Genetic Algorithm Optimization Technique Research Articles

A beamforming comparative study of least mean square, genetic algorithm and grey wolf optimization algorithms for multipath smart antenna system

Multipath environment is a limitation fact in optimized usage of wireless networks. Using smart antenna and beamforming algorithms contributed to that subscribers get a higher-gain signal and better directivity as well as reduce the consumed power for users and the mobile base stations by adjusting the appropriate weights for each element in the antenna array that leads to reducing interference anddirecting the main beam to wanted user. In this paper, the performance of three of beamforming algorithms in multipath environment in terms of directivity and side lobe level reduction has been studied and compared, which are least mean square (LMS), genetic algorithm (GA) and grey wolf optimization (GWO) technique. The simulation result appears that LMS algorithm aids us to get the best directivity followed by the GWO, and we may get most sidelobe level reduction by using the GA algorithm, followed by LMS algorithm in second rank.

Design, modelling and simulation of controlled sepic DC-DC converter-based genetic algorithm

This paper discusses various aspects of a single-ended primary inductance DC-DC converter (SEPIC). The focus is on design, modelling, and simulation results of a SEPIC converter. The study analyses the principle of SEPIC operation when operated in continuous conduction mode (CCM). Additionally, the mathematical equations for the design modules are calculated as per converter requirements. State-space equations are used to formulate the state-space model of the SEPIC converter. To satisfy the best-performance criterion of the system, the parameters for controller (K<sub>p</sub>, K<sub>i</sub>, K<sub>d</sub>) should be tuned or optimized using the genetic algorithm (GA) optimization technique. Controller parameters are determined using an objective function that minimises the integral time absolute error (ITAE). Simulations performed on a closed-loop system reveal that the step response with a PID controlled based GA displayed superior performance. A closed-loop system has a substantially bigger stability region compared to an open-loop system. The simulation optimised performance metrics like maximum overshoot percentage (M<sub>p</sub>), rise time (t<sub>r</sub>), and settling time (t<sub>s</sub>). MATLAB/Simulink R2018a® and m-file code are used for the system modelling, simulation, and optimization of the PID controller parameters based on the GA.

Significant improvement of infrared graphene nanoribbon phototransistor performance: A quantum simulation study

In this paper, we investigate an optimized design of textured graphene nanoribbon phototransistor via quantum simulations. The graphene layer with textured morphology performs dual role, notably as the carrier conduction channel and also as an absorber of light. The proposed design characteristics are numerically investigated via a non-equilibrium Green’s function (NEGF) mode-space formulism in the ballistic edge. The method suggested was tailored to the modern carrier transport model, Electron-photon and Electron-phonon interactions. Besides, a simple analytical model that considers graphene light trapping morphology is developed and introduced into the simulation. This model is regarded as a fitness function for Multi-Objective Genetic Algorithm (MOGA) optimization technique to improve the GNR phototransistor sensing capability. Simulations results show that, compared to a conventional phototransistor, the proposed design has higher responsivity (47.55 mA/W), good sensitivity (2.41 × 103), higher detectivity (6.09 × 1010 Jones), better Photocurrent to Dark Current Ratio (PDCR) (24.15) and keeping better scaling capability.

Design of a compact, wideband, and flexible rhombic antenna using CMA for WBAN/WLAN and 5G applications

AbstractThe design of a compact, wideband flexible monopole antenna printed on a flexible Kapton polyimide substrate is presented. The proposed antenna consists of two rhombic elements embedded within a square plane as the radiating element. The wide bandwidth is aimed at alleviating potential performance degradations due to its application in an on‐body environment. The proposed antenna is designed using two main methods: the method of moments with adaptive cross approximation, and the characteristic mode analysis. The antenna operates from 2.038 to 3.745 GHz, indicating a bandwidth of 1.707 GHz (59.03%). Two dominant resonant frequencies are observed at 2.45 and 3.42 GHz, with realized gain of 3 and 5 dBi, and radiation efficiency of 90% and 83%, respectively. Besides that, the genetic algorithm and particle swarm optimization techniques are used to optimize the antenna and its performance validated via measurements. Besides, in free space, its on‐body performance when placed in proximity of a human body phantom is evaluated. Finally, specific absorption rate assessment is performed to ensure the safety of the proposed antenna when operating on human. Results indicate the antenna's suitability for wearable applications to support different wireless standards.

