Artificial Neural Networks (ANNs) have proved to be useful in approximating non- linear systems in many applications including motion control. ANNs advocated in switched reluctance motor (SRM) control typically have a large number of neurons and several layers which impedes their real time implementation in embedded sys- tems. Real time estimation at high speeds using these ANNs is diffcult due to the high number of operations required to process the ANN controller. An insuffcient availability of time between two sampling intervals limits the available computation time for both processing the neural net and the other functions required for the motor drive. One ideal application of ANNs in SRM control is rotor position estimation. Due to reliability issues, elimination of the rotor position sensors is absolutely required for high volume, high speed and low cost applications of SRM's. ANNs provide a means by which drive designers can implement position sensorless drive technology that is both robust and easily implemented. It is demonstrated that a new and novel ANN configuration can be implemented for accurate rotor position estimation in a sensorless SRM drive. Consisting of just 4 neurons, the neural estimator is the smallest of its kind for SRM rotor position estimation. The breakthrough that provided the reduction was the addition of a non- linear input. Typical input spaces for SRM position neural estimators consist of both current,and fux-linkage. The neural network was trained on-line using these inputs and a third, non-linear input provided by a preprocessed product of the two typical inputs. / Master of Science

This study concerns the contributions of Major Albert Sobey and his educational leadership during the development of the engineering cooperative education program that became the foundation for the General Motors Institute in Flint, Michigan. This study also examines Albert Sobey's contributions to the emergence of industrial cooperative education in America over the past seventy years.

No description available.

Switched Reluctance Motors (SRMs) are gaining recognition due to their robust design, cost-effectiveness, fault tolerance, and reliable high-speed performance, positioning them as promising alternatives to traditional electric motors. However, SRMs face high torque ripples, vibration, acoustic noise, and nonlinear modeling complexities. Through careful geometry design optimization, these drawbacks can be mitigated. Design optimization for SRMs is a multi-objective and nonlinear problem that requires an accurate finite element analysis (FEA) model to relate designable parameters to output objectives. The geometric design process follows a multi-stage and iterative approach, leading to prohibitive computational time until the optimal design is reached. Machine learning algorithms (MLAs) have recently acquired attention in electric machine design. This study introduces an extensive analysis of various MLAs applied to SRM modeling and design. Additionally, it presents a robust framework for a comprehensive evaluation of these MLAs, facilitating the selection of the optimal machine learning topology for SRM design. Existing research on the geometry optimization of SRMs using MLAs has focused only on the machine’s static characteristics. This thesis introduces an advanced optimization method utilizing an MLA to act as a surrogate model for both static and dynamic characteristics of the SRM. The dynamic model incorporates conduction angles optimization to enhance the torque profile. The proposed MLA is applied to map out the SRM geometrical parameters, stator and rotor pole arc angles and their dynamic performance metrics, such as average torque and torque ripples. The optimal design improves the average torque and significantly reduces the torque ripples. Radial forces constitute a critical objective that should be considered alongside average torque, efficiency, and torque ripple in the design optimization of SRMs. Accurate modeling of radial forces is a prerequisite for optimizing motor geometry to mitigate their adverse effects on vibrations and acoustic noise. This work presents an MLA-based surrogate model for the most influential radial force harmonic components, facilitating the integration of radial force reduction into a multi-objective optimization framework. The proposed optimization framework employs two MLA-based surrogate models: the first correlates SRM pole arc angles with average torque and torque ripples, while the second models the most significant radial force harmonics. A genetic algorithm leverages these surrogate models to predict new geometrical parameters that enhance the SRM's torque profile and reduce radial forces. The optimization framework significantly reduced torque ripples and radial forces while slightly increasing average torque. The optimal design candidates were verified using FEA and MATLAB simulations, confirming the effectiveness of the proposed method, which offers significant computational time savings compared to traditional FEA techniques. / Thesis / Doctor of Philosophy (PhD)

