Show: 25 50 75 100 Results

Search results: 25 out of 433

دراسة شاملة عن التحكم ا?تجاهي للمحرك الحثي الخطي == COMPREHENSIVE STUDY ON VECTOR CONTROL OF LINEAR INDUCTION MOTOR

Author name: محمد حميد حسن المرفظ
General topic: Electrical, Electronic and Communications Engineering
Specific topic: Electrical Engineering
Degree: Master
University: University Of Basrah - College Of Engineering
Language: English
University location: Basrah
First pages:

الشبكة العصبية الضوئية ذات البعد الاحادي والثنائي == ONE- and TWO -DIMENSIONAL OPTICAL NEURAL NETWORKS

Author name: علي كامل مرزوق الهلالي
Supervisor name: Raad. S. Fyath |Jassim. M. Abdul-jabbar
General topic: Electrical, Electronic and Communications Engineering
Specific topic: Electrical Engineering
Degree: Master
University: University Of Basrah - College Of Engineering
Language: English
University location: Basrah
First pages:

تحليل الاداء لنظام OSCM/WDM == Some Performance Analysis for Optical Subcarrier Multiplexing/Wavelength Division Multiplexing (SCM/WDM) System

Author name: حسنين محمد مهدي
Supervisor name: Abdul-Jabbar |R. S. Fyath
General topic: Electrical, Electronic and Communications Engineering
Specific topic: Electrical Engineering
Degree: Master
University: University Of Basrah - College Of Engineering
Language: English
University location: Basrah
First pages:

تحسين اداء مغير القدرة المتناوبة الى القدرة المستمرة باستخدام السيطرة الضبابية == AC/DC Converter Performance Improvement using Fuzzy Controller

Author name: سومر صاحب حردان
Supervisor name: رمزي سالم علي | ربيع هاشم ثجيل
General topic: Electrical, Electronic and Communications Engineering
Specific topic: Electrical Engineering
Degree: Master
University: University Of Basrah - College Of Engineering
Language: English
University location: Basrah
First pages:

التحكم بذراع الروبوت باستخدام الشبكات العصبية المضببة

Author name: جلال نزار عبد الباقي
Supervisor name: عبد العظيم عبد الكريم علي
General topic: Electrical, Electronic and Communications Engineering
Specific topic: Electrical Engineering
Degree: Master
University: University Of Basrah - College Of Engineering
Language: English
University location: Basrah
First pages:

دراسة عن منظومة التدوير المكونة من محرك حثي مغذى بمغير تضمين عرض النبضة == Study on Voltage (PWM) Inverter fed Induction Motor Drive

Author name: حيدر خليل ابراهيم الطاهري
General topic: Electrical, Electronic and Communications Engineering
Specific topic: Electrical Engineering
Degree: Master
University: University Of Basrah - College Of Engineering
Language: English
University location: Basrah
First pages:

تصميم وتطبيق النظام الالي لمراقبة الجودة == DESIGN AND IMPLEMENTION OF AUTOMATING QUALITY CONTROL SYSTEM

Author name: فاروق تحسين طه
General topic: Electrical, Electronic and Communications Engineering
Specific topic: Electrical Engineering
Degree: Master
University: University Of Basrah - College Of Engineering
Language: English
University location: Basrah
First pages:

تقنية جديدة مقترحة لكبح المحركات الحثية ثلاثية الطور بطريقة عكس التيار == NEW PROPOSED PLUGGING TECHNIQUE FOR BRAKING THREE-PHASE INDUCTION MOTORS

Author name: وسام اياد كاظم
General topic: Electrical, Electronic and Communications Engineering
Specific topic: Electrical Engineering
Degree: Master
University: University Of Basrah - College Of Engineering
Language: English
University location: Basrah
First pages:

تمييز الحروف العربية المعزولة المكتوبة بخط اليد باستخدام الشبكات العصبية == Recognition Of Isolated Handwritten Arabic Letters Using Neural Networks

Author name: نهلة ظاهر حبيب بهية
Supervisor name: منذر نعمان التكريتي
General topic: Electrical, Electronic and Communications Engineering
Specific topic: Electrical Engineering
Degree: Master
University: Mustansiriyah University - College Of Engineering - Department Of Electrical Engineering
Language: English
University location: Baghdad
First pages:
Abstract: يتناول هذا البحث مشكلة تمييز الحروف العربية المعزولة المكتوبة بخط اليد باستخدام الشبكات العصبية. ان نظام تمييز الحروف العربية المقترح يتكون من ثلاث مراحل وهي : مرحلة مسح صور الحروف العربية ثم مرحلة معالجة الصور واخيرا مرحلة التدريب والتصنيف.ان المرحلة الاخ | This thesis presents an algorithm for recognition of an off - line isolated handwritten Arabic letters using neural networks.The handwritten recognition system present in this work consists of three modules : - 1 - Arabic letter scanning module 2 - Arabic letter preprocessing module 3 - Learning and recognition module.The proposed neural network in the third module is trained in two stages : The first stage employs Self - Organizing Map learning algorithm for clustering the input pattern, which are based on a database of 196 letters collected from 7 independent persons.In the second stage each of the similar output cluster is considered as a subnet introduced to neural network trained by Back - propagation learning algorithm for classification. Several networks architecture are designed in the second stage using 1 - hidden and 2 - hidden layers with 5 and 28 output nodes in the output layer with different number of hidden nodes and learning rate.To examine the efficiency of the system a database of 196 letters collected from other 7 independent persons are used, (in order to test the ability of the trained network to generalize). The results show that clustering the input pattern, using 1 - hidden layer and 28 nodes in output layer improve the network performance. The system was implemented using (IBM - PC) of type Pentium 3.The programming language used to design the system was c++ version 5.02.

