Forecasting Groundwater Levels using a Hybrid of Support Vector Regression and Particle Swarm Optimization

Water Resources Management - Forecasting the groundwater level is crucial to managing water resources supply sustainably. In this study, a simulation–optimization hybrid model was developed...     

Application of Artificial Neural Networks and Particle Swarm Optimization for the Management of Groundwater Resources

Ground management problems are typically solved by the simulation-optimization approach where complex numerical models are used to simulate the groundwater flow and/or contamination transport. These numerical models take a lot of time to solve the management problems and hence become computationally expensive. In this study, Artificial Neural Network (ANN) and Particle Swarm Optimization (PSO) models were developed and coupled for the management of groundwater of Dore river basin in France. The Analytic Element Method (AEM) based flow model was developed and used to generate the dataset for the training and testing of the ANN model. This developed ANN-PSO model was applied to minimize the pumping cost of the wells, including cost of the pipe line. The discharge and location of the pumping wells were taken as the decision variable and the ANN-PSO model was applied to find out the optimal location of the wells. The results of the ANN-PSO model are found similar to the results obtained by AEM-PSO model. The results show that the ANN model can reduce the computational burden significantly as it is able to analyze different scenarios, and the ANN-PSO model is capable of identifying the optimal location of wells efficiently.     

Sustainable Conjunctive Water Use Modeling Using Dual Fitness Particle Swarm Optimization Algorithm

Water Resources Management - In any meta-heuristic algorithm, each search agent must move to the high-fitness areas in the search space while preserving its diversity. At first glance, there is no...     

基于量子行为粒子群优化算法的河道糙率反演

糙率是河道水动力模型的重要参数,在模型中敏感性高,但其在实际工作中难以准确测定。量子行为粒子群算法(QPSO)是粒子群算法的发展,相对于粒子群算法,在全局收敛和收敛率上有很大提高。将量子行为粒子群优化算法与一维河道水动力模型耦合,建立河道糙率反演模型,并在淮河干流蚌埠到花园咀河段进行了模拟,取得了较好的效果。与其他糙率反演算法相比,该算法具有理论简单、参数少、易于编程实现、通用性强等优点。     

水电站机组优化组合的混合粒子群优化算法

机组组合是水电站短期发电计划中一个非常重要的问题,合理的组合运行能带来显著的经济效益,开展对机组优化组合的可行性和有效性研究有重大的现实意义。建立了该问题的数学模型,并提出了混合粒子群算法（Hybrid Particle Swarm Optimization,HPSO）的工程实现方法,采用量子粒子群算法解决机组方案的确立,并采用粒子群算法求解负荷经济分配。设计了粒子的适应度计算方法和速度更新方法,提出了HPSO算法的求解步骤。仿真分析表明：HPSO算法求解机组优化组合问题是可行和有效的,该算法实现简单,具有更快更好的收敛性能。     

A New Groundwater Management Model by Coupling Analytic Element Method and Reverse Particle Tracking with Cat Swarm Optimization

The performance of groundwater management models mostly depends upon the methodology employed to simulate flow and transport processes and the efficiency of optimization algorithms. The present study examines the effectiveness of cat swarm optimization (CSO) for groundwater management problems, by coupling it with the analytic element method (AEM) and reverse particle tracking (RPT). In this study, we propose two coupled simulation-optimization models, viz. AEM-CSO and AEM-RPT-CSO by combining AEM with RPT and CSO. Both the models utilize the added advantages of AEM, such as precise estimation of hydraulic head at pumping location and generation of continuous velocity throughout the domain. The AEM-CSO model is applied to a hypothetical unconfined aquifer considering two different objectives, i.e., maximization of the total pumping of groundwater from the aquifer and minimization of the total pumping costs. The model performance reflects the superiority of CSO in comparison with other optimization algorithms. Further, the AEM-RPT-CSO model is successfully applied to a hypothetical confined aquifer to minimize the total number of contaminant sources, within the time related capture zone of the wells, while maintaining the required water demand. In this model, RPT gets continuous velocity information directly from the AEM model. The performance evaluation of the proposed methodology, illustrates its ability to solve groundwater management problems.     

Short-Term Flood Control and Long-Term Energy Maximization in Multi-reservoir Systems Using Improved Particle Swarm Optimization

Water Resources Management - In this study, a model is developed for short-term flood control of a complex multi-reservoir system located on one of the largest basins in Turkey. The managing body...     

Parameters Estimation for the New Four-Parameter Nonlinear Muskingum Model Using the Particle Swarm Optimization

The Muskingum model is a popular method for flood routing in river engineering. This model has several parameters, which should be estimated. Most of the techniques have applied to estimate these parameters to reduce the distance between observed flow and estimated flows. In this paper, for the first time, the parameters of a novel form of the nonlinear Muskingum model are estimated by the Particle Swarm Optimization (PSO) algorithm. The new Muskingum model, which have four parameters, is applied for three benchmark examples and one real case in Iran. The sum of the squared (SSQ) or absolute (SAD) deviations between the observed and estimated outflows was considered as objective functions. The results showed that although the new Muskingum model became more complex but this model by using PSO technique can improve the fit to observed flow especially in multiple-peak hydrographs.     

