Kind Of Genetic Algorithm Research Articles

Multi-objective shape optimization of Francis runner using metamodel assisted genetic algorithm

Nowadays evolutionary optimization techniques become an integral part of the design process of hydraulic turbine runners. For optimization the shape of the runner is parameterized by a set of geometrical parameters. The objective functions of each runner geometry, being the efficiency and cavitation performance, are evaluated through CFD analysis of 3D flow-field. As usual, some kind of genetic algorithm is used for searching the Pareto front, due to its inherent possibility of parallel evaluation of different runner variants within one generation. This approach requires extensive CFD computations of hundreds and thousands of runner variants. In order to speed up the optimization process a metamodel (approximate, surrogate model) can be utilized. The metamodel provides an approximation of true dependency of the objective functions on the design parameters. The metamodel is built based on the known objective functions for some initial set of points randomly taken form the design space. Once built, it can be used to replace time consuming CFD computations during the optimization process. In the present paper a metamodel, based on Gaussian processes, is implemented and integrated into a multi-objective genetic algorithm. The obtained algorithm is applied for shape optimization of the Francis turbine runner. The number of the design parameters varied from 4 to 24. The objective functions are multi-point efficiency and cavitation characteristics. True values of the objective functions are evaluated using Reynolds averaged Navier-Stokes computations of the flow field in a reduced turbine domain, including wicket gate, runner and draft tube. These values are used to train the initial metamodel. In order to enhance predictive capabilities of the metamodel, it is retrained periodically during the optimization loop. The results of metamodel-assisted optimization runs are compared to that obtained without metamodel. It is shown that application of metamodel significantly reduces the number of actual CFD computations even for high number of design parameters and thus speeds up the overall optimization process.

Optimal Distribution of Workloads in Cloud-Fog Architecture in Intelligent Vehicular Networks

With the fast growth in network-connected vehicular devices, the Internet of Vehicles (IoV) has many advances in terms of size and speed for Intelligent Transportation System (ITS) applications. As a result, the amount of produced data and computational loads has increased intensely. A solution to handle the vast volume of workload has been traditionally cloud computing such that a substantial delay is encountered in the processing of workload, and this has made a serious challenge in the ITS management and workload distribution. Processing a part of workloads at the edge-systems of the vehicular network can reduce the processing delay while striking energy restrictions by migrating the mission of handling workloads from powerful servers of the cloud to the edge systems with limited computing resources at the same time. Therefore, a fair distribution method is required that can evenly distribute the workloads between the powerful data centers and the light computing systems at the edge of the vehicular network. In this paper, a kind of Genetic Algorithm (GA) is exploited to optimize the power consumption of edge systems and reduce delays in the processing of workloads simultaneously. By considering the battery depreciation, the supporting power supply, and the delay, the proposed method can distribute the workloads more evenly between cloud and fog servers so that the processing delay decreases significantly. Also, in comparison with the existing methods, the proposed algorithm performs significantly better in both using green energy for recharging the fog server batteries and reducing the delay in processing data.

A novel MPP-NSGA algorithm and its application in optimization for radiated noises in the aircraft cabin

The paper used the AML method to compute transmission loss of aircraft panels and verifies correctness of the numerical simulation model by experimental test. Finally, this paper used an improved genetic algorithm to conduct a multi-objective optimization for the cabin noise. When the analyzed frequency is less than 250 Hz, transmission loss decreased rapidly with the increased analysis frequency, and decreased from the maximum 63.2 dB to 18.5 dB. Within 250 Hz-4000 Hz, the transmission loss gradually increased with the increased analysis frequency. At 250 Hz, the transmission loss had an obvious valley value. Sound radiation power was then computed based on boundary element method, and panel contribution analysis was conducted to find those panels which had an obvious impact on the cabin noise. Therefore, a multi-objective optimization was conducted on these panels and reinforced ribs. In order to further verify effectiveness of the MPP-NSGA method, it was compared with the traditional GA model and NSGA model. Optimization accuracy using MPP-NSGA model is increased, and optimization time is reduced. Through optimization with traditional GA method, the maximum sound power level decreased by 15.4 %, and the total sound power level decreased by 21.9 %. Through optimization with the NSGA method, the maximum sound power level decreased by 21.7 %, and the total sound power level decreased by 29.0 %. Through optimization with the MPP-NSGA method, the maximum sound power level decreased by 46.3 %, and the total sound power level decreased by 36.0 %. Therefore, compared with other two kinds of genetic algorithms, the MPP-NSGA method is obviously superior in noise optimization in the cabin. In the whole analysis frequency band, noise of the optimized cabin panel at each frequency point was smaller than that of the original structure, fully verifying feasibility of the optimization algorithm proposed in the paper. In addition, in the optimized structure, no panel made obvious contributions to the cabin noise, and each panel showed an equivalent contribution level. Transmission loss of the optimized cabin panel was obviously improved. However, the sound insulation valley still appeared at 250 Hz, but it was not so obvious like the original structure. After optimization, the sound insulation valley was 31.6 dB. The sound insulation valley of the original structure was 18.5 dB. Obviously, the sound insulation valley value of the optimized structure was increased by double compared with the original structure. This paper provided a valuable reference for noise reduction in the aircraft cabin.

