共查询到20条相似文献,搜索用时 157 毫秒
1.
为了提高汽轮机转子故障诊断的识别准确率和效率,提出了一种基于云粒子群算法(cloud particle swarm optimization,简称CPSO)优化支持向量机(support vector machine,简称SVM)的故障诊断方法。首先,将云理论与粒子群算法(PSO)相结合得到CPSO算法;其次,通过CPSO算法优化的SVM得到诊断模型;最后,通过ZT-3转子试验台进行汽轮机转子常见故障模拟实验,获取故障数据后进行故障识别研究。结果表明:与PSO-SVM模型相比,CPSO-SVM的诊断模型可以准确、高效地识别出故障类型,证明了该诊断方法的有效性和可行性。 相似文献
2.
为提高汽轮机转子故障诊断的准确率和识别效率,提出基于云粒子群优化算法(CPSO)优化支持向量机(SVM)的故障诊断方法。首先利用补充总体平均经验模态分解(CEEMD)对转子振动信号进行分解,利用能量法筛选出更为有效的固有模态分量(IMF)并计算对应的排列熵(PE)作为故障特征值;其次将云理论引入到粒子群优化算法(PSO)中得到CPSO算法,通过CPSO算法优化SVM得到诊断模型。在ZT-3试验台对汽轮机转子常见4种故障(正常状态、转子不平衡、转子不对中和动静碰磨状态)状态进行模拟实验,获取故障数据后进行故障识别研究。研究表明:在相同测试样本的条件下,CPSO-SVM诊断模型的识别准确率为95%,比PSO-SVM诊断模型提高了5%,运行时间为22.055 s,比PSO缩短了14.5 s。研究结果验证了CPSO-SVM算法在汽轮机转子故障诊断方面的优越性。 相似文献
3.
为了准确建立汽轮机热耗率预测模型,提出了一种基于变空间Logistic混沌粒子群算法(CPSO)优化最小二乘支持向量机(LSSVM)的汽轮机热耗率软测量模型。采用变空间Logistic混沌搜索策略和粒子镜像越界处理策略来改善粒子群算法(PSO)的全局优化性能,提出了CPSO优化最小二乘支持向量机的超参数以改善模型预测精度,并以某600 MW汽轮机组为研究对象,利用该机组的运行数据建立CPSO-LSSVM的热耗率预测模型。结果表明:CPSO-LSSVM模型具有更高的预测精度和更强的泛化能力,能够准确有效地预测热电厂的汽轮机热耗率。 相似文献
4.
5.
介绍了一种易于实现、参数少且收敛快的集群智能算法——粒子群算法。针对标准PSO算法的缺陷,提出了在位置进化方程中引进动态参数的方法,改进了标准粒子群算法的收敛速度。根据建立的水库优化调度数学模型,将改进的粒子群优化算法运用到水库优化调度计算中,并通过算例验证该算法的可行性和有效性。 相似文献
6.
为解决微电网中随机性、波动性和分布式电源出力管理困难问题,提出一种并网模式下微电网优化调度的综合经济模型。首先,考虑蓄电池和峰谷电价的影响,计及失负荷和蓄电池超容惩罚成本,建立以微电网运行成本和污染物排放成本最低的目标函数;其次,引入白鲨优化算法(WSO)的全局动态捕猎特性,对粒子群优化算法(PSO)进行了Tent混沌映射、动态时间因子和多项式变异的多策略改进,并将其应用于求解多目标多约束非线性的微电网优化问题。仿真结果表明,与PSO算法和麻雀搜索算法(SSA)相比,改进PSO算法用于微电网优化调度可降低微电网运行成本、减少环境污染并提高新能源供电的可靠性。 相似文献
7.
8.
以聚类系数大且平均路径短的NW(Newman-Watts)型小世界网络作为粒子群的拓扑结构,提出NW型小世界邻域粒子群优化算法(SW-PSO)。典型高维基准测试函数表明,该算法能有效避免基本PSO算法因群体间信息交互过快而陷入局部最优解的问题。将新算法应用于含风电场电力系统经济调度问题的优化求解,并根据优化调度的实际需求采用相应的调整策略以修正粒子,保证粒子在可行域中飞行寻优,使新算法的寻优精度显著提高。仿真实例表明,SW-PSO算法比传统PSO算法寻优效果好,是一种求解复杂大规模非线性规划问题的新方法。 相似文献
9.
10.
鉴于梯级水电站优化运行的高复杂度、强非线性、多约束等特点,构建了基于峰谷分时电价下的梯级水电站日最大发电效益模型。针对遗传算法(GA)等传统智能算法对复杂模型求解易陷入局部最优的问题,提出一种水循环算法(WCA)与水位廊道约束耦合、降低约束复杂度、规范寻优空间的方法,并以湖北某梯级短期优化调度为背景进行建模仿真,将计算结果分别与GA和粒子群算法(PSO)所得结果进行比较。实例研究表明,WCA计算的总效益在丰、平、枯典型日分别比GA和PSO计算值约高5.65%、3.15%、0.80%,迭代收敛速度更快,求解能力更强,为解决梯级水电站优化调度问题提供了新思路。 相似文献
11.
