换热网络中大步长与接受差解的关联分析及改进策略

RWCE算法优化换热网络时,较大的优化步长能够在接受差解的作用下促进结构变异,从而有效地跳出局部最优解.但持续的大步长无法保留结构变异后新生成的、换热量较小的换热单元,影响结构进化效率.基于大步长与接受差解的关联性,提出了一种大步长与接受差解联动策略,将接受差解和采用大步长同频,仅当接受差解时才使用大步长,并必定接受大...     

RWCE算法控制参数动态更新促进换热网络结构进化策略

相较其他进化算法,强制进化随机游走（RWCE）算法能够始终保持较高的种群多样性,从而有效地跳出局部最优。然而,目前对于该算法进化过程中的控制参数如最大步长、最小换热量或换热面积以及接受差解概率的设置仍无严格定义,其取值方法和取值范围都将对结构进化的进度、换热单元生成和消去速度以及最终的换热单元数产生直接影响。根据换热单元数设定逐渐变化的控制参数,进行逐级优化尝试。引入logistic函数作为接受差解概率的取值策略,使最大步长和保留系数均随换热单元数线性变化,实现控制参数的动态更新从而促进换热网络结构进化。通过算例验证,该策略能提高RWCE算法优化换热网络的效率,可获得更理想的网络结构。     

联合模糊聚类和多样本SSO优化SVR的配电通信网络流

针对电力通信网络流量预测问题,提出了一种联合模糊聚类和多样本群居蜘蛛优化（social spider optimization,SSO）SVR的配电网络流量预测方案。针对配电网络流量时序非线性、周期性特点,首先采用自适应模糊聚类方法（adaptive fuzzy C-means clustering）对流量样本数据进行预处理,实现了样本数据聚类自动划分,有效降低了流量数据短相关性对预测精度的影响;然后利用多样本SSO优化算法（multi-sample social spider optimization algorithm）对SVR预测模型参数进行优化,通过引入多样本、网格迭代进化策略,从而得到不同流量数据聚类对应的最佳SVR参数组合;最后,运用多样本SSO优化SVR模型对预测数据进行预测分析。仿真结果表明,同ARIMA、神经网络等配电网路流量预测模型相比,提出的预测方案预测精度提高了18.8%~34.1%。     

进化策略在搜索边坡任意形状最危险滑裂面中的应用

将进化策略与边坡稳定极限平衡法相结合，建立了边坡任意形状最危险滑裂面及相应最小安全系数的全局优化搜索方法。该方法适用于任意几何形状、不同岩土材料分层以及多种荷载作用的复杂边坡。通过典型算例分析，证明应用进化策略搜索边坡任意形状最危险滑裂面是可行和高效的，并且具有较强的鲁棒性。     

模拟退火算法在三维气动优化设计中的应用

首先对所开发的模拟退火算法进行函数验证,然后与实验设计、三维流场求解程序、二次多项式响应面近似模型结合,提出了一套省时的气动优化设计体系。应用该优化方法对跨音压气机动叶以绝热效率最大为目标进行气动优化设计,优化后叶片绝热效率提高约1.58%,并且具有良好的变工况性能。结果表明了所开发优化设计体系省时、高效的特点。     

混合微分进化算法应用于换热网络优化

针对微分进化(Differential Evolution,DE)算法应用于换热网络优化存在局部搜索能力不足、收敛速度慢和求解精度低等问题,提出一种混合微分进化(Hybrid Differential Evolution,HDE)算法。当DE算法的变异、交叉和选择操作不再使种群的最优值继续进化时,加入梯度操作使当前种群的最优个体趋向更好的解。为了防止算法早熟收敛,当种群的多样性低于设定的阈值时,引入迁移操作,在最优个体附近区域重新生成新的个体并以此替换旧的个体,增强算法的种群多样性。通过算例验证了该算法可以有效适用于换热网络的优化过程,具有更强的局部搜索能力,更快的收敛速度和更高的优化效率。     

