Genetic Algorithm Parameters Research Articles

Cell loading and shipment optimisation in a cellular manufacturing system: an integrated genetic algorithms and neural network approach

In this paper, cell loading and shipment method optimisation problem in a cellular manufacturing system are studied. A hierarchical methodology that consists of mathematical optimisation model, genetic algorithms (GAs) and artificial neural networks (ANNs) were proposed. The mathematical model is compared with the GA in terms of the optimisation performance. Next, ANN model was developed as decision support tool to study the impact of GA parameters on the solution quality. Several problem sizes were experimented with the proposed GA and the mathematical model, and compared. GA was run to make a total of 648 sample solutions for the 20-job problem. Next, ANN model was built based on the sample solutions' data and the optimal ANN model is identified out of 1,000 networks. The results were also coupled with sensitivity and statistical analyses, which indicated that type of crossover and mutation operators, had the greatest impact on the solution quality.

The Influence of Genetic Algorithm Parameters Over the Efficiency of the Energy Consumption Estimation in a Low–energy Building

Dynamic mathematical models are widely used for estimating the energy consumption in buildings. This paper presents the procedure to follow in order to optimize a simplified gray-box model by using an improved Genetic Algorithm (GA), which can find the nearly-optimum solution to a given problem. The purpose of this paper is to analyze the influence of the parameters required in the implementation of the Genetic Algorithm to estimate the energy consumption in a low–energy building. The variable parameters of the model were modified trough the GA in order to obtain the minimum error between the model and measured data. The influence of the GA parameters variation over the estimation efficiency and convergence time was also analyzed.

A hybrid swarm intelligence of artificial immune system tuned with Taguchi–genetic algorithm and its field-programmable gate array realization to optimal inverse kinematics for an articulated industrial robotic manipulator

This article presents a hybrid swarm intelligence of artificial immune system tuned with Taguchi–genetic algorithm and its field-programmable gate array realization to optimal inverse kinematics for a 5-degree-of-freedom industrial robotic manipulator using system-on-a-programmable-chip technology. This hybridization strategy with a Taguchi–genetic algorithm parameter tuner improves the performance in conventional artificial immune system paradigms. The field-programmable gate array realization of the proposed artificial immune system tuned with Taguchi–genetic algorithm is more effective in practice for real-world embedded applications. This system-on-a-programmable-chip–based artificial immune system tuned with Taguchi–genetic algorithm is then applied to the optimal inverse kinematics redundancy solver of a 5-degree-of-freedom robotic manipulator. The optimal joint configuration is obtained by minimizing the pre-defined affinity function in artificial immune system tuned with Taguchi–genetic algorithm for real-world embedded robotics application. The experimental results and comparative works are presented to show the effectiveness and merit of the proposed system-on-a-programmable-chip–based artificial immune system tuned with Taguchi–genetic algorithm intelligent inverse kinematics redundancy solver for a 5-degree-of-freedom industrial robotic manipulator. This system-on-a-programmable-chip–based artificial immune system tuned with Taguchi–genetic algorithm solver outperforms the conventional solvers, such as geometric solvers, Jacobian-based solvers, hybrid genetic algorithm solvers, and particle swarm optimization solvers.

Optimizing genetic algorithm parameters for multiple sequence alignment based on structural information

Multiple sequence alignments (MSAs) are commonly used approaches in the analysis of sequence structure relationships. MSA is generally the alignment of three or more protein or nucleic acid sequences that maximises the similarities between sequences. In this paper, we use genetic algorithm to compute multiple sequence alignment using the structural information as the scoring scheme implemented in the program Multiobjective Optimizer for Sequence Alignments based on Structural Evaluations (MOSAStrE). We performed numerical experiments on datasets obtained from benchmark alignment database (BAliBASE) to solve multiple sequence alignment. To test the performance of the proposed algorithm, numerical simulations were carried out in deciding the appropriate set of parameter values for the

Transport Aircraft Conceptual Design Optimization Using Real Coded Genetic Algorithm

