Fault Detection, Supervision, and Safety for Chemical Processes: 2020

This paper presents a new method of actuator fault detection based on the functional observer approach for a two‐stage chemical reactor with disturbance, parameter uncertainty, and time‐delay. The mechanism analysis method is used to model the system and the observer is designed. The exogenous disturbance can be decoupled completely from the error dynamic systems by letting some relevant constraints ensure the robustness of the residual generator. Stability conditions are obtained by using the Lyapunov–Krasovskii functional approach in the form of linear matrix inequalities (LMIs). Finally, simulation results are given to illustrate the effectiveness of the proposed method.

Section: Introductionmentioning

confidence: 99%

Actuator fault detection for a two‐stage chemical reactor based on the functional observer approach

Mao

2021

“…The proposed methods have been tested on an industrial E. coli fermentation process, oil‐drilling process, penicillin fermentation process, three‐tank system, large blast furnace, and lithium‐ion battery, respectively. These articles cover a range of topics of interest to the control community at large and to the readers of this journal in particular …”

mentioning

confidence: 99%

Issue Highlights

2020

The purpose of our study was to develop 3D tumour spheroids of human colon cancer cells and to evaluate the therapeutic potential of NP1, a novel cell penetrating peptide (CPP) developed by our group, to assist in siRNA delivery. NP1 elicited significant cellular uptake of siRNA and promoted great siRNA knockdown efficiency of Bcl-2 and VEGF mRNA in 3D spheroids, induced marked apoptosis after silencing HIF mRNA, and 3D spheroids displayed apoptosis resistance compared to 2D cells. Therefore, 3D spheroids provide an improved model for testing siRNA delivery and NP1 has proven to be a powerful in vitro transfection reagent. [1] X-ray diffraction is the most frequent analytical technique that chemical engineers cite as a keyword. The major research clusters include nanoparticles, composites, adsorption phenomena, catalysis, and, in particular, photocatalysis. It measures lattice constants, average grain size, crystallinity, and phase composition. When radiation impinges upon a solid, coherent scattering of the radiation by periodically spaced atoms produces spot patterns from single crystal samples and ring patterns from polycrystalline samples. The intensities of the diffraction maxima (peaks or lines) and their position correlate to a specific crystal structure. Most analysis errors are due to sample type or instrument operation/preparation. [2] Fault Detection, Supervision, and Safety for Chemical Processes: 2020Youqing Wang, Donghua Zhou This special issue section includes ten articles that provide an up-to-date and comprehensive picture of safety methods and techniques for chemical processes. In particular, these articles focus on the following problems: fault diagnosis, remaining useful life prediction, and quality prediction. In terms of methodology, this special issue section involves deep learning, principal component analysis, the observer-based method, and the filter-based method. The proposed methods have been tested on an industrial E. coli fermentation process, oil-drilling process, penicillin fermentation process, threetank system, large blast furnace, and lithium-ion battery, respectively. These articles cover a range of topics of interest to the control community at large and to the readers of this journal in particular. [3] variable 0 5 10 15 20 25 WMK 1 2 3 4 5 WMK thresholdA process monitoring method based on correlation variable classification and vine copula is proposed. The weighted correlation measure is used to divide the variables into related subspace and weakly correlated subspace. C-vine and D-vine copulas are used to model the joint probability distributions of the two subspaces, respectively. Process faults are detected according to the two joint probability distributions. The monitoring results of a numerical system and the Tennessee Eastman (TE) process have been used to demonstrate the effectiveness and validity of the proposed method. [4] REFERENCES[1] K.

Reconstruction error‐based fault detection of time series process data using generative adversarial auto‐encoders

Rani,

Goswami,

Kodamana

et al. 2023

Faults in time series process data are typically difficult to detect due to the complex temporal correlations of data samples. In this context, traditional unsupervised machine learning algorithms, such as principal component analysis, independent component analysis, and so forth, would yield only limited performance. Deep learning‐based methods have been employed in recent years to address these problems. Recently, generative adversarial networks have emerged as a promising generative modelling approach for learning data distributions. Inspired by the above, in this study, we present a novel reconstruction error‐based fault detection method for time series process data using generative adversarial auto‐encoder (GAAE) with Wasserstein loss, cycle consistency loss, and gradient penalty methods. The proposed method is designed to detect abnormal patterns in the time series data by training a GAAE to learn the underlying normal behaviour of the data. GAAEs help in effectively capturing the data's hidden distribution, and Wasserstein loss with gradient penalty is used to improve the accuracy of the latent space representation of the data, while the cycle consistency loss ensures consistency between the input and output data during the reconstruction process. Based on the extent of the reconstruction error metric of the GAAEs, we identify the potential faults in the process data stream. The proposed method is evaluated on two data process sets, namely, the Tennessee Eastman benchmark process dataset and the nuclear power flux real‐time dataset from a pressurized heavy water nuclear reactor, to validate the efficacy of the proposed approach.