A combination of fuzzy Delphi method and hybrid ANN-based systems to forecast ground vibration resulting from blasting

This study presents a new input parameter selection and modeling procedure in order to control and predict peak particle velocity (PPV) values induced by mine blasting. The first part of this study was performed through the use of fuzzy Delphi method (FDM) to identify the key input variables with the deepest influence on PPV based on the experts’ opinions. Then, in the second part, the most effective parameters on PPV were selected to be applied in hybrid artificial neural network (ANN)-based models i.e., genetic algorithm (GA)-ANN, particle swarm optimization (PSO)-ANN, imperialism competitive algorithm (ICA)-ANN, artificial bee colony (ABC)-ANN and firefly algorithm (FA)-ANN for the prediction of PPV. Many hybrid ANN-based models were constructed according to the most influential parameters of GA, PSO, ICA, ABC and FA optimization techniques and 5 hybrid ANN-based models were proposed to predict PPVs induced by blasting. Through simple ranking technique, the best hybrid model was selected. The obtained results revealed that the FA-ANN model is able to offer higher accuracy level for PPV prediction compared to other implemented hybrid models. Coefficient of determination (R2) results of (0.8831, 0.8995, 0.9043, 0.9095 and 0.9133) and (0.8657, 0.8749, 0.8850, 0.9094 and 0.9097) were obtained for train and test stages of GA-ANN, PSO-ANN, ICA-ANN, ABC-ANN and FA-ANN, respectively. The results showed that all hybrid models can be used to solve PPV problem, however, when the highest prediction performance is needed, the hybrid FA-ANN model would be the best choice.

A Comparative Analysis of Data-Driven Based Optimization Models for Energy-Efficient Buildings

Energy-efficient and sustainable buildings have become a specialized research theme for energy planning and the green building industry. Rapidly increased energy consumption in the residential area has attracted researchers' attention to minimize energy consumption without affecting occupant comfort. Accordingly, an effective energy management system is required for energy-efficient buildings to utilize energy resources efficiently and maintain desired comfort. In this paper, data-driven based optimization models are compared to maximize the comfort index and minimize energy consumption simultaneously. Heating-cooling systems are used to maintain indoor thermal comfort, which consumes energy according to the environmental temperature and indoor temperature difference. Therefore, the environmental temperature parameter is optimized using optimization techniques Genetic Algorithm (GA), Bat, Neural Network Algorithm (NNA), Particle Swarm Optimization (PSO), and Artificial Bee Colony (ABC). The main objective of the optimization is to reduce the gap/error between the temperature specified by the user and the environmental temperature. Afterward, the discrepancy between optimized temperature and actual temperature is fed to the designed ML (Machine Learning) based controller. The controller's output is further provided to coordinator agents, which give the power to the actuators accordingly. The linearity of the proposed model improves the performance of the ML algorithm. The obtained dataset from fuzzy temperature controller has been used to design an ML controller. The experimental results show that the designed system significantly improves energy efficiency and occupant comfort for energy-efficient buildings. The BAT model has shown effectiveness in achieving a high comfort index with minimum power consumption compared to other considered models.

Preparation of Sustainable Thermal Insulators from Waste Materials

This work aims to produce an experimental and theoretical analysis of thermal insulator specifications for buildings with sustainable requirements. In the experimental work, three categories of thermal insulators were prepared from composite materials, and each category had ten models. These composites included the addition of two types of waste (sawdust and tyre waste) as fillers for two types of matrices (liquid polyurethane and polyurethane foam) to obtain composite materials for thermal insulation samples. The prepared samples were subjected to tests to show their thermal properties, such as thermal conductivity and specific heat capacity as well as undergoing a hardness test. The theoretical analysis included the discovery of empirical equations for thermal properties as functions of two variables (temperature and mass ratio) and hardness as a function of one variable (mass ratio). A genetic algorithm optimisation technique was used to find the optimum mass ratio of the composite that produced the required insulation specifications. The results showed that thermal conductivity decreased when the sawdust mass ratio and the rubber waste mass ratio increased but remained under the thermal insulation range. Furthermore, the prepared insulator samples showed an improvement in thermal storage and the hardness of tyre waste (liquid polyurethane composites) and sawdust (polyurethane foam composites). Finally, optimum results were obtained using the optimisation technique.

