基于改进混合蛙跳算法的梯级水库优化调度

针对混合蛙跳算法在寻优过程中出现的早熟收敛问题,利用混沌技术的遍历性优势对子群最优个体进行变异操作,形成局部精细搜索策略;根据蛙群相对多样性参数来判断算法是否陷入局部最优,进而对蛙群最优个体进行扰动以提高全局寻优能力,形成全局激励调节策略。耦合2种策略,提出了一种改进混合蛙跳算法。将其应用于李仙江梯级水库优化调度中,结果表明所提算法具有寻优质量高、收敛速度快的特点,有效地克服了标准混合蛙跳算法的早熟缺陷,为水库调度模型的求解提供了一种新方法。     

Multi-Objective Differential Evolution-Chaos Shuffled Frog Leaping Algorithm for Water Resources System Optimization

A multi-objective differential evolution-chaos shuffled frog leaping algorithm (MODE-CSFLA) is proposed for water resources system optimization to overcome the shortcomings of easily falling into local minima and premature convergence in SFLA. The performance of MODE-CSFLA in solving benchmark problems is compared with that of non-dominated sorting genetic algorithm II (NSGA-II) and multi-objective particle swarm optimization (MOPSO). At last, the proposed MODE-CSFLA is used to optimize the water resources allocation plan of the East Route of the South-to-North Water Transfer Project in the normal, dry, and extremely dry years. The results reveal that MODE-CSFLA performs better than NSGA-II and MOPSO under all conditions. Compared with shuffled frog leaping algorithm (SFLA), MODE-CSFLA can result in a 29.39, 27.47 and 22.55% increase in water supply when the single objective is to minimize the water pumpage; and a 41.01, 39.63 and 30.94% decrease in total pumpage when the single objective is to maximize the water supply in the normal, dry, and extremely dry conditions, respectively. Thus, MODE-CSFLA has the potential to be used for solving complex optimization problems of water resources systems.     

基于混合粒子群算法的梯级水电站多目标优化调度

提出多目标混合粒子群算法以解决梯级水电站多目标联合优化调度模型求解的难题。该算法采用混合蛙跳算法的分组－混合循化优化框架,增强算法全局搜索能力,在族群内通过粒子群算法高效灵活的飞行调整策略指导个体进化,同时,引入外部精英集,建立一种基于自适应小生境的外部精英集维护策略,提高算法的收敛性和非劣解集的多样性。最后将该算法应用于三峡梯级水电站多目标优化调度工程应用实例,结果表明,本文算法能够获得计算实时性强、分布均匀、收敛性好的调度方案集,并以此分析明确了调度目标间的耦合关系,为梯级电站的多目标调度决策提供了科学依据。     

鲸鱼优化算法在水库优化调度中的应用

为验证鲸鱼优化算法在水库优化调度求解中的可行性和有效性,采用4个典型测试函数对鲸鱼优化算法进行仿真验证,并与布谷鸟搜索算法、差分进化算法、混合蛙跳算法、粒子群优化算法、萤火虫算法和SCE-UA算法共6种算法的仿真结果进行对比分析;将鲸鱼优化算法与6种对比算法应用于某单一水库和某梯级水库中长期优化调度求解。结果表明:鲸鱼优化算法寻优精度高于其他6种算法8个数量级以上,具有收敛速度快、收敛精度高和极值寻优能力强等特点;鲸鱼优化算法单一水库和梯级水库优化调度结果均优于其他6种算法;鲸鱼优化算法应用于水库优化调度求解是可行和有效的。     

