检索结果-南通市图书馆

Conditional Coverage Estimation for High-Quality prediction intervals

Journal of Systems Science and Systems Engineering 2023年第3期32卷 289-319页

作者： Ziyi Huang Henry Lam Haofeng Zhang School of Engineering and Applied Science Columbia UniversityNew YorkNY 10027USA

Deep learning has been recently studied to generate high-quality prediction intervals(PIs)for uncertainty quantification in regression tasks,including recent applications in simulation *** high-quality criterion requires PIs to be as narrow as possible,whilst maintaining a pre-specified level of data(marginal)***,most existing works for high-quality PIs lack accurate information on conditional coverage,which may cause unreliable predictions if it is significantly smaller than the marginal *** address this problem,we propose an end-to-end framework which could output high-quality PIs and simultaneously provide their conditional coverage *** doing so,we design a new loss function that is both easy-to-implement and theoretically justified via an exponential concentration *** evaluation on real-world benchmark datasets and synthetic examples shows that our approach not only achieves competitive results on high-quality PIs in terms of average PI width,but also accurately estimates conditional coverage information that is useful in assessing model uncertainty.

关键词： Uncertainty quantification prediction intervals conditional coverage neural networks calibrationerror

来源：

维普期刊数据库评论

在线全文

维普期刊数据库

学校读者我要写书评

暂无评论

Nonparametric Probabilistic prediction of Regional PV Outputs Based on Granule-based Clustering and Direct Optimization Programming

引用

Journal of Modern Power Systems and Clean Energy 2023年第5期11卷 1450-1461页

作者： Yonghui Sun Yan Zhou Sen Wang Rabea Jamil Mahfoud Hassan Haes Alhelou George Sideratos Nikos Hatziargyriou Pierluigi Siano College of Energy and Electrical Engineering Hohai UniversityNanjing 210098China College of Water Conservancy and Hydropower Engineering Hohai UniversityNanjing 210098China IEEE Department of Electrical Engineering Tishreen UniversityLatakia 2230Syria National Technical University of Athens Athens 15773Greece Department of Management&Innovation Systems University of SalernoSalerno 84084Italy

Regional photovoltaic(PV) power prediction plays an important role in power system planning and operation. To effectively improve the performance of prediction intervals(PIs) for very short-term regional PV outputs, an efficient nonparametric probabilistic prediction method based on granulebased clustering(GC) and direct optimization programming(DOP) is proposed. First, GC is proposed to formulate and cluster the sample granules consisting of numerical weather prediction(NWP) and historical regional output data, for the enhanced hierarchical clustering performance. Then, to improve the accuracy of samples' utilization, an unbalanced extension is used to reconstruct the training samples consisting of power time series. After that, DOP is applied to quantify the output weights based on the optimal overall performance. Meanwhile, a balance coefficient is studied for the enhanced reliability of PIs. Finally, the proposed method is validated through multistep PIs based on the numerical comparison of real PV generation data.

关键词： Regional photovoltaic output prediction intervals granule-based clustering direct optimization programming nonparametric probabilistic prediction

来源：

维普期刊数据库

同方期刊数据库评论

在线全文

学校读者我要写书评

暂无评论

An integrated new threshold FCMs Markov chain based forecasting model for analyzing the power of stock trading trend

引用

Financial Innovation 2019年第1期5卷 600-618页

作者： Kavitha Ganesan Udhayakumar Annamalai Nagarajan Deivanayagampillai Department of Mathematics Hindustan Institute of Technology&SciencePadurChennaiTamil Nadu 603103India

This paper explores the power of stock trading trend using an integrated New ThresholdFuzzy Cognitive Maps(NTFCMs)Markov chain *** new model captures thepositive as well as the negative jumps and predicts the trend for different stocks over 4months which follow an uptrend,downtrend and a mixed *** mean absolute percent error(MAPE)tolerance limits,the root mean square error(RMSE)tolerance limits aredetermined for various stock indices over a multi-timeframe period and observed for theexisting methods lying within the defined *** results show for every‘n’number ofpredictions made,the predicted close value of the day’s stock price was within tolerancelimit with 0%error and with 100%accuracy in predicting the future trend.

