New Publications are available for Control technology
http://dl-live.theiet.org
New Publications are available now online for this publication.
Please follow the links to view the publication.Control Engineering Solutions: a practical approach
http://dl-live.theiet.org/content/books/ce/pbce054e
<p xmlns="http://pub2web.metastore.ingenta.com/ns/">This book collects together in one volume a number of suggested control engineering solutions which are intended to be representative of solutions applicable to a broad class of control problems. It is neither a control theory book nor a handbook of laboratory experiments, but it does include both the basic theory of control and associated practical laboratory set-ups to illustrate the solutions proposed.</p>Electrical control of dangerous machinery and processes. Part 2: Electrically interlocked guards on machines
http://dl-live.theiet.org/content/journals/10.1049/jiee-2.1951.0098
The dimming of low-pressure discharge lamps
http://dl-live.theiet.org/content/journals/10.1049/jiee-2.1951.0108
Analysis of sine-type non-linearity in control systems
http://dl-live.theiet.org/content/journals/10.1049/pi-b-2.1962.0179
Some aspects of transductor design, with particular reference to discharge-lamp circuits
http://dl-live.theiet.org/content/journals/10.1049/jiee-2.1953.0251
A thermostat with control temperature independent of ambient temperature
http://dl-live.theiet.org/content/journals/10.1049/jiee-2.1954.0095
Decomposition of large optimal-control problems
http://dl-live.theiet.org/content/journals/10.1049/piee.1967.0163
Large systems composed of linearly interconnected linear dynamic subsystems are considered, subject to separable quadratic-performance indices. The theory and results of several computational examples of a decomposition technique are presented, together with a new computing algorithm suitable for this class of problem.Input-transducer errors in binary crosscorrelation experiments
http://dl-live.theiet.org/content/journals/10.1049/piee.1965.0096
The theory of determining the impulse response of a system by crosscorrelating the output of the system with the input is outlined. It is pointed out that pseudorandom binary input codes prove particularly convenient in facilitating interpretation of the results, and some useful properties of a class of these codes known as <i xmlns="http://pub2web.metastore.ingenta.com/ns/">m</i>sequences are enumerated.A source of error will occur in experiments using these sequences when the modulator is unable to follow the sequence exactly, and instead takes a finite time for the transition between the two binary states. The form of an error signal convenient for describing these transition delays is determined, and the cross-correlation function of the perfect sequence with this error signal is evaluated. Distinction is made between two cases, corresponding to reversible and nonreversible interstate transitions (i.e. whether or not the transition from the +1 state to the −1 state is accomplished in the same manner as that from the −1 state to the +1 state).The effect of the finite bandwidth of a perfect <i xmlns="http://pub2web.metastore.ingenta.com/ns/">m</i>sequence input on the measured response is determined by an analysis in the time domain; the results are extended to include the alteration in bandwidth due to reversible-transition errors. The method does not prove convenient for analysing the effect of bandwidth of nonreversible-transition errors to more than a first approximation of the impulse response. However, this result enables criteria to be developed to ensure that nonreversible transitions will give rise to no more serious effects on the measured response than reversible transitions.Method of statistical linearisation. Its application for evaluation of the response of a hysteretic system to Gaussian inputs
http://dl-live.theiet.org/content/journals/10.1049/piee.1966.0362
The paper describes a method of computing the statistical equivalent gain of a nonlinear element in the minimum r.m.s. error sense. The method is based on the facts that:(<i xmlns="http://pub2web.metastore.ingenta.com/ns/">a</i>) A band-limited Gaussian signal <i xmlns="http://pub2web.metastore.ingenta.com/ns/">x</i>(<i xmlns="http://pub2web.metastore.ingenta.com/ns/">t</i>) can be expressed as<i xmlns="http://pub2web.metastore.ingenta.com/ns/">x</i>(<i xmlns="http://pub2web.metastore.ingenta.com/ns/">t</i>)=<i xmlns="http://pub2web.metastore.ingenta.com/ns/">R</i>(<i xmlns="http://pub2web.metastore.ingenta.com/ns/">t</i>) cos (ω<sub xmlns="http://pub2web.metastore.ingenta.com/ns/"><i>m</i></sub><i xmlns="http://pub2web.metastore.ingenta.com/ns/">t</i>+θ)where ω<sub xmlns="http://pub2web.metastore.ingenta.com/ns/"><i>m</i></sub> is the midband frequency, and <i xmlns="http://pub2web.metastore.ingenta.com/ns/">R</i> and θ are independent random variables.(<i xmlns="http://pub2web.metastore.ingenta.com/ns/">b</i>) Ranges of these two variables are (0, ∞) and (0, 2π), respectively.(<i xmlns="http://pub2web.metastore.ingenta.com/ns/">c</i>) Angle θ has uniform distribution, while <i xmlns="http://pub2web.metastore.ingenta.com/ns/">R</i> obeys a definite distribution law obtainable from the known distribution law for <i xmlns="http://pub2web.metastore.ingenta.com/ns/">x</i>(<i xmlns="http://pub2web.metastore.ingenta.com/ns/">t</i>).The method has the advantage that it is equally well applicable to single- and double-valued nonlinearities. This is illustrated by working out the equivalent gains of a number of well known nonlinear characteristics. Finally, the error-response characteristics of a hysteretic system under excitation from a Gaussian random signal are analysed by using the equivalent gain obtained in the paper. These analytical data are checked against simulated results.Ripple instability in closed-loop control systems with thyristor amplifiers