Optimal placement of protective devices and switches in a radial distribution system with distributed generation

Optimal placement of protective devices and switches in the distribution system improves the system reliability at the cost of increased investment. In this study, a new bi-directional analytical model has been developed to solve the problem of optimal placement of protective devices and switches in various zones of a distribution system with distributed generation (DG). The model determines the optimal number and locations of reclosers, switches, ‘fuse-save fuses’ and ‘fuse-blow fuses’ for increasing the system reliability while reducing the investment and outage costs, considering system data uncertainties. For considering uncertainties in load data, failure rates (temporary and permanent) and repair rates, the model utilises three-point estimate method. The formulated problem has been solved for 69-bus distribution system and radial distribution system connected to bus 4 of the reliability busbar test system, using genetic algorithm optimisation technique. After analysing the test results, it is concluded that considerable profit to the utility is achieved by optimal placement of protective devices and switches in various zones of the distribution system with DG units.

Optimization of Specific Power of Surface Mounted Axial Flux Permanent Magnet Brushless DC Motor for Electrical Vehicle Application

Optimization of specific power of axial flux permanent magnet brushless DC (PMBLDC) motor based on genetic algorithm optimization technique for an electric vehicle application is presented. Double rotor sandwiched stator topology of axial flux permanent magnet brushless DC motor is selected considering its best suitability in electric vehicle applications. Rating of electric motor is determined based on vehicular dynamics and application needs. Double rotor sandwiched stator axial flux PMBLDC motor is designed considering various assumed design variables. Initially designed axial flux PMBLDC motor is considered as a reference motor for further analysis. Optimization of the specific power of electric motor for electric vehicle applications is a very important design issue. The Genetic Algorithm (GA) based optimization technique is proposed for optimization of specific power of axial flux permanent magnet brushless DC motor. Optimization with an objective of maximum specific power with the same torque rating is performed. Three-dimensional finite element analysis is performed to validate the proposed GA based specific power optimization. Close agreement between results obtained from finite element analysis and analytical design establishes the correctness of the proposed optimization technique. The performance of the improved motor is compared with the initially designed reference motor. It is analyzed that the specific power of axial flux PMBLDC motor is enhanced effectively with the application of GA based design optimization technique.

Optimal metaheuristic-based sliding mode control of VSC-HVDC transmission systems

The design of classical controllers for Voltage Source Converter High Voltage Direct Current (VSC-HVDC) transmission systems, is load-dependent and has to be adjusted for each operating condition. Thus, the robustness of such controllers becomes necessary to cope with operating condition continuous variations. Therefore, the design of hybrid optimal Artificial Intelligence Based-Sliding Mode Controllers (AI-SMCs) for VSCHVDC transmission systems is crucial research interest. These AI based controllers are proved to improve the system’s dynamic stability over a wide range of operating conditions considering different parameter variations and disturbances. For this purpose, a comprehensive state of the art of the VSC-HVDC stabilization dilemma is discussed. The nonlinear VSC-HVDC model is developed. The problem of designing a nonlinear feedback control scheme via two control strategies is addressed seeking a better performance. For ensuring robustness and chattering free behavior, the conventional SMC (C-SMC) scheme is realized using a boundary layer hyperbolic tangent function for the sliding surface. Then, the Modified Genetic Algorithm (MGA) and Particle Swarm Optimization technique (PSO) are employed for determining the optimal gains for such SMC methodology forming a modified nonlinear MGA-SMC and PSO-SMC control in order to conveniently stabilize the system and enhance its performance. The simulation results verify the enhanced performance of the VSC-HVDC transmission system controlled by both MGA-SMC and PSO-SMC compared to the C-SMC. The comparative dynamic behavior analysis for both the conventional SMC and the two meta-heuristic optimization based SMC control schemes are presented. Through simulation results, the effectiveness of the proposed metaheuristic optimization approaches and their applicability to VSC-HVDC system global stabilization and dynamic behavior enhancement are validated.

Comparison of Genetic Algorithm and Taguchi Optimization Techniques for Surface Roughness of Natural Fiber-Reinforced Polymer Composites

Comparison of Genetic Algorithm and Taguchi Optimization Techniques for Surface Roughness of Natural Fiber-Reinforced Polymer Composites

Optimal working conditions of various city gate stations for power and hydrogen production based on energy and eco-exergy analysis

In this study, the feasibility of producing power and hydrogen from the waste heat of different City Gate Stations (CGSs) is investigated to select the optimal working conditions. A thermodynamic model is developed for a proposed system combined of the CGS station, the Rankin cycle and the extended hydrogen production cycle. Initially, six CGS stations are simulated based on energy, exergy-economic and environmental analysis and then a comparative study is conducted between different stations. The results of numerical modeling show that the Mashhad-old station with 5315 kW and 31.062 ton/year has the highest amount of power and hydrogen production among other stations, respectively. It is also observed that, it is more economic to increase the input gas pressure in order to increase the production rate. In addition, optimal working conditions are determined based on the two important optimization factors of the hydrogen production rate and SUCP (sum unit cost of the product) using genetic algorithm optimization technique. The results of multi-objective optimization indicate that Gonbad, Gorgan and Mashhad-old stations, where the inlet gas mass flow rate is in the range of 8–9 kg/s, are the optimum stations.

