基于改进GWO-SVR的锂电池SOH估计

为了提高锂电池健康状态的估计精度，提出了一种基于IGWO-SVR的锂电池SOH估计方法。针对支持向量回归(SVR)内核参数选择的问题，采用改进灰狼(IGWO)算法优化支持向量回归的内核参数；选取合适的健康特征作为输入，电池SOH作为输出，建立IGWO-SVR估计模型，实现锂电池SOH的估计。基于NASA电池数据集，对该模型进行训练及验证，并与SVR和GWO-SVR方法相比。结果表明，IGWO-SVR方法能有效提高SOH估计的精度和稳定性，最大估计误差不超过2%。     

基于改进的回归支持向量机模型及其在年径流预测中的应用

针对回归支持向量机(SVR)惩罚因子C和核函数参数g的选取对模型性能有着关键性影响以及在实际应用中存在参数选取等困难,提出基于启发式算法的PSO-SVR和GA-SVR年径流预测模型,以云南省落却站年径流预测为例进行实例研究。首先,利用SPSS软件选取年径流影响因子,确定输入向量;其次,基于粒子群算法(PSO)、遗传算法(GA)基本原理,采用PSO、GA优化SVR惩罚因子C和核函数参数g,构建PSO-SVR和GA-SVR多元变量年径流预测模型,并构建基于网格划分(GS)与交叉验证(CV)算法相结合的GS-SVR模型作为对比模型。最后,利用所构建的模型对实例进行预测分析。结果表明:PSO-SVR和GA-SVR模型对实例后16年年径流预测(随机5次平均)的平均相对误差绝对值分别为2.53%、2.79%,最大相对误差绝对值分别为7.11%、6.64%,平均绝对误差分别为0.1394、0.1527,预测精度和泛化能力均优于GS-SVR模型,表明PSO和GA能有效对SVR惩罚因子C和核函数参数g进行优化。PSO-SVR和GA-SVR模型具有预测精度高、泛化能力强以及稳健性能好等特点。相对而言,PSO-SVR模型性能略优于GA-SVR模型。     

T-S模糊模型的辨识与控制

提出了基于支持向量机和遗传算法的T-S模糊模型辨识,支持向量机具有很好的泛化能力,能自动确定T-S模型结构,通过遗传算法优化和估计系统参数。针对辨识出的T-S模型进行控制,控制器包括两个部分,权重最大子系统局部反馈控制和利用滑模控制设计的全局监督控制,能保证系统稳定。辨识和控制仿真结果证明了算法的有效性。     

基于遗传算法与支持向量回归的发电机运行参数趋势预测

针对支持向量回归(SVR)模型在设备运行参数趋势预测中。根据人为经验选取模型参数导致预测精度不高的问题,提出了一种使用遗传算法(GA)优化SVR模型参数的方法(GA-SVR)。将该方法应用于发电机定子线圈出水温度的实时趋势预测中。结果表明,相较于SVR模型,GA-SVR具有更高的预测精度,能够满足电厂对发电机运行参数变化的趋势预测精度要求。     

基于混合算法的短期负荷预测模糊建模

结合最小二乘（LS）辨识以及一种基于进化规划（EP）和粒子群优化（PSO）的混合进化算法EPPSO,针对对温度比较敏感的夏季负荷,提出一种3阶段短期负荷预测（STLF）算法。在第1阶段,应用LS设计模糊基函数网络（FBFN）完成STLF模糊空间划分;第2阶段,首先拓展FBFN成一阶Sugeno模糊模型,然后应用EPPSO调节其前件参数同时训练后件参数,最后将前述模型用于STLF得出的预测误差看做一个新的时间序列,并仅用气象因素对其进行辨识,可以用回归模型表示该辨识模型,进而应用LS进行辨识。文中提出的STLF模糊建模策略主要贡献于受气象因素影响较大的夏季负荷。仿真部分对浙江省电力公司的实际负荷进行了预测,与其他方法的比较结果证明该方法具有良好的预测性能。     

基于Prophet与XGBoost混合模型的短期负荷预测

准确有效的预测电力负荷对电网的安全稳定运行具有重要的参考价值.通过对Prophet框架和XGBboost(eXtreme gradient boosting)机器学习模型的深度分析,提出了基于Prophet与XGBoost的混合电力负荷预测模型,针对大量的历史电负荷数据、日期信息、气象数据,分别构建Prophet电力负...     