استراتيجية السيطرة على المقوم ثلاثي الاطوار المضمن بعرض النبضة (PWM) نوع رافع للجهد ولمصدر تغذية متغير الحالات

Author name: منتظر كاظم عبد الله
Supervisor name: تركي كحيوش حسن
General topic: Electrical, Electronic and Communications Engineering
Specific topic: Electrical Engineering
Degree: Master
University: Mustansiriyah University - College Of Engineering - Department Of Electrical Engineering
Language: English
University location: Baghdad
First pages:
Abstract: In this thesis, a proposed control strategy is presented to improve the performance of the PWM boost type rectifier when operating under different supply voltage conditions (balanced, unbalanced, and distorted three - phase supply voltages). The proposed control strategy is classified into two parts, the first part is voltage controller and the second part is current controller. In the voltage controller, Repetitive Controller (RC) is used to reduce the even order harmonics in the regulated output dc voltage (5.62V (P.P.)) so small output capacitor (480µF) is used instead of large capacitor (920µF) that is used with the conventional controller (PI controller connected in series with a LPF). RC also reduces the even order harmonics (0.014A (P.P)) which appears in the reflected dc current (IMAX), this leads to reduce the odd order harmonics (0.01A) which appears in the input currents. While in the current controller, enhanced Phase Locked Loop (EPLL) technique is used to obtain sinusoidal and balanced three phases, to construct the reference currents, which are in phase with the fundamental supply voltages Therefore, the supply - side power factor is kept close to unity (0.9036 for extremely distorting case). As a result the calculations of the reference currents in the proposed control strategy does not require Clarke or Park transformations, the same for the calculations of the positive or negative sequence components. A proportional controller is used to give excellent tracking between the line and the reference currents. Using this approach makes the control strategy easy to understand and also it saves calculation time of the control algorithms if implements in real time. The complete system with the proposed control strategy is simulated using Matlab/Simulink (Version 2012). The obtained results for the complete system using repetitive voltage controller are compared to the results of the system with using the conventional voltage controller. The results with the repetitive controller show better responses and stable operation in the steady state under different input voltage conditions, as well as in the transient response under different disturbances.

تحقيق دائرة المعدل الثابت للانذار الكاذب باستعمال منظومة البوابات المنطقية المبرمجة موقعيا == Modified GO - CFAR Implementaion Using FPGA

Author name: مصطفى صبحي كمال
Supervisor name: رفعت طالب حسين | حكمت نجم عبد الله
General topic: Electrical, Electronic and Communications Engineering
Specific topic: Electrical Engineering
Degree: Master
University: Mustansiriyah University - College Of Engineering - Department Of Electrical Engineering
Language: English
University location: Baghdad
First pages:
Abstract: هذه الاطروحة تتعلق بتحليل ومعالجة تاثير الضوضاء المحيطية (Clutter) لمنظومة الرادار النبضي، يتطلب رادار الكشف الالي للهدف استعمال عتبة الية (Adaptive threshold) لتحقيق معدل ثابت للانذار الكاذب وذلك لغرض السيطرة على الانذار الكاذب الذي تسببه تغيرات في الضوض | This thesis deals with the analysis and processing of clutter for pulsed radar system. Automatic target detection radar requires adaptive thresholding achieved by Constant False Alarm Rate (CFAR) circuit in order to control the false alarm caused by variations in clutter background.This work focus on worst radar environment that happens when abrupt variation in clutter background merged with multi - interfering target, to detect target in such environments it need robust CFAR algorithm that excise the target spikes and clutter edges from CFAR window in order to give best possible estimation to the noise background. two important algorithms studied which are CA - CFAR and OS - CFAR algorithms in additional to the modified CA - CFAR algorithm. All these algorithms were simulated with mat lab v6.1 and applied them to three different clutter models that represent different environment cases the CA - CFAR family failed to handle model two and three also OS - CFAR family except OSGO - CFAR that handle all models successfully. For modified CA - CFAR family only modified GO - CFAR handle all models successfully and comparing with OSGO - CFAR the modified GO - CFAR need less hardware and processing time because it did not need sorting process that is essential for OSGO - CFAR. Therefore, the modified GO - CFAR is chosen to implement by using FPGA and another important feature in modified GO - CFAR algorithm that is parallel processing since the spike selection process is done at the same time with summing of samples process that make this algorithm much less in processing time from any other algorithm that work in the same environment. The FPGA chip that used to implement modified GO - CFAR algorithm need only three signals from the radar receiver to mach with the receiver circuit correctly which are time base clock signal period reset trigger signal and the pulse duration time. Therefore, the FPGA chip can work effectively with almost any radar receiver system

