Skip to main content
main-content

Über dieses Buch

The use of artificial intelligence, especially in the field of optimization is increasing day by day. The purpose of this book is to explore the possibility of using different kinds of optimization algorithms to advance and enhance the tools used for computer and electrical engineering purposes.

Inhaltsverzeichnis

Frontmatter

Preface

Abstract
Meta-heuristics are one of the most significant decision-making methods concerning especially in the fields of computer and electrical engineering. Most of the real-world optimizations are highly multimodal and nonlinear, under various complex constraints. Different objectives are often conflicting. Even for a single objective, sometimes, optimal solutions may not exist at all. In general, finding an optimal solution or even sub-optimal solutions is not an easy task.
Navid Razmjooy

Jaya Algorithm and Applications: A Comprehensive Review

Abstract
All the population-based optimization algorithms are probabilistic algorithms and require only a few control parameters including number of candidates, number of iterations, elite size, etc. Besides these control parameters, algorithm-specific control parameters are required by different algorithms. Either the computational effort is increased or the local optimal solution is yielded as a result of the improper tuning of algorithm-specific parameters. Thus, the algorithm which works without any algorithm-specific parameters such as Jaya Algorithm (JA) is widespread among the optimization applications and researchers. JA can be easily implemented based on not having to tune any algorithm-specific parameters. In order to prove that JA could be applied to every problem arising in practice, this study presents a comprehensive review of the advances with JA and its variants. On the other hand, the performance of JA was evaluated to solve the Himmelblau function against five well-known optimization algorithms. Therefore, this study is expected to highlight the JA’s capabilities and performances especially for those researchers who are eager to explore the algorithm.
Essam H. Houssein, Ahmed G. Gad, Yaser M. Wazery

World Cup Optimization Algorithm: Application for Optimal Control of Pitch Angle in Hybrid Renewable PV/Wind Energy System

Abstract
In the last decades because of the subsequent rise in the prices of fossil fuels, the application of sustainable energies like solar energy and wind energy has been increased. A combined energy generation system containing a wind turbine and photovoltaic system is studied. Such a hybrid system has more efficiency from the single wind turbine or a single PV array. In this study, the Maximum Power Point Tracking (MPPT) system is adopted to get the maximum possible power from the PV array. The control of the pitch angle in the wind turbine is a popular appliance to adjust the aerodynamic torque of the turbine once the wind speed is superior to the considered speed ratio. Formally, pitch control systems usually employ a PI controller which needs a precise mathematical model of the system. Here, a new applicable optimization algorithm is utilized to optimize the pitch angle controller. Test and validation studies with the proposed algorithm are compared with GA and results are presented with a good agreement.
Navid Razmjooy, Vania V. Estrela, Reinaldo Padilha, Ana Carolina Borges Monteiro

Optimization Techniques in Intelligent Transportation Systems

Abstract
Intelligent Transportation Systems (ITS) refer to a range of transportation applications based on communication and information technology. These systems by the aid of modern ideas, provide comfortable, efficient and safe services for transportation users. They are located in the linkage of information technology, computer science, electrical engineering, system analysis, civil engineering, and optimization. They form a main branch of the smart cities and are fundamental for the development of countries. ITS applications, usually, use the capabilities of sensor networks, electrical devices, and computer processing units to deliver a service, however, they are not limited to the hardware devices. Instead, the modeling of transportation problems and solving them efficiently are more challenging problems. Especially, when they support good ideas for controlling the transportation systems or guiding some users. In this chapter, the application of optimization models for the transportation context are discussed. To this end, the models for data collection by a sensor network are needed. Then, for mining these data and extracting the necessary knowledge for transportation context awareness, some fundamental models such as regression analysis, frequent pattern mining, clustering or classification can be applied. These backgrounds are used to extend the appropriate models of ITS in an integrated architecture. To solve these models, the classical network and combinatorial optimization methods, simulation-optimization techniques, and metaheuristic algorithms will be explained. Then, we categorize the applications of these optimization models in the different subsystems of ITS architecture. The output of this investigation can be used to develop different ITS services for the urban and inter-cities networks.
Mehdi Ghatee

