No description available.

published_or_final_version / Computer Science / Master / Master of Philosophy

Random number generators (RNGs) are widely used in conducting Monte Carlo simulation studies, which are important in the field of statistics for comparing power, mean differences, or distribution shapes between statistical approaches. Statistical results, however, may differ when different random number generators are used. Often older methods have been blindly used with no understanding of their limitations. Many random functions supplied with computers today have been found to be comparatively unsatisfactory. In this study, five multiplicative linear congruential generators (MLCGs) were chosen which are provided in the following statistical packages: RANDU (IBM), RNUN (IMSL), RANUNI (SAS), UNIFORM(SPSS), and RANDOM (BMDP). Using a personal computer (PC), an empirical investigation was performed using five criteria: period length before repeating random numbers, distribution shape, correlation between adjacent numbers, density of distributions and normal approach of random number generator (RNG) in a normal function. All RNG FORTRAN programs were rewritten into Pascal which is more efficient language for the PC. Sets of random numbers were generated using different starting values. A good RNG should have the following properties: a long enough period; a well-structured pattern in distribution; independence between random number sequences; random and uniform distribution; and a good normal approach in the normal distribution. Findings in this study suggested that the above five criteria need to be examined when conducting a simulation study with large enough sample sizes and various starting values because the RNG selected can affect the statistical results. Furthermore, a study for purposes of indicating reproducibility and validity should indicate the source of the RNG, the type of RNG used, evaluation results of the RNG, and any pertinent information related to the computer used in the study. Recommendations for future research are suggested in the area of other RNGs and methods not used in this study, such as additive, combined, mixed and shifted RNGs.

Made available in DSpace on 2014-10-09T12:50:09Z (GMT). No. of bitstreams: 0 / Made available in DSpace on 2014-10-09T14:01:56Z (GMT). No. of bitstreams: 1 10558.pdf: 10529307 bytes, checksum: be6b9ffcdd1b8416d5246549ff0f8d3d (MD5) / Dissertacao (Mestrado) / IPEN/D / Instituto de Pesquisas Energeticas e Nucleares - IPEN/CNEN-SP

Made available in DSpace on 2014-10-09T12:25:21Z (GMT). No. of bitstreams: 0 / Made available in DSpace on 2014-10-09T14:02:57Z (GMT). No. of bitstreams: 1 02288.pdf: 2985772 bytes, checksum: bad64fd8162abc4a61ca0eec6fc0a2cf (MD5) / Dissertacao (Mestrado) / IPEN/D / Instituto de Pesquisas Energeticas e Nucleares - IPEN/CNEN-SP

Made available in DSpace on 2014-10-09T12:50:09Z (GMT). No. of bitstreams: 0 / Made available in DSpace on 2014-10-09T14:01:56Z (GMT). No. of bitstreams: 1 10558.pdf: 10529307 bytes, checksum: be6b9ffcdd1b8416d5246549ff0f8d3d (MD5) / Dissertacao (Mestrado) / IPEN/D / Instituto de Pesquisas Energeticas e Nucleares - IPEN/CNEN-SP

Made available in DSpace on 2014-10-09T12:25:21Z (GMT). No. of bitstreams: 0 / Made available in DSpace on 2014-10-09T14:02:57Z (GMT). No. of bitstreams: 1 02288.pdf: 2985772 bytes, checksum: bad64fd8162abc4a61ca0eec6fc0a2cf (MD5) / Dissertacao (Mestrado) / IPEN/D / Instituto de Pesquisas Energeticas e Nucleares - IPEN/CNEN-SP

Combined electrical and structural models of five types of permanent magnet linear electrical machines suitable for direct-drive power take-off on wave energy applications are presented. Electromagnetic models were developed using polynomial approximation to finite element analysis results. The structural models are based on simple beam theory, other classical techniques, and automated finite element analysis formulations. The machine models have been integrated with a time-domain model of a wave energy converter based on a heaving buoy. They have then been optimised using a genetic algorithm approach, using a score based primarily on the amortised cost per unit of energy production. The optimised designs have then been used for a comparison of the economic performance of the generator types.

published_or_final_version / Electrical Engineering / Doctoral / Doctor of Philosophy

This dissertation describes a system that constructs efficient, retargetable code generators and optimizers. chop reads nonprocedural descriptions of a computer's instruction set and of a naive code generator for the computer, and it writes an integrated code generator and peephole optimizer for it. The resulting code generators are very efficient because they interpret no tables; they are completely hard-coded. Nor do they build complex data structures to communicate between code generation and optimization phases. Interphase communication is reduced to the point that the code generator's output is often encoded in the program counter and conveyed to the optimizer by jumping to the right label. chop's code generator and optimizer are based on a very simple formalism, namely rewriting rules. An instrumented version of the compiler infers the optimization rules as it complies a training suite, and it records them for translation into hard code and inclusion into the production version. I have replaced the Portable C Compiler's code generator with one generated by chop. Despite a costly interface, the resulting compiler runs 30% to 50% faster than the original Portable C Compiler (pcc) and generates comparable code. This figure is diluted by common lexical analysis, parsing, and semantic analysis and by comparable code emission. Allowing for these, the new code generator appears to run approximately seven times faster than that of the original pcc.