Application Of Neural Network Technique And Electrodynamic Sensors In The Identification Of Solid Flow Regimes

2012 ◽  
Author(s):  
Mohd Fua’ad Hj Rahmat ◽  
Hakilo Ahmed Sabit
Keyword(s):  
Neural Network ◽  
Concentration Profile ◽  
Prediction Models ◽  
Flow Regimes ◽  
Industrial Processes ◽  
Concentration Profiles ◽  
Tomography System ◽  
Process Tomography ◽  
Flow Regime Identification ◽  
Profile Flow

Imaging of industrial processes have been accomplished with better efficiency and better control since the introduction of process tomography in several industries. This technique enables a deeper look into the internal conditions of a process without invading the process. In tomographic techniques, process information such as the distribution and velocity of the particles conveying at a particular plane can be obtained by placing sensors around the periphery of the plane. This paper is a continuation of a previous paper entitled Flow Regime Identification Using Neural Network–based Electrodynamic Tomography System in Jurnal Teknologi 40(D). This paper presents the results of sensors output in comparison to that of prediction models, concentration profiles and flow regimes identification obtained from the system described in the previous paper. Key words: Electrodynamic tomography, neural network, concentration profile, flow regimes

Flow Regime Identification Using Neural Network–Based Electrodynamic Tomography System

2012 ◽  
Author(s):  
Mohd. Fua’ad Rahmat ◽  
Hakilo Ahmed Sabit
Keyword(s):  
Neural Network ◽  
Flow Regime ◽  
Low Cost ◽  
Industrial Process ◽  
Solid Particles ◽  
Non Invasive ◽  
Tomography System ◽  
Process Tomography ◽  
Flow Regime Identification ◽  
Cost Efficient

Proses tomografi adalah suatu teknik membina imej yang murah, cekap dan sesuai untuk proses di industri yang kini semakin diguna pakai untuk tujuan pemantauan proses dan pengukuran. Mekanisme pengesanan dalam proses tomografi bergantung kepada bahan aliran dalam paip industri sama ada pepejal, gas atau cecair. Dalam kertas kerja ini, proses yang terlibat adalah pengaliran pepejal kering dalam paip mengikut arah graviti dan mekanisme pengesanan yang digunakan ialah penderia elektrodinamik. Pengenalpastian rejim aliran daripada pengukuran penderia adalah dengan menggunakan rangkaian neural yang akan mengenal pasti aliran pepejal sama ada dalam aliran penuh, suku separuh dan tiga suku. Kata kunci: Proses tomografi, rangkaian neural, penderia elektrodinamik, pengenalpastian Process tomography is a low cost, efficient and non-invasive industrial process imaging technique. It is used in many industries for process imaging and measuring. Provided that appropriate sensing mechanism is used, process tomography can be used in processes involving solids, liquids, gases, and any of their mixtures. In this paper, the process to be imaged and measured involves solid particles flow in gravity drop system. Electrical charge tomography or electrodynamic tomography is a tomographic technique using electrodynamic sensors. This paper presents the flow regime identification using neural network. Keywork: Process tomography; neural network; electrodynamic sensor; identification

Flow Regime Identification of Co-Current Downward Two-Phase Flow With Neural Network Approach

2002 ◽  
Author(s):  
Hiroshi Goda ◽  
Seungjin Kim ◽  
Ye Mi ◽  
Joshua P. Finch ◽  
Mamoru Ishii ◽  
...  
Keyword(s):  
Neural Network ◽  
Flow Visualization ◽  
Flow Regime ◽  
Two Phase Flow ◽  
Flow Regimes ◽  
Phase Flow ◽  
Two Phase ◽  
Neural Network Classification ◽  
The Neural Network ◽  
Flow Regime Identification

Flow regime identification for an adiabatic vertical co-current downward air-water two-phase flow in the 25.4 mm ID and the 50.8 mm ID round tubes was performed by employing an impedance void meter coupled with the neural network classification approach. This approach minimizes the subjective judgment in determining the flow regimes. The signals obtained by an impedance void meter were applied to train the self-organizing neural network to categorize these impedance signals into a certain number of groups. The characteristic parameters set into the neural network classification included the mean, standard deviation and skewness of impedance signals in the present experiment. The classification categories adopted in the present investigation were four widely accepted flow regimes, viz. bubbly, slug, churn-turbulent, and annular flows. These four flow regimes were recognized based upon the conventional flow visualization approach by a high-speed motion analyzer. The resulting flow regime maps classified by the neural network were compared with the results obtained through the flow visualization method, and consequently the efficiency of the neural network classification for flow regime identification was demonstrated.