تحليل الاشارات المشفرة ثنائيا باستخدام الترشيح الموائم وغير الموائم لتقليل الفصوص الجانبية == Analysis Of Binary - Coded Signals Using Matched And Mismatched Filtering For Sidelobe Reduction

Author name: صلاح حسن عباس
Supervisor name: اسماعيل محمد جابر
General topic: Electrical, Electronic and Communications Engineering
Specific topic: Electrical Engineering
Degree: Master
University: Mustansiriyah University - College Of Engineering - Department Of Electrical Engineering
Language: English
University location: Baghdad
First pages:
Abstract: Various types of binary phase - coded signals are studied in terms of their correlation properties. Then, matched filtering technique as well as mismatched filtering techniques is applied to the codes studied previously in order to reduce the sidelobes of the compressed binary phase - coded signal around the mainlobe, and consequently increase the detection of the received signal. Each filtering technique is performed using programs written by the programming language of MATLAB version 7.4.0.287 (R2007a). White Gaussian noise is introduced in the assessment made to both the filtering techniques and the codes that are adopted in this thesis.The MATLAB programs are tested on various types of binary phase codes that are adopted in this thesis to calculate the criteria to choose the best filtering technique. These criteria are peak sidelobe level, the integrated sidelobe level, and the signal - to - noise ratio loss. Later, the results, which are obtained according to these mentioned criteria, are used to test different filtering techniques on the chosen periodical and aperiodical binary phase codes to choose the best one.It is found that, the optimal peak sidelobe level (PSL) filtering technique is the best filtering technique for the aperiodical binary phase codes in terms of the peak sidelobe level, and in terms of an acceptable signal - to - noise ratio loss. The optimal integrated sidelobe level (ISL) filtering technique comes in the second place.While, it is found that, the mismatched filtering technique is the best filtering technique for the periodical binary phase codes in terms of an acceptable signal - to - noise ratio loss.

نظام الاخفاء الصوري المعتمد على تحويل المويجة والاشارات الفوضوية

Author name: سرى فهمي يوسف
Supervisor name: حكمت نجم عبد الله
General topic: Electrical, Electronic and Communications Engineering
Specific topic: Electrical Engineering
Degree: Master
University: Mustansiriyah University - College Of Engineering - Department Of Electrical Engineering
Language: English
University location: Baghdad
First pages:

تحسين التمييز للشبكات العصبية المنعدمة الاوزان == Recognition Enhancement of Weightless Neural Network

Author name: عمار علي مصطفى
Supervisor name: محمود حمزة المفرجي
General topic: Electrical, Electronic and Communications Engineering
Specific topic: Electrical Engineering
Degree: Master
University: University of Technology - Electrical Engineering Department - Electrical Engineering Branch
Language: English
University location: Baghdad
Key words:
  • Pattern Recognition
  • RAM - based Weightless Neural Network
First pages:
Abstract: الفعل الذكي للشبكات المتعلمة في هذا البحث يركز على الشبكات المتعلمة في منظومة تمييز الانماط المعتمدة على الذاكرة. ان الفعل الذكي في هذه المنظومات يكمن في القابلية على تمييز الانماط التي لم تتدرب عليها مسبقا وذلك يتم بالاعتماد على ظاهرة التعميم. الهدف من ا | The intelligent action of the learning network in this research concentrates on the learning network in the pattern recognition systems that are based on the Random Access Memory. The intelligence action in the pattern recognition system is the capability to recognize other patterns that were not learned before in the system depending on the generalization. The objective of the research is oriented to improve the performance of the learning network in pattern recognition system. A pattern recognition system that based on Random Access Memories is developed and evaluated for classifying non - deterministic data with particular reference to unconstrained handwritten Arabic numerals and traffic signs patterns.Various techniques are presented which allow the system to be optimized, giving an increase in the performance and confidence of the results.The results obtained showed the improvements of n - tuple and VG - RAM WNN systems as learning network. Techniques used to enhance the performance are investigated. The averaging of the training patterns for colored images is the proposed system. This technique gives improvements to the system performance in term of required storage size and recognition speed.

تحسين الفولتية وتقليل خسائر منظومة الشبكة العراقية باستخدام افضل جهاز FACTs == Improvement of Voltage Profile And Reduce Iraqi Grid System Losses By Using Optimal FACTs Device

