一种多种群混合搜索微分进化算法优化换热网络

针对基本微分进化算法应用于换热网络优化时易陷入局部极值而出现早熟收敛的问题,提出了一种多种群混合搜索DE(微分进化)算法。首先,利用对立操作进行种群的多轮再生,提高了解空间的多样性;其次,各个按照不同的局部搜索机制实现种群的进化、更新,提高算法的局部搜索能力;最后,经算例研究表明:本文所述方法不仅能够有效提高种群多样性,克服算法易陷入局部极小值的缺点,且算法的稳定性和全局收敛能力均有显著提升。为启发式算法解决复杂换热网络优化问题提供了进一步的研究基础。     

基于动态拓扑结构的分布式Memetic差分进化算法同步综合换热网络

针对标准DE(差分进化)算法在优化换热网络时出现的局部搜索能力弱、易陷入局部最优等问题,本文建立了一种基于动态拓扑结构的分布式Memetic差分进化算法,同步综合换热网络。首先,在子种群内部采用基于欧拉距离的动态拓扑结构,子种群之间采用冯诺依曼拓扑结构,有效地加快了个体之间的信息交流,保持种群多样性,扩大搜索范围。继之,结合Memetic算法思想,将Hooke-Jeeves算法作为局部搜索策略,增强算法局部搜索能力。同时,对于局部搜索获得的新解,提出了一种协作学习机制,平衡算法的全局寻优与快速收敛能力。最后,为处理整型变量,提出了两条简单有效的整型变量优化策略,使算法实现了连续变量与整型变量的同步优化。选取两个经典算例验证了算法的有效性。算法应用于算例一,相对于现有文献的最优值,本文所得结构的费用值下降了1 783$/a,表明算法的性能优于标准DE算法以及其它改进版本的DE算法。算法应用于算例二,相对于现有文献的最优值,本文所得结构的费用值下降了1 209$/a,表明算法能够有效地处理大规模换热网络问题,具有很强的鲁棒性。     

基于种群多样性的改进差分进化算法应用于换热网络优化

针对差分进化算法运用于换热网络优化时后期搜索效率低,容易出现早熟收敛而陷入局部最优的不足,以变量方差为种群多样性定量评价指标。分析了差分进化算法的控制参数对种群多样性的影响及种群多样性与算法寻优能力之间的关系,在此基础上建立缩放因子自适应调整和种群动态更新策略的改进差分进化算法并应用于两个换热网络实例,优化参数以年综合费用为目标,结果分别为5 606 985和2 928 032$/a,较改进前分别减少19 089和18 042$/a。改进后的差分进化算法能够在进化过程中提升种群多样性,增强算法搜索能力,抑制算法早熟收敛。     

基于局部搜索增强策略的微分进化算法应用于换热网络优化

针对微分进化算法应用于换热网络优化时,局部搜索能力不强及易出现"早熟收敛"的现象,提出了一种兼顾局部搜索和全局寻优能力的改进策略。首先,对算法出现早熟的主要原因进行了机理分析;在此基础上,从动态更新角度引入局部增强算子,以当前解为中心进行邻域精细搜索,保证了局部搜索精度;最后,结合强制跳出机制,解决了算法易陷入局部极值的缺陷,对于非凸非线性问题较为严重的换热网络问题,实现了换热网络优化全局寻优能力和局部搜索精度的兼顾。算例结果表明,新的局部搜索增强策略能够进一步改善种群多样性和提高寻优效率及精度,获得比文献中结果更好的解,有效提高微分进化算法应用于与换热网络优化的能力。     

隔代强制进化遗传算法在换热网络优化中应用

在对换热网络分级超结构及其数学模型分析的基础上,对网络综合优化问题进行了研究．针对普通遗传算法及其它优化算法无法保证网络优化质量和效率的缺点,对遗传算法进行了改进,提出了换热网络隔代强制进化遗传算法。该方法将换热网络结构信息转化为种群中染色体信息,利用种群的进化实现网络结构的优化,在进化过程中使用隔代强制策略,使种群向更优方向稳步进化,保证各代优化结果的有效性,降低最优群体的生成代数,并利用最优个体保存技术记录优化过程中最佳换热网络结构。采用此方法对具体换热网络实例进行了优化综合,结果表明：隔代强制进化遗传算法能在网络优化过程中避免早熟收敛而陷入局部最小点的现象,使搜索质量和效率得到有效提高。用隔代强制进化遗传算法对有分流和无分流换热网络进行优化综合,均能获得综合性能良好的网络结构。     

