No description available.

The thesis presents a method of analysis of direct current machines using thyristor assisted commutation. The method is based upon the General Machine Theory and the formulation of voltage and torque equations from the theory is shown to be successful. These equations have predicted the performance of the machine over a wide range of operating conditions. A computer programme has been developed to deal with the most complex operating conditions and, by suitable simplifications, the programme is readily adaptable to compute steady state characteristics. A subsidary theme of the thesis is to show how the output of thyristor assisted machines may be controlled by an electronic method of rocking the brushes. The method is simply demonstrated on no load or light load conditions but loading effects produce some limitations. Sufficient experimental results are given to illustrate the method.

A closed-loop computer-controlled single-sheet test system has been developed to characterise lamination materials and to measure, the iron loss density under any specified flux density waveform. The system has been 'used to validate predictions from a recently developed theoretical model, for the calculation of the excess loss component associated with domaiQ wall movement, under flux density waveforms typical of those encountered in the stator core of brushless permanent magnet dc motors. In addition, an improved expression for the calculation of the iron loss density component, from measured 71 and 7!vectors, due to rotatio~ in non-purely rotating flux conditions, has been derived. A simple analytical model from which the airgap flux density and spread of magnet working points can be determined and which accounts for the effects of curvature for radial-field permanent magnet machines has been developed and validated. The model has been coupled to an analytical technique for the prediction of the open-circuit flux density waveforms in different regions of the stator core, and has subsequently been employed for the prediction of the open-circuit iron loss. In order to predict the iron loss under any specified load condition, a technique which couples a brushless dc drive system simulation to a series of magnetostatic finite element analyses corresponding to discrete instants in a commutation cycle has been developed. It enables the prediction of the local flux density waveforms throughout the stator core under any operating condition, and has been employed to predict the local iron loss density distribution 'and the total iron loss and their variation with both the load and the commutation strategy, Finally, the theoretical findings have been validated against measurements on a representative low power brushless drive system.

No description available.

Established methods of measuring are loss in mercury-are rectifier's are surveyed and examined critically. In particular the difficulty of obtaining a satisfactory instrument deflection is discussed. Two new methods developed by the writer, incorporating a dynamometer instrument are then described. The first of these requires protective gear of lighter rating than is required by the established methods, but provides a smaller instrument deflection. The second also requires protective gear of only light rating, but provides about twice the Instrument deflection obtainable with the established methods. In the second half of the thesis, a survey of established methods of electronic multiplication by analogue circuits is given, followed by a description of a logarithmic multiplier developed by the writer, which incorporates a moving-coil instrument and which is suitable for power measurement, particularly with interrupted d.c. waveforms of the type encountered in the anode circuit of a mercury-are rectifier. The circuit performance of the various units of the multiplier is examined in relation to the degree of accuracy required. Test figures demonstrating the performance of the multiplier with various types of input are given; and a brief description of a proposed extension of the system to deal with alternating quantities is also included. The a.c. mains stabilizer circuit used to regulate the power supply to the electronic circuits of the multiplier is described in outline.

This thesis is concerned with the investigation and development of methods to monitor the condition of electrical machines, with particular reference to induction machines fed from variable frequency, power electronic supplies. A variety of techniques are suggested whereby the faults that most commonly afflict electrical machines can be identified. Techniques involving the detection and analysis of line current, frame vibration, and leakage magnetic field are proposed and developed to take account of the rich spectrum of harmonics that can be injected through the use of power electronic inverters. To do this it was necessary to produce methodologies that could differentiate between spectral components originating in the inverter, and those produced as a result of the fault. As an extension, two further monitoring techniques were developed in an attempt to provide a fully comprehensive strategem for assessing the health of electrical drives. They are sequence component monitoring, and pulsed eddy current monitoring. The former is an adjunct to those techniques previously mentioned, and when used in combination with them leads to a greater security of diagnosis. The use of pulsed eddy currents is aimed at the on line detection of material failure in the rotating shafts of the machines. For each technique suggested, a comprehensive theoretical basis is provided, and its effectiveness verified by suitable experiment.

Several aspects of linear induction devices have been analysed and measured, mainly with regard to their application in tracked ground transport. A linear induction machine (the magnetic river) which can provide lift, lateral guidance and thrust forces in transport schemes has been extensively tested at standstill. The results are compared with calculated predictions obtained by finite element and Fourier harmonic techniques. High lift and thrust forces were found to be incompatible characteristics of such machines. This, coupled with the need for separate control of the three-axis forces, indicates that these forces are best provided by different machines on the same vehicle. The damping of vertical disturbances (heave) at speed in inductively levitated vehicles is theoretically investigated using the above mathematical techniques, very little damping being found. The flux density level in the track backing iron associated with an axial flux linear induction machine at speed is calculated. This quantity is of considerable economic importance as it determines the optimum track iron thickness. A new contactless device for measuring linear velocities, a linear tachometer, is analysed and tested.