多组群教学优化算法-神经网络-支持向量机组合模型在径流预测中的应用

采用5个标准测试函数对多组群教学优化(MGTLO)算法进行仿真验证,并将仿真结果与基本教学优化(TLBO)算法、混合蛙跳算法(SFLA)、差分进化(DE)算法和粒子群优化(PSO)算法的仿真结果进行对比。利用MGTLO算法搜寻基于广义回归神经网络(GRNN)、径向基神经网络(RBF)、支持向量机(SVM)模型单元的组合模型的最佳模型参数和组合权重系数,提出MGTLO-GRNN-RBF、MGTLO-GRNN-SVM、MGTLO-RBF-SVM、MGTLO-GRNN-RBF-SVM 4种组合预测模型,以新疆伊犁河雅马渡水文站和云南省某水文站年径流量预测为例进行了实例分析,并将预测结果与MGTLO-GRNN、MGTLO-RBF、MGTLO-SVM和GRNN、RBF、SVM 6种单一模型的结果进行对比分析。结果表明:MGTLO算法寻优精度优于TLBO、SFLA、DE和PSO算法,具有较好的收敛速度和全局极值寻优能力;组合模型融合了MGTLO算法与GRNN、RBF、SVM模型单元的优点,在预测精度、泛化能力等方面均优于单一模型;MGTLO算法能有效优化各组合模型的相关参数和权重系数,MGTLO-GRNN-RBF-SVM模型预测精度最高。     

梯级水库生态调度多目标混合蛙跳差分算法研究

提出多目标混合蛙跳差分算法求解梯级水库多目标生态调度模型。该算法结合混沌理论生成初始解以提高初始解群体质量,构建基于动态更新机制的外部归档集引导种群进化,提高算法的收敛性与非劣解的多样性,引入自适应差分算法加快子种群个体寻优,提高算法收敛速度。对L河梯级水库多目标生态调度进行实例研究,计算结果表明:本文所提出的算法能够计算得到收敛性与分布性较好的调度方案集,对比典型调度方案下泄径流与物种生态适宜径流,表明生态调度能够较好满足物种的生态需水,生态效益显著。     

基于混合蛙跳算法的混凝土坝加权变形预报模型

利用混合蛙跳算法的优化特点,将大坝安全监控统计模型的求解转换为多目标函数的优化问题;引入调整系数修正回归因子,考虑调整系数与回归因子之间的协调关系,利用混合蛙跳算法同步确定调整系数和回归系数,建立基于混合蛙跳算法的混凝土坝加权变形预报模型。工程算例应用结果表明,该模型具有较优的中长期预报能力,可提高大坝安全监控统计模型的预报精度,在大坝安全监控领域具有一定的工程应用意义。     

SFLA-PP模型在区域水资源利用效率综合评价中的应用

为验证混合蛙跳算法(Shuffled Frog Leaping Algorithm,SFLA)-投影寻踪(Projection Pursuit,PP)模型应用于区域水资源利用效率综合评价中的有效性和可行性。从综合、工业、农业、生活和生态环境5个方面遴选15个指标构建区域水资源利用效率评价指标体系,利用SFLA算法优化PP模型最佳投影方向,提出SFLA-PP水资源利用效率评价模型,与构建的入侵杂草优化(Invasive Weed Optimization,IWO)算法、粒子群优化(Particle Swarm Optimization,PSO)算法和布谷鸟搜索(Cuckoo Search,CS)算法优化的PP模型作对比,并基于各模型投影系列均值及标准差σ构造4个等级的水资源利用效率评价标准,以文山州8县市水资源利用效率评价为例进行实例研究。结果表明:SFLA算法优化PP模型获得的适应度值为715.800 2,均优于IWO,PSO和CS算法PP模型,具有较好的求解精度和极值寻优能力;SFLA-PP模型对文山市、砚山县水资源利用效率评价为“高水平”,对麻栗坡县评价为“低水平”,其余5县评价为“中等水平”,全州水资源利用效率综合评价为“较高水平”;SFLA-PP模型对实例评价及排序结果与IWO-PP模型相同;与PSO-PP模型、CS-PP模型在评价结果及排序上均存在差异。实例验证了4种智能算法PP模型的求解精度对区域水资源利用效率的评价结果起到关键作用。     

