An Oil Fraction Neural Sensor Developed Using Electrical Capacitance Tomography Sensor Data

Zainal-Mokhtar, Khursiah; Mohamad–Saleh, Junita

doi:10.3390/s130911385

Cited by 23 publications

(12 citation statements)

References 48 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…As seen in Table 1, the values of natural frequencies in the damaged rectangular composite plate are less than those in the intact one in all modes. These results are consistent with the results reported in the literature (see [17][18][19][20]) and indicate our results' accuracy. Figures 6, 7, 8, 9, and 10 show the first five-mode shapes of intact and damaged RLCP.…”

Section: Effects Of Single-damage On Modal Characteristicssupporting

confidence: 94%

“…Fig. 3 A typical first-order model of the Sugeno fuzzy system [18] 3 Proposed Model This work proposes a new algorithm to identify the severity of the single damages of RLCPs using FEM and ANFIS. First, the finite element governing equation is formed for the intact RLCPs to calculate natural frequencies and mode shapes.…”

Section: Adaptive Network-based Fuzzy Inferencementioning

confidence: 99%

See 1 more Smart Citation

Adaptive Network-Based Fuzzy Inference for Damage Detection in Rectangular Laminated Composite Plates Using Vibrational Data

Saadatmorad

Jafari-Talookolaei

Pashaei

et al. 2021

Lecture Notes in Civil Engineering

View full text Add to dashboard Cite

A novel model is developed to identify damages severities in rectangular laminated composite plates (RLCPs) based on the finite element method (FEM) and adaptive neuro-fuzzy inference system (ANFIS). Empirical data are generated via FEM, and then the ANFIS algorithm is applied to identify damages' severities of single damage scenarios. In this study, the networks' inputs and outputs are the first five natural frequencies and damages' severities. Also, a multilayer perceptron neural network (MLP) is developed to compare results obtained from ANFIS. Results show that the MLP model can identify the severity of every single damage scenario in RLCPs with a high precision (R > 0.9). Also, results obtained indicate that ANFIS can identify and predict the severity of every single damage scenario in RLCPs better than MLP with perfect precision (R = 0.99999).

show abstract

Section: Effects Of Single-damage On Modal Characteristicssupporting

confidence: 94%

Section: Adaptive Network-based Fuzzy Inferencementioning

confidence: 99%

Adaptive Network-Based Fuzzy Inference for Damage Detection in Rectangular Laminated Composite Plates Using Vibrational Data

Saadatmorad

Jafari-Talookolaei

Pashaei

et al. 2021

Lecture Notes in Civil Engineering

View full text Add to dashboard Cite

show abstract

“…As a result, the resolution changed from 4 to 42 aF. Compared to the approach in paper [9], the sigma-delta converter has a much higher resolution and sensitivity; however, the acquisition rate is much slower. Therefore, this circuit would not be ideal for applications that require real-time monitoring and operations.…”

Section: Ss1mentioning

confidence: 97%

Electronic hardware design of electrical capacitance tomography systems

Saied¹,

Meribout²

2016

Phil. Trans. R. Soc. A.

View full text Add to dashboard Cite

Electrical tomography techniques for process imaging are very prominent for industrial applications, such as the oil and gas industry and chemical refineries, owing to their ability to provide the flow regime of a flowing fluid within a relatively high throughput. Among the various techniques, electrical capacitance tomography (ECT) is gaining popularity due to its non-invasive nature and its capability to differentiate between different phases based on their permittivity distribution. In recent years, several hardware designs have been provided for ECT systems that have improved its resolution of measurements to be around attofarads (aF, 10(-18) F), or the number of channels, that is required to be large for some applications that require a significant amount of data. In terms of image acquisition time, some recent systems could achieve a throughput of a few hundred frames per second, while data processing time could be achieved in only a few milliseconds per frame. This paper outlines the concept and main features of the most recent front-end and back-end electronic circuits dedicated for ECT systems. In this paper, multiple-excitation capacitance polling, a front-end electronic technique, shows promising results for ECT systems to acquire fast data acquisition speeds. A highly parallel field-programmable gate array (FPGA) based architecture for a fast reconstruction algorithm is also described. This article is part of the themed issue 'Supersensing through industrial process tomography'.

show abstract

“…Mohamad-Saleh and Hoyle 23 used artificial neural networks to directly estimate key parameters for characterizing gas-water flows from normalized capacitance measurements. Zainal-Mokhtar and Mohamad-Saleh 24 trained a multilayer perceptron artificial neural network and used it to estimate the oil fraction in a pipeline with various ECT sensor parameters. Wang and Zhang 25 used a model trained by support vector machine to identify flow regime of gas-oil flows.…”

Section: Introductionmentioning

confidence: 99%

A machine learning approach for electrical capacitance tomography measurement of gas–solid fluidized beds

Guo

Yang

et al. 2019

AIChE Journal

View full text Add to dashboard Cite

Electrical capacitance tomography has been widely used to obtain key hydrodynamic parameters of gas-solid fluidized beds, which is normally realized by first reconstructing images and then by analyzing these images. This indirect approach is time-consuming and hence difficult for on-line monitoring. Meanwhile, considering recurrence of similar flow patterns in fluidized beds, most of these calculations are repetitive and should be avoided. Here, we develop a machine learning approach to address these problems. First, superficial gas velocity linear-increasing strategy is used to perform highthroughput experiments to collect a large amount of training samples. These samples are used to train the map from normalized capacitance measurements to key parameters that obtained by an iterative image reconstruction algorithm off-line. The trained model can then be used for on-line monitoring. Preliminary tests revealed that the trained models show good prediction and generality for the estimation of the overall solid concentration and the equivalent bubble diameter.

show abstract

An Oil Fraction Neural Sensor Developed Using Electrical Capacitance Tomography Sensor Data

Cited by 23 publications

References 48 publications

Adaptive Network-Based Fuzzy Inference for Damage Detection in Rectangular Laminated Composite Plates Using Vibrational Data

Adaptive Network-Based Fuzzy Inference for Damage Detection in Rectangular Laminated Composite Plates Using Vibrational Data

Electronic hardware design of electrical capacitance tomography systems

A machine learning approach for electrical capacitance tomography measurement of gas–solid fluidized beds

Contact Info

Product

Resources

About