基于需求响应的居民可控能效负荷优化

为解决居民生活用电需求不断增加,居民生活用电方式不合理造成能源浪费越来越严重的问题,从居民用户的可控能效负荷入手,对典型可控能效负荷空调、热水器和照明负荷进行分析,建立负荷能耗数学模型,根据其运行特性,结合居民用户用电习惯和分时电价,制定居民可控能效负荷优化策略;建立以居民用户用电成本和用电满意度为目标的优化模型。为提高和声搜索算法的求解速度与计算精度,对其参数进行动态调整,并与差分进化算法进行融合,应用于可控能效负荷的优化求解。算例结果表明了改进算法具有较好的收敛性和较高的准确性,验证了居民可控能效负荷优化策略的可行性,实现了从需求响应的角度对可控能效负荷进行优化管理的思想。     

Multi-objective optimization for GPU3 Stirling engine by combining multi-objective algorithms

Stirling engine has become preferable for high attention towards the use of alternate renewable energy resources like biomass and solar energy. Stirling engine is the main component of dish Stirling system in thermal power generation sector. Stirling engine is an externally heating engine, which theoretical efficiency is as high as Carnot cycle's, but actual ones are always far below compared with the Carnot efficiency. A number of studies have been done on multi-objective optimization to improve the design of Stirling engine. In the current study, a multi-objective optimization method, which is a combination of multiple optimization algorithms including differential evolution, genetic algorithm and adaptive simulated annealing, was proposed. This method is an attempt to generalize and improve the robustness and diversity with above three kinds of population based meta-heuristic optimization techniques. The analogous interpreter was linked and interchanged to find the best global optimal solution for Stirling engine performance optimization. It decreases the chance of convergence at a local minimum by powering from the fact that these three algorithms run parallel and members from each population and technique are swapped. The optimization considers five decision variables, including engine frequency, mean effective pressure, temperature of heating source, number of wires in regenerator matrix, and the wire diameter of regenerator, as multiple objectives. The Pareto optimal frontier was obtained and a final optimal solution was also selected by using various multi-criteria decision making methods including techniques for Order of Preference by Similarity to Ideal Solution and Simple Additive Weighting. The multi-objective optimization indicated a way for GPU-3 Stirling engine to obtain an output power of more than 3 kW and an increase by 5% in thermal efficiency with significant decrease in power loss due to flow resistance.     

基于NCPSO的微网群优化调度策略研究

传统的粒子群优化（PSO）算法因在微网优化中不易达到全局最优而导致微网运行成本过高,该文采用小生境混沌粒子群优化（NCPSO）算法对混合微网群的运行策略进行协同优化,以实现区域微网经济性最优、环境治理成本最低、风光等可再生能源利用率高等目的。根据所提出的调度策略,建立的优化调度模型包括动态电价下的负荷模型、经济收益模型以及成本模型等,使用NCPSO算法得到多微网在一个周期内的最佳运行状态,实现微网群系统综合能源的互动调控、空间互补。通过分析微网群的功率交互动态、可控能源的发电以及储能电池的荷电状态等,验证微网群的电力负荷响应动态电价,表明了NCPSO算法优化微网群运行的优越性、有效性。     

Supervisory and optimal control of central chiller plants using simplified adaptive models and genetic algorithm

This paper presents a model-based supervisory and optimal control strategy for central chiller plants to enhance their energy efficiency and control performance. The optimal strategy is formulated using simplified models of major components and the genetic algorithm (GA). The simplified models are used as the performance predictors to estimate the system energy performance and response to the changes of control settings and working conditions. Since the accuracy of the models has significant impacts on the overall prediction results, the models used are linear in the parameters and the recursive least squares (RLS) estimation technique with exponential forgetting is used to identify and update the model parameters online. That is to ensure that the linear models can provide reliable and accurate estimates when working condition changes. The GA, as a global optimization tool, is used to solve the optimization problem and search for globally optimal control settings. The performance of this strategy is tested and evaluated in a simulated virtual system representing the actual central chiller plant in a super high-rise building under various working conditions. The results showed that this strategy can save about 0.73–2.55% daily energy of the system studied, as compared to a reference strategy using conventional settings.     

