Genetic Algorithm Code Research Articles

A Decision Support System to Design Operating Strategies of a Power Plant: Adequate Electricity Generation and Abated Emissions Release

In this study, a mathematical optimization model using Genetic Algorithms (GA) was developed and applied to search for operation modes of a power plant considering the adequate electricity generation for public demand and abated emissions release for the better air quality. Continuous Emissions Monitoring System (CEMS) equipment that reports hourly emissions (i.e., NOx, SO2, and CO) is used to prepare information of emissions regarding to the electricity production (GWh). Emission sources include sixteen stacks of eight units of power generation (four thermal and four combined cycle power plants) using natural gas and fuel oil. This technique shows the beneficial application of optimization model, GA code, CEMS, and power generation database in a unified framework of a Decision Support System (DSS). We find that the best operating strategy is to partially shutdown the medium power units and to abundantly operate the large units for the compensation of both emissions and electricity. By using this approach, the electricity generation can be sufficiently produced while emissions can be reduced by 1.83 % of NOx/kWh, 10.90 % of SO2/kWh, and 7.55% of CO/kWh.

Genetic algorithm for optimisation of A-TIG welding process for modified 9Cr–1Mo steel

It is essential to set up the activated tungsten inert gas (A-TIG) welding process parameters to produce the desired weld bead geometry and heat affected zone (HAZ) width in modified 9Cr–1Mo steel weld joints. Therefore, it becomes necessary to develop a tool for optimisation of A-TIG welding process. Genetic algorithm (GA) based model has been developed to determine the optimum process parameters. In this methodology, first independent ANN models correlating depth of penetration, weld bead width and HAZ width with current, voltage and torch speed respectively were developed. Then, GA code was developed in which the objective function was evaluated using the ANN models. There was good agreement between the target and actual values of bead geometry and HAZ width obtained using the GA optimised process parameters. Thus, a methodology using GA has been developed for optimising the A-TIG process parameters for modified 9Cr–1Mo steel.

An Optimal Control Strategy Combining SVM with RGA for Improving Fermentation Titer

An optimal control strategy is proposed to improve the fermentation titer, which combines the support vector machine (SVM) with real code genetic algorithm (RGA). A prediction model is established with SVM for penicillin fermentation processes, and it is used in RGA for fitting function. A control pattern is proposed to overcome the coupling problem of fermentation parameters, which describes the overall production condition. Experimental results show that the optimal control strategy improves the penicillin titer of the fermentation process by 22.88%, compared with the routine operation.

Mathematical Modeling of Gas Oil HDS and Optimization of Operational Conditions in Trickle Bed Reactor by Genetic Algorithm

The present work aims to employ genetic algorithm (GA) to optimize a HDS process, which is difficult to optimize by conventional methods. The considered chemical process is the three phase catalytic trickle-bed reactor in which hydrodesulphurization reaction occurs. Non-linear kinetics coupled with the transitional mathematical model of the gas, liquid and solid phases are used to describe the dynamic behavior of the multivariable process. The model, based on a two-film theory, was tested with regards to hydrodesulphurization of vacuum gas oil in a high-pressure pilot plant operated under isothermal conditions. Due to the high dimensionality and non-linearity of the model, a rigorous one, the solution of the optimization problem through conventional algorithms does not always lead to the convergence. This fact justifies the use of an evolutionary method, based on the GAs, to deal with this process. In this way, in order to optimize the process, the GA code is coupled with the rigorous model of the reactor. The aim of the optimization through GAs was to search for the optimal conditions that minimize the gas make and sulfur content of the outlet oil. Many simulations are conducted in order to find the maximization of the objective function without violating the constraints. The results show that the GA is used successfully in the process optimization.

Quasi-isentropic compression using functionally graded materials in gas gun and explosive driven systems

Results of hydrodynamic simulations of dynamic compression experiments performed by impact loading of materials are reported. The simulations pertain to a new approach for generating quasi-isentropic compression using functionally graded materials (FGM). First of all, we focus on quasi-isentropic compression waves generated by a constant velocity impactor (similar to that from a gas gun). Quasi-isentropic compression is characterized from the temporal profiles of pressure at target surface and fluid velocity at target-window interface generated from different functional forms of density variation along the FGM flyer. It is shown that quadratic FGM is the best option for increasing rise time of pressure pulse. Secondly, FGM induced quasi-isentropic compressions are studied by accelerating the impactor with high explosive (HE) driven shocks in both the cases when impactor is in contact with the target (contact geometry) and separated from target by air gap (flyer geometry). The study reveals that nearly isentropic pressure profiles can be realized with quadratic FGM impactor in flyer geometry. Contact geometry produces a distinct initial pressure jump which is undesirable for isentropic compression. Finally, quasi-isentropic compressions are observed for spherically symmetric FGM flyers driven by HE. Here also it is seen that quadratic FGM impactor serves better in improving the peak impact velocity at shell target. Thus, based on the study, we conclude that quadratic FGM is the right choice for generating quasi-isentropic compression. The parameters of the FGM impactors, optimized using a genetic algorithm code, are also presented for both planar and spherical systems.

