Pattern recognition in multivariate time series – A case study applied to fault detection in a gas turbine

Fontes, Cristiano Hora; Pereira, Otacílio José

doi:10.1016/j.engappai.2015.11.005

Cited by 54 publications

(40 citation statements)

References 47 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…A sintering process is often used to sinter the blast‐furnace materials and consists of three main parts: blending, burning and breaking. The burning process is mainly considered in , in which there are many signals that can be collected, for example, temperature, pressure, and so on. In this paper, temperatures T 20 , T 21 , T 22 of three west wind boxes and the ignition temperature u are chosen as output and input of the burning process respectively and the size of the data set is 6,000.…”

Section: Resultsmentioning

confidence: 99%

“…Additionally, I/O orders identified are coincident with the given control system (21). Therefore, the proposed method of identifying model orders is feasible.…”

Section: Numerical Examples Of Siso Control Systemsmentioning

confidence: 93%

“…In order to verify the method, firstly, the time series of I/O data are obtained by the system model (21). The size of the I/O data set is 6,000.…”

Section: Numerical Examples Of Siso Control Systemsmentioning

confidence: 99%

See 2 more Smart Citations

Pattern Recognition Based Dynamics Description of Production Processes in Metric Spaces

Wang¹,

Xu²,

Guo³

2017

Asian Journal of Control

View full text Add to dashboard Cite

A new type of modeling method is put forward based on pattern recognition (PR) technology for some industrial production processes. The proposed method is a pure data‐driven modeling method since the model is independent of the controlled plant, and it is based on the measured input and output (I/O) data of the controlled plant in a closed loop. Different from the traditional modeling method, the system dynamics is described by I/O classes, which are obtained from raw I/O data through partitioning of the data space respectively and I/O orders of the model resort to the conditional entropy. The covering algorithm based pattern classification (PC) is used to establish the mapping between input and output of the proposed model in metric spaces. The experimental results illustrate the feasibility of the modeling method.

show abstract

Section: Resultsmentioning

confidence: 99%

“…Additionally, I/O orders identified are coincident with the given control system (21). Therefore, the proposed method of identifying model orders is feasible.…”

Section: Numerical Examples Of Siso Control Systemsmentioning

confidence: 93%

See 1 more Smart Citation

Pattern Recognition Based Dynamics Description of Production Processes in Metric Spaces

Wang¹,

Xu²,

Guo³

2017

Asian Journal of Control

View full text Add to dashboard Cite

show abstract

“…Moreover, a major body of work [38,47] exists in subsequence matching based time series clustering where they identify shorter most identifying portions of time series data also known as shapelets to group them. For the multivariate time series data, same categorizations can be made as modeling based [17,14], and variants of generalizing univariate solutions to multivariate cases [36,13]. Our approach falls into the second category where we extend existing distance functions available for univariate time series data and update centroid finding (i.e.…”

Section: Related Workmentioning

confidence: 99%

Discovering patterns of online popularity from time series

Özer

Sapienza

Abeliuk

et al. 2020

Expert Systems with Applications

View full text Add to dashboard Cite

How is popularity gained online? Is being successful strictly related to rapidly becoming viral in an online platform or is it possible to acquire popularity in a steady and disciplined fashion? What are other temporal characteristics that can unveil popularity of online content? To answer these questions, we leverage a multi-faceted temporal analysis of the evolution of popular online contents. Here, we present dipm-SC: a multi-dimensional shape-based time-series clustering algorithm with a heuristic to find the optimal number of clusters. First, we validate the accuracy of our algorithm on synthetic datasets generated from benchmark time series models. Second, we show that dipm-SC can uncover meaningful clusters of popularity behaviors in a real-world Twitter dataset. By clustering the multidimensional time-series of popularity of contents coupled with other domain-specific dimensions, we uncover two main patterns of popularity: bursty and steady temporal behaviors. Moreover, we find that the way popularity is gained over time has no significant impact on the final cumulative popularity.

show abstract

“…Multivariate time series is an extension of the univariate model to the multivariate case with vector-valued data. 50 The vector autoregression (VAR) model is one of the most successful and easy to use models for the analysis of multivariate time series. The structure of VAR model is that each variable is a linear function of past lags of 250 A. Ozcan & S. G. Oguducu itself and past lags of the other variables that express linear dependencies among multiple time series variables.…”

Section: Univariate and Multivariate Time Seriesmentioning

confidence: 99%

Multivariate Time Series Link Prediction for Evolving Heterogeneous Network

Özcan

Öğüdücü

2019

Int. J. Info. Tech. Dec. Mak.

View full text Add to dashboard Cite

Link prediction is considered as one of the key tasks in various data mining applications for recommendation systems, bioinformatics, security and worldwide web. The majority of previous works in link prediction mainly focus on the homogeneous networks which only consider one type of node and link. However, real-world networks have heterogeneous interactions and complicated dynamic structure, which make link prediction a more challenging task. In this paper, we have studied the problem of link prediction in the dynamic, undirected, weighted/ unweighted, heterogeneous social networks which are composed of multiple types of nodes and links that change over time. We propose a novel method, called Multivariate Time Series Link Prediction for evolving heterogeneous networks that incorporate (1) temporal evolution of the network; (2) correlations between link evolution and multi-typed relationships; (3) local and global similarity measures; and (4) node connectivity information. Our proposed method and the previously proposed time series methods are evaluated experimentally on a real-world bibliographic network (DBLP) and a social bookmarking network (Delicious). Experimental results show that the proposed method outperforms the previous methods in terms of AUC measures in di®erent test cases.

show abstract

Pattern recognition in multivariate time series – A case study applied to fault detection in a gas turbine

Cited by 54 publications

References 47 publications

Pattern Recognition Based Dynamics Description of Production Processes in Metric Spaces

Pattern Recognition Based Dynamics Description of Production Processes in Metric Spaces

Discovering patterns of online popularity from time series

Multivariate Time Series Link Prediction for Evolving Heterogeneous Network

Contact Info

Product

Resources

About