进化概率差异化策略优化有分流换热网络

强制进化随机游走算法优化换热网络时具有程序简单、全局搜索能力强等特点,但在优化过程中不同的进化阶段对进化概率需求不同,算法中设置单一的进化概率难以满足实际进化需求。因此,提出一种进化概率差异化策略,该策略的核心思想是智能识别流股的换热程度,动态调整进化概率,对存在公用工程的流股强制参与进化,使算法在前期具有较强的结构搜索能力;完全换热的流股通过降低进化概率,提升算法全局的精细搜索能力。采用15SP和39SP算例进行验证,所获年综合费用分别为1 494 690和1 894 477$/a,验证了该策略能够提升算法的优化效率与质量。     

基于改进BA算法的光伏发电系统MPPT优化控制

为优化光伏阵列在部分遮蔽情况下的多峰值MPPT控制,保证光伏发电系统实时功率的最大输出,提出了基于改进BA算法的最大功率追踪控制方法,即在基本BA算法的基础上,融入了小生境技术的共享机制与排挤策略,减少相似个体数量,从而增加了BA算法在迭代过程中的种群多样性,提高了BA算法在MPPT控制中的全局搜索能力,增强了最大功率追踪的稳定性,并将该算法与PSO、PO算法在不同光照及温度条件下的MPPT控制效果进行了仿真试验对比。结果表明,与传统算法相比,改进的BA算法具有更好的追踪效果,不仅避免了光伏系统在遮蔽情况下输出功率陷入局部最大值的问题,且提高了发电效率。     

含双馈感应风电机组的配电网无功优化

考虑传统无功调节设备调节次数限制和双馈感应电机无功容量限制等约束条件,提出一种基于双馈感应电机与传统无功调节设备协调控制的分时段分层无功优化策略。首先,该策略采用谱系聚类算法对预测等效负荷曲线进行分段;其次,在每个时段采用分层调控策略进行无功优化,建立以网损和平均电压偏离度之和为目标函数的无功优化模型,上层利用改进粒子群算法计算出包括双馈感应机组在内的各种无功调节设备的优化运行状态,并预先对变压器、电容器动作;在此基础上,下层利用双馈感应机组的无功调节能力对上层优化得出的并网点电压进行自动跟踪控制,由此实现了每个时段内接入点电压控制和全局无功优化相结合,最后以IEEE33节点配电系统为算例来验证上述策略的有效性。     

基于连续蚁群算法的供水水库优化调度

针对供水水库优化调度问题,介绍了一种改进的连续蚁群算法。该算法借鉴蚁群算法的进化思想,主要包括局部搜索、全局搜索和信息素强度更新规则。在随机搜索过程中嵌入确定性搜索,以改善寻优性能,增加经验指导,从而加速收敛。将该算法与离散蚁群算法相比较,结果表明,该算法更优越,具有并行化和较强的全局寻优能力。     

A novel hybrid model based on artificial neural networks for solar radiation prediction

As a kind of clean, substantial and renewable energy, solar energy can reduce environmental pollution with an extensive application potential. Precise prediction of global solar radiation has great significance for the design of solar energy systems and management of solar power plants.In this paper, a new hybrid model combining the SOM-OPELM with time series strategies is presented for predicting the global solar radiation on the horizon. In this model, the SOM divides the original data into distinct clusters and the OPELM establishes the prediction model. Subsequently, three population time series strategies, (i.e. Recursive strategy, DirRec strategy and MISMO strategy) are adopted to accomplish the multi-step prediction. A comparison between the proposed SOM-OPELM model and other conventional methods is carried out to demonstrate its efficiency and estimation performance. The simulation results show that the proposed SOM-OPELM model with DirRec strategy or MISMO strategy outperforms the following models: Recursive-BP, DirRec-BP, MISMO-BP, Recursive-SOM-OPELM and ARIMA.     

