基于DLENN模型的沈阳地区PM2.5浓度预测

受到重工业发展规模、北温带季风气候、秋冬季燃煤取暖、机动车拥堵状况以及微观气象条件等各种因素影响,沈阳地区PM2.5浓度变化具有趋势性、周期性及随机性特征。针对上述三种特征,论文构建了一种集成双向长短期记忆网络的神经网络预测模型DLENN(Double-LSTM Ensemble Neural Network),内含的两个方向LSTM分别刻画PM2.5浓度变化趋势性和周期性,然后采用线性回归杂合神经网络来捕捉PM2.5浓度变化的随机性。基于沈阳地区11个监测站2016至2017年空气质量和气象条件数据,本文将DLENN模型分别与自回归移动平均ARIMA模型、支持向量机SVM模型、随机森林RF模型和梯度提升树GBDT集成学习方法进行对比实验,结果表明DLENN预测模型稳定优于其他方法,其预测误差RMSE相对于ARIMA、SVM和集成模型分别下降了9.23%、3.83%、5.49%.     

Permissible idealizations for the purpose of prediction

Every model leaves out or distorts some factors that are causally connected to its target phenomenon—the phenomenon that it seeks to predict or explain. If we want to make predictions, and we want to base decisions on those predictions, what is it safe to omit or to simplify, and what ought a causal model to describe fully and correctly? A schematic answer: the factors that matter are those that make a difference to the target phenomenon. There are several ways to understand differencemaking. This paper advances a view as to which is the most relevant to the forecaster and the decision-maker. It turns out that the right notion of differencemaking for thinking about idealization in prediction is also the right notion for thinking about idealization in explanation; this suggests a carefully circumscribed version of Hempel’s famous thesis that there is a symmetry between explanation and prediction.     

混凝土自收缩的测定及若干因素对自收缩影响规律的研究

采用非接触式电涡流位移传感器设计了自密实混凝土自收缩试验方法,进行自密实混凝土自收缩试验,考察粉煤灰单掺、粉煤灰与矿渣复掺、胶结料用量、水胶比对自密实混凝土自收缩的影响规律,并在试验的基础上提出自密实混凝土的自收缩计算模型.研究表明,该模型能够准确地对自密实混凝土的自收缩进行有效预测,可供相关研究人员进行参考.     

Prediction in context: On the comparative epistemic merit of predictive success

The considerations set out in the paper are intended to suggest that in practical contexts predictive power does not play the outstanding roles sometimes accredited to it in an epistemic framework. Rather, predictive power is part of a network of other merits and achievements. Predictive power needs to be judged differently according to the specific conditions that apply. First, predictions need to be part of an explanatory framework if they are supposed to guide actions reliably. Second, in scientific expertise, the demand for accurate predictions is replaced with the objective of specifying a robust corridor of estimates. Finally, it is highly uncertain to predict the success of research projects. The overall purpose of the paper is to enlarge the debate about predictions by addressing specifically the roles of predictions in application-oriented research.     

GMPC模型在事故征候预测中的实现

事故征候数据的研究对航空安全管理有至关重要的作用,以科学的算法对其预测更是事故预防的关键。以2003~2012年这十年的事故征候数据和运营飞机数据为基础,通过统计运算处理,结合GM(1,1)模型和马尔科夫模型优点,建立GMPC模型,实现动态预测,可为民航安全管理提供决策依据。实践表明,该模型在预测未来一两年的事故征候数据时较为准确,作为短期预测民航安全趋势是可行的。     

D-FNN模型对中国FDI的预测

建立了动态模糊神经网络模型,并运用此模型对中国经济产生深远影响的FDI进行了预测。结果表明D-FNN在预测精度上表现优异,具有一定的实用性。     

Forecasting time series with increasing seasonal variation

Four options for modeling and forecasting time series data containing increasing seasonal variation are discussed, including data transformations, double seasonal difference models and two kinds of transfer function-type ARIMA models employing seasonal dummy variables. An explanation is given for the typical ARIMA model identification analysis failing to identify double seasonal difference models for this kind of data. A logical process of selecting one option for a particular case is outlined, focusing on issues of linear versus non-linear increasing seasonal variation, and the level of stochastic versus deterministic behavior in a time series. Example models for the various options are presented for six time series, with point forecast and interval forecast comparisons. Interval forecasts from data-transformation models are found to generally be too wide and sometimes illogical in the dependence of their width on the point forecast level. Suspicion that maximum likelihood estimation of ARIMA models leads to excessive indications of unit roots in seasonal moving-average operators is reported.     

针对不平衡数据集的客户流失预测算法

针对客户关系管理中的客户流失预测问题进行探讨,通过对客户流失数据特点的分析,以及现有预测算法的比较,将数据挖掘方法中的随机森林算法引入客户流失预测,建立预测模型,并在实际的银行业贷款客户数据集上进行实验,得到了较好的效果.     

油田水CaCO_3结垢反应的计算与应用

以热力学计算为基础，广泛采用油田水的数据．计算了ＣａＣＯ３的预测条件，并根据油田生产实际，建立了使用防垢剂时的预测模型．根据理论模型编制成计算机程序，并与文献结果和油田观测结果进行了对比，提出了临界浓度和临界稳定常数的计算和应用方法．为防垢剂的选择提供了理论依据，大量计算表明，预测结果与文献和现场观测结果相符．