Faults Forecasting System

This paper presents Faults Forecasting System (FFS) that utilizes statistical forecasting techniques in analyzing process variables data in order to forecast faults occurrences. FFS is proposing new idea in detecting faults. Current techniques used in faults detection are based on analyzing the current status of the system variables in order to check if the current status is fault or not. FFS is using forecasting techniques to predict future timing for faults before it happens. Proposed model is applying subset modeling strategy and Bayesian approach in order to decrease dimensionality of the process variables and improve faults forecasting accuracy. A practical experiment, designed and implemented in Okayama University, Japan, is implemented, and the comparison shows that our proposed model is showing high forecasting accuracy and BEFORE-TIME. Keywords—Bayesian Techniques, Faults Detection, Forecasting techniques, Multivariate Analysis.

[1]  Michael Lawrence,et al.  The effects of structural characteristics of explanations on use of a DSS , 2006, Decis. Support Syst..

[2]  Hsiao-Cheng Yu,et al.  A Litterman BVAR approach for production forecasting of technology industries , 2003 .

[3]  Iain Pardoe,et al.  Sentencing Convicted Felons in the United States: A Bayesian Analysis Using Multilevel Covariates , 2006 .

[4]  C. Granger.  Investigating causal relations by econometric models and cross-spectral methods , 1969 .

[5]  Shengwei Wang,et al.  Multiple ARMAX modeling scheme for forecasting air conditioning system performance , 2007 .

[6]  Raghunathan Rengaswamy,et al.  A fast training neural network and its updation for incipient fault detection and diagnosis , 2000 .

[7]  Volkan Ş. Ediger,et al.  Forecasting production of fossil fuel sources in Turkey using a comparative regression and ARIMA model , 2006 .

[8]  Juergen Hahn,et al.  Fault detection and classification in chemical processes based on neural networks with feature extraction. , 2003, ISA transactions.

[9]  Ronald Bewley.  Forecast accuracy, coefficient bias and Bayesian vector autoregressions , 2002, Math. Comput. Simul..

[10]  An-Sing Chen,et al.  A Bayesian vector error correction model for forecasting exchange rates , 2003, Comput. Oper. Res..

[11]  Albert Jäger.  A note on the informational efficiency of Austrian economic forecasts , 1985 .

[12]  Christophe Croux,et al.  Multivariate Out-of-Sample Tests for Granger Causality , 2006, Comput. Stat. Data Anal..

[13]  V. Haggan,et al.  ON THE SELECTION OF SUBSET AUTOREGRESSIVE TIME SERIES MODELS , 1984 .

[14]  Mattias Villani,et al.  Bayesian prediction with cointegrated vector autoregressions , 2001 .

[15]  E. Martin,et al.  Gaussian process regression for multivariate spectroscopic calibration , 2007 .

[16]  Jin Hyun Park,et al.  Fault detection and identification of nonlinear processes based on kernel PCA , 2005 .

[17]  C. B. Tilanus,et al.  Applied Economic Forecasting , 1966 .

[18]  C. Sims.  MACROECONOMICS AND REALITY , 1977 .

[19]  Brian D. O. Anderson,et al.  A Parametrization for the closed-loop identification of nonlinear time-varying systems , 1996, Autom..

[20]  V. Rao Vemuri,et al.  Artificial Neural Networks - Forecasting Time Series , 1993 .

[21]  Leo H. Chiang,et al.  Process monitoring using causal map and multivariate statistics: fault detection and identification , 2003 .

[22]  C. Granger.  Investigating Causal Relations by Econometric Models and Cross-Spectral Methods , 1969 .

[23]  J. Stock,et al.  A Comparison of Direct and Iterated Multistep Ar Methods for Forecasting Macroeconomic Time Series , 2005 .