Plant-wide oscillation detection using multivariate empirical mode decomposition

Abstract

Plant-wide oscillation detection is an important task in the maintenance of large-scale industrial control systems, owing to the fact that in an interactive multi-loop environment oscillation generated in one loop may propagate to the different parts of the plant. In such a scenario, its is required that different loops oscillating due to a common cause and hence similar frequency may be grouped together. In this paper an adaptive method for plant-wide oscillation detection based on multivariate empirical mode decomposition (MEMD) along with a grouping algorithm is proposed. The method can identify multiple oscillation groups among different variables as well as variables with random noise only. The proposed method is also applicable to both non-linear and non-stationary time series where the techniques based on the conventional Fourier analysis are prone to errors. Within each group that oscillate due to a common cause, the method can also indicate the location of the probable root cause of oscillations. The efficacy of the proposed method is established with the help of both simulation and industrial case studies.

Detecting and Analyzing Nonlinearity-Caused Oscillations in Process Control Systems Using an Improved VNCMD

“…This section considers a more complex case study, where time trends from an Australian refinery separation unit are tested [45]. The simplified structure diagram is shown in Figure 18.…”

Detecting and Characterizing Multiple Oscillations in Process Control Loop Based on PSO-NCMD

“…Having detected the oscillations, the next task is to characterize the oscillations' properties, such as the frequency information and oscillation type. In regard of oscillation frequency, previous studies [41], [42] only consider a few zero-crossing internals, which do not fully mining the information. Although [40] uses the full frequency information to reveal the oscillation properties, its oscillation frequency is calculated as the mean value of mode's instantaneous frequency.…”

Improving Root Cause Analysis by Detecting and Removing Transient Changes in Oscillatory Time Series with Application to a 1,3-Butadiene Process

“…28 This approach inherently removes the non-constant mean (low-frequency oscillation mode) from the signal. Multivariate empirical mode decomposition (MEMD) 29 extends the EMD method to plant wide oscillation detection and benefits from the same advantage i.e. the ability to deal with the slowly varying trends.…”

“…On the other hand, transient changes like spikes and steps affecting the oscillating time series are not discussed in Srinivasan and Rengaswamy 28 , Li et al 30 and Aftab et al 29 . The particularity of these transients is a clear localization in the time domain but no precise localization the frequency domain.…”