Showing items 1 - 12 of 12

Your selections:

Improved steganography through the strategic placement of information and the optimization thereof through the use of evolutionary algorithms

  • Authors: Cilliers, Michael
  • Date: 2015
  • Subjects: Cryptography , Computer algorithms , Genetic algorithms , Image processing - Digital techniques , Evolutionary computation
  • Language: English
  • Type: Masters (Thesis)
  • Identifier: http://hdl.handle.net/10210/84521 , uj:19232
  • Description: Abstract: The development of steganography techniques does not occur in isolation. There is an arms race between steganography and steganalysis techniques. The development of a steganography technique that could adapt when needed could be beneficial. This research begins with a literature study exploring existing methods. Both steganog- raphy and steganalysis approaches are covered in order to get an overview of the environ- ment. Optimization methods are also examined to find a suitable method of optimizing the developed algorithm. The information that is gathered in the literature study was then used to develop a steganography algorithm that aims to decrease detectability through the strategic placement of information. The algorithm is developed in such a way as to allow for optimization. A genetic algorithm is implemented to help optimize the embedding of the information in a specific environment. This should allow the algorithm to be re- optimized when new steganalysis techniques are developed. The algorithm should thus remain relevant as steganalysis advances. The developed algorithms show that the placement of information in the image has an effect on its detectability. The developed algorithm even outperforms the random distribution LSB technique. The optimization that was implemented also produced positive results. The dissertation also includes the development of a novel evolutionary algorithm that drew its inspiration from the aphid life cycle. By switching between reproduction strate- gies the algorithm is able to adjust the balance between exploration and exploitation. , M.Sc. (Computer Science)
  • Full Text:

Non-polarised edge filter design using genetic algorithm and its fabrication using electron beam evaporation deposition technique

  • Authors: Ejigu, Efrem Kebede
  • Date: 2013-11-25
  • Subjects: Fiber optics , Fourier transform optics , Genetic algorithms
  • Type: Thesis
  • Identifier: uj:7791 , http://hdl.handle.net/10210/8685
  • Description: D.Phil. (Electrical & Electronic Engineering Science) , Recent advancement in optical fibre communications technology is partly due to the advancement of optical thin-film technology. The advancement of optical thin-film technology includes the development of new and existing deposition and optical filter design methods. Genetic algorithm is one of the new design methods that show promising results in designing a number of complicated design specifications. The research is entirely devoted to the investigation of the genetic algorithm design method in the design of producible polarised and non-polarised edge filters for optical fibre communication applications. In this study, a number of optical filter design methods such as Fourier Transform and refining are investigated for their potential in designing those kinds of structures. Owing to the serious limitations to which they are subject, they could not yield the kind of results anticipated. It is the finding of this study that the genetic algorithm design method, through its optimisation capability, can give reliable and producible designs. This design method, in this study, optimises the thickness of each layer to get to the best possible solution. Its capability and unavoidable limitations in designing polarised and non-polarised beam splitters, edge filters and reflectors from absorptive and dispersive materials are well demonstrated. It is observed that the optical behaviour of the non-polarised filters designed by this method show a similar trend: as the angle of incidence increases the inevitable increase in the percentage of polarisation, stop bandwidth and ripple intensity is well controlled to an acceptable level. In the case of polarised designs the S-polarised designs show a better response to the optimisation process than the P-polarised designs, but all of them are kept well within an acceptable level. It is also demonstrated that polarised and non-polarised designs from the genetic algorithm are producible with great success. This research has accomplished the task of formulating a computer program using genetic algorithm in a Mathlab® environment for the design of producible polarised and non-polarised filters from materials of absorptive and dispersive nature.
  • Full Text:

Improved evolutionary algorithms with application to smart grid demand response management