混凝土坝变形组合预报模型的蛙跳建模方法

混凝土坝变形的影响因素多而复杂,监测数据包含大量的不确定性信息,因此依据多种建模方法建立的各单一模型的预报效果各不相同.组合模型能够弥补单一模型的局限性,如何在建模过程中使权重系数实时反映单一模型对监测信息的变化,是组合模型建模的关键.从优化极值角度考虑各单一模型中未知量与权重系数之间的相互性,利用蛙跳算法(SFLA)的分布式全局优化性能同步确定其值,提出相应的蛙跳优化建模方法.混凝土坝变形长期监测资料的应用表明,该方法具有良好的预报效果,简化了组合模型的确定过程,提高了模型的预报能力,为混凝土坝变形预报分析提供了新的计算方法.     

Application of the Shuffled Frog Leaping Algorithm for the Optimization of a General Large-Scale Water Supply System

A water supply system is a complex network of pipes, canals and storage and treatment facilities that collects, treats, stores, and distributes water to consumers. Increasing population and its associated demands requires systems to be expanded and adapted over time to provide a sustainable water supply. Comprehensive design tools are needed to assist managers determine how to plan for future growth. In this study, a general large-scale water supply system model was developed to minimize the total system cost by integrating a mathematical supply system representation and applying an improved shuffled frog leaping algorithm optimization scheme (SFLA). The developed model was applied to two hypothetical water communities. The operational strategies and the capacities for the system components including water transport and treatment facilities are model decision variables. An explicit representation of energy consumption cost for the transporting water in the model assists in determining the efficacy of satellite wastewater treatment facilities. Although the water supply systems studied contained highly nonlinear terms in the formulation as well as several hundred decisions variables, the stochastic search algorithm, SFLA, successfully found solutions that satisfied all the constraints for the studied networks.     

基于最大熵投影寻踪模型的云南省近10 a水资源承载力评价

为对云南省2006—2015年水资源承载力进行评价,从水资源、经济社会和生态系统3个方面提出水资源承载力评价指标体系和分级标准,并基于最大熵投影寻踪(MEPP)技术进行区域水资源承载力评价。采用在指标分级标准阈值间随机生成样本的方法构造MEPP目标函数,利用混合蛙跳算法(SFLA)优化MEPP最佳投影方向,提出SFLA-MEPP水资源承载力评价模型,并构建生物地理优化(BBO)算法-MEPP、和声搜索(HS)算法-MEPP和粒子群优化(PSO)算法-MEPP水资源承载力评价模型作对比模型。结果表明SFLA寻优MEPP目标函数获得的最优值、最劣值、平均值和标准差均优于BBO、HS和PSO算法,具有较好的全局极值寻优能力;SFLA-MEPP模型对云南省2006—2007年、2011—2012年水资源承载力评价为“基本可承载”,其他年份评价为“可承载”;2006—2015年间云南省水资源承载力随时间呈提升趋势,但提升趋势不显著;SFLA-MEPP模型对云南省水资源承载力评价结果与BBO-MEPP模型相同,但在排序上存在差异;与HS-MEPP,PSO-MEPP模型在评价结果及排序上均存在差异。     

基于可行域搜索映射的并行动态规划

为解决传统动态规划在处理水库群联合优化调度时面临的约束处理机制选择和计算时间长的问题,引入映射思想,基于映射和集合论知识构建可行域搜索映射模型,并结合动态规划的并行性,提出基于可行域搜索映射的并行动态规划。该算法通过构建时段可行搜索空间和并行模式,以规避无效状态组合计算并充分发挥计算机多核优势,提高计算效率。以李仙江流域三库联合调度为实例进行研究,从年发电量、计算耗时等方面将改进算法与传统动态规划以及逐步优化算法(POA)进行详细的对比分析。结果表明,该算法能在保证解全局收敛性的前提下减少计算耗时,制定梯级水库最优调度策略。     