Due to soaring oil prices, increased air traffic and competition among air transport companies, and environmental concerns, aircraft maximum takeoff weight (MTOW) is becoming a critical aspect, of air transport industry. It is very important to estimate the MTOW of the aircraft in order to determine its performance. However, estimating the weight of an aircraft is not a simple task. The purpose of this paper is to present a simplified method to optimize the aircraft MTOW using a genetic algorithm approach. For the optimization of MTOW of transport aircraft, a MATLAB program consisting of genetic algorithm techniques with appropriate genetic algorithm parameters setting was developed. The objective function for the optimization was a minimization of MTOW. The use of genetic real coded algorithm (GA) as an optimization tool for an aircraft can help to reduce the number of qualitative decisions. Also, using GA approach, the time and the cost of conceptual design can considerably be reduced. The model is applicable to the air transport industry. The proposed model has been validated against the known configuration of an aircraft.

Developing a novel workflow for natural gas lift optimization using advanced support vector machine

Natural Gas Lift (NGL) is one of the most attractive methods to enhance oil recovery. In this method, oil is produced using gas from the gas region either adjacent or far from the oil layer. The reservoir simulation software should be run many times to optimize the NGL process. This is practically impossible due to time-consuming simulation of an actual reservoir. In this study, support vector machine (SVM) was used to overcome the problem. The reservoir simulation software was replaced by the trained SVM. Through this, each run only takes a few seconds. The process was optimized for a real field using particle swarm optimization (PSO) and genetic algorithm (GA). The optimum SVM parameters were determined by the PSO algorithm. Taguchi experiment design was used to determine optimum GA and PSO parameters.

Image Recognition of Maize Leaf Disease Based on Ga-svm

An improved SVM named genetic algorithm support vector machines (GA-SVM) is discussed in this paper for classifying maize disease. According to the disadvantage of manually determining the parameters in traditional SVM, the genetic algorithms is used to automatically obtain the penalty factor and kernel function. The appropriate parameters are selected by rotational orthogonal method. The extracted eigenvalue is entered to the GA-SVM classification model to improve the classification performance. After the comparisons of different genetic operators and different kernel functions, the results show that the appropriate parameters for genetic algorithms is when M=50, Pc=0.7 and Pm=0.05, the average classification rate is at peak when choosing RBF kernel function. The results also demonstrate that the GA-SVM algorithm achieves a better improvement than SVM.

Using a Double Genetic Algorithm for Correlating Thermal Models

Correlating a thermal model could be a challenging problem in global optimization, which often requires employing a genetic algorithm. This work studies choosing the optimal set of the genetic algorithm control parameters when it is used for this purpose. Two practical thermal models are considered. For the small model the optimal set of the genetic algorithm parameters was found by applying a double genetic algorithm where the outer algorithm tried to minimize the best fitness value achieved by the inner one by varying the inner algorithm control parameters. In addition, it was demonstrated that in a practical calculation increasing the population size can easily result in a worse accuracy. The proper choice of the population size was studied for the large model. The double genetic algorithm was modified by starting each run of the inner algorithm with the best solution obtained until then, instead of using the same initial population in all the runs. Doing all the runs with the same population size and choosing the population size randomly were also tried and it was shown that this way a considerable improvement in accuracy could be achieved.

Optimal operation of Mula reservoir with combined use of dynamic programming and genetic algorithm

Genetic algorithm (GA) has been used repeatedly in reservoir operation studies during last two decades. GAs require trying different alternatives, different GA parameter values, and select those which perform best for a particular application. Besides this, there are chances of getting trapped into local optima, since GA starts with randomly generated initial population within the entire search space. Therefore, GA’s search process is slow and time-consuming. GA’s process may be speeded up if initial population is generated in a narrowed search space. This process may save time spent on sensitivity analysis of parameters. Discrete dynamic programming (DP) provides global optimal solutions, but discreteness and dimensionality are the main disadvantages in reservoir operation applications. To overcome these deficiencies a hybrid approach combining DP and GA (DP–GA) is proposed to study a single reservoir operation problem. Where DP provides the narrowed search space within which the optimal solution for the problem is expected. GA then search to achieve the possible optimal solution within this space avoiding any discreteness of variables. Proposed DP–GA approach was found to outperform both GA and DP in terms of less computational requirement and quality of the solution, respectively.

