Use of Vector Autoregressive Model for Anomaly Detection in Utility Gas Turbines

2019 ◽  
Author(s):  
Vipul Goyal ◽  
Mengyu Xu ◽  
Jayanta Kapat
Keyword(s):  
Time Series ◽  
Anomaly Detection ◽  
Autoregressive Model ◽  
Gas Turbines ◽  
Time Series Data ◽  
Combined Cycle ◽  
Series Data ◽  
Vector Autoregressive Model ◽  
Vector Autoregressive ◽  
Multi Stage

Abstract This study is based on time-series data from the combined cycle utility gas turbines consisting of three-gas turbine units and one steam turbine unit. We construct a multi-stage vector autoregressive model for the nominal operation of powerplant assuming sparsity in the association among variables and use this as a basis for anomaly detection and prediction. This prediction is compared with the time-series data of the plant-operation containing anomalies. Granger causality networks, which are based on the associations between the time series streams, are learned as an important implication from the vector autoregressive modelling. Anomaly is detected by comparing the observed measurements against their predicted value.

scholarly journals Soft Computation Vector Autoregressive Neural Network (VAR-NN) GUI-Based

2018 ◽  
Vol 73 ◽  
pp. 13008 ◽  
Author(s):  
Hasbi Yasin ◽  
Budi Warsito ◽  
Rukun Santoso ◽  
Suparti
Keyword(s):  
Neural Network ◽  
Time Series ◽  
Autoregressive Model ◽  
Stock Price ◽  
Time Series Data ◽  
Multivariate Time Series ◽  
Series Data ◽  
Percentage Error ◽  
Vector Autoregressive ◽  
Price Data

Vector autoregressive model proposed for multivariate time series data. Neural Network, including Feed Forward Neural Network (FFNN), is the powerful tool for the nonlinear model. In autoregressive model, the input layer is the past values of the same series up to certain lag and the output layers is the current value. So, VAR-NN is proposed to predict the multivariate time series data using nonlinear approach. The optimal lag time in VAR are used as aid of selecting the input in VAR-NN. In this study we develop the soft computation tools of VAR-NN based on Graphical User Interface. In each number of neurons in hidden layer, the looping process is performed several times in order to get the best result. The best one is chosen by the least of Mean Absolute Percentage Error (MAPE) criteria. In this study, the model is applied in the two series of stock price data from Indonesia Stock Exchange. Evaluation of VAR-NN performance was based on train-validation and test-validation sample approach. Based on the empirical stock price data it can be concluded that VAR-NN yields perfect performance both in in-sample and in out-sample for non-linear function approximation. This is indicated by the MAPE value that is less than 1% .

scholarly journals Asymmetric Effects of the Effect of Oil Price on Stock Markets in Four Asian Countries: Markov Switching Analysis

2018 ◽  
pp. 423-428
Keyword(s):  
Time Series ◽  
Structural Change ◽  
Business Cycle ◽  
Time Series Data ◽  
Markov Switching ◽  
Oil Price ◽  
Series Data ◽  
Vector Autoregressive Model ◽  
Vector Autoregressive ◽  
The Business Cycle

The Markov switching vector autoregressive model is a dynamic stochastic system with stochastic autoregressive parameters. This model able to measure a time varying problem when the variables undergoing regime switching. Structural change or shock is an ordinary fact in time series data. Some shocks have an important role under specific regimes in examining the business cycle contraction. Excluding changes in regime for the measurement of variance decomposition may produce biased results. Moreover, the parameters in the time series model might also have a structural change. Therefore, linear models are no longer suitable to be used in analyzing the financial model; and nonlinear time series models that are Markov switching models are proposed to solve these kinds of problems. A two regimes Markov switching vector autoregressive model is used in this study to analysis the time series data. The regime is dependent heterogeneous with varying the variance to detect every change of the business cycle. The correlations between oil price, Malaysia, Singapore, Thailand and Indonesia stock price are examining using Markov switching model. The result shows that the regimes dependent models suitable to employ in study the asymmetric business cycle; and oil price have a negative relationship with the changes of the four selected Asian stock markets.

scholarly journals Dynamic Modelling and Forecasting of Data Export of Agricultural Commodity by Vector Autoregressive Model

2020 ◽  
Vol 55 (3) ◽  
Author(s):  
L.M. Hamzah ◽  
S.U. Nabilah ◽  
E. Russel ◽  
M. Usman ◽  
E. Virginia ◽  
...  
Keyword(s):  
Time Series ◽  
Statistical Models ◽  
Time Series Data ◽  
Information Criterion ◽  
Series Data ◽  
Var Model ◽  
Vector Autoregressive Model ◽  
Vector Autoregressive ◽  
Coffee Beans ◽  
The Relationship

The Vector Autoregressive Model (VAR) is one of the statistical models that can be used for modeling multivariate time series data. It is commonly used in finance, management, business and economics. The VAR model analyzes the time series data simultaneously to arrive at the right conclusions while dynamically explaining the behavior of the relationship between endogenous variables, as well as endogenous and exogenous variables. From time to time, the VAR model is influenced by its own factors via Granger Causality. In this study, we will discuss and determine the best model to describe the relationship among data export value of Indonesia's agricultural commodities—coffee beans, cacao beans and tobacco—where the monthly data spans the years 2007-2018. Several models are applied to the data, such as VAR (1), VAR (2), VAR (3), VAR (4) and VAR (5) models. As a result, the VAR (2) model was chosen as the best model based on the Akaike’s Information Criterion with Correction, Schwarz Bayesian Criterion, Akaike’s Information Criterion and Hanna-Quinn Information Criterion for selecting statistical models. The dynamic behavior of the three export variables of Indonesian coffee beans, cacao beans and tobacco is explained by Granger Causality. Furthermore, the best model VAR (2) is used to forecast the next 10 months.