Novel Exploration Coefficient Update for the Grey Wolf Optimizer

Avoiding stagnation at local optimum values is one of the greatest challenges faced by computational intelligence techniques when solving nonconvex optimization problems. The transition between global and local search may not be effective and can compromise the performance of optimization algorithms. This work presents a novel manner to update the exploration coefficient of the meta-heuristic known as grey wolf optimizer (GWO), by replacing the linear update of the exploration coefficient by a triangular-shaped function, enabling the algorithm to escape from local optima. In order to validate the proposed grey wolf optimizer (PGWO) methodology, its performance is compared to the original version of GWO and its chaotic version, as well as to the well-known genetic algorithm, bat algorithm and particle swarm optimization techniques, in solving 10 nonconvex benchmark functions. Also, in order to verify the proposed methodology’s ability in solving a more realistic engineering problem, the authors implemented the PGWO to solve the wind farm layout optimization (WFLO) problem, which is a large-sized optimization problem, of combinatorial nature and nonconvex solution region. The results indicate that the PGWO improved the performance of the original GWO, as well as all investigated methodologies for the benchmark functions optimization and for the WFLO problem.

Analysis and Optimization of Composite Kagome Grid Panels Subjected to the Low Velocity Impact

By the present paper, an optimal volume technique of stiffened laminated panel subjected to Low Velocity Impact (LVI) is developed using an analytical model that couples a genetic algorithm optimization technique. Using the first-order transverse-shear deformation theory (FSDT) for anisotropic plates, the stiffness of the Kagome grid stiffeners and the plate are estimated by considering both effects of moments and forces in a basic repetitive cell. Where, the stiffness of the equivalent plate is computed by superposing stiffness the plate and Kagome stiffeners. The contact impactor/panel is idealized by the mass/spring model. The convergence of the impact force and the transverse displacement response at the contact point are plotted. Using the genetic algorithm (GA), the minimization of the panel volume is achieved in two stages. In the first stage, the design variables are the mechanical properties of the equivalent plate and the impactor. But, in the second stage, the stacking sequences were considered as design variables. Finally a parametric study is made and major conclusions are deduced regarding to the panel volume, time and impact velocity on behavior of the stiffened panel. The results show that the increase in stiffeners thickness is traduced by smaller transverse displacement involving a better impact performance, best optimization and panel integrity.

Genetic Algorithm and Particle Swarm Optimization Techniques for Inverted Pendulum Stabilization

Inverted Pendulum is a popular non-linear, unstable control problem where implementation of stabilizing the pole angle deviation, along with cart positioning is done by using novel control strategies. Soft computing techniques are applied for getting optimal results. The evolutionary computation forms the key research area for adaptation and optimization. The approach of finding optimal or near optimal solutions to the problem is based on natural evolution in evolutionary computation. The genetic algorithm is a method based on biological evolution and natural selection for solving both constrained and unconstrained problems. Particle swarm optimization is a stochastic search method inspired by collective behavior of animals like flocking of birds, schooling of fishes, swarming of bees etc. that is suited to continuous variable problems. These methods are applied to the inverted pendulum problem and their performance studied.

Optimization of a T-Shaped MIMO Antenna for Reduction of EMI

In this paper, optimization of a miniaturized multiple-inputs multiple-outputs (MIMO) antenna was performed. This antenna was composed of a T-shape radiating element with stub and reduced ground plane and a compact size of 25 mm × 25 mm × 1.6 mm. The behavior of antenna was evaluated in terms of return loss (S-parameter < −10 dB), electromagnetic interference (EMI), and operation frequency. The antenna design is applicable to many applications. A 2K factorial design combined with a genetic algorithm (GA) optimization technique were used to identify the key design parameters responsible for affecting the performance quality of the antenna. Optimization of the antenna design for EMI reduction was utilized, and the optimal design showed enhanced bandwidth of the antenna and reduced power consumption.

