基于改进果蝇算法优化支持向量机的风电机组性能评估

目前,关于风电机组性能的研究多集中于某一关键部件,文章针对风电机组整体提出一种性能评估的方法。首先采用支持向量回归(SVR)预测正常状态下的评价指标,并使用果蝇优化算法(FOA)来寻找其最优参数。为了克服FOA易陷入局部最优的缺点,引入免疫思想增加种群的多样性,并采用自适应搜索步长,提高其收敛精度;然后通过变权思想对预测误差进行组合,得到风电机组偏离正常状态的劣化度,实现对风电机组的性能评估;最后,采用某风电场的数据进行实验分析,验证了文章所提方法的可行性。     

基于小波深度置信网络的风电爬坡预测方法

为了提高风电爬坡事件预测的准确性,提出一种基于深度学习的具有特征自适应选择的小波深度置信网络(WDBNAFS)算法。首先,分析风电功率混沌特性。然后,对时间序列数据进行小波分解,设计特征自适应选择算法选取建模数据作为预测模型的输入变量。最后,采用深度置信网络构建风电爬坡事件预测模型,设计基于实际生产数据的实验验证所提出算法的有效性。仿真结果表明,所提出算法预测准确率可达90%以上。     

基于ADQPSO-KELM风电功率短期预测模型的研究

为了改善传统风电功率预测方法中误差较大且稳定性较差的问题,引入量子粒子群(QPSO)优化算法、自适应早熟判定准则及混合扰动算子,构建了自适应扰动量子粒子群(ADQPSO)优化算法,通过ADQPSO算法对核极限学习机(KELM)模型进行优化,建立了自适应扰动量子粒子群优化的核极限学习机(ADQPSOKELM)风电功率短期预测模型,并利用内蒙古高尔真风电场采集的风电功率时间序列数据为试验样本进行48h预测分析。结果表明,ADQPSO-KELM风电功率短期预测模型与其他基于KELM优化的风电预测模型及传统风电预测模型相比,其预测的误差更小、准确度更高,且预测稳定性显著增强。     

基于OLS-SFLA-RBF神经网络的风电出力预测

提高风电出力的预测精度,可以减轻风电并网带来的不利影响。利用径向基函数神经网络（RBF）建立风电出力预测模型,并通过正交二乘算法（OLS）对RBF神经网络进行初步训练,以确定网络结构及隐含层各节点中心。在OLS算法训练的网络基础上引入蛙跳算法（SFLA）,进一步对隐含层基函数的宽度值进行优化以提高网络的泛化能力。实例预测表明,在相同的网络结构及隐含层中心下,基函数宽度值优化后的RBF神经网络模型预测精度得到了提升。     

基于果蝇算法与自适应性遗传算法组合优化神经网络的光伏电站短期出力预测

光伏电站的输出功率会随着很多因素发生波动,若能够提高光伏系统出力预测的准确性,则能有效地降低光伏电站并网后对电网造成的冲击,提高电力系统的稳定性。建立了果蝇算法与自适应遗传算法组合优化的BP神经网络的预测模型。从预测结果可以发现,采用组合优化算法的BP神经网络模型能够有效避免地BP神经网络易陷入局部极小值点的缺陷,相比于仅优化权值和阈值的BP神经网络模型提高了预测精度,具有一定的应用价值。     

基于AFOA-BP神经网络的燃气轮机研制风险预测

为提高燃气轮机研制过程中的风险管理能力,针对果蝇算法（Fruit Fly Optimization Algorithm,FOA）及BP神经网络的缺陷,构建了自适应果蝇算法（Adaptive Fruit Fly Optimization Algorithm,AFOA）,提出基于自适应果蝇算法优化BP神经网络的风险预测模型,利用自适应果蝇算法优化BP神经网络的阈值和权值。挖掘燃气轮机研制风险因素及风险事件之间的关系,并根据风险因素的权重预测风险事件的权重。利用燃气轮机研制风险的相关历史数据进行验证,表明该模型具有较高的预测精度和应用价值。     

基于RBF-BP组合神经网络的短期风电功率预测研究

为提高风电输出功率预测精度,提出一种基于RBF-BP组合神经网络模型的短期风电功率预测方法。在考虑尾流等因素影响的基础上,对风速进行预处理。根据相关历史数据,建立RBF-BP组合神经网络短期风电功率预测模型,对风电输出功率进行预测。仿真分析结果表明,该预测方法能有效提高风电输出功率预测精度。     

数据分辨率影响风电功率预测精度分析

目前对风电功率短时预测的研究主要集中在预测方法上,而缺乏对数据本身特性的探讨。从实测数据出发,呈现3种典型分辨率5 min、10 min、15 min,并结合Elman神经网络算法对超短期（4 h）和短期（24 h）的风力发电机输出功率进行预测分析。结果表明：分辨率为10 min的原始数据对风电输出功率的超短期预测具有更好的结果,15 min分辨率的数据对风电功率的短期预测结果更佳。采用合理分辨率的数据后,能够有效地提高风电功率的预测精度。     