Author name: سيف اسعد عبد الحسين
Supervisor name: رشيد حميد الربيعي | انعام ابراهيم علي
General topic: Electrical, Electronic and Communications Engineering
Specific topic: Electrical Engineering
Degree: Master
University: University of Technology - Electrical Engineering Department - Electrical Engineering Branch
Language: English
University location: Baghdad
First pages:
Abstract: الغرض من هذا العمل هو تحسين الجهد والتقليل من خسائرالقدرة للشبكة العراقية ( 400kV ) عن طريق تركيب اجهزة ( FACTs ) في مواقع مختلفة تحت حالة خطا ومقارنة النتائج مع تلك دون وجود (FACTs) تحت نفس الظروف.تم تحديد الموقع الامثل للـ(FACTs) على اساس الخوارزميات | The aim of this work is to improve the voltage profile and reduces power losses of Iraqi (400kV) power grid system by installing Flexible Alternating Current Transmission System devices (FACTs).The optimal location of the FACTs device was specified based on Genetic Algorithm (GA) optimization method and power system analysis toolbox ( PSAT) package, it was utilized to search for optimum FACT parameters setting and location based objective function that depends on the power and voltage as fitness.MATLAB (R2011a) was used for running both GA program and Power System Analysis Toolbox (PSAT) version 2.1.6, Newton Raphson method also used for solving the load flow of the system and the Trapezoidal method for the non - linear differential equations.Two systems have been implemented in this work as follows; • IEEE - machine 9 - bus system.• INSGS 11 - machine 24 - bus, 39(400kV) overhead transmission line.The GA program is implemented on IEEE 9 - bus system, then after getting the appropriate results the same procedure repeated for the Iraqi grid system which is more complicated, its used to find the optimal location of Unified Power Flow Controller (UPFC). The results obtained from GA have been compared with PSAT and showed that the installation of (UPFC) devices at the optimal locations of the Iraqi grid gives an improvement in the power losses and voltage after subjected to the three phase fault to ground at different locations and different cases (temporary fault, permanent fault).

نموذج مقترح للتعامل مع فشل العقدة في شبكة الاستشعار اللاسلكية == Proposed Model To Handle Node Failure In Wireless Sensor Network

Author name: مصطفى خالد مزعل
Supervisor name: ضياء جاسم كاظم
General topic: Electrical, Electronic and Communications Engineering
Specific topic: Electrical Engineering
Degree: Master
University: University of Baghdad - College Of Engineering - Department Of Electrical Engineering
Language: English
University location: Baghdad
First pages:
Abstract: Wireless Sensor Network (WSN) contains many low cost and low power sensor nodes (SNs), these nodes may fail to communicate with each other according to some reasons such as battery lifetime, uncontrolled events or any surrounding environment actions which will lead to partition the network and reduce the Quality of Service (QoS) as well as the reliability and efficiency of the whole network. The motivation of this work was detecting these malfunctions using three approaches considered with two proposed network models : random distributed SNs and real model.Firstly, localization approach in WSN are discussed and verified using four classic localization methods to detect any failed SN, which are : Distance Vector (DV - Hop), Angle of Arrival (AoA), Time of Arrival (ToA) and Received Signal Strength Indicator (RSSI). These methods are analyzed and implemented for three different areas, each with three different numbers of SNs. All of them are implemented using MATLAB programming language. Secondly a known fault detection method called Distributed Fault Detection (DFD) discussed and implemented using MATLAB also, and a new method developed from this method named (MDFD) to overcome the disadvantages of DFD method. Finally the feature of the connectivity of the SNs themselves within the network is exploited by using two Medium Access Control (MAC) protocol standards : Sleep Medium Access Control (SMAC) and Tune Medium Access Control (TMAC) protocols, then proposed a modified MAC protocol (MMAC) to enhance the most affecting factors of these two MAC standards such as energy consumption and latency. The performance analysis of these three MAC protocols is verified and simulated using a real model for Iraq, Baghdad, Al - Kadhumyah City, by Castalia 3.2 simulator also designed and implemented these three protocols using an iOS application by the XCode program for general application sensors especially for explosives detection sensors, Fido® X3 sensor proposed forthis application. Simulation results showed that the localization methods proved their ability to localize the failed SNs, but with impractical errors (about 23%) in results than the other approaches, and additional load effect on the network due to execute the localization programs separately before each node failure detection. The DFD approach applied with homogeneous WSNs only that contain only one type of sensors, its errors (starting from 25% for three SNs) was due to its algorithm limitations in using only half of the neighbor SNs, reduced in MDFD method algorithm in which all neighbor SNs considered to detect the failed SN reaching full detection accuracy but with latency tradeoff. The MAC approach is better than localization approach because there is no need to additional hardware for node failure detection; it is also better than DFD approach because it was applicable with both homogeneous and nonhomogeneous WSNs. The three MAC protocols succeeded to detect the failed SNs, but SMAC and TMAC had their tradeoff in energy consumption and latency, when SMAC had less energy consumption (69.62% than TMAC), while TMAC had less latency (27.58% than SMAC). MMAC is combining the advantages of both SMAC and TMAC that had a less energy consumption than that in SMAC by 54.16% and smaller latency than that in TMAC by 76.19% within a range of division index m : between 1 to 5.

