度下机组组合问题的混沌遗传混合优化方法。采用控制基因与参数基因编码方式对机组发电计划进行编码,通过结合电碳特征优先次序、混沌映射和随机生成3种方法提高初始种群多样性;将混沌迭代搜索引入到遗传算法的进化过程之中,构造新的变异算子,改进遗传算法过早收敛的缺点,并且在变异过程中进行按电碳特征优先权确定的区间偏移,达到了加快算法收敛速度的目的。通过算例验证了混沌遗传混合优化方法具有较好的收敛特性和全局搜索能力。     

求解机组组合问题的帝国竞争算法

提出一种用于解决机组组合问题的改进帝国竞争算法（imperialistic competition algorithm,ICA）。种群个体（国家）分为帝国和殖民地2种类型,两者又组成新的帝国,通过帝国吸收殖民地和帝国之间对殖民地的竞争实现寻优过程。在求解机组组合问题时,首先根据波峰波谷所在时刻将日负荷曲线分割成若干小时间段,然后利用ICA依次求解,与原ICA相比,避免了机组分类不合理对于优化结果的影响,同时采用整数编码的国家个体长度相对减小。最后对10~100机6个算例进行仿真分析,结果表明,算法在较少国家个体的情况下保持了较强的搜索能力,可获得较好的计算结果,是一种有效的优化算法。     

面向启发式调整策略和粒子群优化的机组组合问题

提出一种启发式调整策略和粒子群优化相结合的新方法求解电力系统中的机组组合(UC)问题.算法将UC问题分解为具有整型变量和连续变量的两个优化子问题,采用离散粒子群优化和等微增率相结合的双层嵌套方法对外层机组启、停状态变量和内层机组功率经济分配子问题进行交替迭代优化求解.同时构造了关机调整和替换调整两个启发式搜索策略对优化结果进行进一步局部微调以提高算法解决UC问题的全局寻优能力和计算效率,从而有效改善解的质量.以10～100台机组组成的5个测试系统为算例,通过与其他算法结果进行比较分析,验证了该方法的可行性和有效性.仿真结果表明该方法解决大规模机组组合问题具有求解精度高和收敛速度快的优势.     

A matrix real-coded genetic algorithm to the unit commitment problem

An approach to solving the unit commitment (UC) problem is presented based on a matrix real-coded genetic algorithm (MRCGA) with new repairing mechanism and window mutation. The MRCGA chromosome consists of a real number matrix representing the generation schedule. Using the proposed coding, the MRCGA can solve the UC problem through genetic operations and avoid coping with a suboptimal economic dispatch (ED) problem. The new repairing mechanism guarantees that the generation schedule satisfies system and unit constraints. The window mutation improves the MRCGA searching performance. Numerical results show an improvement in the solution cost compared with the results obtained from other algorithms.     

Ramp rate constrained unit commitment by improved priority list and augmented Lagrange Hopfield network

This paper proposes an improved priority list (IPL) and augmented Hopfield Lagrange neural network (ALH) for solving ramp rate constrained unit commitment (RUC) problem. The proposed IPL-ALH minimizes the total production cost subject to the power balance, 15 min spinning reserve response time constraint, generation ramp limit constraints, and minimum up and down time constraints. The IPL is a priority list enhanced by a heuristic search algorithm based on the average production cost of units, and the ALH is a continuous Hopfield network whose energy function is based on augmented Lagrangian relaxation. The IPL is used to solve unit scheduling problem satisfying spinning reserve, minimum up and down time constraints, and the ALH is used to solve ramp rate constrained economic dispatch (RED) problem by minimizing the operation cost subject to the power balance and new generator operating frame limits. For hours with insufficient power due to ramp rate or 15 min spinning reserve response time constraints, repairing strategy based on heuristic search is used to satisfy the constraints. The proposed IPL-ALH is tested on the 26-unit IEEE reliability test system, 38-unit and 45-unit practical systems and compared to combined artificial neural network with heuristics and dynamic programming (ANN-DP), improved adaptive Lagrangian relaxation (ILR), constraint logic programming (CLP), fuzzy optimization (FO), matrix real coded genetic algorithm (MRCGA), absolutely stochastic simulated annealing (ASSA), and hybrid parallel repair genetic algorithm (HPRGA). The test results indicate that the IPL-ALH obtain less total costs and faster computational times than some other methods.     