Optimal sensors and actuators placement for large-scale unstable systems via restricted genetic algorithm

PurposeOne of the fundamental problems in control systems engineering is the problem of sensors and actuators placement. Decisions in this context play a key role in the success of control process. The methods developed for optimal placement of the sensors and actuators are known to be computationally expensive. The computational burden is significant, in particular, for large-scale systems. The purpose of this paper is to improve and extend the state-of-the-art methods within this field.Design/methodology/approachIn this paper, a new technique is developed for placing sensor and actuator in large-scale systems by using restricted genetic algorithm (RGA). RGA is a kind of genetic algorithm which is developed specifically for sensors and actuator placement.FindingsUnlike its other counterparts, the proposed method not only supports unstable systems but also requires significantly lower computations. The numerical investigations have confirmed the advantages of the proposed methods which are clearly significant, in particular, in dealing with large-scale unstable systems.Originality/valueThe proposed method is novel, and compared to the methods which have already been presented in literature is more general and numerically more efficient.

Optimisation of time window size for wind power ramps prediction

A new long-term wind power prediction approach based on time windows is proposed to improve the accuracy and efficiency of wind power ramp prediction. An optimisation model is built to select the optimal time window size which is the key point of the wind power forecasting. First, a swinging door algorithm is applied to identify historical ramp events, and historical data is divided into several sections by assumed time window size. Then, windows are classified into ramp windows and non-ramp windows, and the non-ramp data of ramp windows is required to be minimal. The variables, parameters, and constraints of the model are investigated in the study, and a kind of genetic algorithm is utilised to achieve the optimal solution. The model presented in this study is validated by the study case of actual wind farms, and evaluation and discussion demonstrate the validity of the proposed approach.

Optimization of a fan-shaped hole to improve film cooling performance by RBF neural network and genetic algorithm

To improve the cooling performance, shape optimization of a fan-shaped film cooling hole was carried out. Three geometric parameters, including incline angle, lateral expansion angle and hole length, were selected as the design parameters. Numerical model of the film cooling system was established, validated, and used to generate training samples. Radial basis function neural network (RBF-NN) was applied for surrogate model, and the optimal design parameters were determined by a kind of genetic algorithms. At low blowing ratio (M=0.5), the area-averaged film cooling effectiveness can reach its maximum value in the design space as incline angle, lateral expansion angle and hole length-to-diameter ratio are 40.5°, 23.97° and 7.43. At M=1.5, the optimal values of length-to-diameter ratio, lateral expansion angle and incline angle are 20.1°, 23.92° and 7.65. RBF-NN coupled with genetic algorithm is an effective scheme for the optimization of fan-shaped film cooling holes.

Filter selection based on representative training samples for multispectral imaging

In the application of spectral reflectance estimation with multispectral imaging, the filters and training samples jointly influence the accuracy of spectral estimation for learning based methods. In this study, by utilizing one kind of genetic algorithm, the selection of optimal filters is implemented with respect to the representative training samples so as to consider the spectral characteristics of filters, imaging system and test samples simultaneously, which also avoids measurement on the detailed spectral characteristics of filters and imaging system. This filter selection method was tested on a filter-wheel based multispectral imaging device along with a set of interference filters. It is verified that a small number of representative training samples are sufficient for selecting the optimal filters, and the filters selected by this method are more effective for spectral estimation than those obtained from the conventional method.

Filter selection based on light source for multispectral imaging

In multispectral imaging, it is necessary to select a reduced number of filters to balance the imaging efficiency and spectral reflectance recovery accuracy. Due to the combined effect of filters and light source on reflectance recovery, the optimal filters are influenced by the employed light source in the multispectral imaging system. By casting the filter selection as an optimization issue, the selection of optimal filters corresponding to the employed light source proceeds with respect to a set of target samples utilizing one kind of genetic algorithms, regardless of the detailed spectral characteristics of the light source, filters, and sensor. Under three light sources with distinct spectral power distributions, the proposed filter selection method was evaluated on a filter-wheel based multispectral device with a set of interference filters. It was verified that the filters derived by the proposed method achieve better spectral and colorimetric accuracy of reflectance recovery than the conventional one under different light sources.