مقترح بروتوكول توجيه الانترنيت الاشياء بالاعتماد على شبكات الاستشعار اللاسلكية == Proposed Routing Protocol For Internet of Things Based On Wireless Sensor Networks

Author name: عمر عبد الستار حماد
Supervisor name: ضياء جاسم كاظم
General topic: Electrical, Electronic and Communications Engineering
Specific topic: Electrical Engineering
Degree: Master
University: University of Baghdad - College Of Engineering - Department Of Electrical Engineering
Language: English
University location: Baghdad
First pages:
Abstract: The routing protocol designed for Internet of Things (IoT) systems has to be adapted with different applications requirements in order to enhance the performance of IoT applications. Therefore, the main objective of this thesis is proposing a routing protocol for wireless sensor network employed to serve IoT systems. A new IoT architecture was proposed with explaining the main functions of each layer forming the proposed architecture. The proposed routing protocol acted according to the tasks specified in this architecture. In addition, an IoT application was proposed to be applicable in some utility organizations. The link quality, node depth and energy were used as metrics to make routing decisions. In this work, different cases were proposed to exhibit the usage of these metrics and show the differences of using each of them, then examine different techniques which deal with link quality and the difference between using node - to - node link quality and end - to - end link quality. Improvements were added to the proposed routing protocol to achieve the perfect integration with suitable Media Access Control (MAC) protocol with taking into consideration the behavior of IoT application. Regarding this orientation, two approaches were suggested; the first includes the proposed routing protocol, application with high data rate and MAC protocol without sleeping capabilities; while the second approach included the proposed routing protocol, application with dynamic data rate and MAC protocol with sleeping capabilities. Comparison with other protocols was essential to show the improvements achieved by this work, thus protocols designed to serve the same purpose such as AODV, REL and LABILE were chosen to compare the proposed routing protocol with. To add integrative and holistic, some of important features such as actuating and mobility were added and tested. These features were greatly required by some of IoT applications and improving the routing protocol to support them made it more suitable for IoT systems. The proposed routing protocol was simulated using Castalia - 3.2 and all proposed approaches were examined to show the enhancement achieved by each approach. The results showed better performance when using end - to - end link quality. The results also showed the effect of MAC protocol if decreasing energy consumption was required. Sleeping MAC was the best choice when used with the proposed routing protocol while reducing the data rate. The proposed routing showed better performance than other protocols did regarding Packet Delivery Ratio (PDR) and latency. It preserved network reliability since it did not generate routing or data packets needlessly. Routing protocol with added features (actuating and mobility) showed good performance. But that performance was affected by increasing the speed of mobile nodes.

تصميم هوائي ضوئي يعمل ضمن نطاق تحت الحمراء الترددي باستخدام تقنية البلازمونك == Optical Antenna Design In Ir Range Using Plasmonic Technology

Author name: هادي كريم شمخي
Supervisor name: محمد ناظم عباس
General topic: Electrical, Electronic and Communications Engineering
Specific topic: Electrical Engineering
Degree: Master
University: University of Baghdad - College Of Engineering - Department Of Electrical Engineering
Language: English
University location: Baghdad
First pages:
Abstract: Optical antennas, resonant structures, which efficiently collect free - space light and focus it into a nanoscale volume using plasmonic, are indispensable in the burgeoning field of nanophotonics. In this work, resonance behaviors of the fundamental resonance mode of diabolo gold metal bar optical antennas were investigated. The relationship between the near - field enhancement, absorption, and scattering spectra of localized plasmonic elements was explored. It was found that as the waist of the diabolo gold metal bar optical antenna was reduced, optical energy absorption cross section and near field enhancement at resonance increased significantly. Also reduction of the diabolo waist width caused redshift of the resonant wavelengths in the spectra of absorption cross - section, scattering cross - section, and the near electric field. Oppositely, it was found that as the length of the diabolo gold metal bar optical antenna was reduced, optical energy absorption cross section and near field enhancement at resonance decreases significantly. Also reduction of the diabolo length caused blue - shift of the resonant wavelengths in the spectra of absorption cross - section, scattering cross - section, and the near electric field.Grating surface diabolo antenna design was proposed. It was found that grating superstructure perturbed the optical properties of SPPs (Surface Plasmon polarions) and enabled the control of resonance wavelength of the antenna. The grating reduced the power that flow out from the antenna and enhanced the localized electromagnetic fields intensity. The grating also increased surface plasmon currents because of the formation of bound surface states. As a result, the absorption spectra and near field intensity were enhanced but at the same time, the scattering spectra and the far field pattern were reduced.Titanium - Gold grating surface diabolo antenna was proposed. It was found that titanium metal strengthen the grating perturbation of the optical properties of SPPs and allowed wide range of wavelength shifting (3.5 ?m). Because Gold and Titanium have difference electrical resonance characteristics, less plasmon damping occurred, resulting in lower scattering efficiency and narrower plasmon line widths.Finally, system modelling and simulations is done by using electrical department super computer (24 parallel MP and 64 GB RAM) and COMSOL multiphysics 4.4 software.