一种应用于换热网络综合的伴随优化策略

为解决强制进化随机游走算法(random walk algorithm with compulsive evolution,RWCE)应用于换热网络综合时进化停滞的问题,提出了一种伴随优化策略(CO-RWCE):对种群中各个体优化进程进行监控,当个体因接受差解陷入长期进化停滞时,将该个体历史最优解回代给个体以调整优化方向;若多次回代后个体仍未进化,则将全局最优解传给个体并摄动以实现重生。在回代或重生后,采用一种游走概率递减技术,控制游走变量个数以提升搜索精度。优化结果表明:改进策略增强了个体自身进化能力,有效提升了算法搜索能力。     

采用新单元换热量生成与分布概率协调策略优化换热网络

强制进化随机游走算法应用于换热网络优化,具有算法程序简单、结构进化能力强等特点,但种群个体进化后期依然很难找到使年综合费用再次降低的进化方向。鉴于此,分析新生成换热单元最大换热量(Q_(max))取值对优化过程及新生成换热单元换热量(Q_n)概率分布的影响,在此基础上采用换热单元换热量生成与分布概率协调的换热网络优化策略,摄动后小概率随机生成换热量较大的换热单元,同时改变Q_n的概率分布情况,用于增强结构进化能力。最后采用15SP和20SP算例验证该策略的可行性,较文献结果分别降低了435 498和42 253$/a,由此证明,该策略可有效提高算法的局部搜索精度和全局搜索能力。     

PSO用于优化换热网络时对全局搜索能力的研究

将PSO(粒子群算法)应用于优化换热网络时,能够快速找到一个全局搜索的最优区域,但同时也会出现局部极值问题。这些问题是由于全局搜索能力的退化和算法的早熟收敛所致。本研究针对该退化现象的机理进行了深入分析,找到了粒子群算法早熟收敛的本质,提出了一种强制跳出的改进策略,通过激活陷入局部极值的粒子,恢复种群多样性并继续搜索全局最优解。算例证明,改进后的粒子群算法的搜索策略适用于换热网络连续变量优化,应用于文献[16]10SP2算例,得到的年综合费用较文献[14]下降了205$/a;应用于文献[18]8SP1算例,得到了目前最小的费用30 793$/a。改进的PSO较标准PSO和文化基因PSO优化后的费用均有所下降。     

采用个体进化能力实时评价和强制更新策略的换热网络优化方法

针对启发式方法在优化换热网络时出现个体进化停滞的现象,提出了一种采用个体进化能力实时评价和强制更新策略的进化算法。首先,建立个体进化能力评价机制,实时考察每个个体进化过程中的性能变化;其次,当个体进化能力退化时,建立强制更新策略,通过随机抽取当前结构中的换热单元,并给予其换热量以随机扰动,提高个体的结构变异能力,促进其整型变量的全局优化,改变个体原先的优化路径从而寻得更优的结构。将该算法应用于无分流换热网络实例中,取得了优于现有文献的结果,表明了该算法具有较强的全局搜索能力。     

换热网络优化中粒子群算法早熟现象的机理分析及改进策略

粒子群算法应用于换热网络综合主要存在的问题是在优化后期经常出现早熟收敛现象,由于算法全局搜索能力的迅速退化导致换热网络优化进程陷入停滞。通过考察种群多样性的变化,并采用灰度图跟踪每个粒子的差异性演化进程,揭示了算法早熟的本质,在此基础上提出了一种随机扰动策略,在粒子群搜索后期选择一部分粒子随机产生新的速度,改善这一阶段粒子群的种群多样性,增强算法的全局搜索能力,通过换热网络优化算例说明该策略的有效性。     

