Determining the propagation path of a disturbance in multi-rate process and electromechanical systems

Abstract This paper proposes a multi-rate method to identify the propagation path of a persistent disturbance in an enlarged system envelope which includes the process plant and its electromechanical equipment. The need to integrate process and equipment diagnosis has been highlighted by industrial commentators. However, process and electromechanical measurements often have different sampling rates. The multi-rate method proposed extends a state-of-the-art propagation path method so that it combines fast-sampled electromechanical measurements and slow-sampled process measurements. The method is based on non-linear mutual prediction, which yields the directionality in the relationship between two time series. The method was demonstrated and validated, giving the expected outcome in an experimental case study, in which the root cause and propagation path of the disturbance were known.

[1]  J. Martinerie,et al.  Nonlinear interdependencies of EEG signals in human intracranially recorded temporal lobe seizures , 1998, Brain Research.

[2]  Biao Huang,et al.  Information transfer methods in causality analysis of process variables with an industrial application , 2013 .

[3]  R. Burke,et al.  Detecting dynamical interdependence and generalized synchrony through mutual prediction in a neural ensemble. , 1996, Physical review. E, Statistical physics, plasmas, fluids, and related interdisciplinary topics.

[4]  Charlotta Johnsson,et al.  A General Method for Handling Disturbances on Utilities in the Process Industry , 2011 .

[5]  Nina F. Thornhill,et al.  Advances and new directions in plant-wide disturbance detection and diagnosis , 2007 .

[6]  Nina F. Thornhill,et al.  Finding the source of nonlinearity in a process with plant-wide oscillation , 2005, IEEE Transactions on Control Systems Technology.

[7]  Ines M. Cecilio,et al.  Nearest neighbors method for detecting transient disturbances in process and electromechanical systems , 2014 .

[8]  H. Kantz,et al.  Nonlinear time series analysis , 1997 .

[9]  Nina F. Thornhill,et al.  Nearest neighbors methods for root cause analysis of plantwide disturbances , 2007 .

[10]  Fan Yang,et al.  Progress in root cause and fault propagation analysis of large-scale industrial processes , 2012 .

[11]  Robert Haber,et al.  Source identification of plant-wide faults based on k nearest neighbor time delay estimation , 2012 .

[12]  Peter Vas,et al.  Sensorless vector and direct torque control , 1998 .

[13]  Ines M. Cecilio,et al.  Removal of transient disturbances from oscillating measurements using nearest neighbors imputation , 2016 .

[14]  A. Horch,et al.  Operational profitability Peak performance Root cause analysis of plant-wide disturbances , 2007 .

[15]  Ines M. Cecilio,et al.  Importance of auxiliary systems for process fault detection and diagnosis , 2011, 2011 19th Mediterranean Conference on Control & Automation (MED).

[16]  Ines M. Cecilio,et al.  Multivariate Detection of Transient Disturbances for Uni- and Multirate Systems , 2015, IEEE Transactions on Control Systems Technology.

[17]  Ute Feldmann,et al.  Predictability Improvement as an Asymmetrical Measure of Interdependence in bivariate Time Series , 2004, Int. J. Bifurc. Chaos.

[18]  Nina F. Thornhill,et al.  Finding the Direction of Disturbance Propagation in a Chemical Process Using Transfer Entropy , 2007, IEEE Transactions on Control Systems Technology.

[19]  Joachim Holtz,et al.  Sensorless control of induction motor drives , 2002, Proc. IEEE.

[20]  Todd Reeves.  Optimising process equipment performance , 2005 .