A Novel Electrical Cpacitance Sensor Design For Dual Modality Tomography Multiphase Measurement

2013 ◽  
Vol 64 (5) ◽  
Author(s):  
Muhammad Jaysuman Pusppanathan ◽  
Fazlul Rahman Yunus ◽  
Nor Muzakkir Nor Ayob ◽  
Ruzairi Abdul Rahim ◽  
Fatin Aliah Phang ◽  
...  
Keyword(s):  
Multiphase Flow ◽  
Flow Measurement ◽  
Ultrasonic Tomography ◽  
Pipe Surface ◽  
Dual Modality ◽  
Tomography System ◽  
Imaging Result ◽  
Process Tomography ◽  
Single Sensor ◽  
Capacitance Tomography

Electrical capacitance tomography (ECT) is one of process tomography technique which is developed rapidly in recent years. ECT is an imaging technique to obtain the internal permittivity distribution of a vessel or pipe by using capacitance electrodes sensor. This method has been integrated with ultrasonic tomography as multimodality system to perform multiphase flow measurement such as crude oil separation and oil process industry. In the present paper, a novel type of ECT sensor was developed using copper FR4 material. The electrode sensors can be flexibly bend or curve to fit the pipe surface for optimum measurement. Thus, every single sensor strip is designed to be functioned independently. Such system has lower sensing capability in the central of the sensing area which often contributes to poor imaging result. This problem can be overcome by combining the ECT with ultrasonic tomography to form a dual modality tomography system. By implementing the new ECT sensor, multiphase flow measurement image results can be achieved. The reconstructed image results are presented in this paper.

scholarly journals Integrating ensemble systems biology feature selection and bimodal deep neural network for breast cancer prognosis prediction

2021 ◽  
Vol 11 (1) ◽  
Author(s):  
Li-Hsin Cheng ◽  
Te-Cheng Hsu ◽  
Che Lin
Keyword(s):  
Breast Cancer ◽  
Neural Network ◽  
Feature Selection ◽  
Systems Biology ◽  
Ensemble Learning ◽  
Microarray Data ◽  
Deep Neural Network ◽  
Prediction Models ◽  
Biological Knowledge ◽  
Prognosis Prediction

AbstractBreast cancer is a heterogeneous disease. To guide proper treatment decisions for each patient, robust prognostic biomarkers, which allow reliable prognosis prediction, are necessary. Gene feature selection based on microarray data is an approach to discover potential biomarkers systematically. However, standard pure-statistical feature selection approaches often fail to incorporate prior biological knowledge and select genes that lack biological insights. Besides, due to the high dimensionality and low sample size properties of microarray data, selecting robust gene features is an intrinsically challenging problem. We hence combined systems biology feature selection with ensemble learning in this study, aiming to select genes with biological insights and robust prognostic predictive power. Moreover, to capture breast cancer's complex molecular processes, we adopted a multi-gene approach to predict the prognosis status using deep learning classifiers. We found that all ensemble approaches could improve feature selection robustness, wherein the hybrid ensemble approach led to the most robust result. Among all prognosis prediction models, the bimodal deep neural network (DNN) achieved the highest test performance, further verified by survival analysis. In summary, this study demonstrated the potential of combining ensemble learning and bimodal DNN in guiding precision medicine.

Experimental study on the factors affecting squeak noise occurrence in automotive suspension bushings

2021 ◽  
pp. 095440702110241
Author(s):  
Byunghyun Kang ◽  
Cheol Choi ◽  
Daeun Sung ◽  
Seongho Yoon ◽  
Byoung-Ho Choi
Keyword(s):  
Neural Network ◽  
Logistic Regression ◽  
Influencing Factors ◽  
Frictional Force ◽  
Prediction Models ◽  
Statistical Tests ◽  
Network Models ◽  
Multiple Logistic Regression ◽  
Neural Network Models ◽  
Automotive Suspension

