Local models-based regression trees for very short-term wind speed prediction

This paper evaluates the performance of different types of Regression Trees (RTs) in a real problem of very short-term wind speed prediction from measuring data in wind farms. RT is a solidly established methodology that, contrary to other soft-computing approaches, has been under-explored in problems of wind speed prediction in wind farms. In this paper we comparatively evaluate eight different types of RTs algorithms, and we show that they are able obtain excellent results in real problems of very short-term wind speed prediction, improving existing classical and soft-computing approaches such as multi-linear regression approaches, different types of neural networks and support vector regression algorithms in this problem. We also show that RTs have a very small computation time, that allows the retraining of the algorithms whenever new wind speed data are collected from the measuring towers.     

一种简单易行的短期负荷预测系统

文章研究一种简单易行的自适应模糊预测系统,系统的规则集经过历史数据样本的自适应学习获得,学习算法每天启动运行一次,文中首先阐述从数据信息中获得模糊规则的一般性方法,然后给具体,易懂的预测模型设计。     

Insolation forecasting by a multi-stage neural network

A single-stage neural network has been proposed to forecast next day insolation. In this paper, a multi-stage neural network is developed to reduce forecasting error further. A first-stage neural network forecasts average atmospheric pressure for the next day from atmospheric pressure data of the previous day. A second-stage neural network forecasts insolation level for the next day from the average atmospheric pressure and weather data of the previous day. A third-stage neural network forecasts next day insolation from the insolation level and weather data of the previous day. Meteorological data of Omaezaki, Shizuoka at April 1994 were chosen as input data. The insolation values forecasted by the multi-stage and the single-stage neural networks are compared with the measurement values. The results show that the forecasting error is reduced to 24% (by the multi-stage) from 33% (by the single-stage). © 1998 Scripta Technica, Electr Eng Jpn, 125(4): 26–33, 1998     

一种基于Kalman的实时话务量预测算法

移动网话务量的预测可以为解决移动网的网络拥塞、覆盖规划、负载均衡和接入控制等方面的问题提供决策支持。日益复杂的移动网络环境对移动网话务量预测的实时性提出了新的要求,针对这一问题进行了研究,结合卡尔曼滤波算法在预测时效性方面的优点,提出了一种基于卡尔曼滤波的移动网实时话务量预测算。该算法通过对呼叫时延的实时预测实现对移动网实时话务量的预测。仿真结果表明,所提算法具有良好的实时性和预测质量。     

基于贝叶斯法和蒙特卡洛仿真的威布尔型装备器材需求预测

为了解决装备器材历史需求数据少、需求规律不明确的问题,提出一种基于贝叶斯法和蒙特卡洛仿真的威布尔型装备器材需求预测方法。针对威布尔分布尺度参数未知以及形状和尺度参数均未知两种情况,分别基于贝叶斯方法通过解析求解和数值模拟的方式进行了参数估计,并引入柯尔莫哥洛夫-斯米尔诺夫检验法对寿命分布模型进行拟合优度检验;综合考虑修复性维修、预防性维修和装备器材的已使用时间,提出了基于蒙特卡洛仿真的部队装备器材年度需求预测方法。算例分析表明：小样本下通过贝叶斯估计得到的寿命分布模型拟合度高,基于仿真的需求预测方法简单、有效。     

Forecasting Winning Bid Prices on an Online Auction Market - Data Mining Approaches -

To solve information asymmetry problem on online auction, this study suggests and validates a forecasting model of winning bid prices. Especially, it explores the usability of data mining approaches, such as neural network and Bayesian network in building a forecasting model. This research empirically shows that, in forecasting winning bid prices on online auction, data mining techniques have showed better performance than traditional statistical analysis, such as logistic regression and multivariate regression.     

一种智能型的PCB产业需求预测方法

目前需求预测在整个印制电路板产业的生产活动控制中正扮演着越来越重要的角色。分析了影响印制电路板需求的因素和现有的预测方法,提出了一种适用于PCB产业需求预测的有效方法——遗传/BP—神经网络。实验表明该方法能够进一步改善印制电路板预测的准确度和减少生产成本的消耗。     

基于粗糙特征量的短期电力负荷预测

针对负荷特征一直是实际电力负荷预测中的重大问题。提出了基于粗糙特征量的约简算法。通过对天气及负荷历史数据进行挖掘,找到负荷的关键特征,并与径向基网络结合建立了负荷预测模型。算例结果表明,与按经验选取输入的传统网络相比,预测准确度有了明显的提高,更适用于电力负荷预测。     

基于复合模型的智能化移动通信话务预测

移动通信话务量作为一种时间序列,具有较强的非线性和随机性,而且易受节假日、旅游等客户行为及天气等其它因素的影响。尤其是话务量长期的发展变化,很难用传统的预测方法进行预测。根据移动通信话务量自身特点,采用复合模型,将话务量分为平稳期趋势分量、平稳期周期分量、节假日话务量,用综合评判的分段一元线性回归及模板匹配算法分别对趋势分量、周期分量和节假日话务进行建模。最后,开发了基于复合模型的智能化预测系统,在广东省某市试运行的结果表明：基于复合模型的预测方法比传统预测方法精度高。     

1995年汛期长江流域长期水文气象预报及其验证分析

长江流域汛期长期水文气象预报成果是通过对气候背景的分析，前期天气气候征兆和重要物理因子的分析，最后经专家会商形成的，在预报方案制定过程中，坚持水文与气象，短中长期预报，科学研究与预报应用相结合，从定性上讲，无论时间和空间都基本吻合，对１９９５年长江汛期防洪发挥了重要的耳目作用。