  • Authors: Essiet, Ima Okon
  • Date: 2019
  • Subjects: Evolutionary computation , Genetic algorithms , Natural computation
  • Language: English
  • Type: Doctoral (Thesis)
  • Identifier: http://hdl.handle.net/10210/296339 , uj:32285
  • Description: D.Ing. (Electrical Engineering) , Abstract: Evolutionary Algorithms (EAs) refer to a group of optimization strategies which are based on Darwin’s theory of natural selection. According to Darwin, attributes of an organism’s genotype are shared among its offspring when mating with another organism occurs. Through this continuous process of mating, combination (or recombination), mutation and selection, it becomes possible to obtain offspring with an optimal balance of gene attributes for both parent organisms. EAs include Evolutionary Strategies (ES), Evolutionary Programming (EP), Genetic Algorithms (GAs), and Differential Evolution (DE). This thesis presents a collection of articles which detail improvements to the performance of existing EAs (mainly based on GAs and DE). The motivation for this research arises from the fact that evolutionary algorithms suffer from inadequacies in handling multiple objectives and constraints which frequently occur in real world problems. In particular, Pareto-based GAs experience difficulty in balancing convergence and diversity in the presence of multiple time-varying objectives and constraints. DE encounters difficulty in finding the global optimum for multiple mutation vectors. This thesis proposes several niching-based strategies to improve the tradeoff between convergence and diversity for GAs. It also presents a 2-archive approach to improve crossover and recombination for DE with multiple mutation vectors. These improved algorithms are tested on both static and dynamic mathematical models representing selected aspects of smart grid, namely: distributed energy resource (DER) allocation, cost function minimization, and demand response management. Results show that the improved EAs provide better optimized parameters for the selected mathematical grid models compared to already existing algorithms. In particular, the improved EAs optimize reference point placement, utilize a pointer-based archiving approach, and adaptively vary crossover rate in order to achieve optimal convergence and diversity of the search population. With respect to GAs, the articles generally adopt a Pareto-based approach to search the solution space. Results obtained from applying a number of improved EA approaches demonstrate their effectiveness. The research in this thesis also details various directions for future research which are discussed at the end of the thesis. , Evolutionary programming (Computer science)
  • Full Text:

Applying genetic algorithms to fly-back converter design

  • Authors: Fivaz, Jean
  • Date: 2012-02-27
  • Subjects: Genetic algorithms , Electric current converters design and construction
  • Type: Thesis
  • Identifier: uj:2073 , http://hdl.handle.net/10210/4420
  • Description: M.Ing. , This thesis investigates how genetic algorithms may be applied to solving for flyback converter design optimization. The genetic algorithm finds the combinations of components and switching frequency required for a capable, efficient and small fly-back solution. Ways of effectively evaluating the proposed solutions are discussed in light of the circuit theories of power electronics, and specifically, fly-back converters. Applying component data effectively to the evaluation process is addressed, especially in the light of the optimization goals. A solution evolved by a genetic algorithm is tested and compared against a prototype designed through conventional methods.
  • Full Text:

Empirical evaluation of optimization techniques for classification and prediction tasks