Low Power Hilbert Transformer Design Using Multi-objective Seeker Optimization Algorithm

Abstract
In this paper, a novel design approach using multi-objective evolutionary seeker optimization algorithm has been proposed, in which the Hilbert transformer is designed using half band FIR filter. The proposed technique has been analyzed for Hilbert transformer model in terms of reducing the power consumption, pass-band error and order simultaneously. The inclusion of power minimization makes the designed Hilbert transformer portable, low power devices thus increasing battery life and less heating effect. The pertinence of the proposed technique was analyzed by comparing the results achieved using the proposed algorithm with other state of the art evolutionary multi-objective algorithms. Using Virtex-7 FPGA and Xilinx X-power analyzer, Power consumption was analyzed. In the present work, a novel EA i.e. hybrid artificial bee colony algorithm has been proposed and further applied for FIR filter design. The filter design task aims at satisfying the dual objectives of meeting the desired frequency domain specifications and power minimization.
Atul Kumar Dwivedi

Metaheuristics Applied to Blood Image Analysis

Abstract
The growing use of digital image processing techniques focused on health is explicit, helping in the solution and improvements in diagnosis, as well as the possibility of creating new diagnostic methods. The blood count is the most required laboratory medical examination, as it is the first examination made to analyze the general clinical picture of any patient, due to its ability to detect diseases, but its cost can be considered inaccessible to populations of less favored countries. In short, a metaheuristic is a heuristic method for generally solving optimization problems, usually in the area of combinatorial optimization, which is usually applied to problems for which no efficient algorithm is known. Digital Image Processing allows the analysis of an image in the various regions, as well as extract quantitative information from the image; perform measurements impossible to obtain manually; enable the integration of various types of data. Metaheuristic techniques have come to be great tools for image segmentation for digitally segmenting containing red blood cells, leukocytes, and platelets under detection and counting optics. Metaheuristics will benefit to computational blood image analysis but still face challenges as cyber-physical systems evolve, and more efficient big data methodologies arrive.
Ana Carolina Borges Monteiro, Reinaldo Padilha França, Vania V. Estrela, Navid Razmjooy, Yuzo Iano, Pablo David Minango Negrete

Optimal Bidding Strategy for Power Market Based on Improved World Cup Optimization Algorithm

Abstract
Power companies in the world-wide have been restructuring their electric power systems from a vertically integrated entity to a deregulated and open-market environment. In the past, electric utilities usually look for maximizing the social welfare of the system with distributional equity as their main operational criterion. The operating paradigm was based on achieving the least-cost system solution while meeting reliability and security margins. This often resulted in investments in generating capacity operating at very low capacity factors. Decommissioning of this type of generating capacity was a natural outcome when the vertically integrated utilities moved over to deregulated market operations. This paper proposes an optimizing base and load demand relative binding strategy for generating the power pricing of different units in the investigated system. Afterward, the congestion effect in this biding strategy is investigated. The described systems analysis is implemented on 5 and 9 bus systems and the optimizing technique in this issue is a new improved version of the world cup optimization algorithm. Simulation results have been compared with the standard world cup optimization algorithm. Finally, examined systems are simulated by using the Power World software. Experimental results show that the proposed technique has a good superiority compared with the world cup optimization algorithm for congestion management purposes.
Navid Razmjooy, Anand Deshpande, Mohsen Khalilpour, Vania V. Estrela, Reinaldo Padilha, Ana Carolina Borges Monteiro

Speed Control of a DC Motor Using PID Controller Based on Improved Whale Optimization Algorithm

