Dynamic neural orthogonal mapping for fault detection

Hu, Zhengwei; Peng, Jingchao; Zhao, Haitao

doi:10.1007/s13042-020-01250-z

Cited by 10 publications

(5 citation statements)

References 25 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…In this paper, we choose the polynomial kernel with

d = 1

and the RBF kernel function with

t = 2

. [ 38 ] Furthermore, to illustrate different kernel functions at each kernel mapping layer to guarantee the diverse capabilities of feature extraction, meanwhile, to demonstrate the effectiveness of the proposed multilayer feature extraction for nonlinear processes, we illustrate it with a numerical example, [ 39 ] as shown in formula ():

\begin{array}{l} z_{2 i - 1} goodbreak= ([], \begin{array}{c} z_{2 i - 1}^{(1)} \\ z_{2 i - 1}^{(2)} \\ z_{2 i - 1}^{(3)} \end{array}) goodbreak= ([], \begin{array}{c} 1.5 a_{i} \\ 1.8 a_{i} \\ 1.3 a_{i} \end{array}) goodbreak+ ([], \begin{array}{c} 0.28 β_{i} \\ 0.385 β_{i} \\ 0.42 β_{i} \end{array}), \\ z_{2 i} goodbreak= ([], \begin{array}{c} z_{2 i}^{(1)} \\ z_{2 i}^{(2)} \\ z_{2 i}^{(3)} \end{array}) goodbreak= ([], \begin{array}{c} 1.57 & 2.37 & 1.8 \\ 2.73 & 1.05 & 1.4 \\ 1.22 & 1.60 & 2.4 \end{array}) z_{2 i - 1} goodbreak+ ([], \begin{array}{c} 0.35 β_{i} \\ 0.385 β_{i} \\ 0.42 β_{i} \end{array}), \end{array}

where

α_{i}

and

β_{i}

are the random variables from the standard normal distribution,

i = 1, 2, \dots, N

. According to for...…”

Section: The Proposed Siamese Denpe Methods (Sdenpe)mentioning

confidence: 99%

“…In this paper, we choose the polynomial kernel with d ¼ 1 and the RBF kernel function with t ¼ 2. [38] Furthermore, to illustrate different kernel functions at each kernel mapping layer to guarantee the diverse capabilities of feature extraction, meanwhile, to demonstrate the effectiveness of the proposed multilayer feature extraction for nonlinear processes, we illustrate it with a numerical example, [39] as shown in formula (20):…”

Section: The Construction Of Sdenpe Networkmentioning

confidence: 99%

See 1 more Smart Citation

SiameseDeNPEnetwork framework for fault detection of batch process

Liu,

Zhao,

Mou

et al. 2023

Can J Chem Eng

View full text Add to dashboard Cite

In batch processes, it is crucial to ensure safe production by fault detection. However, the long batch duration, limited runs, and strong nonlinearity of the data pose challenges. Incipient faults with small amplitudes further complicate the detection process. To achieve safe production, motivated by deep learning strategies, we propose a new fault detection method of batch process called Siamese deep neighbourhood preserving embedding network (SDeNPE). First, the DeNPE network is constructed by means of NPE and kernel functions, which utilizes the different types of kernel functions in the kernel mapping layer to extract diverse deep nonlinear features and overcome strong nonlinearity in the process data. Then, the Siamese network is used to obtain the different features between the data and improve the recognition of incipient faults. In addition, the deep extraction and Siamese network allow for batches of training data reduction without diminishing the performance of fault detection. Finally, we utilize monitoring statistics to complete the fault detection process. Two batch process cases involving the penicillin fermentation process and the semiconductor etching process demonstrate the superior fault detection performance of the proposed SDeNPE over the other comparison methods.

