Browsing by Author "Krishnannair, Syamala"

Now showing 1 - 2 of 2
  • Thumbnail Image
    Item
    Multiscale process monitoring with singular spectrum analysis
    (Stellenbosch : University of Stellenbosch, 2010-12) Krishnannair, Syamala; Aldrich, C.; University of Stellenbosch. Faculty of Engineering. Dept. of Process Engineering.
    ENGLISH ABSTRACT: Multivariate statistical process control (MSPC) approaches are now widely used for performance monitoring, fault detection and diagnosis in chemical processes. Conventional MSPC approaches are based on latent variable projection methods such as principal component analysis and partial least squares. These methods are suitable for handling linearly correlated data sets, with minimal autocorrelation in the variables. Industrial plant data invariably violate these conditions, and several extensions to conventional MSPC methodologies have been proposed to account for these limitations. In practical situations process data usually contain contributions at multiple scales because of different events occurring at different localizations in time and frequency. To account for such multiscale nature, monitoring techniques that decompose observed data at different scales are necessary. Hence the use of standard MSPC methodologies may lead to unreliable results due to false alarms and significant loss of information. In this thesis a multiscale methodology based on the use of singular spectrum analysis is proposed. Singular spectrum analysis (SSA) is a linear method that extracts information from the short and noisy time series by decomposing the data into deterministic and stochastic components without prior knowledge of the dynamics affecting the time series. These components can be classified as independent additive time series of slowly varying trend, periodic series and aperiodic noise. SSA does this decomposition by projecting the original time series onto a data-adaptive vector basis obtained from the series itself based on principal component analysis (PCA). The proposed method in this study treats each process variable as time series and the autocorrelation between the variables are explicitly accounted for. The data-adaptive nature of SSA makes the proposed method more flexible than other spectral techniques using fixed basis functions. Application of the proposed technique is demonstrated using simulated, industrial data and the Tennessee Eastman Challenge process. Also, a comparative analysis is given using the simulated and Tennessee Eastman process. It is found that in most cases the proposed method is superior in detecting process changes and faults of different magnitude accurately compared to classical statistical process control (SPC) based on latent variable methods as well as the wavelet-based multiscale SPC.
  • Thumbnail Image
    Item
    Nonlinear singular spectrum analysis and its application in multivariate statistical process monitoring
    (Stellenbosch : Stellenbosch University, 2016-03) Krishnannair, Syamala; Aldrich, C.; Bradshaw, S. M.; Stellenbosch University. Faculty of Engineering. Dept. of Process Engineering.
    ENGLISH ABSTRACT: Multivariate statistical process control (MSPC) approaches based on principal component analysis (PCA), partial least squares (PLS) and related extensions are now widely used for process monitoring and diagnosis in process systems where observed correlated measurements are readily available. However, highly nonlinear (dynamic) processes pose a challenge for MSPC methods as a large set of nonlinear features are typically required to capture the underlying characteristic behaviour of the process in the absence of faults. Several extensions of basic (PCA) methods have previously been proposed to handle features such as autocorrelation in data, time-frequency localization, and nonlinearity. In this study multivariate statistical process monitoring methods based on nonlinear singular spectrum analysis which use nonlinear principal component analysis, multidimensional scaling and kernel multidimensional scaling are proposed. More specifically, singular spectrum analysis using covariance and dissimilarity scale structure are proposed to express multivariate time series as the sum of identifiable components whose basis functions are obtained from the process measurements. Such an approach is useful for extracting trends, harmonic patterns and noise in time series data. Using nonlinear SSA decomposition of time series data, a multimodal representation is obtained that can be used together with existing statistical process control methods to develop novel process monitoring schemes. The advantages of these approaches are demonstrated on simulated multivariate nonlinear data and compared with those of classical PCA and multimodal SSA on base metal flotation plant data and the Tennessee Eastman process benchmark data. The nonlinear SSA methods better captured the nonlinearities in the observed data. Consequently, this yielded improved detection rates for various faults in nonlinear data over those obtainable by alternative competing multivariate methods.