SEND-OUT PREDICTION MODELING USING REGRESSION TECHNIQUE FOR NATURAL GAS SUPPLY PLANS

机译：使用回归技术的天然气供应计划发送预测模型

获取原文

获取原文并翻译 | 示例

页面导航

摘要
著录项
相似文献
相关主题

摘要

Statistical models have been developed using regression technique for gas send-out predictions in gas industries. For regressions, weather factors such as air temperature, wind speed, solar radiation, and humidity were used to correlate with gas send-out. First, linear regression models were developed using stepwise regression technique and models with and without secondary weather factors (wind speed, solar radiation, humidity) were compared. For further improvement of prediction capability, non-linear models were developed using non-linear multiple regression technique. Finally, prediction capabilities of non-linear models were compared with those of the developed linear models. The results showed that non-linear models were much better than linear models in predicting gas send-out for gas supply plans.

机译：已经使用回归技术开发了统计模型，用于天然气行业中的天然气排放预测。为了进行回归分析，使用了诸如空气温度，风速，太阳辐射和湿度之类的天气因素来与气体排放相关。首先，使用逐步回归技术开发线性回归模型，并比较具有和不具有次级天气因素（风速，太阳辐射，湿度）的模型。为了进一步提高预测能力，使用非线性多元回归技术开发了非线性模型。最后，将非线性模型的预测能力与已开发的线性模型的预测能力进行了比较。结果表明，在预测供气计划的出气量方面，非线性模型要好于线性模型。

著录项

来源
《2001 International gas research conference》|2001年|p.1-8|共8页
会议地点 Amsterdam(NL);Amsterdam(NL)
作者
Seungyong Chang;
展开▼
作者单位

Korea Gas Corporation, R D Center, Korea;

展开▼
会议组织
原文格式 PDF
正文语种 eng
中图分类石油产品;
关键词

相似文献

外文文献
中文文献
专利

1. Contribution of bootstrap techniques to PLS regression:application to the prediction model of gas-oil quality control [J] . S.Aji, S.Tavolaro, F.Lantz, Oil & gas science and technology . 2003,第5期

机译：引导技术对PLS回归的贡献：在油气质量控制预测模型中的应用
2. Meta-Prediction Models for Bullwhip Effect Prediction of a Supply Chain Using Regression Analysis [J] . Chiadamrong Navee, Sarnrak Nont International Journal of Information Systems and Supply Chain Management . 2021,第4期

机译：使用回归分析的牛鞭效应预测牛鞭效应预测的元预测模型
3. Comparative multiple quantitative structure-retention relationships modeling of gas chromatographic retention time of essential oils using multiple linear regression, principal component regression, and partial least squares techniques [J] . Li-Tang Qin, Shu-Shen Liv, Hai-Ling Liu, Journal of chromatography, A: Including electrophoresis and other separation methods . 2009,第27期

机译：使用多元线性回归，主成分回归和偏最小二乘技术对精油的气相色谱保留时间进行比较的多个定量结构-保留关系建模
4. SEND-OUT PREDICTION MODELING USING REGRESSION TECHNIQUE FOR NATURAL GAS SUPPLY PLANS [C] . Seungyong Chang International gas research conference . 2001

机译：利用天然气供应计划回归技术发出预测建模
5. Multi-scale modeling and analysis via surrogate modeling techniques for in vivo knee loading predictions. [D] . Adams, Matthew J. 2014

机译：通过替代建模技术进行多尺度建模和分析，以进行体内膝盖负荷预测。
6. Selecting Modeling Techniques for Outcome Prediction: Comparison of Artificial Neural Networks Classification and Regression Trees and Linear Regression Analysis for Predicting Medical Rehabilitation Outcomes [O] . Deborah K. W. Walters, Richard T. Linn, Margaret Kulas, 1999

机译：预测结果的建模技术选择：人工神经网络分类和回归树的比较以及用于预测医疗康复结果的线性回归分析
7. "A Karnel Regression of Phillips' Data" Abstract: Economists have assumed that the Phillips curve, which shows a positive (negative) relation between inflation and the output ratio (unemployment rate), may be mapped off the aggregate demand -aggregate supply apparatus. The paper shows that the Phillips curve requires that unlikely restrictions be put on the form of the aggregate supply and aggregate demand curves. In this case, it is inappropriate to treat data on inflation and capacity utilization as the basis for estimating an underlying formal model. The paper therefore uses a nonparametric, data-driven method to describe the data. This method, of kernel regression, shows the inflation-unemployment association in Phillips's sample to be negative on a global scale, yet irregular within particular ranges of unemployment. [O] . Nancy J. Wulwick, Y.P. Mack 100

机译：“菲利普斯数据的卡尔内尔回归”摘要：经济学家认为菲利普斯曲线显示了通货膨胀与产出比率（失业率）之间的正（负）关系，可以从总需求 - 聚集供应设备中绘制出来。该文表明，菲利普斯曲线要求对总供给和总需求曲线的形式进行不太可能的限制。在这种情况下，将通货膨胀和能力利用数据作为估算潜在正式模型的基础是不恰当的。因此，本文使用非参数，数据驱动的方法来描述数据。核心回归的这种方法表明菲利普斯样本中的通货膨胀 - 失业关联在全球范围内是负面的，但在特定的失业范围内是不规则的。

SEND-OUT PREDICTION MODELING USING REGRESSION TECHNIQUE FOR NATURAL GAS SUPPLY PLANS

摘要

著录项

相似文献

相关主题

期刊订阅