Abstract
In this paper, a new optimized method is introduced for the optimal control of a DC motor based on a proportional-integral-derivative (PID) controller. In this study, an improved version of the whale optimization algorithm has been adopted for optimal selection of the PID controller parameters for optimal control of the DC motor speed along with minimum settling time. Unlike the other control algorithms, the PID controller can give more accurate and stable control by tuning the process outputs based on the history and rate of change of the error signal. The proposed approach has a premier specification, including easy application, stable convergence characteristics and high-efficiency computational performances. The DC motor designing by optimized PID controller is modeled based on the MATLAB platform. The results of the proposed method are compared with the standard whale optimization algorithm to show the proposed algorithm’s efficiency. Final results show that the proposed approach is better in improving the speed loop response stability, the steady-state error is decreased, and the disturbances do not affect the performances of driving motor with no overtaking.
Navid Razmjooy, Zahra Vahedi, Vania V. Estrela, Reinaldo Padilha, Ana Carolina Borges Monteiro

Skin Color Segmentation Based on Artificial Neural Network Improved by a Modified Grasshopper Optimization Algorithm

Abstract
One of the applications of image processing and computer vision is to detect the skin regions for a wide range of human–computer interaction and content based utilizations. Detecting nude parts in the movies, face detection, tracking of human body parts, and people recognizing in multimedia databases are a small part of its applications. Therefore, designing an efficient technique for skin area detection can help a lot to the determined applications. The grasshopper optimization algorithm is a new optimization algorithm evolutionary algorithm which is recently introduced to solve optimization problems. The main purpose of this paper is to propose a newly developed version of this algorithm to optimize the weights of the backpropagation neural network to design a good segmentation tool for skin area segmentation. The method has been compared with the traditional multi-layer perception and the ICA-MLP to declare the proposed method’s efficiency.
Navid Razmjooy, Saeid Razmjooy, Zahra Vahedi, Vania V. Estrela, Gabriel Gomes de Oliveira

A New Design for Robust Control of Power System Stabilizer Based on Moth Search Algorithm

Abstract
This paper presents a new optimal design for the stability and control of the synchronous machine connected to an infinite bus. The model of the synchronous machine is 4th order linear Philips-Heffron synchronous machine. In this study, a PID controller is utilized for stability and its parameters have been achieved optimally by minimizing a fitness function to removes the unstable Eigen-values to the left-hand side of the imaginary axis. The considered parameters of the PID controller are optimized based on a new nature-inspired, called moth search algorithm. The proposed system is then compared with the particle swarm optimization as a high-performance and popular algorithm for different operating points. Final results show that using a moth search algorithm gives better efficiency toward the compared particle swarm optimization.
Navid Razmjooy, Saeid Razmjooy, Zahra Vahedi, Vania V. Estrela, Gabriel Gomes de Oliveira

Design of Self-adaptive Fuzzy Logic Droop Controller for Hybrid Units in Islanded Microgrids

Abstract
In the present paper, a Multi-Segment P/f Droop Control Strategy is developed to achieve localized control, coordination and power management of different sources in an islanded microgrid which consists of two PV/Battery storage Hybrid units and a droop unit. Separate control loops are designed, each for PV power production control, Battery charging and discharging level control, and Droop unit operation control based on multi-segment P/f Droop characteristics. The conventional PI controllers can be used in the control loop design as they are very simple for implementation and give a better dynamic response. But their performance deteriorates when the complexity in the system increases due to disturbances like load variations and intermittent sources. Fuzzy Logic Control (FLC) technique is model-independent and a flexible tool. It can deal with complex systems such as microgrids with different types of imprecise inputs and variables particularly if the power is supplied by intermittent sources and is also consumed by varying and unpredictable loads during sudden disturbances. The Membership functions (MF) of a conventional FLC are determined by trial and error method. This method is time-consuming and does not guarantee an optimal controller. In the present paper, a Self-Adaptive Fuzzy Logic Droop Controller (SAFLDC) is proposed, whose input–output MFs are made adaptive by obtaining the cluster centers using Self-Organizing Maps (SOM) Algorithm. The paper aims to follow a systematic approach to propose an effective SAFLDC with optimal MFs whose parameters are tuned using Hybrid GA-PSO. The developed strategy has been validated using detailed switching models in MATLAB/SIMULINK tool and the behavior of key variables in the strategy is illustrated.
V. S. Vakula, B. Damodar Rao

Skin Melanoma Segmentation Using Neural Networks Optimized by Quantum Invasive Weed Optimization Algorithm