风速时间序列非线性短期预测

针对风速时间序列复杂的非线性特征,根据C-C算法确定重构参数(嵌入维数及延迟时间)并对风速重构相空间,建立径向基函数神经网络(RBF网络)及Volterra自适应预测模型对风速时间序列进行预测,以Lorenz方程数值解为例验证了两种预测方法的可行性。结果表明:RBF神经网络模型和Volterra自适应预测模型都能对实测风速时间序列进行较为准确的预测,预测误差分别在0.3和0.1 m/s内;Volterra自适应预测模型预测结果总体较RBF神经网络模型预测精度更高,且随着预测时间的增大,预测误差呈增大趋势,这与混沌存在初值敏感性的特征相符。     

基于混沌RBF神经网络的汽油机瞬态工况油膜参数辨识研究

提出了基于混沌径向基(RBF)神经网络的汽油机瞬态工况油膜参数辨识方法。利用混沌优化算法确定隐含层高斯函数径向基中心和输出层连接权值,使其达到全局最优,有效地提高RBF神经网络的收敛速度;同时,利用混沌算法训练RBF神经网络,使目标函数取全局最小值或逼近全局最小值,有效地提高辨识模型的辨识精度,并与BP神经网络模型及最小二乘法辨识进行了分析和比较。仿真结果表明:混沌RBF神经网络模型收敛速度快,具有更强的非线性辨识能力,能够有效地提高油膜动态参数的辨识精度,进而得出不同工况下的油膜参数动态特征。     

基于改进粒子群算法的多目标无功优化

提出了一种基于粒子群算法的多目标优化方法,该算法采用Pareto支配关系来更新粒子的个体最优和全局最优值,用存储池保存搜索过程中发现的非支配解;采用聚类算法裁剪非支配解,以保持解的分散性;采用动态惯性权重来平衡粒子的局部和全局搜索能力,并将该算法应用于IEEE14节点系统的多目标无功优化。     

Power system optimization

Long-term gas purchase contracts usually determine delivery and payment for gas on the regular hourly basis, independently of demand side consumption. In order to use fuel gas in an economically viable way, optimization of gas distribution for covering consumption must be introduced. In this paper, a mathematical model of the electric utility system which is used for optimization of gas distribution over electric generators is presented. The utility system comprises installed capacity of 1500 MW of thermal power plants, 400 MW of combined heat and power plants, 330 MW of a nuclear power plant and 1600 MW of hydro power plants. Based on known demand curve the optimization model selects plants according to the prescribed criteria. Firstly it engages run-of-river hydro plants, then the public cogeneration plants, the nuclear plant and thermal power plants. Storage hydro plants are used for covering peak load consumption. In case of shortage of installed capacity, the cross-border purchase is allowed. Usage of dual fuel equipment (gas–oil), which is available in some thermal plants, is also controlled by the optimization procedure. It is shown that by using such a model it is possible to properly plan the amount of fuel gas which will be contracted. The contracted amount can easily be distributed over generators efficiently and without losses (no breaks in delivery). The model helps in optimizing of fuel gas–oil ratio for plants with combined burners and enables planning of power plants overhauls over a year in a viable and efficient way.     

Design optimization of shell-and-tube heat exchanger using particle swarm optimization technique

Shell-and-tube heat exchangers (STHEs) are the most common type of heat exchangers that find widespread use in numerous industrial applications. Cost minimization of these heat exchangers is a key objective for both designer and users. Heat exchanger design involves complex processes, including selection of geometrical parameters and operating parameters. The traditional design approach for shell-and-tube heat exchangers involves rating a large number of different exchanger geometries to identify those that satisfy a given heat duty and a set of geometric and operational constraints. However, this approach is time-consuming and does not assure an optimal solution. Hence the present study explores the use of a non-traditional optimization technique; called particle swarm optimization (PSO), for design optimization of shell-and-tube heat exchangers from economic view point. Minimization of total annual cost is considered as an objective function. Three design variables such as shell internal diameter, outer tube diameter and baffle spacing are considered for optimization. Two tube layouts viz. triangle and square are also considered for optimization. Four different case studies are presented to demonstrate the effectiveness and accuracy of the proposed algorithm. The results of optimization using PSO technique are compared with those obtained by using genetic algorithm (GA).     