تبؤ الاحمال الكهربائية للشبكة العراقية للمدة القصيرة باعتماد نظام المنطق الضبابي == Short - Term Electrical Load Forecasting For Iraqi Power System Based On Fuzzy Logic System

Author name: هدى منهي عبد العباس
Supervisor name: Firas Mohammed Tuaimah
General topic: Electrical, Electronic and Communications Engineering
Specific topic: Electrical Engineering
Degree: Master
University: University of Baghdad - College Of Engineering - Department Of Electrical Engineering
Language: English
University location: Baghdad
First pages:
Abstract: Load forecasting is used by participants in electric energy generation, transmission, distribution, and marketing for a variety of decision - making processes, such as economic dispatch, unit commitment, hydro - thermal coordination, transaction evaluation, and expansion planning. However, the need for accurate forecasts has intensified in the last decade due to the energy industry deregulation. Taking this into account as well as the rapid fluctuations in demand and abrupt changes in weather condition, access to reliable models for accurate forecast of load demand is essential. Due to the need for accurate load forecasts, numerous statistical and artificial intelligence methods have been proposed for the short - term load forecasting problem.In this study, the Multiple Linear Regression (MLR) method, which is one of the statistical methods, and an Interval Type - 2 Takagi - Sugeno - Kang Fuzzy Logic System (IT - 2 TSK FLS), which is one of the artificial methods and an extension of the conventional fuzzy logic system, were applied. Developed models were trained using the genetic algorithm. With the purpose of an objective assessment, the available dataset was split into training samples (80%, ????????????) and test samples (20%, ??????????). The training data used in this study covered the period from January 1, 2012 to February 1, 2012 for winter season and the period from July 1, 2012 to August 1, 2012 for summer season. The actual load forecasting period started from January 22, till 28, 2012 for winter model and from July 22 till 28, 2012 for summer model.This work suggested two models; the first model was for hourly (24 hour) load forecasting for one day ahead and the second model was for one week ahead in hourly forecasting (from one until 168 hour). For each model, winter and summer seasons were presented. The Main average percentage error (MAPE) term is an index that provides information about the bias of the model and how close forecasts or predictions are to the eventual outcomes. Experiments conducted with real datasets for the Iraqi power system showed that IT2 TSK FLS models precisely approximated future load demands with an acceptable accuracy.The real data for Iraqi power system were taken from Iraqi Operation and Control Office which belongs to the Ministry of Electricity.A computer program, written in MATLAB programming languages, was developed to represent the proposed method.

تحسين تخطيط المسار للروبوتات المتنقلة والمستقلة بالحركة باستخدام خوارزميات الامثلية القائمة على مبدا التجمعات == Improvement of Path Planning For Autonomous Mobile Robots Using Population - Based Optimization Algorithms

Author name: فرح مهدي علي
Supervisor name: نزار هادي عباس
General topic: Electrical, Electronic and Communications Engineering
Specific topic: Electrical Engineering
Degree: Master
University: University of Baghdad - College Of Engineering - Department Of Electrical Engineering
Language: English
University location: Baghdad
First pages:
Abstract: Autonomous mobile robots are used in various applications such as dangerous regions. These applications demand robust and adaptable methods for path planning.Path planning optimization problem is a fundamental problem for robot navigation; its purpose is to find a collision - free path from the initial position to a target position. Many real - world optimization problems have become increasingly large, complex and dynamic, which require the development of methods and solutions whose efficiency ismeasured by their ability to find acceptable results within a reasonable amount of time.The Artificial Bee Colony (ABC) and Bacterial Foraging Optimization (BFO) algorithms are used for solving hard optimization problems, including path planning. BFO algorithm is a simple and powerful searching technique. On the other hand, Artificial Potential Field (APF) as a real time obstacle avoidance method is quite simple in theory, but it is easy to bring local minimum problem and failing to find a path in the narrow channel. Therefore, a method for hybridizing APF with BFO is introduced to make use of both methods' advantages. This thesis presents efficient and reliable four optimization algorithms to solve the path planning problem. The first two algorithms are developed based on ABC algorithm as global path planning; they specifically are modified version of ABC algorithm called Directed ABC (DABC) algorithm and a new fitness function added to ABC algorithm named Minimum Angles (MAABC) algorithm was developed. The other two algorithms were proposed based on BFO and APF algorithms as local path planning; they were namely improved versions of BFO algorithm with adaptive step size (ASBFO) algorithm and a proposed version of BFO algorithm with an adaptive tumble (ATBFO) algorithm.The developed algorithms were simulated using MATLAB R2011b package. The simulation results showed that these algorithms could find the shortest paths for even crowded environments. Moreover, these algorithmswere compared with other researchers’ work to evaluate their performance. Additionally, a comparison was made between the executing time of the proposed local path planning algorithms (ASBFO and ATBFO algorithms) and the real tracked military robot (e.g., Talon) to verify the capability of the proposed methods. Both algorithms achieved fast run time, approximately three times faster than Talon for complex environment, which made them practically efficient.

تمثيل وتنفيذ العاكس ثلاثي الطور رباعي السيقان بواسطة طريقة المتجه الفضائي بتضمين عرض النبضة == Modeling And Implementation of Space Vector PWM Four - Leg Three Phase Inverter