Recent advances in space‐mapping‐based modeling of microwave devices

We review the latest developments in space‐mapping‐based modeling techniques with applications in microwave engineering. We discuss the two techniques that utilize a combination of standard space mapping and function approximation methodologies, in particular fuzzy systems and support vector regression (SVR). In both cases, the initial space‐mapping model is enhanced by an additional term that approximates the differences between the fine model and the initial space‐mapping surrogate. We compare the standard and enhanced space‐mapping models, as well as the fuzzy systems and SVR directly used for modeling fine model data. A discussion of the advantages and disadvantages of the presented methods is also given. Copyright © 2010 John Wiley & Sons, Ltd.     

Cyclic electric load forecasting by seasonal SVR with chaotic genetic algorithm

Application of support vector regression (SVR) with chaotic sequence and evolutionary algorithms not only could improve forecasting accuracy performance, but also could effectively avoid converging prematurely (i.e., trapping into a local optimum). However, the tendency of electric load sometimes reveals cyclic changes (such as hourly peak in a working day, weekly peak in a business week, and monthly peak in a demand planned year) due to cyclic economic activities or climate seasonal nature. The applications of SVR model to deal with cyclic electric load forecasting have not been widely explored. This investigation presents a SVR-based electric load forecasting model which applied a novel hybrid algorithm, namely chaotic genetic algorithm (CGA), to improve the forecasting performance. With the increase of the complexity and the larger problem scale of tourism demands, genetic algorithm (GA) is often faced with the problems of premature convergence, slowly reaching the global optimal solution or trapping into a local optimum. The proposed CGA based on the chaos optimization algorithm and GA, which employs internal randomness of chaos iterations, is used to overcome premature local optimum in determining three parameters of a SVR model. A numerical example from an existed reference is used to elucidate the forecasting performance of the proposed SSVRCGA model. The forecasting results indicate that the proposed model yields more accurate forecasting results than ARIMA and TF-ε-SVR-SA models. Therefore, the SSVRCGA model is a promising alternative for electric load forecasting.     

基于改进支持向量回归的IGBT老化预测

为了准确预测绝缘栅双极型晶体管（IGBT）的老化状态,提出了一种基于改进鲸鱼优化算法（IWOA）优化支持向量回归（SVR）的IGBT老化预测方法。该方法提取IGBT集电极-发射极电压信号的时频域特征,通过核主成分分析（KPCA）降维将时频域特征融合成一个综合指标来表征IGBT的老化状态;针对鲸鱼优化算法（WOA）不足,在WOA的基础上引入Sobol序列种群初始化、惯性权重和反向学习策略,增强WOA的局部搜索能力和收敛速度;利用IWOA优化SVR的惩罚因子和核参数,并构建一种基于综合指标的IGBT预测模型。利用NASA Ames实验室的IGBT老化数据集对IWOA-SVR方法进行验证,结果表明,所构建IWOA-SVR预测模型可以更准确实现对IGBT的老化预测。     

Particle swarm optimization-least squares support vector regression based forecasting model on dissolved gases in oil-filled power transformers

This paper presents a forecasting model based upon least squares support vector machine (LS-SVM) regression and particle swarm optimization (PSO) algorithm on dissolved gases in oil-filled power transformers. First, the LS-SVM regression model, with radial basis function (RBF) kernel, is established to facilitate the forecasting model. Then a global optimizer, PSO is employed to optimize the hyper-parameters needed in LS-SVM regression. Afterward, a procedure is put forward to serve as an effective tool for forecasting of gas contents in transformer oil. The application of the proposed model on actual transformer gas data has given promising results. Moreover, four other forecasting models, derived from back propagation neural network (BPNN), radial basis function neural network (RBFNN), generalized regression neural network (GRNN) and support vector regression (SVR), are selected for comparisons. The experimental results further demonstrate that the proposed model achieves better forecasting performance than its counterparts under the circumstances of limited samples.     