Diploid code genetic algorithm used to solve UTP

The University Timetable Problem (UTP) is a typical NP-hard problem. Using the layered method, the common combined programming model of five dimensions was decompounded into two three-dimensional combinations, cutting down the problem scale. Then in view of the disadvantages of the traditional genetic algorithm for UTP, the diploid code was put forward and alternate evolution was used to optimize multi-targets one by one. Experimental results show that the method realizes the pattern theorem and quickens the evolving velocity.

Genetic algorithms for simultaneous variable and sample selection in metabonomics

Metabolic profiles derived from high resolution (1)H-NMR data are complex, therefore statistical and machine learning approaches are vital for extracting useful information and biological insights. Focused modelling on targeted subsets of metabolites and samples can improve the predictive ability of models, and techniques such as genetic algorithms (GAs) have a proven utility in feature selection problems. The Consortium for Metabonomic Toxicology (COMET) obtained temporal NMR spectra of urine from rats treated with model toxins and stressors. Here, we develop a GA approach which simultaneously selects sets of samples and spectral regions from the COMET database to build robust, predictive classifiers of liver and kidney toxicity. The results indicate that using simultaneous sample and variable selection improved performance by over 9% compared with either method alone. Simultaneous selection also halved computation time. Successful classifiers repeatedly selected particular variables indicating that this approach can aid defining biomarkers of toxicity. Novel visualizations of the results from multiple computations were developed to aid the interpretability of which samples and variables were frequently selected. This method provides an efficient way to determine the most discriminatory variables and samples for any post-genomic dataset. GA code available from http://www1.imperial.ac.uk/medicine/people/r.cavill/

Correlation of Activity Coefficients in Aqueous Solutions of Ammonium Salts Using Local Composition Models and Stochastic Optimization Methods

In this study, several local composition models and stochastic optimization methods have been used and compared in data fitting of activity coefficients in aqueous electrolytes. We have utilized the electrolyte-NRTL model of Chen et al. and the modified Wilson models proposed by Xu and Macedo, and Zhao et al. to fit the activity coefficients of several quaternary ammonium salts in water at 25°C. These electrolytes have interesting properties for their application as ionic liquids. However, the modeling of their thermodynamic behavior using local composition models is a global optimization problem. In this study, several stochastic optimization methods have been used to solve this optimization problem and their numerical performances have been compared. Specifically, we have tested the classical Simulated Annealing and the hybrid methods: Direct Search Simulated Annealing, Simplex Coding Genetic Algorithm, Simulated Annealing Heuristic Pattern Search, and Directed Tabu Search. Our results show that Simulated Annealing is a suitable tool for data fitting of the activity coefficients of aqueous electrolytes. Finally, the tested models can satisfactorily correlate the mean activity coefficients of electrolytes treated in this study, and are suitable for process design.

A data embedding scheme for color images based on genetic algorithm and absolute moment block truncation coding

Recently, embedding a large amount of secret data into gray-level and color images with low distortion has become an important research issue in steganography. In this paper, we propose a data embedding scheme by using a well-known genetic algorithm, block truncation code and modification direction techniques to embed secret data into compression codes of color images to expand the variety of cover media. In the scheme, the common bitmap generation procedure of GA-AMBTC has been modified to speed up the hiding procedure. Two embedding strategies are proposed to hide secret data into the common bitmap and the quantization values in each block of the cover image. Experimental results confirm that the proposed scheme can provide high data capacity with acceptable image quality of the stego-images. Moreover, the compression ratio of the scheme is exactly the same as that of GA-AMBTC so that attackers cannot detect any trace of hidden data from the size of the modified compressed result.

Modifications of real code genetic algorithm for global optimization

A series of modifications to the real coded genetic algorithm for the task of locating the global minimum of multidimensional functions are presented here. These modifications include a new stopping rule, a novel mutation mechanism and a periodically application of a local search procedure. The proposed modifications are tested on a series of optimization problems and the results are reported.

