基于组合预测的变压器故障诊断模型

为了快速准确地预测出变压器的故障类型,及时做好维修工作,本文提出了一种基于非线性规划的组合预测模型.首先,利用改进的鲸鱼算法优化BP神经网络建立IWOA-BP预测模型;然后,在IWOA-BP预测模型和梯度提升树的基础上,利用非线性规划与遗传算法相结合的方法确定各算法的权系数,再将各算法的结果加权得出组合模型的最终预测结果.通过实例验证,IWOA-BP预测模型的变压器故障预测效果强于BP神经网络、随机森林等多种预测模型,并且利用IWOA-BP预测模型和梯度提升树建立的组合模型,其预测准确率高于组合中任意一种算法.     

新组合模型在光电功率预测中的应用

为提高光伏预测要求的精准性,文章提出一种新算法将神经网络和ARMA算法改进组合,构成NEW ARMA-BP模型算法.以某30兆瓦的光伏电站采集的输出功率为输入样本,基于ARMA和BP神经网络算法在Matlab环境下依次搭建了相应的预测模型,预估光伏短期输出量.采用"误差正态检验图"判断基于两种不同算法的误差水平,依据两种单模型预测误差,运用所提出的新方法计算权值并获得新的预测值.基于Matlab的仿真结论验证了组合预测在光伏输出预测领域的适用性.     

基于粒子群优化的神经网络预测模型

BP学习算法多采用梯度下降法调整权值,针对其易陷入局部极小、收敛速度慢和易引起振荡的固有缺陷,提出了一种改进粒子群神经网络算法.其基本思想是:首先采用改进粒子群优化算法反复优化BP神经网络模型的权值参数组合,再用BP算法对得到的网络参数进一步精确优化,最后用得到精确的最优参数组合进行预测.实验结果表明,该算法在股指预测中的预测性能明显提高.     

基于时序分解的自适应在线学习电力负荷预测算法研究

传统的机器学习方法无法捕捉到电力负荷需求的不确定性以及动态变化规律.本文将最新提出的隐马尔可夫模型在线学习算法应用于电力负荷预测研究,充分提取历史数据中的不确定性特征和动态变化规律,并结合分解算法,更精确利用数据中的动态变化特征,从而提高预测精度.算法基于隐马尔可夫概率预测模型,在获得新样本时对模型进行在线更新,适应最新数据;利用STL时序分解算法对负荷数据进行分解,使具有不同不确定性和动态变化规律的分量分离开,再分别使用在线学习算法对不同特征的分量进行预测,构造电力负荷预测组合算法.基于三个公开电力负荷数据集的测试结果表明,相比于单一的在线学习模型,本文提出的组合算法提高了预测精度,预测相对误差最高减少了27%.     

基于蚁群算法的灰色组合预测模型

分别利用灰色GM(1,1)模型、GM(1,1)优化模型和新息GM(1,1)模型建立三个单项预测模型,进一步建立了组合灰色预测模型,组合模型的权系数利用蚁群算法确定.最后给出了一个我国人口数量组合预测模型,计算结果表明,基于蚁群算法的灰色组合预测模型的拟合和预测精度要优于传统组合预测模型.     

基于EWT-PSO-SVM误差校正组合模型的中国股市预测研究

鉴于股市预测的复杂性.遵循"先分解后集成"的总体建模思路.文章基于EWT分解算法和SVM支持向量机模型.同时结合PSO粒子群优化算法和误差校正组合预测方法,构建了一种中国股票市场建模及预测的EWT-PSO-SVM误差校正组合预测模型.先基于EWT算法将原始价格序列分解成若干分量,再根据频率将其重组成高、中、低频3个分量,对它们分别建立PSO-SVM误差校正组合模型.最后集成各个分量的预测结果.与其他预测模型相比较,文章所构建预测模型的MSE、MAE、MAPE、RMSE、Theil不等系数、确定性系数DC和方向性指标DS 7个指标均优于其他基准预测模型,MCS检验结果同样表明本文构建模型的预测性能最优.稳健性检验结果进一步证实了文章构建的模型预测性能所具备的稳健性.     