变尺度混沌蚁群算法在NOx排放优化中的应用

为了控制燃煤电厂NOx排放,应用支持向量回归建立了大型四角切圆燃烧电站锅炉NOx 排放特性模型。利用大样本量的热态实炉NOx 排放试验数据对模型进行了训练和验证,结合NOx排放模型采用一种变尺度混沌蚁群算法对锅炉运行参数进行优化, 定量分析优化算法参数对优化结果的影响。计算结果表明,相对于BP神经网络,支持向量回归模型能更好地预测锅炉NOx排放;变尺度混沌蚁群算法能明显降低NOx排放,且具有较高的稳定性与鲁棒性,1.8 min的优化时间也便于在线应用;支持向量回归与变尺度蚁群混合算法能有效降低燃煤锅炉NOx排放,是锅炉NOx排放控制的有效工具。     

Nonlinear adaptive filtering using kernel‐based algorithms with dictionary adaptation

Nonlinear adaptive filtering has been extensively studied in the literature, using, for example, Volterra filters or neural networks. Recently, kernel methods have been offering an interesting alternative because they provide a simple extension of linear algorithms to the nonlinear case. The main drawback of online system identification with kernel methods is that the filter complexity increases with time, a limitation resulting from the representer theorem, which states that all past input vectors are required. To overcome this drawback, a particular subset of these input vectors (called dictionary) must be selected to ensure complexity control and good performance. Up to now, all authors considered that, after being introduced into the dictionary, elements stay unchanged even if, because of nonstationarity, they become useless to predict the system output. The objective of this paper is to present an adaptation scheme of dictionary elements, which are considered here as adjustable model parameters, by deriving a gradient‐based method under collinearity constraints. The main interest is to ensure a better tracking performance. To evaluate our approach, dictionary adaptation is introduced into three well‐known kernel‐based adaptive algorithms: kernel recursive least squares, kernel normalized least mean squares, and kernel affine projection. The performance is evaluated on nonlinear adaptive filtering of simulated and real data sets. As confirmed by experiments, our dictionary adaptation scheme allows either complexity reduction or a decrease of the instantaneous quadratic error, or both simultaneously. Copyright © 2015 John Wiley & Sons, Ltd.     

气象数据弱相关的光伏出力短期预测

光伏出力受气象因素影响,气象数据的有效程度影响着预测结果的准确性。本文提出了气象数据与光伏出力弱相关时短期光伏出力的预测方法。首先采用Pearson关联系数分析法得到影响光伏发电的主要因素,而后采用模糊聚类理论构建相似日,建立了具有优秀小样本学习能力的支持向量回归机预测模型。针对该模型,提出了两阶段确定模型参数的方法,首先采用全局网格搜索确定核参数p和正则化参数C的取值范围,再通过自适应差分进化算法寻找最优核参数p和正则化参数C,以提高参数ε选取范围设置较大时的预测精度。实例测试表明,使用本文提出的SVR方法预测的平均RMSE为5.551%,满足预测要求,比常规BP预测方法提高精度1.238%,在气象数据弱相关时对光伏短期出力有更好的预测能力。     

A hybrid model of EMD and multiple-kernel RVR algorithm for wind speed prediction

In this paper, the hybrid model of empirical mode decomposition and multiple-kernel relevance vector regression algorithm (EMD-MkRVR) is presented for wind speed prediction. The multiple-kernel relevance vector regression (MkRVR) model includes radial basis function (RBF) kernel and polynomial kernel whose proportions are determined by a controlled parameter. Grid method is used to select the kernel parameters and controlled parameter in this study. In addition, wind speed can be regarded as a signal and decomposed into several intrinsic mode functions (IMFs) with different frequency range by empirical mode decomposition (EMD), the prediction models of these decomposed signals can be established by MkRVR with their respective appropriate embedding dimension. The experimental results show that the EMD-MkRVR model has a better prediction ability for wind speed than the RBF kernel RVR (RBFRVR) model and the polynomial kernel RVR (PolyRVR) model.     

基于MCEA-KPCA和组合SVR的滚动轴承剩余使用寿命预测

为了准确预测滚动轴承的剩余使用寿命(RUL),提出一种多评价标准有效性分析(MCEA)、核主成分分析(KPCA)和组合支持向量回归(SVR)相结合的滚动轴承RUL预测方法。该方法对提取的特征计算每个评价标准的有效性得分,自适应地确定每个评价标准的权重,筛选出有效性总得分高于其整体平均值的特征,进一步利用KPCA去除已筛选特征之间的信息冗余,建立约简后的特征矩阵。将多个轴承约简后的特征分别作为SVR的输入,当前使用寿命与全寿命的比值p即RUL作为输出,建立多个SVR模型,并采用自适应的方法确定各模型的权重,最终构建组合SVR预测模型。最后,对与训练不同的轴承进行测试,将约简后特征输入到组合SVR预测模型中,预测轴承的p值,实验结果表明,所提方法可准确地对滚动轴承进行RUL预测。     