Abstract
Early detection of skin cancer makes a high chance for the patient to escape from the malady and cure him/her at initial stages. In other words, by early detection of skin cancer, the quality of human life improves. In recent years, a wide range of dermatology clinics and hospitals employed systems based on image processing and computer vision for early detection of skin cancer. In this paper, a new method based on the optimized artificial neural network is presented to recognize the malignant lesion of skin cancer from benign lesions. To do this purpose, at first, a number of pre-processing operations are applied to the input image to filter noise and unwanted parts. Afterward, the proposed optimized neural network based on Quantum Invasive Weed Optimization Algorithm is applied to the filtered image for separating the skin lesion regions. To analysis the system performance, it has been applied to the DermIS Database and Dermquest Database. Experimental results show that the proposed method has a good efficiency for the skin lesion segmentation.
Navid Razmjooy, Saeid Razmjooy

A Computational Intelligence Perspective on Multimodal Image Registration for Unmanned Aerial Vehicles (UAVs)

Abstract
Remote Sensing (RS) applications generally require robustness, stability, accuracy, promptness, and a high autonomy level to simplify the Big Data (BD) processing in real-time. Image Registration (ImR) is among the most employed RS tasks. ImR transforms different groups of images into a coordinate system that allows overlaying two or more images from the same scene acquired with various sensors and/or taken at different times and angles. The original imageries must be normalized and geometrically aligned to create an ample image containing information from all the separate images. ImR is a crucial step when one has several views and a myriad of sensors that must be fused. The BD aspect of Multimodal Image Registration (MIR) is related to the idea of multispectral and hyperspectral imaging, which involve a vast amount of frequency bands. BD from different sources assist the decision-making processes and create additional more massive datasets for the long-term tracking of various phenomena. This chapter focuses on the MIR from infrared and optical sensors relying on the Particle Swarm Optimization (PSO) class of algorithms. These computational intelligence procedures circumvent problems related to multiresolution methods and the high computational cost of hard optimization methods.
Vania V. Estrela, Navid Razmjooy, Ana Carolina Borges Monteiro, Reinaldo Padilha França, Maria A. de Jesus, Yuzo Iano

Using Metaheuristics in Discrete-Event Simulation

Abstract
Metaheuristics have proven to be a powerful tool for roughly solving optimization problems, applied to find answers to problems about which there is little information. In general, meta-heuristics use a combination of random choices and historical knowledge of the previous results acquired by the method to guide and search the search space in neighborhoods within the search space, avoiding premature stoppages in optimal locations. A strategy that guides or modifies a heuristic to produce solutions that surpass the quality of those commonly encountered. The discrete event simulation (DES) is a representation of a system as a sequence of operations by state transactions (entities), where these entities are discrete and may be relative to various types depending on the context of the problem that is being sought. In this way is brought to the eyes of interest, the union of discrete events simulation with metaheuristic science, whether direct or not, is successful.
Reinaldo Padilha França, Ana Carolina Borges Monteiro, Vania V. Estrela, Navid Razmjooy

An AWGN Channel Data Transmission Proposal Using Discrete Events for Cloud and Big Data Environments Using Metaheuristic Fundamentals

Abstract
Cloud computing consists of providing computing services and re-sources, including servers, storage, databases, networking, software, analytics, and intelligence, over the Internet (“the cloud”) to deliver faster innovation, flexible capabilities, and economies of scale. Big Data refers to a large amount of data, coming from controlled or uncontrolled sources, in accordance with structured or unstructured data, its potential comes from this volume where through its analysis it allows extracting insights from analyzes of this information. Metaheuristics have proven to be a powerful tool for roughly solving optimization problems, applied to find answers to problems about which there is little information. This research aims to propose modeling to improve the transmission of content in wireless communication systems, employing the pre-coding process of bits based on the application of a discrete event in signals before the modulation process, using DQPSK modulation in an AWGN channel. The results show improvement achieving 74.61% in memory utilization and 131.29% at runtime with respect to information compression.
Reinaldo Padilha, Yuzo Iano, Ana Carolina Borges Monteiro, Rangel Arthur
Weitere Informationen