Optimization of thin conical frusta for impact energy absorption

Experimental and numerical investigations were carried out to optimize thin-walled conical shells for their use in design for energy absorption. Geometrical parameters, such as bottom diameter, height, and semi-apical angle were considered to obtain the design space. The numerical analysis and impact experiments were designed using design of experiments (DOE). A three-level, second-order Box–Bhenken technique was used to select the design points from the design space. Various set of numerical simulations were carried out using LS-DYNA. To investigate the influence of flow stress of the material on the energy absorption, numerical simulations were carried out using frusta made of aluminium, zinc, and mild steel. From the numerical results, mathematical models were created using response surface methodology (RSM). With the help of impact experiments carried out on specimens made of zinc on a drop mass test rig, a mathematical model has been developed using RSM. The mathematical models developed using experimental data and the numerical data were used as objective functions for optimization of the design. The non-dominated sorting genetic algorithm code NSGA II was used to optimize the design. The mathematical models were also used to predict the energy absorbed and deformation. The influence of various design parameters on energy absorption has been analysed and is discussed.

Optimization of a large scale industrial reactor by genetic algorithms

The present work aims to employ genetic algorithms (GAs) to optimize an industrial chemical process, characterized by being difficult to be optimized by conventional methods. The considered chemical process is the three phase catalytic slurry reactor in which the reaction of the hydrogenation of o-cresol producing 2-methyl-cyclohexanol occurs. In order to describe the dynamic behavior of the multivariable process, a non-linear mathematical model is used. Due to the high dimensionality and non-linearity of the model, a rigorous one, the solution of the optimization problem through conventional algorithms does not always lead to convergence. This fact justifies the use of an evolutionary method, based on the GAs, to deal with this process. In this way, in order to optimize the process, the GA code is coupled with the rigorous model of the reactor. The aim of the optimization through GAs is the searching of the process inputs that maximizes the productivity of 2-methyl-cyclohexanol subject to the environmental constraint of conversion. Many simulations are conducted in order to find the maximization of the objective function without violating the constraint. The results show that the GAs are used successfully in the process optimization. The selection of the most important GA parameters making use of a factorial design approach by fractional factorial design is proposed. A factorial design approach by a central composite design is also proposed in order to determine the best values of the GA parameters that lead to the optimal solution of the optimization problem.

A Proposed Genetic Algorithm Coding for Flow-Shop Scheduling Problems

A new genetic algorithm coding is proposed in this paper to solve flowshop scheduling problems. To show the efficiency of the considered approach, two examples, in pharmaceutical and agro-food industries are considered with minimization of different costs related to each problem as a scope. Multi-objective optimization is thus, used and its performances proved.

New genetic algorithms (GA) to optimize PWR reactors: Part III: The Haling power depletion method for in-core fuel management analysis

The Haling Power Distribution (HPD) has been applied in a unique process to greatly accelerate the in-core fuel management optimization calculations. These calculations involve; the arrangement of fuel assemblies (FAs) and the placement of Burnable Poisons (BPs) in the fresh FAs. The HPD deals only with the arrangement of FAs. The purpose of this paper is to describe past uses of the HPD, provide an example selected from many similar calculations to explain why and how it can be used, and also to show its effectiveness as a filter in the GARCO GA code. The GARCO (Genetic Algorithm Reactor Core Optimization) is an innovative GA code that was developed by modifying the classical representation of the genotype and GA operators. A reactor physics code evaluates the LPs in the population using the HPD Method, which rapidly depletes the core in a single depletion step with a constant power distribution. The HPD is used basically in GARCO as a filter to eliminate invalid LPs created by the genetic operators, to choose a reference LP for BP optimization, and to create an initial population for simultaneous optimization of the LP and BP placement into the core. The accurate depletion calculation of the LP with BPs is done with the coupled lattice and reactor physics CASMO-4/SIMULATE3 package. However, the fact that these codes validate safety of the core with the added BP placement design also validates the use of the HPD method. The calculations are applied to the TMI-1 core as an example PWR providing concrete results.