基于支持向量回归的300MW电站锅炉再热汽温建模

为解决某电厂300MW电站锅炉再热汽温异常的问题,提出一种基于支持向量回归的建模方法,采用现场数据进行数据建模。建立在数据统计特性基础上的模型具有高的回归相关度,能反映出再热汽温与操作参数之间的内在联系。针对机组存在的再热器出口汽温偏低而部分管壁温度过高的问题进行了回归分析,结果表明模型具有较高的相关系数,且模型复杂度较低,具有好的鲁棒性。作为现场试验辅助手段,对进一步进行参数优化和再热汽温调节具有重要指导意义和参考价值。     

基于v-支持向量回归的T-S模糊模型辨识

结论参数对T-S模糊模型的泛化能力有重要影响。该文引入v-支持向量回归机(v-SVRM),把T-S模型结论参数的辨识问题转化为一个约束优化问题,并推导了新的迭代求解算法。该方法通过一个参数v控制支持向量的数目和落在ε不灵敏带外样本点的数目,并自动计算合适的ε。针对典型负荷被控对象的仿真结果表明:该方法比通常采用最小二乘法进行结论参数辨识的方法具有更好的泛化能力;此外,由于采用了ε不灵敏损失函数,该方法具有更好的噪声适应能力。     

基于熵权与混合代理模型的永磁驱动器的优化设计

针对永磁驱动器(PMD)的结构设计问题,提出一种基于改进熵权法结合混合代理模型的优化设计方法。首先利用基于交叉验证误差的最优加权法,将响应曲面法、克里金法以及支持向量机回归结合起来,构建PMD的参数变量与响应变量之间的混合代理模型;然后引入改进的熵权法,将PMD的多指标转化为单一综合指标,并建立其优化的数学模型,通过自适应权重粒子群优化算法求解;最后对结果进行有限元仿真分析和实验室仿真平台验证。研究结果表明,所提出的优化设计方法优于其它方法,得到的PMD结构参数合理有效,较好的实现了PMD的多目标优化设计。     

基于 CEEMDAN 与改进核极限学习机的S700K 转辙机健康状态诊断

针对S700K转辙机健康状态分类过于粗放、诊断速度慢、效率低的问题,提出一种基于CEEMDAN与改进核极限学习机(kernel based extreme learning machine, KELM)的诊断方法。首先,对S700K转辙机功率数据进行自适应噪声完备集合经验模态分解,得到6个本征模态函数(intrinsic mode function, IMF);然后,计算本征模态函数的模糊熵值(fuzzy entropy, fuzzyEn, FE)作为表征转辙机健康状态的特征参数;最后,利用麻雀算法(sparrow search algorithm, SSA)改进的核极限学习机对9种健康状态进行健康诊断,并与SVR和ELM模型进行对比。仿真结果表明,改进核极限学机模型准确率、精确率、召回率等指标分别达到97.8%、98.0%、97.8%,相较于SVR和ELM模型,SSA-KELM模型在保证运行速度的基础上,将诊断准确率至少提高2.2%。     

基于ALO-SVR的锂离子电池剩余使用寿命预测

锂离子电池(Lithium-ion batteries,LIBs)的剩余使用寿命(remaining useful life,RUL)预测在电池故障预测与健康管理(prognostics and health management,PHM)中起着十分重要的作用。准确预测电池RUL可以提前对存在安全隐患的电池进行维护和更换,以确保储能系统安全可靠。文章提出一种基于蚁狮优化和支持向量回归(ant lion optimization and support vector regression,ALO-SVR)的方法,可有效提高锂离子电池RUL预测的准确性。SVR方法在处理小样本数据和时间序列分析上具有优势,但SVR方法在内核参数选择上存在困难。因此,文章利用ALO算法优化SVR核参数,随后采用PCoE(NASA ames prognostics center of excellence)和CALCE(center for advanced life cycle engineering)电池数据集对所提方法进行仿真验证。通过对比SVR方法,ALO-SVR方法可以提供更精确的电池RUL预测结果,能有效提高锂离子电池剩余使用寿命预测的准确性和鲁棒性。