show abstract

“…In this paper, we choose the polynomial kernel with

d = 1

and the RBF kernel function with

t = 2

\begin{array}{l} z_{2 i - 1} goodbreak= ([], \begin{array}{c} z_{2 i - 1}^{(1)} \\ z_{2 i - 1}^{(2)} \\ z_{2 i - 1}^{(3)} \end{array}) goodbreak= ([], \begin{array}{c} 1.5 a_{i} \\ 1.8 a_{i} \\ 1.3 a_{i} \end{array}) goodbreak+ ([], \begin{array}{c} 0.28 β_{i} \\ 0.385 β_{i} \\ 0.42 β_{i} \end{array}), \\ z_{2 i} goodbreak= ([], \begin{array}{c} z_{2 i}^{(1)} \\ z_{2 i}^{(2)} \\ z_{2 i}^{(3)} \end{array}) goodbreak= ([], \begin{array}{c} 1.57 & 2.37 & 1.8 \\ 2.73 & 1.05 & 1.4 \\ 1.22 & 1.60 & 2.4 \end{array}) z_{2 i - 1} goodbreak+ ([], \begin{array}{c} 0.35 β_{i} \\ 0.385 β_{i} \\ 0.42 β_{i} \end{array}), \end{array}

where

α_{i}

and

β_{i}

are the random variables from the standard normal distribution,

i = 1, 2, \dots, N

. According to for...…”

Section: The Proposed Siamese Denpe Methods (Sdenpe)mentioning

confidence: 99%

Section: The Construction Of Sdenpe Networkmentioning

confidence: 99%

SiameseDeNPEnetwork framework for fault detection of batch process

Liu,

Zhao,

Mou

et al. 2023

Can J Chem Eng

View full text Add to dashboard Cite

show abstract

“…Dynamic linking and dynamic typing combine with their built-in high-level data structures to provide a perfect scripting language for quickly combining existing pieces during implementation. Readability is one of Python's primary focuses, which lowers the cost of maintaining programmers [55].…”

Section: Toolmentioning

confidence: 99%

A New Novel Approach for Sentiments Analysis using Contextual Mining and Supervised Learning

Vyawhare,

Bhushan,

Tayde

et al. 2024

CJACSIT

View full text Add to dashboard Cite

Textual data mining is used to anticipate the sentiment of a user based on a similar book. Using conditional probability distributions, the rating similarity between books can be quantified by textual mining. In supervised learning, the input and output data are sent to the machine learning model in tandem to maximize accuracy. In this paper, a cloudbased suggested system is incorporated which detects and recommends similar types of content based on the ranking of books. A cloud recommendation system is a means of determining which service is best suited to the user's tastes and requirements. The work uses specific formats such as graphics or text, and the networks built throughout the process may help uncover the links between the words. Textual mining is used to calculate book-ranging similarities and recommendations. From the comparison result, the proposed logistic regression technique has maximum accuracy which is 82% as compared to the existing technique The model shows all recommended books that are like the input book are displayed, while all other types of books are hidden from view. Also, positive, or negative masking can be determined by comparing the intensity of the extracted images.

show abstract

“…Python supports using packages and modules to facilitate the reuse of code and the modularization of projects. Each general platform may download and distribute the Python interpreter and its standard library in source or binary form (Hu et al, 2021).…”

Section: Toolsmentioning

confidence: 99%

English language analysis using pattern recognition and machine learning

Moolchandani,

Singh

2023

The Scientific Temper

View full text Add to dashboard Cite

Pattern identification and classification in complicated systems are difficult. This study uses optical character recognition (OCR) to digitize handwritten data. OCR segments and categorizes characters using online and offline methods for different input sources. Hindi and Bangladeshi categorization results unite linguistic studies. Handwriting recognition systems create editable digital documents from touchscreens, electronic pens, scanners, and photographs. Statistical, structural, neural network and syntactic methods improve online and offline recognition. In “english language analysis using pattern recognition and machine learning,” the accuracy of various approaches is examined, showing deep convolution neural networks (DCNN) 98% accuracy in recognizing subtle linguistic patterns. Nave Bayes, a trustworthy language analysis approach, has 96.2% accuracy. Table recognition (TR) algorithms retrieve structured information at 97%. This method outperforms others with 98.4% accuracy. This unique strategy could improve english language analysis using cutting-edge pattern recognition and machine learning techniques.

show abstract

Dynamic neural orthogonal mapping for fault detection

Cited by 10 publications

References 25 publications

SiameseDeNPEnetwork framework for fault detection of batch process

SiameseDeNPEnetwork framework for fault detection of batch process

A New Novel Approach for Sentiments Analysis using Contextual Mining and Supervised Learning

English language analysis using pattern recognition and machine learning

Contact Info

Product

Resources

About