Tuned reactive power dispatch through modified differential evolution technique

This paper explores the capability of modified differential evolution (MDE) technique for solving the reactive power dispatch (RPD) problem. The proposed method is based on the basic differential evolution (DE) technique with a few modifications made into it. DE is one of the strongest optimization techniques though it suffers from the problem of slow convergence while global minima appear. The proposed modifications are tried to resolve the problem. The RPD problem mainly defines loss minimization with stable voltage profile. To solve the RPD problem, the generator bus voltage, transformer tap setting and shunt capacitor placements are controlled by the MDE approach. In this paper, IEEE 14-bus and IEEE 30-bus systems are chosen for MDE implementation. The applied modification show much improved result in comparison to normal DE technique. Comparative study with other soft-computing technique including DE validates the effectiveness of the proposed method.     

Multipopulation differential evolution algorithm based on the opposition-based learning for heat exchanger network synthesis

Multipopulation differential evolution combined with opposition-based learning is developed to improve the convergence efficiency and optimization accuracy for heat exchanger network synthesis. The algorithm is based on a stagewise superstructure simultaneous optimization model without considering stream splitting. The candidate population and its opposite population are searched in parallel. Mutation operations are implemented on both populations to provide a full information exchange among populations at each generation. A regrouping schedule is introduced to avoid premature convergence. The algorithm is applied to five heat exchanger network cases of different sizes. More economic networks are found using this method with less computational time.     

Solving the economic dispatch problem with a modified quantum-behaved particle swarm optimization method

In this paper, a modified quantum-behaved particle swarm optimization (QPSO) method is proposed to solve the economic dispatch (ED) problem in power systems, whose objective is to simultaneously minimize the generation cost rate while satisfying various equality and inequality constraints. The proposed method, denoted as QPSO-DM, combines the QPSO algorithm with differential mutation operation to enhance the global search ability of the algorithm. Many nonlinear characteristics of the generator, such as ramp rate limits, prohibited operating zones, and nonsmooth cost functions are considered when the proposed method is used in practical generator operation. The feasibility of the QPSO–DM method is demonstrated by three different power systems. It is compared with the QPSO, the differential evolution (DE), the particle swarm optimization (PSO), and the genetic algorithm (GA) in terms of the solution quality, robustness and convergence property. The simulation results show that the proposed QPSO–DM method is able to obtain higher quality solutions stably and efficiently in the ED problem than any other tested optimization algorithm.     

基于改进灰狼算法的独立微电网容量优化配置

为降低独立微电网的综合发电成本,提高供电可靠性,研究基于改进灰狼优化算法的独立微电网电源容量优化配置方法.针对基本灰狼算法在进化后期由于种群多样性的缺失而易出现局部收敛或算法早熟的问题,提出一种具有全局寻优性能的改进灰狼优化算法.改进算法首先利用Tent混沌序列产生初始种群,以增强种群的多样性;其次,通过对收敛因子设置...     

基于改进AWNN的风电功率超短期多步预测

为提高风电功率超短期多步预测精度,针对梯度修正学习算法采用随机初始化网络参数训练自适应小波神经网络(AWNN)易陷入局部最优的缺点,将粒子群(PSO)算法和差分进化(DE)算法相结合,提出利用IPSO-DE算法优化AWNN的初始化网络参数,得到改进AWNN模型(IAWNN)并将其用于风电功率超短期多步预测。仿真结果表明:IPSO-DE算法优化AWNN初始化网络参数的性能优于IPSO算法、DE算法和梯度修正学习算法,所提改进模型的多步预测性能优于AWNN模型、持续法(PM)模型和BP神经网络(BPNN)模型。     