This work presents HVDC/FACTS control device implementation framework in the Gulf cooperative council’s countries. It comprises of five layers of FACTS control devices (STATCOM, SSSC, UPFC, HVDC and centralized/De-centralized Control). This five-layer architecture is designed in order to configure and produce the desired results; based on these outcomes, GCC power system network control and operational problems can be identified and addressed within the control architecture on the GCC power grid. In the context of power FACTS-FRAME, this work is to identify and determine a number of power systems operational and control problems which are persistent on the GCC power grid e.g. poor voltage quality (SAG-Swell), poor load flow control, and limited power transfer capacity issues. The FACTS-FRAME is configured and synthesized by integrating multiple FACTS control devices (STATCOM, SSSC, UPFC) in parallel at different locations on the GCC power grid in order to meet stringent power system control and operational requirements with improved power transfer capacity, controllability and reliability. The mathematical models are derived to indentify and determine operational constraints on the GCC power grid by incorporating real-time and estimated data and the acquired desired results. Herein, FACTS-FRAME is designed to handle distributed computation for intensive power system calculation by integrating multiple FACTS devices on multiple networks within the GCC power network. Distributed power flow algorithms are also derived in order to understand and implement centralized and decentralized control topologies as appropriate. The simulation results indicate the feasibility of FACTS devices implementation and their potential benefits under current operating conditions on the GCC power grid.

The following chapters give details of the construction of a new equivalent circuit for induction motors used in variable and modulated frequency supplies and one of its advantages for the determination of the additional losses in the machine due to harmonic frequencies. Chapter 1 includes the aim of the work and briefly explains the reason for obtaining such a model and its applications. In Chapter 2 the technique of constructing the new equivalent circuit is given. This is done as follows. With the aid of a digital computer and circle fit technique the frequency responses of the open and short circuit test results taken at various supply frequencies are used to find the values of the new equivalent circuit parameters. The new equivalent circuit takes care of the parameter changes arising from different supply frequencies and produces the correct input and output performance characteristics. The new equivalent circuit also produces the exact responses for mixed and modulated frequencies. In Chapter 3 the new equivalent circuit is analysed and performance equations are derived. This enables changes in the performance characteristics at different supply frequencies to be calculated directly from the performance equations. Using a digital computer the non-linear model of the induction motor is constructed from the test results for variable speed applications. The magnitude of the errors introduced in calculating the machine's performance characteristics are determined using the linear model in the non-linear region. Their magnitude becomes greater with the increasing value of the rotor speed or the supply frequency. An accurate torque speed plotting machine for extremely high speed induction machines is designed and constructed. Highly accurate results are achieved by sensing the rotor torque and speed without any physical contact with the rotor. In Chapter 4 the Fourier series analysis is used to analyse and derive the output harmonic contents of two popular types of inverter. The additional losses in the machine due to these harmonic frequencies of both inverter types are determined and compared. The new equivalent circuit is used to determine the magnitude of the errors induced in the calculation of these additional losses if the standard equivalent circuit is used.

The investigation described, in this is designed to produce a digital closed loop speed control system for a d.c. motor using a thyristor power amplifier and a digital, minicomputer as integral components of the system. The thyristor amplifier is direct digitally controlled and also has digital current limiting protection. A digital form of a conventional analogue proportional controller, and a scheduled controller based on pre-recorded performance characteristics for the motor are first discussed, but rejected as impractical because of the complexity of the resulting system. Control based on a bang-bang principle with a variable mode switching control action designed to minimise limit cycling is then introduced. This method is shown to produce very acceptable performance using quite simple system software and appears to be immediately applicable to microprocessor implementation. A further refinement, that of dual or adaptive sampling designed to further reduce limit cycling effects in steady state operation, is then introduced. A control or switching matrix is developed as a design method for the switching controller which functions when the control computer recognises the required operating region for the motor as defined by signum functions of error and error rate within the system. The final controller uses multi-mode operation in which the loop gain and feedback constants are controlled as a function of error and error rate. It also incorporates an adaptive sampling period, and results in a digital speed control system which has both simple hardware and software. Suggestions for methods of further improving performance using a system of parallel processing with a view to using a microprocessor as the control computer, but still retaining system simplicity, are also discussed.