<p> This thesis defends the use of genetic algorithms (GA) to solve the maximum number of repetitions in a binary string. Repetitions in strings have significant uses in many different fields, whether it is data-mining, pattern-matching, data compression or computational biology 14]. Main extended the definition of repetition, he realized that in some cases output could be reduced because of overlapping repetitions, that are simply rotations of one another [10]. As a result, he designed the notion of a run to capture the maximal leftmost repetition that is extended to the right as much as possible. Franek and Smyth independently computed the same number of maximum repetition for strings of length five to 35 using an exhaustive search method. Values greater than 35 were not computed because of the exponential increase in time required. Using GAs we are able to generate string with very large, if not the maximum, number of runs for any string length. The ability to generate strings with large runs is an advantage for learning more about the characteristics of these strings. </p> / Thesis / Master of Science (MSc)

No / Abstract: This paper is to apply the species conserving genetic algorithm (SCGA) to search multiple solutions of truss topology optimization problems in a single run. A real-vector is used to represent the corresponding cross-sectional areas and a member is thought to be existent if its area is bigger than a critical area. A finite element analysis model has been developed to deal with more practical considerations in modeling, such as existences of members, kinematic stability analysis and the computation of stresses and displacements. Cross-sectional areas and node connections are taken as decision variables and optimized simultaneously to minimize the total weight of trusses. Numerical results demonstrate that some truss topology optimization examples have many global and local solutions and different topologies can be found by using the proposed algorithm in a single run and some trusses have smaller weight than the solutions in the literature.

The multi-response optimization (MRO) problem in response surface methodology (RSM) is quite common in industry and in many other areas of science. During the optimization stage in MRO, the desirability function method, one of the most flexible and popular MRO approaches and which has been utilized in this research, is a highly nonlinear function. Therefore, we have proposed use of a genetic algorithm (GA), a global optimization tool, to help solve the MRO problem. Although a GA is a very powerful optimization tool, it has a computational efficiency problem. To deal with this problem, we have developed an improved GA by incorporating a local directional search into a GA process. In real life, practitioners usually prefer to identify all of the near-optimal solutions, or all feasible regions, for the desirability function, not just a single or several optimal solutions, because some feasible regions may be more desirable than others based on practical considerations. We have presented a procedure using our improved GA to approximately construct all feasible regions for the desirability function. This method is not limited by the number of factors in the design space. Before the optimization stage in MRO, appropriate fitted models for each response are required. The parametric approach, a traditional RSM regression technique, which is inflexible and heavily relies on the assumption of well-estimated models for the response of interests, can lead to highly biased estimates and result in miscalculating optimal solutions when the user's model is incorrectly specified. Nonparametric methods have been suggested as an alternative, yet they often result in highly variable estimates, especially for sparse data with a small sample size which are the typical properties of traditional RSM experiments. Therefore, in this research, we have proposed use of model robust regression 2 (MRR2), a semi-parametric method, which combines parametric and nonparametric methods. This combination does combine the advantages from each of the parametric and nonparametric methods and, at the same time, reduces some of the disadvantages inherent in each. / Ph. D.

Predicting the performance of complex structures exposed to harsh thermal environments is a crucial issue in many of today's aerospace and space designs. To predict the thermal stresses a structure might be exposed to, the thermal properties of the independent materials used in the design of the structure need to be known. Therefore, a noninvasive estimation procedure involving Genetic Algorithms was developed to determine the various thermal properties needed to adequately model the Outer Wing Subcomponent (OWS), a structure located at the trailing edge of the High Speed Civil Transport's (HSCT) wing tip. Due to the nature of the nonlinear least-squares estimation method used in this study, both theoretical and experimental temperature histories were required. Several one-dimensional and two-dimensional finite element models of the OWS were developed to compute the transient theoretical temperature histories. The experimental data were obtained from optimized experiments that were run at various surrounding temperature settings to investigate the temperature dependence of the estimated properties. An experimental optimization was performed to provide the most accurate estimates and reduce the confidence intervals. The simultaneous estimation of eight thermal properties, including the volumetric heat capacities and out-of-plane thermal conductivities of the facesheets, the honeycomb, the skins, and the torque tubes, was successfully completed with the one-dimensional model and the results used to evaluate the remaining in-plane thermal conductivities of the facesheets, the honeycomb, the skins, and the torque tubes with the two-dimensional model. Although experimental optimization did not eliminate all correlation between the parameters, the minimization procedure based on the Genetic Algorithm performed extremely well, despite the high degree of correlation and low sensitivity of many of the parameters. / Master of Science