基于改进混合差分进化算法的水轮机调速系统空载PID参数优化

为优化水轮发电机组调速系统空载PID相关参数,提出一种改进混合差分进化算法(IHDE),即先根据某电站混流式水轮发电机组建立相应空载工况数学模型,其次引入IHDE对空载工况数学模型PID参数进行优化,由于DE算法易出现早熟收敛问题,对该算法缩放因子的取值进行优化;同时,为了增强差分进化算法(DE)的全局搜索性能,在DE的选择阶段加入模拟退火算法的个体更新机制进行选择操作;然后,提出一种改进的综合ITAE指标用于评价空载扰动的仿真结果,使得目标函数的评价方式更加准确且符合相关标准要求;最后,通过试验仿真与不同算法进行对比,验证了IHDE优化水轮发电机组调速系统PID参数的有效性。     

Parameter identification of an SOFC model with an efficient,adaptive differential evolution algorithm

An efficient, adaptive differential evolution (DE) algorithm is proposed in which DE parameter adaptation is implemented. A ranking-based vector selection and crossover rate repairing technique are also presented. The method is referred to as IJADE (Improved Jingqiao Adaptive DE). To verify the performance of IJADE, the parameters of a simple SOFC electrochemical model that is used to control the output performance of an SOFC stack are identified and optimized. The SOFC electrochemical model is built to provide the simulated data. The results indicate that the proposed method is able to efficiently identify and optimize model parameters while showing good agreement with both simulated and experimental data. Additionally, when compared to other DE variants and other evolutionary algorithms, IJADE obtained better results in terms of the quality of the final solutions, robustness, and convergence speed.     

Application of the improved chaotic grey wolf optimization algorithm as a novel and efficient method for parameter estimation of solid oxide fuel cells model

In this paper, important functional parameters of solid oxide fuel cells are identified by introducing a novel high-speed optimization method, namely adaptive chaotic grey wolf optimization algorithm. The suggested optimization method is obtained by combining the adaptive grey wolf optimization and chaotic grey wolf optimization algorithms. The chaotic algorithm is applied to the basic grey wolf optimization to achieve higher convergence speed, keep the population's diversity, and provide an initial population with uniform distribution. Besides, a nonlinear convergence factor is defined for balancing the global and local exploration abilities. Employing the improved convergence factor resulted in a new version of the grey wolf optimization algorithm, namely adaptive grey wolf optimization algorithm. Adaptive chaotic grey wolf optimization algorithm adopts the advantages of both chaotic grey wolf optimization and adaptive grey wolf optimization methods simultaneously. The adaptive grey wolf optimization algorithm is applied to a 5 kW dynamic tubular stack. The results of the simulation report the lowest values of mean squared error, higher accuracy, higher robustness, and high convergence speed for the adaptive grey wolf optimization algorithm compared to some well-known optimization methods. Besides, the proposed method shows a good agreement with experimental results with lower computational difficulty.     

The synthesis of cost optimal heat exchanger networks with unconstrained topology

The optimization of heat exchanger network (HEN) synthesis still remains an open problem because of the complexity of the space comprising all the possible solutions, and most of the proposed methods introduce simplifying assumptions that mainly affect the topological features of the candidate solutions considered and thus artificially limit the boundaries of the search space.This work is devoted to the pursuit of cost-optimal HENs with unconstrained topology, exploiting the advantages deriving from two graph representations of a HEN. One representation is used by an evolutionary algorithm to manage HEN topology and the other is used by a NLP algorithm to manage heat load distribution among the exchangers. The proposed two-level hybrid optimization method is applied to four test cases taken from the literature about HEN synthesis, among which the well-known Aromatics Plant problem.     

基于启发式免疫遗传算法的配电网广义电源规划

在分布式电源接入配电网的规划中将广义电源的有功功率作为控制变量,考虑了广义电源无功功率对配电网分布式电源优化配置的影响,从而合理规划广义电源接入,降低系统的网络损耗,进一步提高系统运行的电压水平。针对传统优化算法在局部搜索能力和收敛性能等方面的缺陷,根据累加优化原理对系统种群进行初始优化以提高收敛速度,在疫苗接种时利用矢量矩浓度的概念进行抗体选择,依据抗体浓度和抗体适应性原则进行个体优选,提出了启发式免疫遗传算法。对IEEE-33节点系统的计算分析表明,该方法能够对广义电源在配电网中的选址和定容进行有效配置和优化,在寻优能力和收敛速度上优于传统算法。