A Forecasting Tool for Predicting Australia's Domestic Airline Passenger Demand Using a Genetic Algorithm

This study has proposed and empirically tested for the first time genetic algorithm optimization models for modelling Australia's domestic airline passenger demand, as measured by enplaned passengers (GAPAXDE model) and revenue passenger kilometres performed (GARPKSDE model). Data was divided into training and testing datasets; 74 training datasets were used to estimate the weighting factors of the genetic algorithm models and 13 out-of-sample datasets were used for testing the robustness of the genetic algorithm models. The genetic algorithm parameters used in this study comprised population size (n): 200; the generation number: 1,000; and mutation rate: 0.01. The modelling results have shown that both the quadratic GAPAXDE and GARPKSDE models are more accurate, reliable, and have greater predictive capability as compared to the linear models. The mean absolute percentage error in the out of sample testing dataset for the GAPAXDE and GARPKSDE quadratic models are 2.55 and 2.23%, respectively.

A hybrid genetic algorithm based heuristic for an integrated supply chain problem

Supply chain management is becoming an essential component of efficient decision-making for companies, as they are increasingly implementing optimization strategies in order to improve their business performance. In this paper we develop a capacitated multi-echelon joint location-inventory model, according to which a single product is distributed from a manufacturer to retailers through a set of warehouses, the locations of which are to be determined by the model. Each retailer is assigned exactly one warehouse, while each warehouse can serve multiple retailers. The model decides where to locate warehouses, assigns retailers to the warehouses and decides the times between orders at the warehouses and the retailers, so as to minimize the cost of operating the supply chain. Notably, the model considers capacity constraints for each warehouse, ensuring that the reorder quantity is below the capacity limit. We develop a genetic algorithm (GA) based heuristic to solve the problem and the GA is validated on small size problems by comparing its solution to the optimal solution obtained by the General Algebraic Modeling System (GAMS). We focus our attention on specifically customizing the GA and thus an experimental analysis is carried out to find the optimal parameter setting for the GA as well as to obtain insights on the effect of the various GA parameters. Finally, a sensitivity analysis is conducted to show the effect of the capacity constraints.

Uncertainty Analysis of Looped Water Distribution Networks Using Linked EPANET-GA Method

The hydraulic analysis of water distribution network is generally done by the conventional method. The obtained results such as pipe discharges, hydraulic gradient level of nodes, nodal concentrations etc., are normally crisp values assuming crisp input parameters. But, in real networks, there are many uncertainties in nodal demands, roughness, length, diameters of pipes, valve operations, water levels in reservoirs, head-discharge characteristics of pumps, etc. The results obtained by the conventional method may not be satisfactory in practice. Hence, in this study, an attempt is made to consider pipe roughness as uncertain parameter and to evaluate the pipe discharges and nodal hydraulic gradient levels accordingly. The Genetic Algorithm Optimization based methodology has been used to obtain the unknown parameters at each α-cut level. The hydraulic simulations are done by using EPANET 2 in MATLAB environment. The maximum and minimum values of pipe discharges and nodal hydraulic gradient levels are obtained in each simulation run. The results show that the required time to run the simulation is same for all networks irrespective of the size after adopting a particular Genetic Algorithm parameters tuning. The obtained results are compared with past studies and it is found that present method is effective in analyzing the uncertainty problem, particularly for large scale networks. The results of pipe discharges are found to vary at 30, 48.3, 58 and 9 % and of the hydraulic heads at nodes at 3.17 m, 4.3 m, 2.2 m and 12.64 m for the selected problems taken from the literature. This study would help to analyze the pipe network under the conditions of uncertainty in input parameters.