During the course of the experimental work included in this investigation, it was found that certain adjustments and conditions had a major effect on the performance of the tested engine, whereas certain others had a very minor effect. In order of decreasing importance, the conditions which affected the engine's operation were: (1) restriction of the air supply to the carburetor; (2) restriction of the flow of exhaust products from the engine; (3) a change in the carburetor float level; (4) a variation of the ignition timing within five degrees of that recommended; (5) the use of spark plugs which had been run a larger mileage than that recommended by the manufacturer; and (6) a change of the distributor breaker point gap within 0.007" limits of that recommended by the manufacturer. Restricting the air supply to the engine had a very large adverse effect on the economy of the engine, and a smaller effect on the power. Restriction of the exhaust line had approximately the same effect on the power and economy of the engine. Raising the carburetor float level decreased economy characteristics markedly, and power characteristics to a much smaller extent. Lowering the carburetor float level increased both the power and economy of the engine. A five degree advance in the ignition timing resulted in a slight increase in power and economy in the low speed range, this increase being at the expense of violent detonation. The performance in the high speed range was adversely affected. Retarding the spark timing five degrees had a detrimental effect on both power and economy, although this effect was comparatively small. The use of spark plugs beyond the recommended mileage limit showed a decrease in power and economy when this effect was not counteracted by the increase in the size of the gap due to normal usage. Spark plugs which had been cleaned at frequent intervals showed up much better on test than did those which had not. A change in the spacing of the distributor breaker points through a 0.014” range showed an effect of less than three percent when the natural effect of this change on the ignition timing was corrected. Things such as air cleaner and muffler restrictions, changes in the carburetor float level, and fouled spark plugs caused enough effect on the engine's performance to indicate the advisability of checking these things often. Other conditions which would normally be checked in a minor engine tune-up do not have an appreciable effect on the performance of the engine, providing that their effect is not sufficient to make the engine miss. / M.S.

Thesis (PhD (Electric and Electronic Engineering))--University of Stellenbosch, 2009. / Keywords: position sensorless control, torque control, synchronous machines The work in this thesis deals with energy e cient torque control and rotor position estimation in the full speed range, for a family of synchronous machines that should be used more often in the near future. This family consists of the permanent magnet synchronous machine (PMSM), the reluctance synchronous machine (RSM), the interior-PMSM and the PMassisted- RSM. By designing and controlling these synchronous machines correctly, better performance and higher energy e ciency can be expected compared to the performance and e ciency of an industry standard induction machine. However, applications are limited to variable speed drives (VSD) in a certain power range, e.g. below 100kW. With the growing concern and necessity of a better utilization of energy, it is becoming standard to use electronically controlled power converters between the electricity grid and electrical machines. Therefore, there is a very large scope for the implementation of this synchronous machine technology. For traction applications like electrical vehicles, the optimally controlled synchronous machine technology has a very strong position. Very compact and robust synchronous machines with a very high power density can be designed that may out-perform the induction machine by far. However, one major requirement for most applications is position sensorless control, i.e. rotor position estimation in the whole speed range. To achieve energy e cient torque control, maximum torque per Ampere (MTPA) control should be implemented. It is possible to achieve MTPA control at low speed, but above the rated speed of the machine, eld weakening needs to be performed. The question is how to implement MTPA and e ective eld weakening for any value of speed and DC bus voltage and for any machine within this family of synchronous machines. In this thesis a method is explained to achieve this goal using results from nite element (FE) analysis directly. The scheme may be implemented within a very short period of time. The contribution of this thesis is a general understanding of the problems at hand, with an in-depth view into the mathematical representation of synchronous machines, a generic method of energy e cient torque control and a thorough study of rotor position and speed estimation methods.