Identification of Pollution Sources in Transient Groundwater Systems

A methodology using a nonlinear optimization model is presentedfor estimating unknown magnitude, location and duration ofgroundwater pollution sources under transient flow and transportconditions. The proposed optimization model incorporates thegoverning equations of flow and solute transport as binding equalityconstraints, and thus essentially simulates the physical processes oftransient flow and transient transport in the groundwater systems.The proposed inverse model identifies unknown sources of pollution byusing measured values of pollutant concentration at selectedlocations. Performance of the proposed model for the identificationof unknown groundwater pollution sources is evaluated for anillustrative study area in a hypothetical confined aquifer underdifferent cases of data availability. The effect of observation welllocation vis-à-vis pollution source location on identificationaccuracy is also investigated. Performance of the developedidentification model is also evaluated for a condition whenconcentration measurements are missing during few initial timeperiods after the pollution sources become active. The effect ofspecified initial guesses of the variable values on the optimalsolutions are also investigated. These performance evaluation resultsdemonstrate the limitations and potential applicability of theproposed optimization model for identifying the sources of pollutionin transient groundwater systems.     

改进粒子群优化算法在水文模型参数优化中的应用研究

引入改进的粒子群优化算法,对垂向混合产流模型计算参数进行优化,并对比参数优化前后水文模拟精度。研究结果表明：改进的粒子群优化算法模型可较快完成参数优化,相比于参数优化前,垂向混合产流模型年尺度模拟相对误差减少6.15%,模拟的过程确定性系数平均提高0.11;在次洪模拟中,模拟相对误差平均减少3.03%,模拟的洪水过程确定性系数平均提高0.19,水文模拟精度得到较大程度提高。研究成果对于区域水文模型参数优化提供参考价值。     

Multiobjective Calibration of Reservoir Water Quality Modeling Using Multiobjective Particle Swarm Optimization (MOPSO)

Water resource management encounters large variety of multi objective problems that require powerful optimization tools in order to fully characterize the existing tradeoffs between various objectives that can be minimizing difference between forecasted physical, chemical, and biological behaviors of model and measured data. Calibration of complex water quality models for river and reservoir systems may include conflicting objectives addressed by various combinations of interacting calibration parameters. Calibration of the two dimensional CE-QUAL-W2 water quality and hydrodynamic model is an excellent example where the model must be calibrated for both hydrodynamic and water quality behavior. The aim of the present study is to show how multiobjective particle swarm optimization (MOPSO) can be implemented for automatic calibration of water quality and hydrodynamic parameters of a 2-dimensional, hydrodynamic, and water quality models (CEQUAL-W2) to predict physical, chemical, and biological behaviors of a water body, and then focus on a relevant case study. So MOPSO is utilized to generate Pareto optimal solutions for two conflicting calibration objectives. A combined measure of thermal and reservoir water level is considered as the first calibration objective. The second objective is formulated to forecast the best physical, chemical, and biological behavior of the model. Realizing the strong interactions between water quality and hydrodynamic issues of water bodies and their dependencies on the same set of calibration parameters, the proposed multiobjective approach may provide a wide version of all possible calibration solutions for better decision making to select best solution from pareto front.     

Determining Rainwater Harvesting Storage Capacity with Particle Swarm Optimization

Water Resources Management - A reduction in the amount of available clean water is a universal problem, and the harvest of rainwater is one of the methods that can solve this issue. In this study,...     

粒子群算法及其在暴雨强度公式参数优化中的应用

暴雨强度公式中含有多个参数,用传统的方法难以直接优化,或拟合误差较大.通过把粒子群优化算法(PSO)应用到暴雨强度公式参数优化中,经实例分析表明该算法与遗传算法相比具有较好应用效果.     

改进粒子群优化算法在水库优化调度中的应用

为解决粒子群优化算法寻优过程中易出现种群趋同化而导致早熟收敛的问题,引入粒子群的进化速度和种群多样性适应度方差两个因素,构建了自适应的动态的惯性因子取值机制,并讨论了惯性因子的收敛性及参数的独立性,从而改进了传统粒子群优化算法的惯性因子线性取值机制。将改进的粒子群优化算法应用于某水库的优化调度中,验证了该算法能以较快的速度收敛得到全局极值,克服了易陷入局部最优的缺点,为水库优化调度问题提供了一条新途径。     