Cloud-Based Multidimensional Parallel Dynamic Programming Algorithm for a Cascade Hydropower System

The joint optimal operation of cascade reservoir system can greatly improve the utilization of water resources. However, the complex high-dimensional and non-linear features and calculated costs often hinder the refined operation and management of reservoirs. Recently, the local parallel computing has become an effective way to alleviate the "curse of dimensionality". Current local parallel computing has hardware limitations, which is difficult to adapt to large-scale computing. This study proposes a novel parallel dynamic programming algorithm based on Spark (PDPoS) via cloud computing. The simulation experiments are carried out for a comparative analysis of the solution efficiency, influence factors and stability of cloud computing. The results are as follows: (1) The efficiency of the cloud-based PDPoS is related to some factors; the number of CPU cores is the main influencing factor, followed by the operator, and the architecture has the least influence. (2) The runtime variance of cloud computing is 2.03, indicating cloud computing has high stability. (3) Under the same configuration (i.e., CPU and memory), the runtime of cloud computing is 41.5%?~?110.3% longer than that of physical machines. However, cloud computing has rich resources, good scalability, and good portability of online operations, which is an attractive alternative for optimal operation of large-scale reservoir system.

     

Integration of Shuffled Frog Leaping Algorithm and Support Vector Regression for Prediction of Water Quality Parameters

Quality of surface water is a serious factor affecting human health and ecological systems. Accurate prediction of water quality parameters plays an important role in the management of rivers. Thus, different methods such as (support vector regression) SVR have been employed to predict water quality parameters. This paper applies SVR to predict eight water quality parameters including (sodium (Na⁺), potassium (K⁺), magnesium (Mg⁺²), sulfates (SO₄ ^?2), chloride (Cl^?), power of hydrogen (pH), electrical conductivity (EC), and total dissolved solids (TDS)) at the Astane station in Sefidrood River, Iran. To achieve an efficient SVR model, the SVR parameters should be selected carefully. Commonly, various techniques such as trial and error, grid search and metaheuristic algorithms have been applied to estimate these parameters. This study presents a novel tool for estimation of quality parameters by coupling SVR and shuffled frog leaping algorithm (SFLA) . Results of SFLA-SVR compared with genetic programming (GP) as a capable method in water quality prediction. Using SFLA-SVR, average of RMSE for training and testing of six combinations of data sets for all of the water quality parameters improved 57.4 % relative to GP. These results indicate that the new proposed SFLA-SVR tool is more efficient and powerful than GP for determining water quality parameters.     

改进差分进化算法在水库优化调度中的应用研究

差分进化算法在求解水库优化调度时,进化后期种群多样性急剧下降,导致算法无法跳出局部最优解而出现“早熟”收敛。针对该问题,该文对算法的贪婪选择策略进行改进,使其以一定的概率动态接受稍差解作为子代个体,从而提高算法的种群多样性;同时,提出种群基因重生策略,进一步改善种群进化的基因信息结构。将改进的差分进化算法应用于清江梯级发电调度问题,并与差分进化算法、模拟退火算法求解结果进行对比。模拟结果表明,改进算法具有更强的全局搜索能力,求解梯级水库优化调度问题更具有优势。     

基于并行混沌量子粒子群算法的梯级水库群防洪优化调度研究

梯级水库群防洪优化调度问题规模庞大、结构复杂,涉及大量的决策变量和复杂的约束条件,各水库、各时段之间的水位、流量存在复杂的耦合关系,呈现出高维度、非线性、强约束特性,传统的优化方法难以直接求解或者计算效率低,存在早熟收敛问题。研究工作试图将量子粒子群算法(QPSO)引入到水库群防洪优化调度问题中,为了提高算法的全局搜索能力和收敛性能,对标准QPSO做了改进,包括利用混沌思想初始化种群、自适应激活机制和精英粒子混沌局部搜索策略3个方面,并引入多核并行计算技术以降低计算时间,提出了并行混沌量子粒子群算法(PCQPSO),函数测试证明了PCQPSO的可行性、稳定性和高效性。将PCQPSO应用到水库群防洪优化调度问题中,与POA、QPSO进行对比分析,结果表明PCQPSO收敛效率快、求解精度高,为解决梯级水库群防洪优化调度问题提供了一种有效的新思路。     

