No description available.

No description available.

No description available.

No description available.

No description available.

This thesis describes two different techniques for efficient control of slip energy in a slip-ring induction machine. The static Kramer system merely recovers slip power and returns it to the a.c. supply. As a result only sub-synchronous motoring or super-synchronous generating is possible. In the static Scherbius system, however, the slip power can be controlled both into and out of the secondary circuit. This allows the machine to operate as a motor and generator at both sub- and super-synchronous speeds. For wide speed range operation a current source inverter was used as this can inherently provide reversal of power flow. The operating requirements for the current source inverter operating in the secondary circuit of an induction machine have been determined. These considerations show that the current source inverter control signal must be synchronised to the secondary e.m.f. of the machine. The machine can then operate in a stable manner over a very wide speed range. The conventional analysis of the current source inverter has been developed to include the effect of the secondary slip e.m.f. which is shown to have a major effect on the commutation behaviour of the inverter. The action of the commutation circuit is affected by the phase angle between the secondary current and the slip e.m.f. This angle can be controlled electronically and the effect of this has been predicted and observed. A detailed study of the Kramer system has included analysis of the d.c. link current waveform including Fourier harmonic prediction in terms of the circuit parameters and the operating slip. The operation of the Kramer and Scherbius systems has been studied for both motoring and generating modes of the induction machine and their relative merits have been compared. In particular the novel idea of using the Scherbius system for variable speed wind energy recovery has been considered and reported in a published paper. Finally suggestions have been made for further work particularly for application to wind energy recovery.

D.Ing. (Electrical and Electronic Engineering) / The non-linear resonant pole (NLRP) inverter is part of the family of soft switching topologies based on resonant phenomena. The sequence of commutation that occurs between the semiconductors of a conventional voltage source inverter is modified through the mechanisms of energy exchange between added passive energy storage components. The NLRP inverter, through its psuedo resonant behaviour (resonant transition), gives rise to zero voltage and zero current turn-on of the switching devices as well as soft turn-off. The switching device voltage stresses are around 1 p.u, while the current stresses are reduced to around 1.3 p.u, by feeding back a portion of the load current. The rms current flowing through the inductor and switches is greatly reduced by driving the inductor into saturation (non-linear mode of operation). The advantages of soft switching, such as high switching frequency which allows greater dynamic response and higher power densities, along with reduced EMI, are achieved with this topology. Detailed analysis at multi- and sub-cycle levels is carried out, resulting in circuit equations and the criteria for commutation success. The commutation boundaries of the inverter are defined and methods discussed on how to extend them. The modulation of the NLRP inverter and some aspects regarding its use as part of both low and high performance induction motor drives are presented.

This dissertation presents theoretical and experimental results from a research program that was aimed at finding practical ways of transferring energy to various loads, mainly from an inductive energy store fluxed by a primary store such as a capacitor bank. The main obj ecti ves of the work were to investigate and develop high power opening and closing switches, together with the transfer circuits needed to generate the fast (less than lOOns duration) high energy, pulses required in many applications. The study was to include a feasibility study of the use of the Plasma Erosion Opening Switch (PEOS) in such a system. To produce the large fast pulses required, an opening switch is required that: * Carry a current of the order of several kA during the inductor storage time. It should also be able to interrupt this current and to withstand the high voltage it will experience as the current is subsequently transferred to a load. * Conduct for as long as possible (up to one quarter period of the current waveform), to maximise the inductively stored energy which can be transferred to xhe load when the switch opens. * Open to an impedance that is large compared to the load impedance . This ensures that most of the inductively stored energy is transferred to the load. Open sufficiently rapidly to produce the required sharp pulse of voltage. In pulsed-power applications, energy is usually supplied from slow and relatively inexpensive power sources such as a capacitor bank, or an explosive flux-compression generator, which deliver large quantities of energy in the lO-lOO time range. Although no single switch is currently available which has such a long conduction time, together with a nanosecond opening time, the PEOS is a potential candidate. To overcome its short conduction time, while still obtaining an opening time of less than lOOns, the PEOS is used together with an additional slower stage or stages of switching. The key to this method is that each successive switching stage produces a considerably increased voltage. Various different types of switch were investigated and these are described in the thesis. Particular consideration is paid to the performance of the PEOS, as the final conditioning stage. Exploding foils are also investigated, together with a novel Automatic Exploding Foil Change-Over Switch, since an exploding foil opening switch is needed to condition the output of the capacitor bank before the PEOS. The initial resistance of the PEOS is very low, and the change-over switch is required to ensure that the current transfer takes place when the voltage across the fuse approaches its peak value. An important part of the investigation was to develop a mathematical model of the PEOS, as a part of the power condi tioning circuit, in order to simulate the system for different load conditions. The thesis explains the design, operation and performance characteristics of the various pulsed-power components, such as capacitor banks, closing and opening switches, pulse transformer, the vacuum system required for a PEOS, and high voltage and current measurement techniques.

A major problem that arises in the inverter-fed induction motor drives is the inevitable introduction of harmonics into the alternating output voltage of the inverter. These harmonics, particularly the low order ones, produce harmful effects such as torque pulsations, excessive motor heating etc. Various techniques for controlling inverters have been proposed in order to reduce the harmonic content of the inverter output waveforms and to effect the voltage/frequency relationship for optimum utilisation of the motor. The latest forms of inverter control employ sinusoidal Pulse-Width-Modulation (PWM) for producing an AC supply with variable voltage and frequency.

To increase transmission capability and to improve transient stability of long distance transmission systems, use has been made of series capacitors and high-speed excitation systems. Unfortunately, their use has resulted in sub synchronous resonance (SSR) and reduced machine damping.