  • Authors: Leke, Collins Achepsah
  • Date: 2014-03-27
  • Subjects: Genetic algorithms , Statistical decision - Data processing , Decision support systems , Missing observations (Statistics) , Computational intelligence
  • Type: Thesis
  • Identifier: uj:4520 , http://hdl.handle.net/10210/9858
  • Description: M.Ing. (Electrical and Electronic Engineering) , Missing data is an issue which leads to a variety of problems in the analysis and processing of data in datasets in almost every aspect of day−to−day life. Due to this reason missing data and ways of handling this problem have been an area of research in a variety of disciplines in recent times. This thesis presents a method which is aimed at finding approximations to missing values in a dataset by making use of Genetic Algorithm (GA), Simulated Annealing (SA), Particle Swarm Optimization (PSO), Random Forest (RF), Negative Selection (NS) in combination with auto-associative neural networks, and also provides a comparative analysis of these algorithms. The methods suggested use the optimization algorithms to minimize an error function derived from training an auto-associative neural network during which the interrelationships between the inputs and the outputs are obtained and stored in the weights connecting the different layers of the network. The error function is expressed as the square of the difference between the actual observations and predicted values from an auto-associative neural network. In the event of missing data, all the values of the actual observations are not known hence, the error function is decomposed to depend on the known and unknown variable values. Multi Layer Perceptron (MLP) neural network is employed to train the neural networks using the Scaled Conjugate Gradient (SCG) method. The research primarily focusses on predicting missing data entries from two datasets being the Manufacturing dataset and the Forest Fire dataset. Prediction is a representation of how things will occur in the future based on past occurrences and experiences. The research also focuses on investigating the use of this proposed technique in approximating and classifying missing data with great accuracy from five classification datasets being the Australian Credit, German Credit, Japanese Credit, Heart Disease and Car Evaluation datasets. It also investigates the impact of using different neural network architectures in training the neural network and finding approximations for the missing values, and using the best possible architecture for evaluation purposes. It is revealed in this research that the approximated values for the missing data obtained by applying the proposed models are accurate with a high percentage of correlation between the actual missing values and corresponding approximated values using the proposed models on the Manufacturing dataset ranging between 94.7% and 95.2% with the exception of the Negative Selection algorithm which resulted in a 49.6% correlation coefficient value. On the Forest Fire dataset, it was observed that there was a low percentage correlation between the actual missing values and the corresponding approximated values in the range 0.95% to 4.49% due to the nature of the values of the variables in the dataset. The Negative Selection algorithm on this dataset revealed a negative percentage correlation between the actual values and the approximated values with a value of 100%. Approximations found for missing data are also observed to depend on the particular neural network architecture employed in training the dataset. Further analysis revealed that the Random Forest algorithm on average performed better than the GA, SA, PSO, and NS algorithms yielding the lowest Mean Square Error, Root Mean Square Error, and Mean Absolute Error values. On the other end of the scale was the NS algorithm which produced the highest values for the three error metrics bearing in mind that for these, the lower the values, the better the performance, and vice versa. The evaluation of the algorithms on the classification datasets revealed that the most accurate in classifying and identifying to which of a set of categories a new observation belonged on the basis of the training set of data is the Random Forest algorithm, which yielded the highest AUC percentage values on all of the five classification datasets. The differences between its AUC values and those of the GA, SA, PSO, and NS algorithms were statistically significant, with the most statistically significant differences observed when the AUC values for the Random Forest algorithm were compared to those of the Negative Selection algorithm on all five classification datasets. The GA, SA, and PSO algorithms produced AUC values which when compared against each other on all five classification datasets were not very different. Overall analysis on the datasets considered revealed that the algorithm which performed best in solving both the prediction and classification problems was the Random Forest algorithm as seen by the results obtained. The algorithm on the other end of the scale after comparisons of results was the Negative Selection algorithm which produced the highest error metric values for the prediction problems and the lowest AUC values for the classification problems.
  • Full Text:

Performance prediction of trace metals and cod in wastewater treatment using artificial neural network.

Design of comminution circuits for improved productivity using a multi-objective evolutionary algorithm (MOEA)

  • Authors: Mhlanga, Samson , Ndlovu, Jabulani , Mbohwa, Charles , Mutingi, Michael
  • Date: 2011
  • Subjects: Comminution circuits , Evolutionary algorithms , Multi-objective optimisation , Multi-objective evolutionary algorithms , Genetic algorithms
  • Type: Article
  • Identifier: uj:5170 , http://hdl.handle.net/10210/14411
  • Description: The performance of a processing plant has a large impact on the profitability of a mining operation, yet plant desig optimisation decisions are based on past experience and intuition rather than on scientific analysis. Genetic algorithms as a tool for circuit analysis in plant design and optimisation was considered. The multi-objective evolutionary algorithm initialises the plant design and optimisation based on experimental results, which are used to formulate and determine the objective function values. A simulation was conducted to assess the performance of candidate solutions. The two optima are then traded-of using cost objective, which is sought to be minimized. Once an optimum was selected, the circuit mass balance and equipment design was performed, bringing the theory of network design and genetic algorithms into unison. Results of the study provide financial benefits, optimal parameter settings for the comminution equipment and ultimate better plant performance.
  • Full Text:

South African inflation forecasting using genetically optimised neural networks

  • Authors: Molwantoa, Lebogang
  • Date: 2014-03-03
  • Subjects: Neural networks (Computer science) , Artificial intelligence , Inflation (Finance) - Forecasting , Genetic algorithms , Inflation (Finance) - South Africa
  • Type: Thesis
  • Identifier: uj:4219 , http://hdl.handle.net/10210/9577
  • Description: M.Com. (Financial Economics) , Forecasting inflation is an important concern for economists and business alike throughout the world. Despite the relative success of macroeconomic forecasting models in forecasting inflation, there is potential to improve these models to account for nonlinear relationships between inflation and the chosen independent variables. Artificial neural networks (ANNs) have found increased applicability as a potential nonlinear forecasting tool that accounts for nonlinearity found in data. In this study, we investigate the ability of genetically optimised neural networks to forecast South African inflation. The results were compared to economic forecasts obtained from traditional econometric models as well as macroeconomic structural models. The results obtained show that the genetically optimised neural networks indicate some ability to be used as potential forecasting tools. Their biggest advantage over the traditional forecasting techniques is that they do not impose the restriction of linearity on the data to be forecasted.
  • Full Text:

A group genetic algorithm for the fleet size and mix vehicle routing problem

  • Authors: Mutingi, M. , Mbohwa, Charles
  • Date: 2012
  • Subjects: Logistics , Vehicle routing , Genetic algorithms , Vehicle fleets - Management
  • Type: Article
  • Identifier: uj:5181 , http://hdl.handle.net/10210/14423
  • Description: In logistics management, the use of vehicles to distribute products from suppliers to customers is a major operational activity. Optimizing the routing of vehicles is crucial for providing cost-effective services to customers. This research addresses the fleet size and mix vehicle routing problem (FSMVRP), where the heterogeneous fleet and its size are to be determined. A group genetic algorithm (GGA) approach, with unique genetic operators, is designed and implemented on a number of existing benchmark problems. GGA demonstrates competitive performance in terms of cost and computation time when compared to other heuristics.
  • Full Text:

Grouping genetic algorithm for industrial engineering applications

  • Authors: Mutingi, Michael , Mapfaira, Herbert , Dube, Partson
  • Date: 2013
  • Subjects: Grouping problems , Grouping genetic algorithms , Genetic algorithms , Metaheuristics , Industrial engineering
  • Type: Article
  • Identifier: uj:4734 , http://hdl.handle.net/10210/11560
  • Description: Industry is inundated with grouping problems concerned with formation of groups or clusters of system entities for the purpose of improving the overall system efficiency and effectiveness. Various extant grouping problems include cell formation problem, vehicle routing problem, bin packing problem, truck loading, home healthcare scheduling, and task assignment problem. Given the widespread grouping problems in industry, it is important to develop a tool for solving such problems from a common view point. This paper seeks to identify common grouping problems, identify their common grouping structures, present an outline of group genetic algorithm (GGA), and map the problems to the GGA approach. The practicality of the GGA tool in is highly promising in Industrial Engineering applications.
  • Full Text:

Healthcare staff scheduling in a fuzzy environment : a fuzzy genetic algorithm approach