The search for rebel interdependence

2017 ◽  
Vol 54 (1) ◽  
pp. 16-30 ◽  
Author(s):  
Ezra Schricker
Keyword(s):  
Time Series ◽  
Time Series Data ◽  
Series Data ◽  
Entire System ◽  
Triangular System ◽  
Vector Autoregressive Model ◽  
Vector Autoregressive ◽  
Rebel Groups ◽  
Qualitative Knowledge ◽  
The Relationship

The existing conflict literature tends to treat interdependence between rebel groups as a binary category: either groups are allied or unallied, fragmented or unified, interdependent or independent. Yet much of our qualitative knowledge suggests that interdependence is better understood as a matter of degree where certain groups exert a disproportionate influence over their counterparts. The challenge is how to identify the degree of interdependence in practice. As a solution, I conceptualize interdependence as a property of a system of interactions between rebel groups and government forces within and across borders. My approach is to model the entire system of interactions in order to test hypotheses related to the directionality of influence and the potential for military coordination between groups. I demonstrate the utility of this approach by examining the relationship between Pakistan and the two major factions which make up the Taliban organization – the Afghan and Pakistani Taliban. I analyze the triangular system with a vector autoregressive model and monthly time series data on violent actions initiated by each group from January 2008 to February 2013. The substantive findings support much of the received wisdom concerning Pakistan’s disparate relationship to both groups, which is characterized by antagonism with the Pakistani Taliban and collusion with the Afghan Taliban. The results also suggest that the claims of interdependence between the two Taliban groups have been overstated.

An Anomaly Detection Method with Exemplar Subsequence for Time Series Data

2016 ◽  
Vol 136 (3) ◽  
pp. 363-372
Author(s):  
Takaaki Nakamura ◽  
Makoto Imamura ◽  
Masashi Tatedoko ◽  
Norio Hirai
Keyword(s):  
Time Series ◽  
Anomaly Detection ◽  
Time Series Data ◽  
Detection Method ◽  
Series Data

scholarly journals Change Point Enhanced Anomaly Detection for IoT Time Series Data

Water ◽  
2021 ◽  
Vol 13 (12) ◽  
pp. 1633
Author(s):  
Elena-Simona Apostol ◽  
Ciprian-Octavian Truică ◽  
Florin Pop ◽  
Christian Esposito
Keyword(s):  
Time Series ◽  
Anomaly Detection ◽  
Change Point ◽  
Time Series Data ◽  
Multivariate Time Series ◽  
Change Point Detection ◽  
Change Points ◽  
Series Data ◽  
Prediction And Forecasting ◽  
Point Detection

Due to the exponential growth of the Internet of Things networks and the massive amount of time series data collected from these networks, it is essential to apply efficient methods for Big Data analysis in order to extract meaningful information and statistics. Anomaly detection is an important part of time series analysis, improving the quality of further analysis, such as prediction and forecasting. Thus, detecting sudden change points with normal behavior and using them to discriminate between abnormal behavior, i.e., outliers, is a crucial step used to minimize the false positive rate and to build accurate machine learning models for prediction and forecasting. In this paper, we propose a rule-based decision system that enhances anomaly detection in multivariate time series using change point detection. Our architecture uses a pipeline that automatically manages to detect real anomalies and remove the false positives introduced by change points. We employ both traditional and deep learning unsupervised algorithms, in total, five anomaly detection and five change point detection algorithms. Additionally, we propose a new confidence metric based on the support for a time series point to be an anomaly and the support for the same point to be a change point. In our experiments, we use a large real-world dataset containing multivariate time series about water consumption collected from smart meters. As an evaluation metric, we use Mean Absolute Error (MAE). The low MAE values show that the algorithms accurately determine anomalies and change points. The experimental results strengthen our assumption that anomaly detection can be improved by determining and removing change points as well as validates the correctness of our proposed rules in real-world scenarios. Furthermore, the proposed rule-based decision support systems enable users to make informed decisions regarding the status of the water distribution network and perform effectively predictive and proactive maintenance.

scholarly journals Anomaly Detection in COVID-19 Time-Series Data

2021 ◽  
Vol 2 (4) ◽  
Author(s):  
Hajar Homayouni ◽  
Indrakshi Ray ◽  
Sudipto Ghosh ◽  
Shlok Gondalia ◽  
Michael G. Kahn
Keyword(s):  
Time Series ◽  
Anomaly Detection ◽  
Time Series Data ◽  
Series Data

scholarly journals Deep Learning for Anomaly Detection in Time-Series Data: Review, Analysis, and Guidelines

IEEE Access ◽  
2021 ◽  
Vol 9 ◽  
pp. 120043-120065
Author(s):  
Kukjin Choi ◽  
Jihun Yi ◽  
Changhwa Park ◽  
Sungroh Yoon
Keyword(s):  
Time Series ◽  
Deep Learning ◽  
Anomaly Detection ◽  
Time Series Data ◽  
Series Data ◽  
Review Analysis
Sign in / Sign up

Export Citation Format

Share Document