12.
13.
求解水火电力系统短期发电计划的粒子群优化算法研究 总被引:3,自引:1,他引:3
提出求解水火电力系统短期发电计划的新方法——粒子群优化算法。实例研究表明,该方法优化能力高于遗传算法和进化规划。同时对全局搜索粒子群优化算法和局部搜索粒子群优化算法在求解水火电力系统短期发电计划时的动态收敛特性和计算效率等进行了对比研究,结果显示.局部搜索粒子群优化算法更适合求解水火电力系统短期发电计划的复杂问题。 相似文献
14.
15.
In this paper, a new approach is proposed to solve the economic load dispatch (ELD) problem. Power generation, spinning reserve and emission costs are simultaneously considered in the objective function of the proposed ELD problem. In this condition, if the valve-point effects of thermal units are considered in the proposed emission, reserve and economic load dispatch (ERELD) problem, a non-smooth and non-convex cost function will be obtained. Frequency deviation, minimum frequency limits and other practical constraints are also considered in this problem. For this purpose, ramp rate limit, transmission line losses, maximum emission limit for specific power plants or total power system, prohibited operating zones and frequency constraints are considered in the optimization problem. A hybrid method that combines the bacterial foraging (BF) algorithm with the Nelder-Mead (NM) method (called BF-NM algorithm) is used to solve the problem. In this study, the performance of the proposed BF-NM algorithm is compared with the performance of other classic (non-linear programming) and intelligent algorithms such as particle swarm optimization (PSO) as well as genetic algorithm (GA), differential evolution (DE) and BF algorithms. The simulation results show the advantages of the proposed method for reducing the total cost of the system. 相似文献
16.
针对梯级电站优化调度具有多阶段、非线性和组合性的特点,采用改进粒子群算法来求解。针对目前采用的基本粒子群算法在求解时存在易陷入局部最优和早熟的缺点,改进粒子群算法以混沌变量生成机制来增加种群的多样性,以逐步优化和随机生成相结合的方法生成初始种群,以增加粒子生成的有效性。实例计算结果表明,改进粒子群算法可以取得较好的效果,并为梯级电站优化调度提供了一种有效的方法。 相似文献
17.
18.
引入微粒群算法建立了以各工序实际开始时间作为微粒坐标的微粒群计算模型,并对相关参数及工程多资源均衡问题进行了分析,结合算例验证了该方法在工程多资源均衡优化中的可行性和有效性. 相似文献
19.
Alireza Askarzadeh Mahdi Gharibi 《Energy Sources, Part A: Recovery, Utilization, and Environmental Effects》2018,40(24):2986-2999
Accurate parameter estimation of the input-output characteristics in thermal power plants is an important issue in power system because these characteristics directly affect the economic dispatch calculations. Parameter estimation is an optimization problem in which the optimal values of the unknown parameters should be estimated by an optimization technique. By considering the valve-point effect, the parameter estimation will be more difficult since the fitness function of this optimization problem turns into a non-smooth and non-convex function in which finding the global optimal is a challenging task. In this paper, a recently proposed metaheuristic approach, crow search algorithm (CSA), is proposed for accurate estimation of the input-output characteristics of thermal power plants with and without valve-point effect. Simulation results show that CSA finds more promising results than least squares method (LSM), particle swarm optimization (PSO), genetic algorithm (GA) and artificial bee colony (ABC). 相似文献
20.
Based on particle swarm optimization (PSO), an optimization platform for offshore wind farm electrical system (OWFES) is proposed in this paper, where the main components of an offshore wind farm and key technical constraints are considered as input parameters. The offshore wind farm electrical system is optimized in accordance with initial investment by considering three aspects: the number and siting of offshore substations (OS), the cable connection layout of both collection system (CS) and transmission system (TS) as well as the selection of electrical components in terms of voltage level and capacity. Because hundreds of optimization variables, continuous or discrete, are involved in the problem, a mix integer PSO (MIPSO) is required to obtain the solution. The fuzzy C‐means clustering (FCM) algorithm is used to partition the wind farm into several sub regions. The collection system layout in each sub region as well as the connection scheme between offshore substations are optimized by an adaptive PSO‐minimum spanning tree algorithm (APSO‐MST) which has been proposed in a previous work. The simulation results show that the proposed optimization platform can find an optimized layout that save 3.01% total cost compared with the industrial layout, and can be a useful tool for OWFES design and evaluation. Copyright © 2016 John Wiley & Sons, Ltd. 相似文献