摘要:

The loop or ring-type data networks using full duplex-communication links between the nodes have been investigated. The results presented in the paper concern two specific modes for handling the messages. These are termed ‘store and forward’ and ‘check and forward’. Both modes have been analysed using extensions of the single-server queueing theory and also simulated extensively using GPSS software.The message traffic is Poisson distributed and can arise from any node with equal probability. The destination is selected to be any other node again with equal probability. The study has also considered the performance of the two modes for traffic having priorities. One conclusion arrived at in this investigation is that the traffic with priorities can be better handled by a network which handles messages in either store and forward or check and forward mode depending on the priority of the message.

DOI:10.1049/piee.1977.0036

出版商:IEE    

年代:1977

数据来源: IET

摘要:

Network analysis, in terms ofS-parameters, is invariably carried out in modern textbooks by employing the power-wave concept. It is shown that this approach is not relevant to practical measurement; furthermore, the complication of square roots, real parts, conjugates, moduli and conjugate transpose matrices is carried throughout the mathematics. These difficulties are overcome by employing the voltage-wave concept with complex normalisation, whereby the analysis is carried out in terms of port-voltage and port-current relationships, from which, power flow relationships are readily obtained in a straightforward manner. The new method of network analysis presented is not only completely general and rigorous, but simple to apply and faster than any other known technique. It is based on a once-for-all solution of the most general problem of all in 3-port analysis, namely, a 3 × 3 renormalising matrix, each element of which embodies three renormalising factors. Examples are given of the derivation of some 2-port and 3-port network equations, but in a more general form in which the port normalising impedances are not only different but complex. It is concluded that the voltage-waveS-parameter approach, with complex normalisation, should not only be revived and taught in academic institutions, but that it should take precedence over the power-wave approach which has a very restricted application. In a companion paper, a reappraisal is made of computer-corrected network analyser design and calibration.

DOI:10.1049/piee.1977.0037

出版商:IEE    

年代:1977

数据来源: IET

摘要:

A rigorous analysis is carried out of a typical dual-directional coupler network analyser to demonstrate the application of a new circuit analysis technique based on the principle of matrix renormalisation, described in a companion paper, to cascaded multiport networks. The manner in which previously published calibration equations were derived is explained in detail. The analysis is carried out external to the network in terms of voltage excitation and the voltage responses at the accessible ports of the network under mismatched conditions. Each section of the network is represented by a generalised scattering matrix defined in terms of voltage-waveS-parameters with complex normalisation. In this manner, a knowledge of the internal flow paths of the network is unnecessary, in contrast to the flowgraph method employing the nontouching loop rule. The analysis shows that current methods of network analysis, based on the power-flow concept, have given rise to certain misconceptions because source mismatch does not contribute to systematic error-as is generally believed. In addition, these misconceptions appear to have led to the adoption of general design principles for network analysers that are unnecessarily complicated because the effects of mismatches, discontinuities and poor directivity can be completely eliminated by calibration in terms of air-dielectric line standards, provided that all port mismatches are invariant between the reflection and transmission modes of operation.

DOI:10.1049/piee.1977.0038

出版商:IEE    

年代:1977

数据来源: IET

摘要:

A new form of inverter-fed brushless synchronous motor is described, having a cylindrical rotor with internal flux barriers and a 3-phase rotor winding connected to two diodes. An improved analysis and calculation method is presented for this and related types of machine. Very good synchronous performance and correlation with theoretical predictions are demonstrated by practical tests.

DOI:10.1049/piee.1977.0039

出版商:IEE    

年代:1977

数据来源: IET

摘要:

Numerical solutions are presented for temperature distributions in the coils of a disc-type transformer winding. The numerical method was applied to plane coils corresponding to those used in a full-scale model of an oil-cooled disc-type transformer winding. The agreement between theoretical and experimental results was found to be very good.

DOI:10.1049/piee.1977.0040

出版商:IEE    

年代:1977

数据来源: IET

摘要:

The paper details a method by which the equations describing a 3-phase synchronous machine may be transformed into the frequency domain to provide a means of digitally simulating faults at the generator terminals. Evaluation of the Fourier integrals involved in determining the machine response is achieved by means of the fast-Fourier-transform algorithm, and numerical results are presented together with the criteria necessary to achieve an accuracy comparable with that attainable using direct-phase quantities. The methods provide a basis for combining synchronous sources with transmission lines having frequency-variant parameters, and which are not therefore amenable to simulation using other than frequency-domain techniques. Studies using the new method indicate that, when predetermining the armature currents of a faulted machine, considerable economies in computing time may be effected, and the possibility of improving the simulation to simulate more accurately rotor eddycurrent effects by utilising the direct- and quadrature-axis frequency response emerges.

DOI:10.1049/piee.1977.0041

出版商:IEE    

年代:1977

数据来源: IET

摘要:

Three methods of transient analysis of a d.c. separately excited motor, controlled by a chopper with square-wave output voltage, are derived. These methods are compared with regard to accuracy and usefulness. The methods require much less computation time than necessary for a step-by-step solution of the system differential equations, and, in some cases, can be used for deriving a transfer function of the pulse controlled motor.

DOI:10.1049/piee.1977.0042

出版商:IEE    

年代:1977

数据来源: IET

摘要:

With the replacement of steam locomotives by diesel and electric tractors, it was natural to look to electrical energy as a heat source for passenger trains. When substantial parts of the main lines were electrified at 25 kV, the opportunity to change from steam heating to electric heating was taken. To maintain compatibility between electric and diesel locomotives some of the latter were required to be capable of providing a power source for electric train heating. The train-heating-supply systems which became standard are discussed, together with the various means of providing these supplies from a diesel locomotive. As an example of the method generally adopted for British railways, the train-heating installation on one particular type of locomotive is described in some detail, together with some of the practical problems encountered and the solutions adopted to overcome these. As a more recent example, the heating system fitted to British Rail's new high-speed train is discussed. In conclusion, the paper attempts to indicate likely directions for future developments.

DOI:10.1049/piee.1977.0045

出版商:IEE    

年代:1977

数据来源: IET

摘要:

The large number of published circuits for forced commutation inverters has created a selection problem for the user and designer. The cost of commutation components, with reference to rating, is often disregarded when selecting a particular circuit. The efficiency of the commutation circuit and its effect on the efficiency of the main circuit, is also often disregarded. The paper compares on a per unit basis the commutation, and powercircuit components, for five typical inverter circuits, and provides guidance on which configuration is likely to suit a particular application. While the comparison is restricted to a single-phase bridge invertor, results can frequently be extrapolated to polyphase circuits.

DOI:10.1049/piee.1977.0046

出版商:IEE    

年代:1977

数据来源: IET

摘要:

The decomposition of the constrained allocation problem for active power generation is considered, and a problem formulation developed which enables the grouping of the constraints to form a structure suitable for linear decomposition. Dantzig and Wolfe showed that a finite convergence algorithm is a suitable solution of the decomposed primal problem in which a series of linear, constrained subproblems are co-ordinated by a global problem of transformed variables. The solution of a large dispatch problem is presented and the computational requirements contrasted with conventional primal and dual linear-programming techniques. The technique is concluded to permit significant computational savings and also to be suitable for implementation on a multilevel or multiprocessor computer system.

DOI:10.1049/piee.1977.0047

出版商:IEE    

年代:1977

数据来源: IET