Optimization of a Large Scale Industrial Reactor Towards Tailor Made Polymers Using Genetic Algorithm

Abstract This paper presents a computational procedure for producing tailor made polymer resins, satisfying customers’ needs while operating with maximum profit. The case study is an industrial large-scale polymerization reactor. The molecular properties considered are melt index (MI), which measures the molecular weight distribution, and stress exponent (SE), which is related to polydispersity. An economic objective function is associated to a deterministic mathematical model and the resulting optimization problem is solved by genetic algorithm (GA), a stochastic method. The GA parameters for both binary and real codifications are tuned by means of the design of experiments. Attempting to achieve the global optimum, a hybrid method, which introduces process knowledge into GA random initial population, is proposed. The binary codification performs better than the real GA, especially with hybridization. Results show that the GA can satisfactorily predict tailor made polymer resins with profits up to 25% higher than the industrial practice.

Genetic Algorithm Parameter Optimization using Taguchi Robust Design for Multi-response Optimization of Experimental and Historical Data

This paper presents a methodology for robust optimization of Genetic Algorithm (GA) involving complex interactions among the control parameters. Finding the Optimum GA parameters to solve an optimization problem for producing best results with least variability is still an open area of research. The proposed research approach primarily covers the robust optimization of Genetic Algorithm control parameters using Taguchi Design of Experiment (DOE) with a special set of L25 orthogonal array (OA). The experimental design and the study is conducted with MATLAB Genetic Algorithm internal control parameters using real-coded Genetic Algorithm fitness functions operates directly on real values of two different case studies. One of them is based on experimental data for the development of a new product and the other one is based on the historical data of an existing product under large scale complex manufacturing system. The performance characteristics (Best fitness value) of GA are evaluated. The main effect for signal-tonoise ratios (SNR), main effect for means, response tables and analysis of variance (ANOVA) are employed and optimum parameter settings are obtained. The confirmation experiments are carried out with the optimum setting parameters and the results are compared with the predicted & default setting values of GA. The experimental results show an increase of signal-to-noise ratio (SNR) by around 54 % and mean by 12 to 25 % from the default setting of GA to the optimum settings arrived during the experimental process, which are extremely significant. Further, cross over was found to be the most influential parameter for both the case studies, followed by other parameters like population size, selection process for case study-1 and mutation rate, population size for case study-2. It is evident that the GA performances are also sensitive to the objective function (transfer function in GA terms). The study clearly shows the robust optimization of GA parameters leading to best level of performance characteristics with least variability for the solution of real life optimization problem, using Taguchi Experimental Design Techniques.

Intelligent Approaches to Real Time Level Control

Liquid level control is important for ensuring energy and material balance in many installations but it also difficult as the plant is nonlinear, inertial and with model uncertainties. Fuzzy logic controllers (FLCs) are successfully applied to ensure system stability and robustness by simple means and a model-free design. This paper suggests a procedure for off-line tuning of the many FLC parameters based on optimization of a suggested multi-objective function defined on several system performance indices using genetic algorithms (GAs). First, a model-free FLC is empirically tuned, then applied for real time control of the plant and the necessary data recorded and used to GA parameter optimize a TSK plant model of an accepted structure. The validated on different set of experimental data model is employed in FLC closed loop system simulation experiments to evaluate the fitness function in the GA optimization of the FLC pre-processing and post- processing parameters. The procedure is applied for the real time PI/PID FLC level control in a laboratory-scale tank system. The improvement of the system performance indices due to the GA optimization is estimated in level real time control.

Optimizing the construction procedures of large-span structures based on a real-coded genetic algorithm

The erection phase is the most vulnerable period in the life of a structure. However, no mature and effective method for optimizing the construction procedures of large-span steel structures is yet available. Thus, proposing an optimization method for constructing such structures is critical. A real-coded genetic algorithm (GA) was proposed in the present work, and the selection, crossover, and mutation operators were programmed. The proposed algorithm is effective in optimizing integer combinations and sequences, such as an unloading sequence. A series of problems that occurred during the erection of conchshaped complex large steel structures was optimized using the proposed method. Optimum analysis was conducted using MATLAB and the general finite element software ANSYS. The optimal layouts of the lifting points, false work, and unloading sequence were derived. The influence of the GA parameters on its convergence efficiency was also studied. Therefore, the proposed GA method is reliable and effective for optimizing the construction procedures of large-span steel structures.