No / Objectives The presence of microcalcifications (MCs), clusters of tiny calcium deposits that appear as small bright spots in a mammogram, has been considered as a very important indicator for breast cancer diagnosis. Much research has been performed for developing computer-aided systems for the accurate identification of MCs, however, the computer-based automatic detection of MCs has been shown difficult because of the complicated nature of surrounding of breast tissue, the variation of MCs in shape, orientation, brightness and size. Methods and materials This paper presents a new approach for the effective detection of MCs by incorporating a knowledge-discovery mechanism in the genetic algorithm (GA). In the proposed approach, called knowledge-discovery incorporated genetic algorithm (KD-GA), the genetic algorithm is used to search for the bright spots in mammogram and a knowledge-discovery mechanism is integrated to improve the performance of the GA. The function of the knowledge-discovery mechanism includes evaluating the possibility of a bright spot being a true MC, and adaptively adjusting the associated fitness values. The adjustment of fitness is to indirectly guide the GA to extract the true MCs and eliminate the false MCs (FMCs) accordingly. Results and conclusions The experimental results demonstrate that the incorporation of knowledge-discovery mechanism into the genetic algorithm is able to eliminate the FMCs and produce improved performance comparing with the conventional GA methods. Furthermore, the experimental results show that the proposed KD-GA method provides a promising and generic approach for the development of computer-aided diagnosis for breast cancer.

Feature selection is a pre-processing step in data-mining and machine learning, and plays an important role for analyzing high-dimensional data. Appropriately selected features can not only reduce the complexity of the mining or learning process, but also improve the accuracy of results. In the past, the concept of performing the task of feature selection by attribute clustering was proposed. If similar attributes could be clustered into groups, attributes could be easily replaced by others in the same group when some attribute values were missed. Hong et al. also proposed several genetic algorithms for finding appropriate attribute clusters. Their approaches, however, suffered from the weakness that multiple chromosomes would represent the same attribute clustering result (feasible solution) due to the combinatorial property, thus causing a larger search space than needed. In this thesis, we thus attempt to improve the performance of the GA-based attribute-clustering process based on the grouping genetic algorithm (GGA). Two GGA-based attribute clustering approaches are proposed. In the first approach, the general GGA representation and operators are used to reduce the redundancy of chromosome representation for attribute clustering. In the second approach, a new encoding scheme with corresponding crossover and mutation operators are designed, and an improved fitness function is proposed to achieve better convergence speed and provide more flexible alternatives than the first one. At last, experiments are made to compare the efficiency and the accuracy of the proposed approaches and the previous ones.

This thesis identifies some problems, the genetic algorithm (GA) is facing in the area of vehicle routing and proposes various methods to address those problems. Those problems arise from the unavailability of suitable chromosomal representation and evaluation schemes of GA for the Vehicle Routing Problem (VRP). The representation and evaluation schemes already in use have problems of high computational cost, illegal chromosomes (chromosomes not representing a legal tour) and wrong fitness assignment (fitness not truly representing chromosome genetic makeup). These problems are addressed by several proposed new schemes, namely the Self Imposed Constraints Evaluation scheme, the Contour and Reverse Contour Evaluation schemes and the Order Skipping Evaluation scheme, which are specifically tailored for various objectives, problems and situations. Apart from this, a methodology, which has previously being used in other meta-heuristics, is incorporated into GA i.e., the independent application of GA on various sub-localities of the problem. We call this GA, a Localized Genetic Algorithm (LGA). LGA is an iterative procedure between optimization and controlled de-optimization. The procedure of controlled de-optimization is also novel. It brings the solution into a new search space while controlling its cost effectively. LGA is introduced with various search techniques, i.e. intensive, extensive and selective, the use of which depends on the problem size and the availability of computational resources. Furthermore, search reduction techniques (Fitness Approximation Methods) are also introduced into the LGA, which has enabled the LGA to be applied to large scale problems. Due to the implementation of those proposals, LGA is the first GA-driven approach to be applied to very large scale CVRP problems of up to 1200 customers, i.e. datasets presented by Feiyue in 2005 and large scale VRPTW problems of up to 1000 customers, datasets presented by Gehring and Homberger in 1999. Lastly, a standard unit for computational comparison, i.e., Bellman's Evaluation Units BEUs, is also introduced to facilitate computational comparisons for future researchers. LGA has shown promising results on CVRP and VRPTW problems. It is flexible and also has the potential to be extended to not only other vehicle routing problems, but also to other ordering problems.