  • Authors: Mutingi, Michael , Mbohwa, Charles
  • Date: 2014
  • Subjects: Healthcare staff scheduling , Nurse scheduling , Fuzzy modeling , Genetic algorithms , Jarosite precipitate
  • Type: Article
  • Identifier: uj:4970 , http://hdl.handle.net/10210/13071
  • Description: In the presence of imprecise management targets, staff preferences, and patients’ expectations, the healthcare staff scheduling problem becomes complicated. The goals, preferences, and client expectations, being humanistic, are often imprecise and always evolving over time. We present a Jarosite precipitate (FGA) approach for addressing healthcare staff scheduling problems in fuzzy environments. The proposed FGA-based approach can handle multiple conflicting objectives and constraints. To improve the algorithm, fuzzy set theory is used for fitness evaluations of alternative candidate schedules by modeling the fitness of each alternative solution using fuzzy membership functions. Furthermore, the algorithm is designed to incorporate the decision maker’s choices and preferences, in addition to staff preferences. Rather than prescribing a sing solution to the decision maker, the approach provides a population of alternative solutions from which the decision maker can choose the most satisfactory solution. The FGA-based approach is potential platform upon which useful decision support tools can be developing for solving healthcare staff scheduling problems in a fuzzy environment characterized with multiple conflicting objectives and preference constraints.
  • Full Text:

Asymmetrical three-phase fault evaluation in a distribution network using the genetic algorithm and the particle swarm optimisation

  • Authors: Shambare, Chikomborero
  • Date: 2020
  • Subjects: Genetic algorithms , Asymmetric synthesis , Short circuits , Electrical engineering , Pattern recognition systems
  • Language: English
  • Type: Masters (Thesis)
  • Identifier: http://hdl.handle.net/10210/413252 , uj:34808
  • Description: Abstract: Modern electric power systems are made up of three main sub-systems: generation; transmission; and distribution. The most common faults in distribution sub-systems are asymmetrical three-phase short circuit faults due to the fact that asymmetrical three-phase faults can be: line-to-line faults; two lines-to-earth faults; and single line-to-earth faults. This increases their probability of occurrence, unlike symmetrical three-phase faults which can only occur when all the three phases have been simultaneously shorted. Standard IEC 60909 and IEC 61363 provide all the basic information that is used for the detection of short circuit faults. However, the two standards use numerous estimates in their faults evaluation procedures. They estimate voltage factors (c), impedance correction factors (k), resistance to reactance ratios (R/X), resistance to impedance ratios (R/Z) and various other scaling factors for rotating machines. These IEC estimates are not evenly distributed throughout the 550kV and as such, they do not sufficiently cater for every nominal voltage. When the need arises, the user has to estimate these values accordingly. This research presents a genetic algorithm (GA) and a particle swarm optimisation (PSO) for the detection of asymmetrical three-phase short circuit faults within electric distribution networks of power systems with nominal voltages less than 550kV. GA and PSO are nature-inspired optimisation techniques. Although PSO has quick convergence, it suffers from partial optimism and premature stagnation. Some innovative coding adjustments were made in the creation of initial positions and particle distribution within the swarm. The GA struggles with: survival rates of individuals; stalling during optimisation; and proper gene replacements. Coding adjustments were also made to GA with regards to: strategic gene replacements; crossover when combining the properties of parents; and the arrangement of scores and expectation. Pattern search and Fmincon algorithms were also added to both algorithms as minimisation functions that commence after the evolutionary algorithms (EAs) terminate. The EAs were initially tested on the Rastrigin and Rosenbrock functions to ensure their efficiencies. During fault detection, the developed EAs were used to stochastically determine some of the most crucial estimates (R/X and R/Z ratios). The proposed methodology would compute these values on a case-to-case basis for every optimisation case with regards to the parameters and unique specifications of the power system. The EAs were tested on a nominal voltage that is properly catered for by Standard IEC. They obtained ratios, impedances and currents that were within an approximate range to the IEC values for that nominal voltage. This further implies that EAs can be reliably used to: stochastically determine these ratios; compute impedances; and detect fault currents for all the nominal voltages including those that are not sufficiently catered for by Standard IEC. Since R/X and R/Z ratios play a key role in determining the upstream and fault point impedances, the proposed methodology can be used to compute much more precise fault magnitudes at various network levels thereby setting up and repairing power systems sufficiently. , M.Ing. (Electrical and Electronic Engineering Science)
  • Full Text:
  • 1
  • Disclaimer
  • Privacy
  • Copyright
  • Contact