A good contribution of computational fluid dynamics (CFD) and GA-ANN methods to find the best type of helical wire inserted tube in heat exchangers

Researchers from all over the places have been reporting their experimental and simulation studies on thermal analysis of enhanced heat-exchangers, such as coiled wire inserts. If those researchers decide to optimize the heat-exchangers' performances under specific conditions, inevitably they will be in need of extracting an empirical or semi-empirical correlation to facilitate their calculations. Developing a correlation for a complex thermal system with loads of variables such as coiled wire inserts is truly hard and somewhat an impossible task. This paper steps forward to reveal the success of artificial neural network (ANN) and genetic algorithm (GA) together in this especial case. To do so, a three dimensional numerical simulation of the fluid flow under non-isothermal condition was initially proposed using computational fluid dynamics (CFD) method. After validation of the numerical model, twelve wire coil inserted tubes were tested through the validated model to get their heat transfer and friction coefficients at specific Reynolds ranges. Then, a prosperous ANN configuration was chosen for simulating the heat-exchangers and then obtaining a continuous function to get valuable output data with any values of input variables. Finally, to answer which type of the heat-exchangers works better, the optimization technique of GA was used along with the ANN's model. Results reveal the fact that while a suitable helical wire empowers the heat transfer efficiency, wrong choice of wire inserts may decrease the overall enhancement efficiency of the heat-exchanger unexpectedly.

Considering soil–structure interaction effects on performance-based design optimization of moment-resisting steel frames by an engineered cluster-based genetic algorithm

ABSTRACT An engineered cluster-based genetic algorithm (ECGA) has been used for performance-based design optimization of two dimensional moment-resisting steel frames (MRSF) considering effects of the soil-structure interaction (SSI). Based on the engineering judgments and via using simple specified filters on defined constrained of the problem, an engineered population is generated. To impose this concept for increasing efficiency of the genetic algorithm (GA), first, engineered populations are generated, and then linear analysis of the structure is performed to check some of the constraints, finally in the case of satisfying the first stage constraints, nonlinear static analysis is carried out. The nonlinear spring model has been used to model SSI of the frames in OpenSees software. The constructed model in the OpenSees has been coupled with the GA optimization technique in MATLAB software to minimize weight of the structural elements. Efficiency and accuracy of the proposed method are demonstrated.

Application of ANN Modelling and GA Optimization for Improved Creep and Corrosion Properties of Spin-Arc Welded AA5083-H111 Alloy

In this work, an attempt is made to identify the optimized parameter combination for improved creep and corrosion properties of AA5083-H111 alloy weldments processed by Spin-Arc gas metal arc welding (SA-GMAW) process. For this, the Artificial Neural Network (ANN) coupled with Genetic Algorithm (GA) was used as a statistical tool. Experiments were conducted by considering the input parameters namely welding current, filler spinning speed and filler spin diameter. The weld quality was assessed by measuring microhardness, corrosion resistance and steady-state creep rate. Initially, ANN was used to establish the relationship between input and output process variables. Different learning algorithms such as quick propagation (QP), back batch propagation (BBP) and incremental batch propagation (IBP) were used for predicting the output parameters. Prediction accuracy of the different learning algorithms was compared, and the best algorithm was used for the GA optimization technique. The optimum parameters were found to be 134 A of welding current, 1050 rpm of filler spinning speed and 1 mm of filler spin diameter. Among the input parameters, the filler spinning speed was identified as the most influential factor (40.11%) that effected the formation and distribution of second phase particles through which improved corrosion and creep properties were achieved.

Prediction of Optimized Color Design for Sports Shoes Using an Artificial Neural Network and Genetic Algorithm

Product design is a complicated activity that is highly reliant on individual impressions, feelings and emotions. Back-propagated neural networks have already been applied in Kansei engineering to solve difficult design problems. However, artificial neural networks (ANNs) have a slow rate of convergence, and find it difficult to devise a suitable network structure and find the global optimal solution. This study developed an ANN-based predictive model enhanced with a genetic algorithm (GA) optimization technique to search for close-to-optimal sports shoe color schemes for a given product image. The design factors of the sports shoe were set as the network inputs, and the Kansei objective value was the output of the GA-based ANN model. The results show that a model built with three hidden layers (28 × 38 × 19) could predict the object value reliably. The R2 of the preference objective was equal to 0.834, suggesting that the developed model is a feasible and efficient tool for predicting the objective value of product images. This study also found that the prediction accuracy for shoes with two colors was higher than that for shoes with only one color. In addition, the prediction accuracy for shoes with a relatively familiar shape was also higher. However, the prediction of color preferences is relatively difficult, because the respondents had different individual color preferences. Exploring the sensitivity and importance of the visual factors (form, color, texture) for various image words is a worthy topic for future research in this field.