Parameter identification of the glazed photovoltaic thermal system using Genetic Algorithm–Fuzzy System (GA–FS) approach and its comparative study

In this paper, Genetic Algorithm–Fuzzy System (GA–FS) approach is used to identify the optimized parameters of the glazed photovoltaic thermal (PVT) system and to improve its overall exergy efficiency. The fuzzy knowledge base is used to improve the efficiency of Genetic Algorithm (GA). It is observed that three GA parameters, namely: (i) crossover probability (Pcross), (ii) mutation probability (Pmut) and (iii) population size are changing dynamically during the program, according to fuzzy knowledge base to maximize the efficiency of the GA. Here, overall exergy efficiency is considered as an objective function during the optimization process for GA–FS approach. The effort has been made to identify the different optimized parameters like; length and depth of the channel, velocity of flowing fluid, overall heat transfer coefficient from solar cell to ambient and flowing fluid and overall back loss heat transfer coefficient from flowing fluid to the ambient to maximize the overall exergy efficiency using GA–FS approach. Performance of glazed PVT using GA–FS approach has been compared with performance using GA approach and without GA. It has also been observed that the GA–FS approach is a better approach as compared to GA approach because it converges faster as compare to GA because the use of the fuzzy knowledge base with GA and take less time for identification of optimized system parameters.

A robust model for computing pressure drop in vertical multiphase flow

The pressure drop in tubing constitutes a major part of the total pressure drop. However, its calculation is very complex due to the possibility of presence of multiple phases. A number of empirical and mechanistic methods have been proposed for multiphase flows. Each method had its own assumptions, which often reduce the calculation accuracy of the actual pressure drop. In this study, the best probable model was developed by employing the Artificial Neural Network (ANN) and training this network using Levenberg–Marquardt (LM), Genetic algorithm (GA), and Particle Swarm Optimization (PSO). A total number of 1740 data collected from the wells in Middle East region were used to train and test the ANN. Taguchi design was used to optimize GA and PSO parameters. The trend analysis test carried out to prove the model stability and its ability to simulate the physical process of the problem. The comparison between the best obtained results and the results of the available methods indicated the higher accuracy of the proposed method.

Improving solution of discrete competitive facility location problems

We consider discrete competitive facility location problems in this paper. Such problems could be viewed as a search of nodes in a network, composed of candidate and customer demand nodes, which connections correspond to attractiveness between customers and facilities located at the candidate nodes. The number of customers is usually very large. For some models of customer behavior exact solution approaches could be used. However, for other models and/or when the size of problem is too high to solve exactly, heuristic algorithms may be used. The solution of discrete competitive facility location problems using genetic algorithms is considered in this paper. The new strategies for dynamic adjustment of some parameters of genetic algorithm, such as probabilities for the crossover and mutation operations are proposed and applied to improve the canonical genetic algorithm. The algorithm is also specially adopted to solve discrete competitive facility location problems by proposing a strategy for selection of the most promising values of the variables in the mutation procedure. The developed genetic algorithm is demonstrated by solving instances of competitive facility location problems for an entering firm.

Odhady parametrů ekonomických modelů pomocí genetických algoritmů

The paper concentrates on capability of genetic algorithms for parameter estimation of non-linear economic models. In the paper we test the ability of genetic algorithms to estimate of parameters of demand function for durable goods and simultaneously search for parameters of genetic algorithm that lead to maximum effectiveness of the computation algorithm. The genetic algorithms connect deterministic iterative computation methods with stochastic methods. In the genteic aůgorithm approach each possible solution is represented by one individual, those life and lifes of all generations of individuals run under a few parameter of genetic algorithm. Our simulations resulted in optimal mutation rate of 15% of all bits in chromosomes, optimal elitism rate 20%. We can not set the optimal extend of generation, because it proves positive correlation with effectiveness of genetic algorithm in all range under research, but its impact is degreasing. The used genetic algorithm was sensitive to mutation rate at most, than to extend of generation. The sensitivity to elitism rate is not so strong.