Thermal placement optimization of multichip modules using a sequential metamodeling-based optimization approach

The study attempts to seek the optimal thermal design of planar multichip module (MCMs) under natural convection through optimal chip placement design. To attain the goal, a sequential metamodeling-based optimization approach is introduced. This approach incorporates a response surface methodology (RSM)-based design of experiment (DOE), three-dimensional (3D) thermal finite element analysis (FEA) and an updating scheme. Essentially, the RSM is used to construct, via quadratic polynomial approximation, the global RS of the chip junction temperature in terms of design variables. For speeding up the DOE and the solution of the optimization, several dynamic experimental design strategies using move limits and different proposed sampling techniques are introduced. The feasibility of the strategies is demonstrated, and their solution accuracy and efficiency are also compared with each other. By the explicit RS-based performance function together with geometry constraints, a constrained thermal optimization subproblem is formed. The optimum of the subspace optimization is sought, which is considered as the nominal starting point of next iteration. The iterative process continues with a new defined design subspace and factorial design plan until convergence is attained. The applicability of the proposed design optimization technique is demonstrated through several design case studies involving various planar MCMs.     

Vortex tube optimization theory

The Ranque–Hilsch vortex tube splits a single high pressure stream of gas into cold and warm streams. Simple models for the vortex tube combined with regenerative precooling are given from which an optimization can be undertaken. Two such optimizations are needed: the first shows that at any given cut or fraction of the cold stream, the best refrigerative load, allowing for the temperature lift, is nearly half the maximum loading that would result in no lift. The second optimization shows that the optimum cut is an equal division of the vortex streams between hot and cold. Bounds are obtainable within this theory for the performance of the system for a given gas and pressure ratio.     

Distributionally robust optimization of home energy management system based on receding horizon optimization

This paper investigates the scheduling strategy of schedulable load in home energy management system (HEMS) under uncertain environment by proposing a distributionally robust optimization (DRO) method based on receding horizon optimization (RHO-DRO). First, the optimization model of HEMS, which contains uncertain variable outdoor temperature and hot water demand, is established and the scheduling problem is developed into a mixed integer linear programming (MILP) by using the DRO method based on the ambiguity sets of the probability distribution of uncertain variables. Combined with RHO, the MILP is solved in a rolling fashion using the latest update data related to uncertain variables. The simulation results demonstrate that the scheduling results are robust under uncertain environment while satisfying all operating constraints with little violation of user thermal comfort. Furthermore, compared with the robust optimization (RO) method, the RHO-DRO method proposed in this paper has a lower conservation and can save more electricity for users.     

基于粒子群算法的储能电站经济优化调度策略

储能电站的功率分配方案直接影响其调度成本,合理的功率分配方案是保障电站运行经济性的基础。为了合理有效地进行储能电站功率分配,文章以单位周期内调度成本最低为优化目标,搭建了考虑电池容量损失的储能电站调度成本模型,并利用粒子群优化算法(PSO)寻求储能电站调度任务的最优分配方案。在保证完成储能电站调度任务的同时,最大限度地降低调度成本。案例仿真结果表明,文章所提出的PSO优化分配方案比传统等比例功率分配方案具有明显的优越性。当储能电站在传统调度方法下寿命终止(1 597次)时,PSO优化调度方法同比节约调度成本约19.7%,且在此基础上可继续工作370次,使储能电站的运行寿命延长23.2%。     

基于改进灰狼优化算法的孤岛微网系统容量优化分析

孤岛微网的日前调度安排对岛屿配电网的安全稳定运行意义重大。为了对孤岛微网的日前调度进行安排,更好地优化微网结构,文章针对含有光伏和抽水蓄能电站的孤岛微网进行建模分析,将光伏阵列数量、额定功率和抽水蓄能电站上水库的容积作为决策变量,等年值平均投资成本和负荷失电率为目标函数,使用Matlab对某岛屿的天气和负荷数据进行分析。为了使光电-抽水蓄能协同运行的效益最大化,采用灰狼优化算法求解微网模型。优化结果表明,灰狼优化算法在孤岛微网模型的容量优化问题中可以节约投资成本,实现环境友好型供电。     

Thermodynamic optimization of cross flow plate-fin heat exchanger using a particle swarm optimization algorithm

This study explores the use of particle swarm optimization (PSO) algorithm for thermodynamic optimization of a cross flow plate-fin heat exchanger. Minimization of total number of entropy generation units for specific heat duty requirement under given space restrictions, minimization of total volume, and minimization of total annual cost are considered as objective functions and are treated individually. Based on the applications, heat exchanger length, fin frequency, numbers of fin layers, lance length of fin, fin height and fin thickness or different flow length of the heat exchanger are considered for optimization. Heat duty requirement constraint is included in the procedure. Two application examples are also presented to demonstrate the effectiveness and accuracy of the proposed algorithm. The results of optimization using PSO are validated by comparing with those obtained by using genetic algorithm (GA). Parametric analysis is also carried out to demonstrate the effect of heat exchanger dimensions on the optimum solution. The effect of variation of PSO parameters on convergence and optimum value of the objective has also been presented.