纳米零价铁用于地下水污染修复时存在的问题与对策

介绍了纳米零价铁(Nanoscale Zero-Valent Iron,NZVI))用于地下水污染修复时存在的问题及改进技术。NZVI可以去除水体中的各种卤代烃(主要是氯代烃)以及Cr、Pb、As等重金属,但NZVI在水体中会由于团聚、沉淀和钝化等而降低其修复效率。将超声波技术和添加微生物方法与NZVI技术进行协同,以及对NZVI进行固体负载和表面改性是解决NZVI修复地下水时存在问题的有效途径。今后,探讨地下复杂环境因素对应用NZVI进行地下水修复的影响,以及改进NZVI水污染修复技术是利用NZVI修复地下水污染的重要研究方向。     

Large scale reservoir operation by Constrained Particle Swarm Optimization algorithms

In this paper two adapted versions of Particle Swarm Optimization (PSO) algorithm are presented for the efficient solution of large scale reservoir operation problems with release volumes taken as the decision variables of the problem. In the first version, exploiting the sequential nature of the solution building procedure of the PSO, the continuity equation is used at each period to define a new set of bounds for the decision variable of the next period which satisfies storage volume constraints of the problem. Particles of the swarm are, therefore, forced to fly in the feasible region of the search space except for very rare cases and hence the name of the Partially Constrained Particle Swarm Optimization (PCPSO) algorithm. In the second, the periods of the operations are treated in a reverse order prior to the PCPSO search to define a new set of bounds for each storage volume such that partially constrained particles are not given any chance of producing infeasible solutions and, hence, the name of Fully Constrained Particle Swarm Optimization (FCPSO) algorithm. These methods are used here to solve two problems of water supply and hydropower operation of “Dez” reservoir in Iran and the results are presented and compared with those of the conventional unconstrained PSO and a genetic algorithm. Three cases of short, medium and long-term operations are considered to illustrate the efficiency and effectiveness of the proposed methods for the solution of large scale operation problems. The methods are shown to be superior to the original PSO and genetic algorithm in locating near optimal solutions and convergence characteristics. Proposed algorithms are also shown to be relatively insensitive to the swarm size and initial swarm compared to the original unconstrained PSO and genetic algorithm.     

Efficient Identification of Unknown Groundwater Pollution Sources Using Linked Simulation-Optimization Incorporating Monitoring Location Impact Factor and Frequency Factor

This study aims to improve the accuracy of groundwater pollution source identification using concentration measurements from a heuristically designed optimal monitoring network. The designed network is constrained by the maximum number of permissible monitoring locations. The designed monitoring network improves the results of source identification by choosing monitoring locations that reduces the possibility of missing a pollution source, at the same time decreasing the degree of non uniqueness in the set of possible aquifer responses to subjected geo-chemical stresses. The proposed methodology combines the capability of Genetic Programming (GP), and linked simulation-optimization for recreating the flux history of the unknown conservative pollutant sources with limited number of spatiotemporal pollution concentration measurements. The GP models are trained using large number of simulated realizations of the pollutant plumes for varying input flux scenarios. A selected subset of GP models are used to compute the impact factor and frequency factor of pollutant source fluxes, at candidate monitoring locations, which in turn is used to find the best monitoring locations. The potential application of the developed methodology is demonstrated by evaluating its performance for an illustrative study area. These performance evaluation results show the efficiency in source identification when concentration measurements from the designed monitoring network are utilized.     

改进粒子群算法在梯级水库优化调度中的应用

针对粒子群优化算法搜索精度不高和易陷入局部最优解的缺陷,结合梯级水库联合优化调度的实际问题,提出了惯性权重的余弦处理机制和自适应选择学习对象策略,并结合分层交叉思想对粒子群算法进行了改进。实例计算表明:该算法合理、可靠,有效地提高了计算精度与效率。     

基于粒子群算法的灌溉水泵系统优化设计与运行

文章提出了一种相对较新的用于灌溉抽水系统优化设计和运行的管理模式。该管理模式利用粒子群优算法化建立并求解了一个两步优化模型。新提出的模型通过对所有可行的泵机组组合进行详尽的枚举搜索后,在所需时间段内处理给定的需求曲线,然后调用粒子群优化算法搜索每个集合的最优解。在优化机组的运行问题后,计算所有机组的运行总成本和初始投资折旧,确定最优的机组组合,并制定相应的运行策略。研究将所提出的模型用于实际泵站系统的设计和运行后,将结果与优化算法的结果进行比较。结果表明,所提出的模式与粒子群优化算法相结合是一种用于实际灌溉泵系统设计和运行的通用管理模型。     

基于粒子群算法的水利类试算问题求解

基于粒子群算法,提出水利工程试算问题的求解方法,与传统的迭代算法相比,此方法可以避免一些非线性复杂问题迭代公式的推算.最后应用该方法对实例进行了分析,结果证明了这种方法可以提高试算问题解的精度及求解效率,由于具有通用性,编制相关计算程序后,只需要更换其中的目标函数,即可完成复杂试算问题的求解,故具有良好的推广应用前景.