Statistical fault detection techniques are able to detect and diagnose root-cause(s) from the monitored process variables. For complex operations, it is not feasible screen all variables due monitoring cost flooding of alarms. Thus, if a originated variable that monitored, conventional statistical incapable locating true root-cause. To relax this limitation, two-stage diagnosis technique proposed for operations. In first-stage, modified independent component analysis used identify faulty...

Pearson's correlation measure is only able to model linear dependence between random variables. Hence, conventional principal component analysis (PCA) based on not suitable for application modern industrial processes where process variables are often nonlinearly related. To address this problem, a nonparametric PCA proposed nonlinear measures, including Spearman's and Kendall tau's rank correlation. These two measures also less sensitive outliers comparing correlation, making the robust...

Safety analysis of rare events with potentially catastrophic consequences is challenged by data scarcity and uncertainty. Traditional causation-based approaches, such as fault tree event (used to model event), suffer from a number weaknesses. These include the static structure causation, lack occurrence data, need for reliable prior information. In this study, new hierarchical Bayesian modeling based technique proposed overcome these drawbacks. The can be used flexible risk major accidents....

A self-organizing map (SOM) based methodology is proposed for fault detection and diagnosis of processes with nonlinear non-Gaussian features. The SOM trained to represent the characteristics a normal operation as cluster in two-dimensional space. dynamic behavior process system then mapped trajectory on SOM. dissimilarity index deviation from center derived classify operating condition system. Furthermore, coordinate each best matching neuron used compute loading variable. For diagnosis,...

Scale-invariant principal component analysis (PCA) is prevalent in process monitoring because of its simplicity and efficiency. However, a number limitations are associated with this technique underlying assumptions. This article attempts to relax these by introducing three key elements. First, semiparametric Gaussian transformation proposed make the data follow multivariate distribution, such that standard PCA can be directly applied explain majority variance. The function preserves both...

Gd2O3/PEEK (poly ether ketone) composites were prepared on a twin-screw extruder by the incorporation of Gd2O3 as shield against X-ray to PEEK matrix. The influence addition and surface treatment particles with sulfonated (SPEEK) morphology, thermal mechanical properties was investigated SEM, DSC, TGA tensile tests respectively. DSC results showed that both crystallization temperature (Tc) melting (Tm) decreased compared pure at random filler content, which suggested hindered process...

The intramolecular synergistic effects of two dendritic antioxidants between hindered phenol groups and tertiary amine were investigated using the DPPH˙ method oxygen uptake method.

Masked Image Modeling (MIM) has garnered significant attention in self-supervised learning, thanks to its impressive capacity learn scalable visual representations tailored for downstream tasks. However, images inherently contain abundant redundant information, leading the pixel-based MIM reconstruction process focus excessively on finer details such as textures, thus prolonging training times unnecessarily. Addressing this challenge requires a shift towards compact representation of...

Abstract As military reforms continue to develop, the battlefield environment is becoming increasingly complex, and traditional single-service combat methods have evolved into integrated joint collaborative information operations that break down service boundaries on land, sea, air. The level of weapon system confrontation has also a system-to-system confrontation. Traditional document-based architecture design can no longer address complexity emergent challenges construction. In this paper,...

Abstract The conventional dynamic risk assessment technique does not consider the effect of nonlinear interactions among process variables in its operational estimation. Thus, this type fails to provide a realistic estimation complex industrial processes. To address issue, multivariate risk‐based monitoring is proposed. This takes advantage powerful dimensionality reduction and visualization power self‐organizing map identify origin propagation path fault. Through integration with inverted...

A series of poly(aryl ether ketone) polymers (m-PAEK-CN) containing phthalonitrile were synthesized by a direct solution polycondensation and characterized Fourier-transform infrared spectroscopy hydrogen nuclear magnetic resonance. Thermal crosslinking m-PAEK-CN, catalyzed p-BAPS, was then performed via heating their films up to 350oC. Dynamic rheology results showed that the rate diamine-catalyzed crosslink reaction could be easily controlled varying content cyano groups in polymer. The...

Abstract In this article, we intend to investigate the performance of channel access protocols in multi-hop underwater acoustic sensor networks, which are characterized by long propagation delays and limited bandwidth. An analytical model specifically designed for contention-based networks is identified validated. The based on an network model, called string topology provides a method computing expected throughput probability packets’ delivery gateway from arbitrary sensor. This study...

The emitting distributions of a hidden Markov model (HMM) are normally constructed using the cross moments process variables. Similar to mean univariate probability distribution, moment is most fundamental statistic multivariate which not capable capturing high-order statistical features data. To alleviate this limitation, equivalence demonstrated in paper, as complete dependence structure, used construct distribution for HMM. structure among variables modeled Gaussian copula. A...

ABSTRACT We prepared mixed‐matrix membranes (MMMs) composed of carboxylated single‐walled carbon nanotubes (f‐SWCNTs) and a sulfonated biphenyl poly(ether sulfone) (S‐PPSU) polymer matrix. The thermal stability properties the pores S‐PPSU f‐SWCNTs were characterized by thermogravimetric analysis sorption isotherm curves, respectively; these showed that surface pore diameter decreased after introduction carboxyl groups to (SWCNTs), did not restore original values even when preheated 350 °C...