基于交叉验证方法的海洋经济发展的组合预测

海洋经济的发展是我国经济发展的重要组成部分,因此科学地对海洋经济进行预测分析具有重要的现实意义.文章选取全国海洋生产总值作为研究对象,分别利用灰色模型,ARIMA模型和组合模型进行预测.提出了一种基于交叉验证方法确定组合预测权重的新算法,研究显示该方法相对传统组合预测具有更小的平均绝对百分比误差(MAPE).     

基于Bass-BP组合模型的微信扩散研究

微信扩散具备非线性变化特征,仅采纳单一预测算法无法有效描述微信扩散的规律.因此,提出了Bass-BP扩散组合模型.首先利用经典Bass模型对微信数据进行初步预测,再利用BP神经网络对Bass模型预测结果进一步非线性逼近.结果表明,Bass-BP组合模型相较于Bass经典模型和BP神经网络这两种单一的预测模型,具有更好的预测效果.     

一种线性ARIMA基础上的非线性BP神经网络修正组合方法在粮食产量预测中的运用

对我国粮食产量预测工作作出了重大贡献.通过线性规律和非线性规律的区分,构建了自回归移动平均(ARIMA)和BP人工神经网络的组合预测模型,结论显示:ARIMA和BP算法能够各自对粮食产量序列中的线性和非线性规律实现充分挖掘,拟合精度几乎达到完美的程度,克服了以往单一预测和组合预测中信息挖掘能力不足的缺陷.     

基于GAAA-RBF神经网络的工程估价方法

利用遗传-蚁群混合算法(GAAA),对RBF神经网络的主要结构参数中心矢量、基宽向量和网络权重进行组合优化,建立了GAAA-RBF神经网络组合算法的工程估价模型.将55个工程造价案例,随机抽取10个作为预测样本,剩下的45个作为训练样本.通过与相同结构的RBF神经网络相比较,结果表明算法克服了RBF神经网络易陷于局部极值、搜索质量差和精度不高的缺点,改善了RBF神经网络的泛化能力,收敛速度快,输出稳定性好,提高了工程造价的预测精度.     

Multi-variable grey model based on dynamic background algorithm for forecasting the interval sequence

The multi-variable grey model based on dynamic background algorithm improves the forecasting performance of the multi-variable grey model on the precise number sequence. In order to make this model suitable for the interval sequence, the matrix form of the multi-variable grey model based on dynamic background algorithm is proposed in the paper. In the modeling process, the interval is treated as a two-dimensional column vector, the parameters of the multi-variable grey model are replaced by matrices, and the dynamic background algorithm for interval sequences is proposed. The analysis results of the matrix algorithm for the dynamic background value and the prediction formula show that the new model is essentially a way to predict one of the two bounds of an interval by combining them, reflecting the integrity and interaction between the lower and upper bounds. The interval predictions of industrial electricity consumption of Zhejiang Province, China national electricity consumption and consumer price index show that the new model can well predict the minimum and maximum values of the interval sequence and has better prediction performance compared with the method of predicting each boundary sequence separately.     

Lazy lasso for local regression

Locally weighted regression is a technique that predicts the response for new data items from their neighbors in the training data set, where closer data items are assigned higher weights in the prediction. However, the original method may suffer from overfitting and fail to select the relevant variables. In this paper we propose combining a regularization approach with locally weighted regression to achieve sparse models. Specifically, the lasso is a shrinkage and selection method for linear regression. We present an algorithm that embeds lasso in an iterative procedure that alternatively computes weights and performs lasso-wise regression. The algorithm is tested on three synthetic scenarios and two real data sets. Results show that the proposed method outperforms linear and local models for several kinds of scenarios.     

改进型LOBNN & AR-GARCH模型在股票预测中的应用

为实现对股票价格的短期预测,本文在Laguerre正交基神经网络 (LOBNN)模型的基础上,提出了一种新的组合预测模型来预测短期股价的变化。该模型先通过改进LOBNN模型的权值求解算法,用以增强模型的通用性。接着在其基础上设计新的迭代算法,进一步提高模型的预测精度,进而得到新的LOBNN模型。之后将股价数据分别代入AR-GARCH模型和改进后的LOBNN模型,得到输入数据的两组预测值。最后通过不同的权重来组合两种预测结果,生成最终股价的预测结果。文末的仿真结果表明该组合模型在预测精度与通用性上较原始模型有较大的提升,是一种高效的预测模型。     