A presente pesquisa viabilizou, possivelmente, a expansão do arcabouço teórico desenvolvido por Suarez (2004) e Lynn et al. (1996), seguindo a mesma direção dos resultados obtidos na pesquisa longitudinal desenvolvida por Yu et al. (2011). Desenvolvemos um estudo descritivo de modo a analisar um período de 42 anos com o objetivo de entender os fatores que influenciaram a trajetória do desenvolvimento da tecnologia do veículo elétrico na GM de modo que utilizamos o arcabouço teórico desenvolvido por Suarez (2004) como norteador das observações empíricas. Coletamos dados secundários da tecnologia de baterias referente a fase um (realização de P&D) e dados secundários da tecnologia do veículo elétrico referente às fases dois (protótipos) e fase três (criação de mercado) para a construção do estudo de caso. As observações dessas fases viabilizaram os seguintes resultados: a possível expansão do arcabouço conceitual do Suarez (2004) em relação ao nível da firma (sondagem e aprendizagem tecnológica, sondagem e aprendizagem mercadológica e transferência de conhecimento) e ao nível do ambiente (setores industriais associados, pressão social, política internacional e questão econômica); e a possível ampliação da vertente teórica de sondagem e aprendizagem mercadológica desenvolvida por Lynn et al. (1996) no sentido de constatarmos a sondagem e aprendizagem tecnológica em um momento anterior a sondagem e aprendizagem mercadológica. Na década passada, iniciou-se um processo de efervescência de novos entrantes na indústria do veículo elétrico e, nessa década, duas das principais montadoras mundiais (GM e Nissan) lançaram os seus primeiros veículos elétricos para serem produzidos em larga escala. Essas evidências empíricas indicam uma possibilidade real da tecnologia do veículo elétrico estar em um estágio de maturidade que viabilize a realização do processo de transição tecnológica indicada no arcabouço conceitual desenvolvido por Utterback e Abernathy, em 1975. Nesse sentido, os resultados teóricos obtidos nessa pesquisa podem apoiar o processo de tomada de decisão no contexto de incentivo a essa tecnologia tanto da perspectiva da firma quanto da perspectiva do governo. / In alignment with the results obtained in the longitudinal research conducted by Yu et al. (2011), the present study possibly enabled the expansion of the theoretical framework developed by Suarez (2004) and Lynn et al. (1996). A descriptive study was developed so as to analyse a 42 year period with views to understanding the factors which influenced the course of development of electric vehicle technology at GM. Thus, to guide empirical observations, the theoretical framework developed by Suarez (2004) was employed. To build the study case, phase one (execution of R&D) battery technology and phase two (prototypes) and three (market shaping) electric vehicle technology secondary data was collected. Findings observed in these phases gave rise to the following results: the possible expansion of Suarez´s (2004) conceptual framework at corporate level (technological survey and learnings, market survey and learnings and knowledge transfer) and at environment level (associated industrial segments, social pressure, international politics and economic issues); and the possible extension of the theoretical market probe and learning line of thought developed by Lynn et al. (1996) in as much as the identification of the occurrence of technological probe and learning, at a moment preceding that of market probe and learning, is concerned. Over the last decade, a new entrant commotion process sprung within the electric vehicle industry and during this decade, two of the top world manufacturers (GM and Nissan) launched their first electrical vehicles for mass production purposes. This empirical evidence indicates that currently, there is an effective possibility of electric vehicle technology being at a stage of maturity whereby the technological transition mentioned in the conceptual framework developed by Utterback and Abernathy in 1975, is feasible. To this effect, the theoretical results obtained in this study may support decision processes in as much as incentives are concerned, both from a corporate and government perspective.

