基于SCGM(1,1)的三种负荷预测模型及其组合

振动法故障诊断是近年来兴起的变压器故障诊断新方法,正成为常规法故障诊断的有力补充,在大型变压器故障诊断中发挥着越来越重要的作用。对振动法故障诊断的原理、作用、技术要求等方面进行了分析,结合相关电力规范提出了振动法故障诊断的要点,并阐述了振动法故障诊断在今后一段时期内发展趋势及值得关注的若干问题。     

Data-driven approach to machine condition prognosis using least square regression tree

Machine fault prognosis techniques have been profoundly considered in the recent time due to their substantial profit for reducing unexpected faults or unscheduled maintenance. With those techniques, the working conditions of components, the trending of fault propagation, and the time-to-failure are precisely forecasted before they reach the failure thresholds. In this work, we propose the least square regression tree (LSRT) approach, which is an extension of the classification and regression tree (CART), in association with one-step-ahead prediction of time-series forecasting techniques to predict the future machine condition. In this technique, the number of available observations is first determined by using Cao’s method and LSRT is employed as a prediction model in the next step. The proposed approach is evaluated by real data of a low methane compressor. Furthermore, a comparative study of the predicted results obtained from CART and LSRT are carried out to prove the accuracy. The predicted results show that LSRT offers the potential for machine condition prognosis. This paper was recommended for publication in revised form by Associate Editor Eung-Soo Shin Van Tung Tran is a lecturer at the Hochiminh City University of Technology in Vietnam. He received his B.S. and M.S. degrees in mechanical engineering from Hochiminh City University of Technology, Vietnam, in 1997 and 2003, respectively, and Ph.D. from Pukyong National University, South Korea in 2009. His research interests include machine fault diagnosis and condition prognosis. Bo-Suk Yang is a professor at the Puyong National University in Korea. He received his Ph.D. degree in mechanical engineering from Kobe University, Japan in 1985. His main research fields cover machine dynamics and vibration engineering, intelligent optimum design, and condition monitoring and diagnostics in rotating machinery. He has published well over 190 research papers in the research areas of vibration analysis, intelligent optimum design and diagnosis of rotating machinery. He is listed in Who’s Who in the World, Who’s Who in Science and Engineering, among others.     

基于运行可靠性和结构重要度的连锁故障预测

基于运行可靠性理论和多属性决策理论,综合考虑当前系统运行状态和线路重要程度的基础上,提出了一种反映电网状态变化和结构变化的连锁故障预测方法。此方法引入考虑了外界环境因素的线路综合停运概率模型确定线路故障概率,并结合线路结构重要度构建了连锁故障综合预测指标。采用离差最大化的灰色关联度算法综合考虑线路结构重要度。通过对IEEE30系统仿真,搜索出了较严重的连锁故障模式,验证了该方法的有效性。     

基于混沌时间序列法的微网短期负荷预测

针对微网中居民小区用电量较低、负荷波动大的特点,提出了结合混沌理论重构相空间并建立最大Lyapunov指数模型的方法。该方法不直接考虑影响负荷的气候、电价等因素,输入数据参数较少,采用C-C方法求延迟时间与嵌入维,运用改进的最大Lyapunov指数方法进行预测。将此方法用于安徽某一小区的实际负荷数据预测,预测结果表明该算法的预测精度较高,可以为微网的优化运行提供负荷依据,仿真结果验证了算法的有效性和实用性。     

基于混沌理论和Legendre正交基神经网络的短期负荷预测

考虑到短期负荷所具有的混沌特性和神经网络的非线性映射能力,提出了一种基于混沌理论的Legendre神经网络预测方法。该方法运用混沌理论对短期负荷数据进行向空间重构,并以欧式距离选取最佳训练样本,而后采用以Legendre正交多项式为隐含层神经元激励函数的三层神经网络进行训练,并运用训练好的网络进行预测。训练网络时,为了确定网络的最佳拓扑结构,文中引入了衍生算法来确定隐含层神经元的最佳个数。实例分析表明了该方法的可行性,且能得到较高的预测精度和良好的预测效果。     

Service levels,system cost and stability of production–inventory control systems

Production-inventory control system models have been analysed in the literature either in terms of their stability against demand fluctuations or in terms of their service level and cost performance under uncertain demand. This article analyses the production-inventory system performance in terms of service level (i.e. order fill rate) and average system costs, under stable settings of the control parameters. The classical automatic pipeline variable inventory and order-based production control system has been modified by explicitly modelling safety stock to help achieve higher services levels in the face of random demand. The stability of the system is affected by the control parameters: fractional rates of adjustment of work-in-process and inventory. However, the service level and average cost are affected by the control parameters as well as the smoothing factor in demand forecasting. This article puts forward five propositions which give light to general system performance based on the parameters selection. Intensive simulation experiments have also been carried out to reveal the performance variations within the stable region, leading to further insights on the system behaviour. The managerial insights which can assist proper tuning of systems to help achieve the desired performances have been discussed.     

A multi-model approach to determine early manufacturing parameters for small-data-set prediction

Constructing an accurate prediction model from a small training data set is an important but difficult task in the field of forecasting. This is because when the data size is small, the incomplete data may mean that the model produced cannot sufficiently represent the true data structure or cause the model training to be overfitted. To address this issue, this paper presents an approach that combines multiple prediction models to extract data information in multiple facets. In the multi-model approach, a compromise weight method is proposed to determine the relative reliability of each of the prediction model. The methods used include multiple regression, artificial neural network, and support vector machines for regression. A thin-film transistor liquid crystal display manufacturing case study is used to illustrate the details of this research. The empirical results not only show that the proposed multi-model can reduce the manufacturing variation and increase the production yield, but also can propose a robust and reliable parameter interval to the online engineers in the early manufacturing stage.     

灰关联模式识别与预测方法在河南省黄河中游流域夏季雨量中长期预报中的应用

基于三门峡、郑州、卢氏、孟津、济源和孟州站6~8月累积雨量资料,采用灰关联模式识别与预测方法进行了中长期雨量预测,预报验证期为5年。结果表明,6个站点的预报合格率均大于等于60%,总体上看丰水年的预报合格率大于枯水年,其中卢氏站的预报合格率最好,达到100%;三门峡站2006年、孟津站2002年和2006年预报相对误差较大,原因在于其实测值在各自站点从小到大排序中排位靠前,可见灰关联模式识别与预测方法在预测极小值时效果较差。     

基于相空间重构的BP神经网络月径流预测模型及应用

针对径流时间序列的非线性特点,利用重构相空间的嵌入维数确定神经网络结构,建立了基于相空间重构理论的BP神经网络月径流预测模型,并采用该模型预测了瀑布沟水库的径流时间序列。结果表明,该BP神经网络预测模型收敛速度快、预测精度较高。     

系统动力学模型在电网公司经营决策支持中的应用

对电网公司关键经营统计指标进行预测及敏感性分析,先利用加权移动平均、多元线性回归、灰色模型预测售电量指标得到三种预测结果,将GDP、人口作为神经网络的输入条件,借助Matlab软件,基于RBF神经网络的综合预测模型将三种预测结果拟合后预测出核心指标售电量;再借助STELLA软件用图形表示因素之间的相互关系和影响,进而构建系统动力学模型,定量分析了各指标间的关系,获得了所有关键指标的预测值,并选取售电量和售电均价做敏感性分析,分析售电量和售电均价变化对电网投资能力的影响程度。