Author name: لؤي كمال عبد القادر
Supervisor name: فاضل عباس مهدي القرملي
General topic: Electrical, Electronic and Communications Engineering
Specific topic: Electrical Engineering
Degree: Master
University: University of Baghdad - College Of Engineering - Department Of Electrical Engineering
Language: English
University location: Baghdad
First pages:
Abstract: The thesis considers a three - phase four - leg inverter, which is used to supply unequal loads without using a transformer. In the literatures, a three - dimensional modulation technique, requiring complex mathematical algorithms were proposed for this kind of inverter. In this research a different approaches are discussed through two kinds of controller; first type is based on the separation control method of the fourth - leg of the inverter from the other phases for one type for modulation. This research applied to the three - phase inverter of the traditional SVM techniques avoiding the employment of complex procedures. The voltage modulation method based on a triangular carrier waveform for the three - phase four - leg voltage source converter. The four - leg converter can produce three - phase output voltages independently from the additional forth - leg. The proposed modulation method for the four - leg inverter is implemented with a single carrier by a simple useful “offset voltage” concept. The second type is equivalent to the so - called three - dimensional space vector PWM method, but its implementation is much easier. The simultaneously synthesized equations of the maximum magnitude of unbalanced three - phase voltage.The feasibility of the proposed modulation technique types verified by computer simulation and experimental results. These results show that both types of “Carrier - Based Pulse Width Modulation (PWM) technique” and “Three Dimensional Space Vector Modulation Technique” are easier implemented than conventional type by using microcontroller component (PIC16F874A/877A) instead of digital components (DSP & FPGA). So this thesis offers a prototype for this circuit with low cost and low harmonics, but with more complex control circuit.

الموقع الامثل للمعوض التزامني الثابث لشبكة الضغط الفائق العراقية باستخدام الخوارزمية الجينية == Optimal Location of Static Synchronous Compensator (STATCOM) For Iraqi National (400kV) Super High Voltage Grid Using Genetic Algorithm

Author name: حسن علي عبد المجيد سلبي البياتي
Supervisor name: Firas Mohammed Tuaimah
General topic: Electrical, Electronic and Communications Engineering
Specific topic: Electrical Engineering
Degree: Master
University: University of Baghdad - College Of Engineering - Department Of Electrical Engineering
Language: English
University location: Baghdad
First pages:
Abstract: Transmission networks of modern power systems are becoming increasingly stressed because of the growing demand and restrictions on constructions new lines. One of the consequences of such a stressed system is the threat of losing stability following adisturbance. Flexible AC Transmission System (FACTS) devices are found to be very effective in a transmission network for better utilization of its existing facilities without sacrificing the desired stability margin.The Static Synchronous Compensator (STATCOM) is a shunt device which employs one of the latest technologies of FACTS and power electronic switching devices in electric power transmission systems to control the voltage and power flow. The STATCOMregulates the voltage at its terminal by controlling the amount of reactive power injected into or absorbed from the power system.Whilst the Iraqi National (400 kV) Super High Voltage Grid System suffers from voltage stability problems because of the high variation in the reactive power conditions all over the year seasons. Therefore, this thesis proposed an application of the STATCOM devices to maintain the voltages within the specified limits and enhancing voltage stability for all seasons in the year, as well as reducing the apparent power losses. And in order to reduce therequired installation cost of the STATCOM devices, this performance has been done by using minimum possible size of the reactive power injected or absorbed by the STATCOM devices, while satisfying the stability limits.The benefit of the STATCOM devices depends greatly on how these devices can be placed in the system. The general problem focused in this thesis was how to optimally determine the locations and the sizes of the STATCOM devices that to be installed. For thatreason, it was decided to follow an optimization approach such as a Genetic Algorithm (GA), which is one of the heuristic methods to find the optimum location and the value of installed STATCOM devices.The Newton - Raphson load flow method, with the proposed approach has been examined and tested first on the 5 - bus IEEE test system; the results obtained encouraged us to implement the same approach on the 27 - bus Iraqi National (400kV) SHV Grid System.The real data for Iraqi network have been taken from Iraqi National Control Center (INCC), in the Ministry of Electricity.computer program, written in MatLab environment, was developed to represent the proposed method.

تقنيات النمذجة الذكية في تنبوء حمل الطاقة الكهربائية == Intelligent Modeling Techniques of Electric Power Load Forecasting

Author name: يونس محي نصيف
Supervisor name: حنان ميخائيل داود
General topic: Electrical, Electronic and Communications Engineering
Specific topic: Electrical Engineering
Degree: Master
University: University of Baghdad - College Of Engineering - Department Of Electrical Engineering
Language: English
University location: Baghdad
First pages:
Abstract: As far as electrical power system is concerned, there has been a need to find out the future load in advance. Load forecasting has played an important role in the generation, transmission and distribution system planning. Load forecasting of future load demand is significant for an economic and secured operation of power systems. In general, the objective of high - precision load forecasting is difficult to achieve due to complex effects on load by a variety of factors.This thesis focuses on the study of long term load forecasting by using multiple linear regression (MLR) method and using different types of computational intelligence methods such as feed forward Neural Network with back - propagation (BP) tuning algorithm (FNN - BP), feed forward Neural Network with particle swarm optimization (PSO) tuning algorithm (FNN - PSO) and Elman Neural Network with back - propagation (BP) tuning algorithm (ENN - BP). Such forecasts will be highly useful in proper system planning and operations. The algorithms have been demonstrated using simulation programs in MATLAB environment.The usefulness of the five forecasting techniques was tested on a test system data for a Big Utility Company (Egyptian Unified Network). The obtained results showed that the ENN - PSO method takes advantage of accuracy and efficiency in prediction.The ENN - PSO method was performed using population growth data on the Iraqi National Grid for the duration from 2014 to 2030.