Research and Analysis on Gray Genetic Algorithm

In view of the slow match speed of the image, the article proposes gray genetic algorithm (GGA), one kind of new fast image match method, combining the gray connection theory with genetic algorithm. This method, firstly, determines question's parameter space to obtain several of initial points required to be match through coding the parameter space and the string collection initialized. Then the reference sequence and the comparison sequence separately are to be constructed by means of the template chart and the histogram searching for current subgraph's information. Lastly, fitness function are established on these two sequences between pessimistic interrelatedness as reference Based on this, the string collection initialized evolves gradually optimizing region of the search space after many kinds of genetic algorithm's operation, such as the choice operation, the overlapping operation and the variation operation and so on. Finally, it infinitely approaches the optimum matching position. Because the GGA law has fully applied the small sample and genetic algorithm computation parallelism characterized in the gray connection theory, the timeliness of the image match have been distinctly enhanced with certain match precision.

Intrusion Detection Engineering Management Based on Optimization of BP Neural Network

In order to short the modelling time of BP neural network, this paper designs a kind of genetic algorithm to optimize it. By encoding the individual components, initializing the number of populations, and designing proper fitness function, a binary coding genetic algorithm is provided for BP neural network. And it is used to optimize input variables of BP neural network and reduce its dimension. The experiment is carried out based on KDD Cup 99 data set. The results show that the optimized model has shorter modelling time.

Optimization for Coal Heavy Haul Transportation Assembly Scheme Problem Using Genetic Algorithm

Thispaper presents aCoal Heavy Haul Transportation Assembly Scheme Problem (CHASP), in which the time consuming functions, assembly number constraints and assembly weight constraints etc are considered. The time consuming costs consist of residence time and disassembly time. The disassembly time functions are usually nonlinear functions of unit train departure directions. Then, a nonlinear 0-1 programming is formulated for the problem and solved by lingo mathematical solver. Considering the complexity of the problem, a kind ofGenetic Algorithm is proposed to solve it. Extensive computational experiments are taken on randomly generated data, the detailed results are given and the genetic algorithm is shown to be efficient.

OPTIMIZATION OF A NONLINEAR ELECTRICAL–THERMAL MODEL OF THE SKIN

This paper presents modified equations for a nonlinear electrical–thermal model of the skin, and optimizes the parameters of the model by a genetic algorithm. In publications, an electrical–thermal model for the skin has been presented. Although, the model is sophisticated and includes the most of physical phenomena occurring in the skin; however, mathematical equations used in this model for the capacitance and dissipated power are incompletely stated so that they sometimes cause instability in solving the equations. In this paper, the equations given previously for the capacitance current and the dissipated power are first modified. We have considered different inward and outward current in the model, since there are different ions which make this two currents in the skin. Then, a special kind of genetic algorithm is applied to estimate the parameters of the model. The used algorithm is a powerful optimization tool against high-dimensional problems and can overcome the curse of dimensionality. In each iteration, it divides variable sets into two groups: a small set still variable and a large set frozen on their best values. Finally, modeling results are compared with experimental results. The comparison shows there is a good agreement between the modeling and experimental results especially in the steady state regime.

Intrusion Detection Based on Improved SOM with Optimized GA

In order to improve the effectiveness of supervised self-organizing map (SSOM) neural network, a kind of genetic algorithm is designed to optimize it. To improve its classification rate, a real number encoding genetic algorithm is provided and used to optimize the learning rate and neighbor radius of SSOM. To speed up the modeling speed, a binary encoding genetic algorithm is provided to optimize input variables of SSOM and reduce its dimension of input sample. Finally, intrusion detection data set KDD Cup 1999 is used to carry out experiment based on the proposed model. The results show that the optimized model has shorter modeling time and higher intrusion detection rate.

Urban water resources planning by using a modified particle swarm optimization algorithm

A new optimization algorithm by coupling the mutation process to the particle swarm optimization (PSO) is developed in this paper. This algorithm, entitled particle swarm optimization with mutation similarity (PSOMS), is successfully applied to an urban water resources management problem for the large city of Tabriz, Iran. The objective functions of the optimization problem are to minimize the cost, maximize water supply and minimize the environmental hazards. The constraints are physical limits such as pipelines capacity, ground water, the demand and the impact of conservation tools. Due to the parameters uncertainty, the water supply objective is modeled with fuzzy set theory and the objectives are then combined with compromise programming. The resulted single objective is solved using PSOMS, and its efficiency is then compared with the basic PSO and two kinds of genetic algorithms. Among them, PSOMS shows rapid convergence and suitable results compared to other methods. PSOMS is also improved to provide the Pareto frontier, which is needed to proper selecting of the optimal solutions in the uncertain conditions. Finally, the diversity of solutions is checked based on an indicator of the distances between different solutions, which show the efficiency of the PSOMS algorithm with respect to the genetic algorithm. Then by using the non-symmetric Kalai–Smorodinsky method a guideline is provided for comfort selection of the most preferred solution in the Pareto frontier. Based on these outcomes, the multi-objective PSOMS provides more appropriate results needed for urban systems management.