Mooring optimization of floating platforms using a genetic algorithm

This paper presents a new procedure for the optimization of the mooring design of floating platforms, in which an automatic design sequence is also established. Regarding the optimization philosophy, the following aspects are dealt with: • The optimization of the platform heading and its mooring pattern, taking into account the environmental force spreading; • optimum line length or line tension for each mooring line, associated to the optimization of the mooring line materials and sizes. Basically, the main goal of this paper is to introduce a new method, which will provide the quickest way to find the best mooring system, defined here as that which minimizes platform responses. A genetic algorithm (GA) is applied in this contribution, and this paper describes exactly the procedure of developing a GA code directed toward the solution of mooring design optimization problems. In order to prove the efficiency and the vast potential of the proposed algorithm as a design tool, sample moorings are analyzed for different environmental conditions and the final results, including the time required to run them, are presented.

Study of automatic designing of line heating technique parameters

Based on experimental data of line heating, the methods of vector mapping, plane projection, and coordinate converting are presented to establish the spectra for line heating distortion discipline which shows the relationship between process parameters and distortion parameters of line heating. Back-propagation network (BP-net) is used to modify the spectra. Mathematical models for optimizing line heating techniques parameters, which include two-objective functions, are constructed. To convert the multi-objective optimization into a single-objective one, the method of changing weight coefficient is used, and then the individual fitness function is built up. Taking the number of heating lines, distance between the heating lines' border (line space), and shrink quantity of lines as three restrictive conditions, a hierarchy genetic algorithm (HGA) code is established by making use of information provided by the spectra, in which inner coding and outer coding adopt different heredity arithmetic operators in inherent operating. The numerical example shows that the spectra for line heating distortion discipline presented here can provide accurate information required by techniques parameter prediction of line heating process and the technique parameter optimization method based on HGA provided here can obtain good results for hull plate.

APPLICATION OF GENETIC ALGORITHMS TO THE OPTIMIZATION OF AN INDUSTRIAL REACTOR

Genetic Algorithms (GAs) have shown great potential and ability to solve complex problems of optimization in diverse industrial fields, including chemical engineering process. In this paper, the main objective is to develop and implement a GA code in an industrial reactor of Cyclic Alcohol (CA) production for the optimization of operational parameters. The intention is to show that this technique is suitable for the maximization of Cyclic Alcohol production, obtaining good results with operational improvements (reduction of catalyst, reduction of the temperature of the process). The results show that the best performance of the process was achieved with the application of GAs. The developed procedure works very well in all the considered conditions which cover the most usual operating range for the considered process.

Genetic Algorithm Coding Methods for Leather Nesting

The problem of placing a number of specific shapes in order to minimise waste is commonly encountered in the sheet metal, clothing and shoe-making industries. The paper presents genetic algorithm c...

GHHAGA for Environmental Systems Optimization

The global optimization of complicated nonlinear systems is mathematically intractable and such an optimization extensively exists in science and engineering. Once an objective function has many local extreme points, the traditional optimization methods may not obtain the global optimization efficiently. A genetic algorithm (GA) based on the genetic evolution of a species provides a robust procedure to explore broad and promising regions of solutions and to avoid being trapped at the local optimization. However, the computational amount is very large. To reduce computations and to improve the computational accuracy, a method based on the two-point crossover and two-point mutation of the hybrid accelerating genetic algorithm with Hooke-Jeeves searching operator is developed for systems optimization. With the shrinking of searching range, the method gradually directs to optimal result by the excellent individuals obtained by Gray code genetic algorithm embedding with Hooke-Jeeves searching operator and Hooke-Jeeves algorithm. The efficiency of the new algorithm is verified by application of several test functions. The comparison of our GA with six existing other algorithms is presented. This algorithm overcomes the Hamming-cliff phenomena in other existing genetic methods, and is proved to be very efficient for the given environmental systems optimization.

Reliable Shape Optimization of Structures Subjected to Transient Dynamic Loading Using Genetic Algorithms

This paper presents a reliable method of solution of two dimensional shape optimization problems subjected to transient dynamic loads using Genetic Algorithms. Boundary curves undergoing shape changes have been represented by B‐splines. Automatic mesh generation and adaptive finite element analysis modules are integrated with Genetic algorithm code to carry out the shape optimization. Both space and time discretization errors are evaluated and appropriate finite element mesh and time step values as obtained iteratively are adopted for accurate dynamic response. Two demonstration problems have been solved, which show convergence to the optimal solution with number of generations. The boundary curve undergoing shape optimization shows smooth shape changes. The combinations of automatic mesh generator with proper boundary definition capabilities, analysis tool with error estimation and Genetic algorithm as optimization engine have been observed to behave as a satisfactory shape optimization environment to deal with real engineering problems.