No description available.

No description available.

No description available.

No description available.

No description available.

This thesis studies dynamic routing in circuit-switched non-hierarchical networks based on learning automata algorithms. The application of a mathematical model for a linear reward penalty algorithm is explained. Theoretical results for this scheme verified by simulations shows the accuracy of the model. Using simulation and analysis, learning automata algorithms are compared to several other strategies on different networks. The implemented test networks may be classified into two groups. The first group are designed for fixed routing and in such networks fixed routing performs better than any dynamic routing scheme. It will be shown that dynamic routing strategies perform as well as fixed routing when trunk reservation is employed. The second group of networks are designed for dynamic routing and trunk reservation deteriorates the performance. Comparison of different routing algorithms on small networks designed to force dynamic routing demonstrates the superiority of automata under both normal and failure conditions. The thesis also considers the instability problem in non-hierarchical circuit-switched networks when dynamic routing is implemented. It is shown that trunk reservation prevents instability and increases the carried load at overloads. Finally a set of experiments are performed on large networks with realistic capacity and traffic matrices. Simulation and analytic results show that dynamic routing outperforms fixed routing and trunk reservation deteriorates the performance at low values of overload. At high overloads, optimization of trunk reservation is necessary for this class of networks. Comparison results show the improved performance with automata schemes under both normal and abnormal traffic conditions. The thesis concludes with a discussion of proposed further work including expected developments in Integrated Service Digital Networks.

The work described in this part of the thesis is involved mainly with uprating a 55W air-turbine driven permanent magnet generator to 250W. It is shown that the electrical performance of the generator can be predicted from a simple equivalent circuit comprising an, induced emf source with a series inductance and resistance. When matching the generator and turbine characteristics iron losses are included as an additional torque requirement. Analysis has identified that the most important parameters which determine the rating-of the generator are the stator flux, linkage, stator inductance, and number of pole pairs. Investigations have therefore centred around the calculation of these quantities. Previous design methods for calculating the parameters have been dependent on experimental data for particular magnet geometries. They are not sufficiently general to permit design calculations for magnets having radically different shapes or properties. Therefore the finite element method is used to predict the magnetic field distribution, from which the stator winding flux linkages and inductance, and the saliency torque, are predicted, an important parameter when matching the turbine and generator at starting. The finite element method offers significant advantages over analytical methods because it can account accurately for leakage flux, it can handle complex configurations of magnetic circuit and the directional properties of the magnet, and it allows different parts of the magnet to operate at different flux density levels. The performance of the generator is predicted with reasonable accuracy. Alternative rotors for the 250W generator have been designed and tested. Results have shown that the existing generator can be adapted to the 250W design simply by using a rare-earth magnet rotor, and selecting a suitable stator winding turns factor for matching the generator and bulb. Two methods of matching the load characteristics of the turbine, generator and bulb, to produce an acceptable system have been developed. A simple electronic protection circuit has been designed to prevent the 250W generator from overspeeding in the event of bulb failure. Additionally it can limit over-voltages, caused by variations in the pressure of the air supply, which would otherwise decrease the life of the bulb.

A comprehensive analytical technique is developed for predicting the instantaneous magnetic field distribution in radial-field, surface-mounted permanent magnet brushless DC motors under any load condition and commutation strategy. It is based on a 2-dimensional analysis in polar coordinates and accounts implicitly for the corresponding stator winding current waveforms and the effect of stator slot openings. In addition, a 2-dimensional analytical method for calculating the back-emf waveform is presented, whilst the analytical technique is applied to the prediction of the cogging torque waveform and the calculation of the self- and mutual-winding inductances. Also developed and validated is an analytical model for predicting the steady-state dynamic performance of a 3-phase brushless DC drive, by exploiting the periodicity in the stator winding voltage and current waveforms, with due account of the influence of commutation events in the inverter bridge, the back-emf waveform, current limiting, and commutation timing etc. The model is developed further to couple with the motion equation of the rotor to enable the transient and steady-state dynamic performance of brushless DC drives to be predicted. The effect of end-shields on the vibrational behaviour of stators is investigated by the modem modal analysis technique, and new formulae for the calculation of the acoustic power radiated by a cylindrical stator of finite length, using an analytical method, are presented. A technique which combines the finite element method and Fourier analysis to account for the effects of end-shields on the acoustic radiation is developed, and the spherical acoustic radiation model of motors has been improved by the application of finite elements. Finally, a systematic analytical approach to the estimation and analysis of the acoustic noise from a radial-field, internal rotor, brushless DC motor is presented.

No description available.

The principle of operation dictates that the output of a cycloconverter contains some harmonics. For drive applications, the harmonics at best increase losses in the motor and may well cause instability. Various methods of analysing the output waveform have been considered. A Fortran 77 program employing a modified Fourier series, making use of the fact that the input waveforms are sinusoidal, was used to compute the individual harmonic amplitudes. A six pulse three phase to single phase cycloconverter was built and a Z-80 microprocessor was used for the control of firing angles. Phase locked loops were used for timing, and their effect upon the output with changing input frequency and voltage were established. The experimental waveforms are analysed by a FFT spectrum analyser. The flexibility of the control circuit enables the following investigations not easily carry out using traditional analog control circuit. The phase relationship between the cosine timing and reference wave in the cosinusoidal control method was shown to affect the output waveform and hence the harmonic content. There is no clear optimum value of phase and the T.H.D. up to 500Hz remains virtually constant. However, the changes of individual harmonic amplitudes is quite significant. In practice it may not be possible to keep the value of phase constant but it should be considered when comparing control strategies. Another investigation involves the changing of the last firing angle in a half cycle. It shows that the value of firing angles produced by the cosinusoidal control method is desirable. Operation at theoretical maximum output frequency was also demonstrated.