Cellular genetic algorithms without additional parameters

Cellular genetic algorithms (cGAs) are a kind of genetic algorithms (GAs) with decentralized population in which interactions among individuals are restricted to close ones. The use of decentralized populations in GAs allows to keep the population diversity for longer, usually resulting in a better exploration of the search space and, therefore, in a better performance of the algorithm. However, it supposes the need of several new parameters that have a major impact on the behavior of the algorithm. In the case of cGAs, these parameters are the population and neighborhood shapes. We propose in this work two innovative cGAs with new adaptive techniques that allow removing the neighborhood and population shape from the algorithm's configuration. As a result, the new adaptive cGAs are highly competitive (statistically) with all the compared cGAs in terms of the average solutions found in the continuous and combinatorial domains, while finding, in general, the best solutions for the considered problems, and with less computational effort.

Modifiable Attribute Cell Problem and Solution Method for Population Synthesis in Land Use Microsimulation

Land use microsimulation requires the preparation of a set of microdata for the base year. Most existing procedures used for the synthesis of population data are based on the iterative proportional fitting method, in which the number of individuals in each cell of the cross-classification table is estimated. Such a procedure is referred to as the cell-based approach in this study. The approach is based on predefined categories of individuals. Originally, however, these individuals have continuous attributes. Therefore, a different type of categorization would yield a different classification table, which would change the end results of the analysis. In this paper, this phenomenon is referred to as the modifiable attribute cell problem (MACP). It is similar to the modifiable area unit problem that arises when spatial data are aggregated into zones. This paper addresses MACP and proposes a method to determine the best combination of the categories. The solution of MACP is considered to be the minimization of the number of cells in a table with respect to the key output variable that has been defined and used as an evaluation criterion. Because of the computational difficulty resulting from the combination explosion, symbiotic evolution, which is a kind of genetic algorithm, is used. Finally, a case study is presented for the Sapporo metropolitan area of Japan.

Random Search Algorithm for the Generalized Weber Problem

In this paper, we consider the planar multi-facility Weber problem with restricted zones and non-Euclidean distances, propose an algorithm based on the probability changing method (special kind of genetic algorithms) and prove its efficiency for approximate solving this problem by replacing the continuous coordinate values by discrete ones. Version of the algorithm for multiprocessor systems is proposed. Experimental results for a high-performance cluster are given.

Application of quantum neural networks in localization of acoustic emission

Due to defects of time-difference of arrival localization, which influences by speed differences of various model waveforms and waveform distortion in transmitting process, a neural network technique is introduced to calculate localization of the acoustic emission source. However, in back propagation (BP) neural network, the BP algorithm is a stochastic gradient algorithm virtually, the network may get into local minimum and the result of network training is dissatisfactory. It is a kind of genetic algorithms with the form of quantum chromosomes, the random observation which simulates the quantum collapse can bring diverse individuals, and the evolutionary operators characterized by a quantum mechanism are introduced to speed up convergence and avoid prematurity. Simulation results show that the modeling of neural network based on quantum genetic algorithm has fast convergent and higher localization accuracy, so it has a good application prospect and is worth researching further more.

Fuzzy Parallel Machines Scheduling Problem Based on Genetic Algorithm

A kind of unrelated parallel machines scheduling problem is discussed. The memberships of fuzzy due dates denote the grades of satisfaction with respect to completion times with jobs. Objectives of scheduling are to maximize the minimum grade of satisfaction while makespan is minimized in the meantime. Two kind of genetic algorithms are employed to search optimal solution set of the problem. Both Niched Pareto Genetic Algorithm (NPGA) and Nondominated Sorting Genetic Algorithm (NSGA-II) can find the Pareto optimal solutions. Numerical simulation illustrates that NSGA-II has better results than NPGA.

Multi-objective Optimization of Fuzzy Parallel Machines Scheduling Problem Using Nondominated Genetic Algorithms

A kind of unrelated parallel machines scheduling problem is discussed. The memberships of fuzzy due dates denote the grades of satisfaction with respect to completion times with jobs. Objectives of scheduling are to maximize the minimum grade of satisfaction while makespan is minimized in the meantime. Two kind of genetic algorithms are employed to search optimal solution set of the problem. Both Niche Pareto Genetic Algorithm (NPGA) and Nondominated Sorting Genetic Algorithm (NSGA-II) can find the Pareto optimal solutions. Numerical simulation illustrates that NSGA-II has better results than NPGA.