Multidimensional Parallel Dynamic Programming Algorithm Based on Spark for Large-Scale Hydropower Systems

The “curse of dimensionality” is a major problem in dynamic programming (DP) algorithms for large-scale hydropower systems. This study proposes a parallel DP algorithm based on Spark (PDPoS) to alleviate the “curse of dimensionality”. Parallel computing experiments are formulated by varying the number of reservoirs, the number of discrete water levels and the number of CPU cores to analyze the quality and efficiency of PDPoS. The methodologies were applied to a cascade reservoir system made up of eight reservoirs in the Yuanshui River Basin in China. The results are as follows. (1) The number of discrete water levels is the dominant factor in the solution quality, while the number of reservoirs is the dominant factor in the solving efficiency. (2) The runtime of PDPoS is markedly affected by the calculational scale (determined by the number of reservoirs and discrete water levels), and the relationship between the number of CPU cores and the runtime is triphasic with increasing calculational scale. (3) The larger the calculational scale is, the better the parallel performance (i.e., the parallel speedup and parallel efficiency). The proposed PDPoS method has strong generality, high parallel performance, and high practical value.

     

A Demand Management Based Crop and Irrigation Planning Using the Simulation-Optimization Approach

Coupled simulation-optimization models are useful tools for solving optimum water allocation and crop planning problems. In this study, the optimum crops pattern in the Arayez plain in the Karkheh river basin in Iran is determined by integration of a network flow programming (NFP) based simulation model and the shuffle frog leaping optimization algorithm (SFLA) in the form of a simulation-optimization approach. MODSIM applies NFP for finding water allocations which by use of its customization ability, the benefit of water supply for the agricultural crops is calculated based on the agronomic equations. The objective function is to maximize the total net benefit gained from crops production where the decision variables which are the irrigation depths and the cultivation areas are optimized by SFLA. Results show that by use of the coupled SFLA-NFP model, the net benefit increases 12% comparing the present situation in the plain. Also, the sensitivity analyses on effective parameters indicate that the potential maximum yield and the net price of the crops yield in the market have a direct impact on the crops optimum cultivation area.     

粗粒度并行自适应混合粒子群算法及其在梯级水库群优化调度中的应用

为了充分利用现今普及的多核配置计算机,提高大规模梯级水库群优化调度问题的求解效率,提出了梯级水库群优化调度的粗粒度并行自适应混合粒子群算法。该方法以自适应混合粒子群算法为求解基础,采用粗粒度并行设计模式,利用Fork/Join多核并行框架的分治策略,将其初始种群递归划分为多个子种群,平均分配到不同的内核逻辑线程中实现并行计算,并在各子种群优化结束后,合并优化结果集从而输出全局最优解。以澜沧江下游梯级水库群发电优化调度为例,利用该方法进行计算。结果表明,该方法能充分发挥多核配置的计算性能,在4核环境下最大加速比达到3.97,缩短计算耗时1 787.2 s,计算效率显著提高,为我国不断扩张的大规模梯级水库群优化调度提供了一种切实可行的高效求解途径。     

边坡临界滑动面确定的蛙跳算法

将蛙跳算法引入到边坡极限平衡稳定性分析中,详细介绍了蛙跳算法的流程,并对典型土坡进行了计算分析,结果表明:蛙跳算法搜索到的最危险滑动面上较多的点位于软弱夹层内,所得到的安全系数比遗传算法结果更为合理。通过正交试验分析了蛙跳算法各个参数对计算结果的影响,并给出了推荐的参数取值。