research centers


Search results: Found 3

Listing 1 - 3 of 3
Sort by

Article
Feedforward Controller for Nonlinear Systems Utilizing a Genetically Trained Fuzzy Neural Network
مسيطر ذو تغذيه أماميه للأنظمه اللاخطية بإستخدام شبكة ضبابية عصبية مد ربة جينيا

Author: Omar F. Lutfy Al-Karkhy
Journal: Engineering and Technology Journal مجلة الهندسة والتكنولوجيا ISSN: 16816900 24120758 Year: 2007 Volume: 25 Issue: suppl.of No.3 Pages: 475-494
Publisher: University of Technology الجامعة التكنولوجية

Loading...
Loading...
Abstract

This paper presents an intelligent controller that acts as a FeedForwardController (FFC). utilizing the benefits of Fuzzy Logic (FL), Neural Networks(NNs) and Genetic Algorithms (GAs), this controller is built to controlnonlinear plants, where the GA is used to train this Fuzzy Neural Controller(FNC) by adjusting of its parameters based on minimizing the Mean Squareof Error (MSE) criterion.These parameters of the FNC include the input and output scaling factors,the centers and widths of the membership functions (MFs) for the inputvariable and the quantisation levels of the output variable, that are subjectedto constraints on their values by the expert. The GA used in this work is areal-coding GA with hybrid selection method and elitism strategy. To showthe effectiveness of this FNC several invertable (open-loop stable) nonlinearplants have been selected to be controlled by this FNC through simulation.

يقدم هذا البحث مسيطر ذكي والذي يعمل كمسيطر ذو تغذيه أماميه . بإستغلال فوائدالمنطق المضبب والشبكات العصبية والخوارزميات الجينية , تم بناء هذا المسيطر ذو التغذيهالأماميه للسيطره على أنظمه لا خطية , حيث تستخدم الخوارزمية الجينية لتدريب المسيطرالضبابي العصبي من خلال ضبط عدة معاملات فيه إعتمادا على تقليل معيار معدل مربع الخط أ.وهذه المعاملات للمسيطر الضبابي العصبي تشمل عوامل التقييس للإدخال والإخراج , المراكزوالامتدادات لدوال العضوية لمتغير الإدخال ومستويات الكميات لمتغير الإخراج والتي تخضعلقيود على قيمها من قبل الخبير . الخوارزمية الجينية المستخدمة في هذا العمل هي خوارزميةجينية ذات معاملات ترميز بالقيم الحقيقية وطريقة اختيار مهجنة وحكم النخبة.ولإظهار فعالية هذا المسيطر الضبابي العصبي تم إختيار عدة أنظمة لاخطية قابلة للعكسودارتها المفتوحة مستقرة ليتم السيطرة عليها من قبل هذا المسيطر الضبابي العصبي من خلالالتمثيل.


Article
Intelligent Modeling of Metal Oxide Gas Sensor

Authors: Omar F. Lutfy --- Alaa A. Abdul-Hamead
Journal: Engineering and Technology Journal مجلة الهندسة والتكنولوجيا ISSN: 16816900 24120758 Year: 2018 Volume: 36 Issue: 7 Part (A) Engineering Pages: 777-783
Publisher: University of Technology الجامعة التكنولوجية

Loading...
Loading...
Abstract

Due to the complexity of the gas detection process, traditional modeling techniques cannot provide accurate modeling performance to reproduce the behavior of this difficult process. In this paper, an intelligent modeling technique is utilized to develop an accurate model to represent the complex and nonlinear gas detection process. In particular, in this study nickel Oxide NiO gas sensor, which was specifically fabricated by a simple chemical spray pyrolysis technique. In the process, the nickel chloride hexahydrate salt was used at a concentration of (0.05 M) and a temperature of 350 ºC. Because of this process, the thickness of NiO was 0.1μm. Inspection was done using three different testing techniques; X-ray diffraction, scanning electron microscopy, and the sensitivity test of NiO for Methane gas CH4 in the range of (0-500) ppmv. Inspection results show that the film was crystalline, has a cubic system, and without cracks or open pores. On the other hand, the sensitivity results were disparate and low in value within the considered range. From the real-time experiment described above, training samples were gathered to develop the desired process model. The considered modeling technique was based on exploiting the wavelet network (wavenet) to represent the nonlinear function of the nonlinear autoregressive with exogenous input (NARX) structure. In model development process, the experimental data were utilized as the training samples for the wavenet-based NARX model. As the modeling accuracy, the proposed wavenet-based NARX model attained a value of 1.895 × 10-12 for the root mean square of error (RMSE) criterion.


Article
A Simplified Recurrent Neural Network Trained by Gbest-Guided Gravitational Search Algorithm to Control Nonlinear Systems

Authors: Omar F. Lutfy --- Ahmed L. Jassim
Journal: Engineering and Technology Journal مجلة الهندسة والتكنولوجيا ISSN: 16816900 24120758 Year: 2018 Volume: 36 Issue: 12 Part (A) Engineering Pages: 1290-1301
Publisher: University of Technology الجامعة التكنولوجية

Loading...
Loading...
Abstract

This paper presents a feedback control strategy using a SimplifiedRecurrent Neural Network (SRNN) for nonlinear dynamical systems. As anenhancement for a previously reported modified recurrent network (MRN),the proposed SRNN structure is used as an intelligent Proportional-IntegralDerivative(PID)-like controller. More precisely, the enhancement in theSRNN structure was realized by employing unity weight values between thecontext and the hidden layers in the original MRN structure. The newlydeveloped Gbest-guided Gravitational Search Algorithm (GGSA) wasadopted for optimizing the parameters of the SRNN structure. To show theefficiency of the proposed PID-like SRNN controller, three differentnonlinear systems were considered as case studies, including a control valve,and a complex difference eq.. From an extensive set of evaluation tests, whichincludes a control performance test, a disturbance rejection test, and ageneralization test, the proposed PID-like SRNN controller demonstrated itseffectiveness with regards to precise control and good robustness andgeneralization abilities. Furthermore, compared to other Neural Network(NN) structures, including the original MRN and the Multilayer Perceptron(MLP) NN, the SRNN structure attained superior results due to the utilizationof a reduced set of parameters. From another study, the GGSA accomplishedthe best optimization results in terms of control precision and convergencespeed compared to the original Gravitational Search Algorithm (GSA).

Listing 1 - 3 of 3
Sort by
Narrow your search

Resource type

article (3)


Language

English (3)


Year
From To Submit

2018 (2)

2007 (1)