Research on optimal operation control based on the equivalent model of VRFB system

全钒电池的建模和运行参数是影响电池性能的重要因素,现有的等效电路模型往往由于缺乏和流体力学模型的耦合,无法研究流量对全钒液流电池性能和效率的影响,因而在实际项目应用中不够完善;另外纯粹的流体力学模型对于电池本身的电气特性过于简化,无法充分反映全钒液流电池系统的特性。因此本文利用钒电池基本原理和等效损耗建立了等效电路模型,根据机械损耗和钒电池结构参数,建立钒电池流体力学模型,根据搭建的钒电池系统等效模型分析电池运行参数对电池性能的影响规律。仿真分析表明,充放电期间的最优流量是关于荷电状态的函数,根据该现象利用流量随着荷电状态SOC的变化进行分段控制。通过仿真分析,根据SOC变化而分段控制流量的优化运行策略可以有效地提高钒电池效率。     

Numerical studies on the combustion properties of char particle clusters

Particle clustering is an important phenomenon in dense particle–gas two-phase flow. One of the key problems worth studying is the reacting properties of particle clusters in coal particle combustion process in the dense particle region. In this paper, a two-dimensional mathematical model for the char cluster combustion in airflow field is established. This char cluster consists of several individual particles. The comprehensive model includes mass, momentum, and energy conservation equations for both gas and particle phases. Detailed results regarding velocity vector, mass component, and temperature distributions inside and around the cluster are obtained. The micro-scale mass and heat transfer occurred inside and around the char cluster are revealed. By contrastively studying the stable combustion of char particle clusters consisting of different particles, the combustion properties of char clusters in various particle concentrations are presented and discussed.     

基于改进径向移动算法的含风电场电力系统优化调度

为应对风电接入对电力系统稳定运行带来的影响,考虑风电高估低估成本、阀点效应、旋转备用约束和网络损耗等常需因素,建立计及风电不确定性的通用经济调度模型。为求解此模型,提出一种改进的径向移动算法(IRMO),该算法针对基本径向移动算法易陷入局部最优解的不足,一方面结合遗传算法中种群变异的思想,在迭代过程中随机对一部分粒子进行突变,改善种群多样性,使算法能够跳出局部最优;另一方面引入凹抛物线式的惯性权值非线性递减策略,以进一步增强算法中后期的搜索精度,更易找到全局最优解。最后对含风电场的电力系统进行算例分析和算法对比,验证模型的合理性以及IRMO的优越性。     

粒子群算法惯性权重策略对参数反分析优化性能的影响

以某混凝土重力坝挡水坝段为例,针对惯性权重为粒子群算法中平衡全局搜索能力和局部搜索能力的关键参数,分析了不同的惯性权重策略影响粒子群算法在材料参数反分析领域的优化性能,并比较了四种惯性权重策略。结果表明,线性微分递减惯性权重策略最优,可使材料参数反分析过程收敛速度更快、稳定性更强。     

基于改进鸡群优化算法的质子交换膜燃料电池模型参数辨识

针对当前大部分智能算法在求解质子交换膜燃料电池模型参数辨识问题时易陷入局部最优,导致参数辨识精度低、模型泛化能力差等问题,提出一种基于改进鸡群优化算法的质子交换膜燃料电池模型参数辨识方法。首先,引入Tent映射策略初始化种群,提高种群的均匀性和遍历性;其次,设计基于个体进食速度的自适应惯性权重,改善母鸡个体寻优效率,平衡算法的开发与探索能力;然后,利用Levy飞行策略的长短跳跃特点对小鸡位置进行随机更新,增强算法的全局最优搜索能力。最后,通过4组测试函数验证了该算法的优越性,并将算法应用于H-12电堆的参数辨识问题中。结果表明：相比于鲸鱼优化算法、花卉授粉算法等算法,该算法具有更高的参数辨识精度,所辨识出的模型具有更强的泛化能力。