Detection of gross erros in data reconciliation by principal component analysis

Tong, Hongwei; Crowe, C. M.

doi:10.1002/aic.690410711

Cited by 160 publications

(92 citation statements)

References 24 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…For each life stage, PCA was performed on the whole spectral reflectance dataset that included all fish from all light treatments, and the loadings of the various principal components (PCs) were obtained. Thereafter, the number of substantially contributing PCs was determined using a threshold criterion that compared the standard deviation of all PCs with the standard deviation of the first PC (Tong and Crowe, 1995). The loadings for the substantially contributing PCs were divided according to their respective treatments, and were used to test for differences between treatments.…”

Section: Discussionmentioning

confidence: 99%

Modulation of environmental light alters reception and production of visual signals in Nile tilapia

Hornsby

Sabbah

Robertson

et al. 2013

Journal of Experimental Biology

View full text Add to dashboard Cite

SUMMARYSignal reception and production form the basis of animal visual communication, and are largely constrained by environmental light. However, the role of environmental light in producing variation in either signal reception or production has not been fully investigated. To chart the effect of environmental light on visual sensitivity and body colouration throughout ontogeny, we measured spectral sensitivity, lens transmission and body pattern reflectance from juvenile and adult Nile tilapia held under two environmental light treatments. Spectral sensitivity in juveniles reared under a broad-spectrum light treatment and a red-shifted light treatment differed mostly at short wavelengths, where the irradiance of the two light treatments differed the most. In contrast, adults held under the same two light treatments did not differ in spectral sensitivity. Lens transmission in both juveniles and adults did not differ significantly between environmental light treatments, indicating that differences in spectral sensitivity of juveniles originated in the retina. Juveniles and adults held under the two environmental light treatments differed in spectral reflectance, and adults transferred to a third, white light treatment differed in spectral reflectance from their counterparts held under the two original treatments. These results demonstrate that environmental light plays a crucial role in shaping signal reception in juveniles and signal production throughout ontogeny, reinforcing the notion that environmental light has the capacity to influence animal communication, and suggesting that the characteristics of environmental light should be considered in models of ecological speciation.Key words: ontogenetic variation, visual sensitivity, body colouration, sensory drive, animal communication. (Kocher, 2004), and are a growing presence in laboratory research (El-Sayed, 2006). Cichlid fish show variation in signal reception, as visual pigment complements show a great range of natural variation (Carleton and Kocher, 2001;Parry et al., 2005;Sabbah et al., 2010;O'Quin et al., 2010). Additionally, cichlids are notorious for their variation in signal production, showing myriad behaviours (Barlow, 2000) and body colourations (Kornfield and Smith, 2000), many of which are species specific (Hofmann et al., 2009). When these sources of variation are coupled with changes in the spectral quality of available environmental light, which, for example, occur as water depth varies (e.g. Sabbah et al., 2011), the basic requirements for sensory drive are met. However, the role of environmental light in these sources of variation, particularly signal production, remains somewhat elusive.Nile tilapia have been shown to differentially express seven visual pigments throughout ontogeny: SWS1 (360nm), SWS2b (425nm), SWS2a (456nm), Rh2b (472nm), Rh2aβ (518nm), Rh2aα (528nm) and LWS (561nm) (Spady et al., 2006). Juveniles predominantly express SWS2b, SWS2a, Rh2a (a combination of Rh2aβ and Rh2aα, as their maximum absorbance wavelengt...

show abstract

Section: Discussionmentioning

confidence: 99%

Modulation of environmental light alters reception and production of visual signals in Nile tilapia

Hornsby

Sabbah

Robertson

et al. 2013

Journal of Experimental Biology

View full text Add to dashboard Cite

show abstract

“…2,[15][16][17][18][19] Both methods project data onto lowerdimensional subspaces and use the Hotelling's T 2 or squared predicted error (SPE) indices to isolate the normal and abnormal conditions. 20,21 Other complementary MSPM approaches, including Fisher discriminant analysis (FDA), canonical variate analysis (CVA), independent components analysis (ICA), have been used to overcome some limitations in PCA/PLS-based monitoring schemes. [22][23][24][25][26][27][28] Meanwhile, machine learning techniques, e.g., discriminant analysis (DA), neural network (NN), expert systems, support vector machines (SVM), Bayesian belief network (BBN), and mutual information, have been explored to address the complex process monitoring problems with some success.…”

Section: Introductionmentioning

confidence: 99%

Multimode process monitoring with Bayesian inference‐based finite Gaussian mixture models

2008

View full text Add to dashboard Cite

in Wiley InterScience (www.interscience.wiley.com).For complex industrial processes with multiple operating conditions, the traditional multivariate process monitoring techniques such as principal component analysis (PCA) and partial least squares (PLS) are ill-suited because the fundamental assumption that the operating data follow a unimodal Gaussian distribution usually becomes invalid. In this article, a novel multimode process monitoring approach based on finite Gaussian mixture model (FGMM) and Bayesian inference strategy is proposed. First, the process data are assumed to be from a number of different clusters, each of which corresponds to an operating mode and can be characterized by a Gaussian component. In the absence of a priori process knowledge, the Figueiredo-Jain (F-J) algorithm is then adopted to automatically optimize the number of Gaussian components and estimate their statistical distribution parameters. With the obtained FGMM, a Bayesian inference strategy is further utilized to compute the posterior probabilities of each monitored sample belonging to the multiple components and derive an integrated global probabilistic index for fault detection of multimode processes. The validity and effectiveness of the proposed monitoring approach are illustrated through three examples: (1) a simple multivariate linear system, (2) a simulated continuous stirred tank heater (CSTH) process, and (3) the Tennessee Eastman challenge problem. The comparison of monitoring results demonstrates that the proposed approach is superior to the conventional PCA method and can achieve accurate and early detection of various types of faults in multimode processes.

show abstract

“…Confounding statistics may be misleading and may result in throwing away good data while keeping corrupted ones. Tong and Crowe (1995) presented a method based on principal component analysis (PCA) that appropriately handles the correlation among the variables.…”

Section: Statistical Tests For Gross Errorsmentioning

confidence: 99%

Detecting persistent gross errors by sequential analysis of principal components

Tong

Crowe

1997

AIChE Journal

Self Cite

View full text Add to dashboard Cite

Measurements such as flow rates from a chemical process violate conservation laws and other process constraints because they are contaminated by random errors and possibly gross errors such as process disturbances, leaks, departures from steady state, and biased instrumentation. Data reconcilation is aimed at estimating the true values of measured variables that are consistent with the constraints, at detecting gross errors, and at solving for unmeasured variables. An approach to constructing sequential principal‐component tests for detecting and identifying persistent gross errors during data reconciliation by combining principal‐component analysis and sequential analysis is presented. The tests detect gross errors as early as possible with fewer measuremennts. They were sharper in detecting and have a substantially greater power in correctly identifying gross errors than the currently used statistical tests in data reconciliation.

show abstract

Detection of gross erros in data reconciliation by principal component analysis

Abstract: Statistical testing prouides a tool for engineers and operators to judge the validity

Cited by 160 publications

References 24 publications

Modulation of environmental light alters reception and production of visual signals in Nile tilapia

Modulation of environmental light alters reception and production of visual signals in Nile tilapia

Multimode process monitoring with Bayesian inference‐based finite Gaussian mixture models

Detecting persistent gross errors by sequential analysis of principal components

Contact Info

Product

Resources

About