No / A genetic algorithm (GA) augmented with knowledge-based methods has been developed for solving the unit commitment economic dispatch problem. The GA evolves a population of binary strings which represent commitment schedules. The initial population of schedules is chosen using a method based on elicited scheduling knowledge. A fast rule-based dispatch method is then used to evaluate candidate solutions. The knowledge-based genetic algorithm is applied to a test system of ten thermal units over 24-hour time intervals, including minimum on/off times and ramp rates, and achieves lower cost solutions than Lagrangian relaxation in comparable computational time.

Operators of parking guidance and information (PGI) systems often have difficulty in providing the best car park availability information to drivers in periods of high demand. A new PGI configuration model based on the optimized combination method was proposed by analyzing of parking choice behavior. This article first describes a parking choice behavioral model incorporating drivers perceptions of waiting times at car parks based on PGI signs. This model was used to predict the influence of PGI signs on the overall performance of the traffic system. Then relationships were developed for estimating the arrival rates at car parks based on driver characteristics, car park attributes as well as the car park availability information displayed on PGI signs. A mathematical program was formulated to determine the optimal display PGI sign configuration to minimize total travel time. A genetic algorithm was used to identify solutions that significantly reduced queue lengths and total travel time compared with existing practices. These procedures were applied to an existing PGI system operating in Deqing Town and Xiuning City. Significant reductions in total travel time of parking vehicles with PGI being configured. This would reduce traffic congestion and lead to various environmental benefits.

Stricter regulations and evolving environmental concerns have been exerting ever-increasing pressure on the automotive industry to produce low carbon vehicles that reduce emissions. As a result, increasing numbers of alternative powertrain architectures have been released into the marketplace to address this need. However, with a myriad of possible alternative powertrain configurations, which is the most appropriate type for a given vehicle class and duty cycle? To that end, comparative analyses of powertrain configurations have been widely carried out in literature; though such analyses only considered limited types of powertrain architectures at a time. Collating the results from these literature often produced findings that were discontinuous, which made it difficult for drawing conclusions when comparing multiple types of powertrains. The aim of this research is to propose a novel methodology that can be used by practitioners to improve the methods for comparative analyses of different types of powertrain architectures. Contrary to what has been done so far, the proposed methodology combines an optimisation algorithm with a Modular Powertrain Structure that facilitates the simultaneous approach to optimising multiple types of powertrain architectures. The contribution to science is two-folds; presenting a methodology to simultaneously select a powertrain architecture and optimise its component sizes for a given cost function, and demonstrating the use of multi-objective optimisation for identifying trade-offs between cost functions by powertrain architecture selection. Based on the results, the sizing of the powertrain components were influenced by the power and energy requirements of the drivecycle, whereas the powertrain architecture selection was mainly driven by the autonomy range requirements, vehicle mass constraints, CO2 emissions, and powertrain costs. For multi-objective optimisation, the creation of a 3-dimentional Pareto front showed multiple solution points for the different powertrain architectures, which was inherent from the ability of the methodology to concurrently evaluate those architectures. A diverging trend was observed on this front with the increase in the autonomy range, driven primarily by variation in powertrain cost per kilometre. Additionally, there appeared to be a trade-off in terms of electric powertrain sizing between CO2 emissions and lowest mass. This was more evident at lower autonomy ranges, where the battery efficiency was a deciding factor for CO2 emissions. The results have demonstrated the contribution of the proposed methodology in the area of multi-objective powertrain architecture optimisation, thus addressing the aims of this research.