基于SP/A决策思想的发电公司报价决策模型及求解方法

以安全、潜力和期望(security, potential and aspiration, SP/A)风险决策理论为基础,该文对发电公司面对风险决策时应该考虑的因素进行了分析和数学描述,其中在报价方案潜力性的分析上有效地结合了经济学中机会成本理论。在此基础上,考虑发电公司自身发电计划约束,建立起一种发电公司计及风险因素的竞价决策模型,并结合矩阵实数编码遗传算法(matrix real-coded genetic algorithm, MRCGA)对该模型的优化求解进行了探讨。该文所建的竞价决策模型既考虑了自身发电计划安排,又顾及了在面对风险发电公司竞价决策时应该考虑的一些因素,因此模型比较贴近于实际的发电竞价情况。通过算例的模拟分析表明,文中所提出的基于SP/A的发电公司竞价决策模型是合理的,其求解方法也是切实可行的。     

基于电力系统日发电计划的混合智能messy遗传算法

机组组合是电力系统日发电计划中主要的优化任务,在满足各种约束条件下求得全局最优解是一个比较困难的问题.传统遗传算法的二进制编码和随机遗传操作不适合于求解大规模机组组合问题.针对电力系统日发电计划的特点,提出了一种混合智能messy遗传算法(HIMGA),该算法实现简单,大大减小了求解问题的规模,保证了群体的多样性,提高了算法的搜索效率,改善了算法的收敛性.仿真计算结果表明了该算法的有效性和实用性.     

Unit commitment using Lagrangian relaxation and particle swarm optimization

Unit commitment (UC) is a NP-hard nonlinear mixed-integer optimization problem. This paper proposes ELRPSO, an algorithm to solve the UC problem using Lagrangian relaxation (LR) and particle swarm optimization (PSO). ELRPSO employs a state-of-the-art powerful PSO variant called comprehensive learning PSO to find a feasible near-optimal UC schedule. Each particle represents Lagrangian multipliers. The PSO uses a low level LR procedure, a reserve repairing heuristic, a unit decommitment heuristic, and an economic dispatch heuristic to obtain a feasible UC schedule for each particle. The reserve repairing heuristic addresses the spinning reserve and minimum up/down time constraints simultaneously. Moreover, the reserve repairing and unit decommitment heuristics consider committing/decommitting a unit for a consecutive period of hours at a time in order to reduce the total startup cost. Each particle is initialized using the Lagrangian multipliers obtained from a LR that iteratively updates the multipliers through an adaptive subgradient heuristic, because the multipliers obtained from the LR tend to be close to the optimal multipliers and have a high potential to lead to a feasible near-optimal UC schedule. Numerical results on test thermal power systems of 10, 20, 40, 60, 80, and 100 units demonstrate that ELRPSO is able to find a low-cost UC schedule in a short time and is robust in performance.     

面向节能发电调度的日前机组组合优化方法

节能发电调度是对电网优化调度机制的重大修改。机组组合是电网调度的重要环节,随着节能发电调度的逐步推广,需要结合中国国情研究新形势下机组组合模型与优化方法。提出一种求解电力系统机组组合的新方法,将机组组合问题分解为末状态和状态改变时间优化2个过程。基于节能发电调度通过多贪婪因子完善机组排序指标,利用贪婪算法确定机组组合初始解,进而结合深度优先算法遍历机组组合方案以保证问题优化的深度。10机24时段系统算例表明,该方法可有效处理机组组合各类约束条件及保证节能调度效果。     

A solution to the ramp rate and prohibited operating zone constrained unit commitment by GHS-JGT evolutionary algorithm

Unit commitment (UC) problem on a large scale with the ramp rate and prohibited zone constraints is a very complicated nonlinear optimization problem with huge number of constraints. This paper presents a new hybrid approach of ’Gaussian Harmony Search’ (GHS) and ’Jumping Gene Transposition’ (JGT) algorithm (GHS-JGT) for UC problem. In this proposed hybrid GHS-JGT for UC problem, scheduling variables are handled in binary form and other constants directly through optimum conditions in decimal form. The efficiency of this method is tested on ten units, forty units and hundred units test system. Simulation results obtained by GHS-JGT algorithm for each case show a better generation cost in less time interval, in comparison to the other existing results.     

Unit commitment by an enhanced simulated annealing algorithm

A new simulated annealing (SA) algorithm combined with a dynamic economic dispatch method has been developed for solving the short-term unit commitment (UC) problem. SA is used for the scheduling of the generating units, while a dynamic economic dispatch method is applied incorporating the ramp rate constraints in the solution of the UC problem. New rules concerning the tuning of the control parameters of the SA algorithm are proposed. Three alternative mechanisms for generating feasible trial solutions in the neighborhood of the current one, contributing to the reduction of the required CPU time, are also presented. The ramp rates are taken into account by performing either a backward or a forward sequence of conventional economic dispatches with modified limits on the generating units. The proposed algorithm is considerably fast and provides feasible near-optimal solutions. Numerical simulations have proved the effectiveness of the proposed algorithm in solving large UC problems within a reasonable execution time.     

HSA based solution to the UC problem

Unit-commitment (UC) as a complicated problem needs powerful methods to solve. This paper presents the use of harmony search algorithm (HSA), a recently developed meta-heuristic algorithm, in order to obtain optimal solution for the UC problem. The proposed algorithm has simple implementation and provides optimal solutions in a reasonable time. The method is tested using small and large scale test cases in the literature. Numerical results show that the proposed algorithm can find better solutions in comparison with conventional methods and it is an efficient way to solve UC problems especially in large-scale power systems.