In this study, friction tests are performed, via a custom-built friction tester, on specimens of natural rubber used in automotive suspension bushings. By analyzing the problematic suspension bushings, the eleven candidate factors that influence squeak noise are selected: surface lubrication, hardness, vulcanization condition, surface texture, additive content, sample thickness, thermal aging, temperature, surface moisture, friction speed, and normal force. Through friction tests, the changes are investigated in frictional force and squeak noise occurrence according to various levels of the influencing factors. The degree of correlation between frictional force and squeak noise occurrence with the factors is determined through statistical tests, and the relationship between frictional force and squeak noise occurrence based on the test results is discussed. Squeak noise prediction models are constructed by considering the interactions among the influencing factors through both multiple logistic regression and neural network analysis. The accuracies of the two prediction models are evaluated by comparing predicted and measured results. The accuracies of the multiple logistic regression and neural network models in predicting the occurrence of squeak noise are 88.2% and 87.2%, respectively.

scholarly journals Corrosion Behavior of LENS Deposited CoCrMo Alloy Using Bayesian Regularization-Based Artificial Neural Network (BRANN)

2021 ◽  
Vol 7 (3) ◽  
Author(s):  
Nagoor Basha Shaik ◽  
Kedar Mallik Mantrala ◽  
Balaji Bakthavatchalam ◽  
Qandeel Fatima Gillani ◽  
M. Faisal Rehman ◽  
...  
Keyword(s):  
Neural Network ◽  
Artificial Neural Network ◽  
Corrosion Rate ◽  
Prediction Models ◽  
Test Time ◽  
Bayesian Regularization ◽  
Cobalt Chromium ◽  
Corrosion Properties ◽  
Cocrmo Alloy ◽  
Artificial Neural

AbstractThe well-known fact of metallurgy is that the lifetime of a metal structure depends on the material's corrosion rate. Therefore, applying an appropriate prediction of corrosion process for the manufactured metals or alloys trigger an extended life of the product. At present, the current prediction models for additive manufactured alloys are either complicated or built on a restricted basis towards corrosion depletion. This paper presents a novel approach to estimate the corrosion rate and corrosion potential prediction by considering significant major parameters such as solution time, aging time, aging temperature, and corrosion test time. The Laser Engineered Net Shaping (LENS), which is an additive manufacturing process used in the manufacturing of health care equipment, was investigated in the present research. All the accumulated information used to manufacture the LENS-based Cobalt-Chromium-Molybdenum (CoCrMo) alloy was considered from previous literature. They enabled to create a robust Bayesian Regularization (BR)-based Artificial Neural Network (ANN) in order to predict with accuracy the material best corrosion properties. The achieved data were validated by investigating its experimental behavior. It was found a very good agreement between the predicted values generated with the BRANN model and experimental values. The robustness of the proposed approach allows to implement the manufactured materials successfully in the biomedical implants.

Autism Spectrum Disorder Identification Using Polynomial Distribution based Convolutional Neural Network

2021 ◽  
Vol 19 (2) ◽  
pp. 19-30
Author(s):  
G. Nagarajan ◽  
Dr.A. Mahabub Basha ◽  
R. Poornima
Keyword(s):  
Neural Network ◽  
Prediction Models ◽  
Autism Spectrum ◽  
Spectrum Disorder ◽  
Machine Learning Techniques ◽  
Hybrid Technique ◽  
Learning Techniques ◽  
Chicken Swarm Optimization ◽  
Sensitivity Specificity ◽  
Polynomial Distribution

One main psychiatric disorder found in humans is ASD (Autistic Spectrum Disorder). The disease manifests in a mental disorder that restricts humans from communications, language, speech in terms of their individual abilities. Even though its cure is complex and literally impossible, its early detection is required for mitigating its intensity. ASD does not have a pre-defined age for affecting humans. A system for effectively predicting ASD based on MLTs (Machine Learning Techniques) is proposed in this work. Hybrid APMs (Autism Prediction Models) combining multiple techniques like RF (Random Forest), CART (Classification and Regression Trees), RF-ID3 (RF-Iterative Dichotomiser 3) perform well, but face issues in memory usage, execution times and inadequate feature selections. Taking these issues into account, this work overcomes these hurdles in this proposed work with a hybrid technique that combines MCSO (Modified Chicken Swarm Optimization) and PDCNN (Polynomial Distribution based Convolution Neural Network) algorithms for its objective. The proposed scheme’s experimental results prove its higher levels of accuracy, precision, sensitivity, specificity, FPRs (False Positive Rates) and lowered time complexity when compared to other methods.

Sign in / Sign up

Export Citation Format

Share Document