النظام المضبب للتحكم بمشكلة التزاحم لبروتوكول (التحكم في الارسال) في الشبكات المختلطة == A Fuzzy Based TCP Congestion Control For Hybrid Networks

Author name: سارة رعد قاسم
Supervisor name: زينب توفيق باقر
General topic: Electrical, Electronic and Communications Engineering
Specific topic: Electrical Engineering
Degree: Master
University: University of Baghdad - College Of Engineering - Department Of Electrical Engineering
Language: English
University location: Baghdad
First pages:
Abstract: Transmission Control Protocol (TCP) is designed to deliver seamless and reliable end - to - end data transfer across unreliable networks and is offered connection - oriented, byte - stream service. The Internet’s TCP is probably the most widely used protocol; it is also the most carefully tuned since performance of Internet is largely based on TCP. However, throughput in TCP degrades notably when operated in erroneous networks. In erroneous networks, TCP misinterprets random packet losses and the subsequent packet reordering due to high bit error rate as congestion and invokes congestion control by triggering to fast retransmission and fast recovery, leading to under - utilization of the network resources and affecting TCP performance critically.Many proposals have been used to improve TCP performance in wireless networks. Most of these proposals were based on changing the TCP header and add more complications to the main mechanism. Means of improving performance of TCP over erroneous links were proposed using fuzzy controller in a simple manner (without changing the header) by tracking the delay and the frequencies of the occurrences of timeouts and triple duplicate acknowledgement (3dupacks) to differentiate congestion loss from bit error loss. The first proposal was FReno (Fuzzy+NewReno) but this new algorithm failed to achieve good efficiency. The second proposal was FVegas (Fuzzy+Vegas) and it also failed to improve TCP’s resource utilization. The third proposal was FWestwood (Fuzzy+Westwood) which significantly improved TCP performance. The proposed algorithm was compared against standard TCP variants for different error rates using OMNET++4 version 4.3.1 IDE simulator. FWestwood algorithm helped TCP to distinguish between congestion and packet loss and increase the performance in erroneous wired network, heterogeneous networks, wireless networks and mobile network. Results from simulations showed that in a congested network with (1% to 10%) of random packet loss rate, FWestwood achieved significant improvements in different erroneous networks over other TCP schemes. Also it achieved friendliness fair share of the link with other TCP standards.

التوزيع الامثل للمتحسسات في شبكة المتحسسات اللاسلكية باستخدام خوارزمية الامثلية المستوحاة من نظرية الاسراب == Optimum Sensors Deployment In Wireless Sensor Networks Using Swarm Based Bio - Inspired Optimization Algorithms

Author name: هيثم سعدون عفتان
Supervisor name: نزار هادي عباس
General topic: Electrical, Electronic and Communications Engineering
Specific topic: Electrical Engineering
Degree: Master
University: University of Baghdad - College Of Engineering - Department Of Electrical Engineering
Language: English
University location: Baghdad
First pages:
Abstract: The increased demand for Wireless Sensor Networks (WSNs) in different areas of application has intensified studies dedicated to the deployment of sensor nodes in the recent past. The deployment of sensor nodes required that some of the key objectives should be satisfied, which are the coverage ratio of the monitoring area and the lifetime of the network.In this thesis, a mathematical model to optimize the coverage ratio and the lifetime of network is developed to ensure a better utility of the WSN. The model is formulated based on several parameters such as the size of the monitoring area, network models, the total number of sensor nodes, visibility requirements, sensing/communication radius, etc.Popular swarm based bio inspired algorithms have been used to optimize the WSN deployment. The coverage optimization process has been carried out by single objective optimization algorithms such as the Ant Colony Optimization (ACO) algorithm, modified version of Particle Swarm Optimization called Discrete PSO (DPSO), Discrete Artificial Bee Colony (DABC) algorithm and a new proposed algorithm called Quantum Artificial Bee Colony (QABC). QABC is based on quantum physical concepts to improve the local search capability of standard ABC algorithm. Thereafter, a multi objective optimization algorithm has been utilized to optimize the coverage ratio and lifetime of WSN. The multi objective optimization has been carried out by Non - dominated Sorting Genetic Algorithm (NSGA - II). All of these algorithms are simple, effective and computationally efficient optimization techniques.The WSN deployment has been simulated using MATLAB 7.12.0 (R2011a) package, NetBeans 7.4 Java integrated development environments, JMETAL 4.5 and Java Universal Network/Graph (JUNG 2.0.1) frameworks. The computer simulation results showed that the proposed algorithm for coverage ratio maximization was up to two times faster than the others. Furthermore, the conducted simulation indicated that the QABC algorithm offered (6%) better solution in terms of coverage in comparison with the others in some cases. Also the results showed that the deterministic deployment can achieve up to (25%) better coverage ratio than the random deployment. In addition, QABC outperformed GA, PSO and ABC algorithms when applied to several test problems. Additionally, the results showed that the NSGA - II algorithm could effectively optimize the network lifetime and coverage ratio and produced good convergent solutions to the Pareto front and was uniformly distributed along it.
1 ... 9 10 11 12 13 ... 18