基于改进PSO-BP神经网络的贮存可靠性预测

贮存可靠性是军事储备质量监测的重要环节,科学准确地预测贮存可靠度是现代化军事评估的必然要求。针对历史贮存数据,建立可靠度与年限的贮存可靠性预测模型,采用进化策略改进粒子群算法(PSO)优化BP神经网络进行贮存可靠性预测。通过数据扩充提高样本质量和数量,应用改进后的PSO算法优化BP神经网络的初始权值和阈值,提高网络的泛化能力。PSO算法较好的全局搜索能力与BP网络很强的局部搜索能力相结合,能够避免早熟现象,提高算法的收敛速度及预测精度。实验结果表明,改进的PSO-BP网络模型比PSO-BP和BP神经网络获得更好的预测性能。     

Condition number based complexity estimate for computing local extrema

In this paper, we present a new algorithm for computing local extrema by modifying and combining algorithms in symbolic and numerical computation. This new algorithm improves the classical steepest descent method that may not terminate, by combining a Sturm’s theorem based separation method and a sufficient condition on infeasibility. In addition, we incorporate a grid subdivision method into our algorithm to approximate all local extrema. The complexity of our algorithm is polynomial in a newly defined condition number, and singly exponential in the number of variables.     

A neural fuzzy control system with structure and parameter learning

A general connectionist model, called neural fuzzy control network (NFCN), is proposed for the realization of a fuzzy logic control system. The proposed NFCN is a feedforward multilayered network which integrates the basic elements and functions of a traditional fuzzy logic controller into a connectionist structure which has distributed learning abilities. The NFCN can be constructed from supervised training examples by machine learning techniques, and the connectionist structure can be trained to develop fuzzy logic rules and find membership functions. Associated with the NFCN is a two-phase hybrid learning algorithm which utilizes unsupervised learning schemes for structure learning and the backpropagation learning scheme for parameter learning. By combining both unsupervised and supervised learning schemes, the learning speed converges much faster than the original backpropagation algorithm. The two-phase hybrid learning algorithm requires exact supervised training data for learning. In some real-time applications, exact training data may be expensive or even impossible to obtain. To solve this problem, a reinforcement neural fuzzy control network (RNFCN) is further proposed. The RNFCN is constructed by integrating two NFCNs, one functioning as a fuzzy predictor and the other as a fuzzy controller. By combining a proposed on-line supervised structure-parameter learning technique, the temporal difference prediction method, and the stochastic exploratory algorithm, a reinforcement learning algorithm is proposed, which can construct a RNFCN automatically and dynamically through a reward-penalty signal (i.e., “good” or “bad” signal). Two examples are presented to illustrate the performance and applicability of the proposed models and learning algorithms.     

中国能源消费的EMD分析与预测

基于EMD分析,结合SVM方法构建了中国能源消费的预测模型,提供了一种新的能源消费预测思路.研究结果表明,到2020年我国能源消费量3027百万吨油当量,低于IEA2010能源展望中的预测结果.在低碳经济、节能减排已成为国际政治领域热点议题的背景下,采用定量的方法对中国能源消费状况进行分析与预测,有利于分析中国的国际政治压力及中国能源企业海外投资.     

Framework for efficient feature selection in genetic algorithm based data mining

We present the design of more effective and efficient genetic algorithm based data mining techniques that use the concepts of feature selection. Explicit feature selection is traditionally done as a wrapper approach where every candidate feature subset is evaluated by executing the data mining algorithm on that subset. In this article we present a GA for doing both the tasks of mining and feature selection simultaneously by evolving a binary code along side the chromosome structure used for evolving the rules. We then present a wrapper approach to feature selection based on Hausdorff distance measure. Results from applying the above techniques to a real world data mining problem show that combining both the feature selection methods provides the best performance in terms of prediction accuracy and computational efficiency.     

Interactive visualization in multiclass learning: integrating the SASSC algorithm with KLIMT

The paper deals with multiclass learning from the perspective of analytically interpreting the results of the analysis as well as that of navigating into them by using interactive visualization tools. It is showed that by combining the Sequential Automatic Search of Subset of Classifiers (SASSC) algorithm with the interactive visualization of classification trees provided by the Klassification—Interactive Methods for Trees (KLIMT) software it is possible to highlight important information deriving from the knowledge extraction process without neglecting the prediction accuracy of the classification method. Empirical evidence from two benchmark datasets demonstrates the advantages deriving from the joint use of SASSC and KLIMT.