关键词： Financial markets prediction intervals Price forecasting Comparative studies Decision making Fuzzy cognitive maps(FCMs) Markov chain

来源：

维普期刊数据库评论

在线全文

维普期刊数据库

学校读者我要写书评

暂无评论

Statistical and environmental analyses for the definition of a regional rainfall threshold system for landslide triggering in Tuscany （Italy）

引用

Journal of Geographical Sciences 2012年第4期22卷 617-629页

作者： ROSI Ascanio SEGONI Samuele CATANI Filippo CASAGLI Nicola Department of Earth Science University of Florence Florence Italy

The aim of this work is the determination of regional-scale rainfall thresholds for the triggering of landslides in the Tuscany Region （Italy）. The critical rainfall events related to the occurrence of 593 past landslides were characterized in terms of duration （D） and intensity （I） I and D values were plotted in a log-log diagram and a lower boundary was clearly noticeable： it was interpreted as a threshold representing the rainfall conditions associated to landsliding. That was also confirmed by a comparison with many literature thresholds, but at the same time it was clear that a similar threshold would be affected by a too large approximation to be effectively used for a regional warning system. Therefore, further analyses were performed differentiating the events on the basis of seasonality, magnitude, location, land use and lithology. None of these criteria led to discriminate among all the events different groups to be characterized by a specific and more effective threshold. This outcome could be interpreted as the demonstration that at regional scale the best results are obtained by the simplest approach, in our case an empirical black box model which accounts only for two rainfall parameters （I and D）. So a set of thresholds could be conveniently defined using a statistical approach： four thresholds corresponding to four severity levels were defined by means of the prediction interval technique and we developed a prototype warning system based on rainfall recordings or weather forecasts.

关键词： landslide threshold rainfall black box models prediction intervals

来源：

维普期刊数据库

同方期刊数据库评论

在线全文

学校读者我要写书评

暂无评论

Scenario-based robust MPC for energy management systems with renewable generators

Scenario-based robust MPC for energy management systems with...

引用

第37届中国控制会议

作者： Shotaro Sato Toru Namerikawa Department of System Design Engineering Keio University

This paper addresses photovoltaics（PV） power prediction and energy storage problem which are known to be a key technology in energy management systems（EMS）. Extending results of the point prediction of PV power, we first describe a prediction interval（PI） method using a copula, which can express the relation between a multivariable joint distribution and each marginal distribution. Then, resorting to the PI method, the energy storage optimization problem in a building is developed. A scenario robust（SR） optimization theorem, which calculates the robustness of the optimal solution, is applied to the proposed PI method,and hence we obtain an optimal energy storage solution taking the robustness of the solution into account. Additionally, we propose a method which combines a model predictive control（MPC） technique and SR to reduce the total electricity costs. The simulation results finally illustrate the cost reduction and robustness of the proposed method.

关键词： prediction intervals Robust Optimization Energy Management

来源： cnki会议评论

在线全文

cnki会议

学校读者我要写书评

暂无评论

欢迎您,

建议与咨询留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

时间限定

文献类型

馆藏选择

核心期刊

语言

文献类型

帮助

文字说明：

检索规则说明：

检索范例：

分类表

所选分类

限定检索结果

文献类型

馆藏范围

日期分布

学科分类号

主题

机构

作者

语言

在线全文

在线全文

在线全文

在线全文

在线全文

请选择保存的检索档案：

请选择收藏分类：

通借通还

欢迎您,

建议与咨询 留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

时间限定

文献类型

馆藏选择

核心期刊

语言

文献类型

帮助

文字说明：

检索规则说明：

检索范例：

分类表

所选分类

限定检索结果

文献类型

馆藏范围

日期分布

学科分类号

主题

机构

作者

语言

在线全文

在线全文

在线全文

在线全文

在线全文

请选择保存的检索档案： 新增检索档案 确定 取消

请选择收藏分类： 新增自定义分类 确定 取消

通借通还

建议与咨询留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

请选择保存的检索档案：

请选择收藏分类：