Quan analitzem el funcionament de les màquines elèctriques dubtem en l'elecció d'un o altre conjunt de variables, models o joc de paràmetres. En el treball que segueix hem trobat, molt sovint, resultats que es contradiuen amb les mesures experimentals, o temps de càlcul molt diferents segons el model escollit o el tipus de simulació realitzada.La necessitat d'una bona modelització es planteja com un requisit previ per a obtenir un bon control i un adequat dimensionat dels accionaments; amb aquest fi es planteja el treball amb el contingut i ordre que segueix.En primer lloc (Capítol 1) es fa un repàs de l'estat de l'art. Segueix l'estudi (Capítol 2) de models de màquines asíncrones amb caracterització de la saturació; podem plantejar diversos conjunts de variables que ho fan possible amb més o menys dificultat de càlcul i resolució. La línia escollida planteja un model transformat que redueix al mínim el nombre de paràmetres a estimar; en altres dels casos plantejats es redueix la càrrega de càlcul per a la seva resolució. Tanmateix es planteja l'anàlisi de sensibilitat de la resposta (Capítol 3) respecte les variacions dels paràmetres; òbviament hi haurà paràmetres que tindran poca o nul.la influència en la resposta del sistema, així com d'altres que tindran influència directa i immediata en la mateixa.Una vegada escollit el model o models a emprar s'han de trobar els paràmetres dels mateixos. Una primera aproximació es dedica a la determinació dels paràmetres mitjançant assajos fora de línia (Capítol 4). La determinació dels paràmetres mitjançant assajos en línia i tècniques d'estimació en temps real (Capítol 5), juntament amb la consideració d'observadors per a variables auxiliars, per exemple el flux, permeten tancar el tema d'estimació dels paràmetres. La determinació dels parells electromagnètics i de càrrega (Capítol 6), és l'objectiu final del treball. L'aplicació d'observadors i estimadors permet l'obtenció de les variables desitjades resultant un model més precís i exacte.A modus d'aplicació final s'ha construït un mesurador de parell en règim transitori; aspecte residual del treball que resulta força important ja que no disposem, actualment, d'un mesurador d'aquestes característiques. L'equip desenvolupat és format a partir d'una màquina de corrent continua i un sistema de control.Tanmateix es tracta la modelització del conjunt convertidor-màquina-càrrega (Capítol 7). No pretenem analitzar a fons la modelització del convertidor, això podria constituir un treball més específic; malgrat això la influència que presenta el tipus de senyal aplicat a la màquina aconsella la seva consideració a fi i efecte d'obtenir resultats més acurats que els que permeten la suposició d'una alimentació sinusoïdal.Un capítol final de discussions (Capítol 8) sobre els resultats obtinguts, als que segueixen aquelles aplicacions que s'entenen millorades, així com un recull de concussions tanca el cos central del treball.Es tanca el document amb una relació bibliogràfica rellevant (Bibliografia) i un apartat d'annexes on es troben desenvolupaments matemàtics: equacions de sensibilitat (Annex I), càlcul dels guanys de l'observador (Annex II), els assajos realitzats i les característiques de les màquines assajades (Annex III) i el mètode dels mínims quadrats. Aplicacions (Annex IV). / Al analizar el funcionamiento de las máquinas eléctricas dudamos acerca de la elección de, uno u otro, conjunto de variables, modelos, o conjunto de parámetros. En este trabajo se han encontrado, frecuentemente, resultados que contradicen lo medido experimentalmente, o tiempos de cálculo muy diversos según el modelo adoptado o el tipo de simulación realizada; cosa por otra parte lógica en un ámbito que admite aportaciones de mejora.La necesidad de una buena modelización actualmente se plantea como una premisa para obtener un mejor control y un adecuado dimensionado de los accionamientos; con este fin se aborda este trabajo determinando la secuenciación que sigue.En primer lugar (Capítulo 1) se repasan los antecedentes y estado actual. Seguidamente se ha planteado (Capítulo 2) el estudio de modelos de máquinas asíncronas con caracterización de la saturación; existen diversos modelos que lo posibilitan a costa de una mayor o menor dificultad de cálculo y resolución. La línea escogida toma un modelo transformado que reduce el número de parámetros a encontrar o estimar; en algunos de los modelos planteados se reduce la carga de cálculo necesaria para su resolución. Asimismo se plantea el análisis de sensibilidad de la respuesta (Capítulo 3) respecto a las variaciones de los parámetros; es evidente que habrá parámetros que tendrán escasa influencia en la respuesta del sistema, así como otros que influirán de forma directa e inmediata en la misma.Una vez establecido el modelo, o modelos a utilizar, deben encontrarse los parámetros de los mismos. Una primera parte se dedica a la determinación de parámetros mediante ensayos fuera de línea (Capítulo 4), lo que permite aproximarnos a la solución deseada. La determinación de parámetros mediante ensayos en línea y técnicas de estimación en tiempo real (Capítulo 5), junto con la consideración, en su caso, de observadores para variables auxiliares como el flujo, permiten cerrar el capítulo de estimación de parámetros.La determinación de variables auxiliares como los pares electromagnético y de carga (Capítulo 6), es el objetivo final del trabajo. La aplicación de observadores y estimadores permite la obtención de las variables deseadas, resultando un modelo mas preciso y exacto.A modo de aplicación se ha obtenido un medidor de par en régimen transitorio; aspecto colateral del trabajo que resulta relevante dado que no se dispone, actualmente, de un medidor de par con estas características. El equipo desarrollado se basa en un conjunto que comprende máquina de corriente continua y sistema de control.Finalmente se aborda la modelización conjunta del sistema convertidor-máquina-carga (Capítulo 7). No se pretende analizar a fondo la modelización del convertidor, ello podría ser objeto de un estudio más específico; no obstante la influencia que tiene el tipo de señal aplicada a la máquina, aconseja su consideración con el fin de obtener resultados más reales que una suposición de alimentación sinusoidal permitirían.Se finaliza con un capítulo final de discusiones (Capítulo 8) de los resultados obtenidos, a los que siguen aquellas aplicaciones que se entienden como mejoradas por este trabajo, así como las pertinentes conclusiones.Cierra el documento una relación bibliográfica (Bibliografía) relevante y un apartado de anexos donde se añaden deducciones matemáticas prolíficas tales como las de las ecuaciones de sensibilidad (Anexo I), el cálculo de las ganancias del observador (Anexo II) las características de las máquinas ensayadas (Anexo III) y .el método de los mínimos cuadrados. Aplicaciones (Anexo IV). / When analyzing the operation of the electric machines we doubt about the election of, one or other, group of variables, models, or group of parameters. In this work they have been, frequently, results that they contradict that measured experimentally, or very diverse times of calculation according to the adopted model or the type of carried out simulation; it sews logic on the other hand in an environment that admits contributions of improvement.The necessity of a good modelización at the moment thinks about like a premise to obtain a better control and an appropriate dimensionado of the drives; with this end this work is approached determining the sequence that continues.In the first place (Chapter 1) the antecedents and current state are reviewed. Subsequently he has thought about (Chapter 2) the study of models of asynchronous machines with characterization of the saturation; diverse models that facilitate it to coast of a bigger or smaller calculation difficulty and resolution exist. The line chosen taking a transformed model that it reduces the number of parameters to find or to estimate; in some of the outlined models he decreases the necessary calculation load for their resolution. Also he thinks about the analysis of sensibility of the answer (Chapter 3) regarding the variations of the parameters; it is evident that there will be parameters that will have scarce influence in the answer of the system, as well as others that will influence in a direct and immediate way in the same one.Once established the pattern, or models to use, they should be the parameters of the same ones. A first one leaves he dedicates to the determination of parameters by means of test outside of line (Chapter 4), what allows to approach to the wanted solution. The determination of parameters by means of on-line and real time test (Chapter 5), together with the consideration, in their case, of observers for auxiliary variables as the flux, they allow to close the chapter of estimate of parameters.The determination of auxiliary variables as the electromagnetic and load torque (Chapter 6), it is the final objective of the work. The application of observers and observers allow the obtaining of the wanted variables, being a model but I specify and exact.By way of application a meter has been obtained of even in transitory régime; colateral aspect of the work that is outstanding since he doesn't prepare, at the moment, of a meter of even with these characteristics. The developed team is based on a group that he understands machine of D.C. current and control system.Finally the combined modelización of the system converter-machine-load is approached (Chapter 7). It is not thoroughly sought to analyze the modelización of the converter, it could be it object of a more specific study; nevertheless the influence that has the sign type applied to the machine, advises its consideration with the purpose of obtaining realler results that a supposition of feeding sinusoidal would allow.You concludes with a final chapter of discussions (Chapter 8) of the obtained results, to those that follow those applications that understand each other as improved by this work, as well as the pertinent conclusions.He closes the document a bibliographical relationship (Bibliography) outstanding and a section of annexes where such fruitful mathematical deductions are added as those of the equations of sensibility (Annex I), the calculation of the observer's gains (Annex II) the characteristics of the tested machines (Annex III) and the method of the recursive least-squared method and applications (Annex IV).

This dissertation presents an investigation of miniaturized magnetic induction and permanent magnet (PM) machines, intended for use in a microengine. Similar to a macroscale turbogenerator, a microengine comprises a small, gas-fueled turbine engine for converting chemical fuel energy into mechanical power and an integrated electrical generator for converting mechanical power to electrical power. The microengine system is proposed as a revolutionary, high power-density source for portable electronics. In this research, miniaturized magnetic induction machines and PM machines were designed, fabricated, and characterized. Both types of machines used axially directed magnetic fields and were nominally 10 mm in diameter and 1.5-2.3 mm in thickness. Innovative microfabrication techniques were developed to demonstrate the feasibility of integrating magnetic machines within a bulk-micromachined, silicon-based microengine system. Two-phase, eight-pole induction machines were constructed within silicon substrates using Cu coils in a laminated, slotted ferromagnetic NiFe or CoFeNi stator core. Silicon etching, wafer bonding, and electrodeposition were used to form all of the magnetic machine components. The induction machines were characterized in motoring mode using tethered rotors and demonstrated motoring torques of up to 2.5 uN-m. Also, three-phase, eight-pole, surface wound PM machines were built using a hybrid microfabrication/assembly approach. The stators were fabricated by electroplating Cu coils on ferromagnetic NiFeMo (Moly Permalloy) substrates. The rotors were formed by assembling a magnetically patterned SmCo PM with a FeCoV (Hiperco 50) back iron. The PM machines were tested as generators with free-spinning rotors, powered by an air-driven spindle, and demonstrated 2.6 W of mechanical-to-electrical power conversion with continuous DC power generation of 1.1 W at 120 krpm rotor speed. The primary contributions of this work are (1) the demonstration of microfabricated magnetic machines integrated within bulk-micromachined silicon and (2) the demonstration of multi-watt power conversion from a microfabricated PM generator. These achievements represent progress in the ongoing development of silicon-based microengines, but in addition, the fabrication technologies and device structures may find application in other microsystems.

published_or_final_version / Environmental Management / Master / Master of Science in Environmental Management