http://dl-live.theiet.org/content/journals/10.1049/piee.1967.0022
In a simplified analysis of thyristor amplifiers in closed-loop control systems, the amplifier is represented at all frequencies by its d.c. gain, and the effect of the output ripple voltage is neglected. This approach is inadequate for wide-bandwidth systems, and fails to predict the occurrence of ripple instability. A more accurate analysis is made in the paper, taking into account the sampling action of the amplifier in the presence of alternating component voltages in the loop. Ripple instability is characterised by the generation of continuous oscillations at a subharmonic of the balanced-amplifier ripple frequency, and manifests itself in the amplifier as a periodic variation in the firing pattern. The behaviour of an <i xmlns="http://pub2web.metastore.ingenta.com/ns/">m</i>-phase amplifier in an otherwise linear system of any order is investigated, and an exact analysis is given for subharmortic oscillations of order ½. At the ½-order, and other, subharmonic frequencies, describing functions are derived, and the amplifier response is seen to differ markedly from the simple assumption of constant gain at all frequencies. The describing-function technique is extended to evaluate phase margins which must be satisfied at the various subharmonic frequencies to prevent ripple instability, thus allowing a straightforward design technique based on conventional frequency-response methods. The results of the analysis are compared with experimental results from regulated power-supply systems, and good agreement is reached.Note on Whiteley's a.c. compensating network
http://dl-live.theiet.org/content/journals/10.1049/piee.1967.0023
Whiteley's demodulating compensating network for use in a.c. servomechanisms is analysed in a phase-lag mode of operation. The transfer function is derived, and the notch frequency is shown to be related to the reference carrier frequency. The criterion that (<i xmlns="http://pub2web.metastore.ingenta.com/ns/">ω<sub>s</sub>T</i>)<sup xmlns="http://pub2web.metastore.ingenta.com/ns/">2</sup> ≫ 1 must be satisfied is discussed.Design of multivariable feedback systems
http://dl-live.theiet.org/content/journals/10.1049/piee.1967.0086
A canonical form for a multivariable linear control system is described. This canonical form is important because it enables a linear-feedback law to be chosen to produce arbitrary characteristic modes in the closed-loop system. The computational difficulties and methods suggested for minimising then are discussed.Discussion on “Electrical control of dangerous machinery and processes. Part 2: Electrically interlocked guards on machines” before the North-Western Utilization Group, at Manchester, 30th January, 1951
http://dl-live.theiet.org/content/journals/10.1049/pi-2.1953.0069
Chairman's address: Control & Automation Division. Automation and industry: the technical and economic approach to wider utilisation
http://dl-live.theiet.org/content/journals/10.1049/piee.1967.0019
Relay control of the Universities' research reactor
http://dl-live.theiet.org/content/journals/10.1049/piee.1967.0341
The 2-zone core of the Universities' research reactor was simulated on an analogue computer, and the average flux signal from the analogue was used to drive an ‘on-off’ autocontroller with a deadband. On completion of the analogue tests, the controller was installed on the reactor, and the same series of tests repeated. Good agreement was obtained between the analogue and reactor results, and the equipment is now in permanent use on the plant.The author's reply to the discussion on “The dimming of low-pressure discharge lamps”
http://dl-live.theiet.org/content/journals/10.1049/pi-2.1951.0139
The dimming of low-pressure discharge lamps
http://dl-live.theiet.org/content/journals/10.1049/pi-2.1951.0137
The fundamental requirements for the satisfactory dimming of low-pressure hot-cathode and cold-cathode fluorescent lamps are discussed. D.C. operation is first considered, and the treatment is then developed to cover a.c. circuits. Lamp stability is considered from the points of view of smooth control of current, supply-voltage variation, and intermittent extinction. Lamp circuits are generalized in terms of Thévenin's theorem, and the ideas of voltage control and impedance control are introduced. Practical circuits are discussed, the difficulties of multiple-lamp control, and the means for overcoming these, being mentioned. The separate-impedance method of control is described by reference to several circuits, and methods of unifying the control are suggested.A thermostat with control temperature independent of ambient temperature
http://dl-live.theiet.org/content/journals/10.1049/pi-2.1954.0096
A circuit is described for the precise temperature control of the oven of a quartz-crystal frequency standard. A temperature-sensitive bridge of copper and manganin resistors is used as the amplitude control of an oscillator, power from the oscillator heating the oven. It is shown that the temperature of an isothermal oven is maintained independent of ambient temperature and depends only on the gain of the maintaining amplifier. An application of the thermostat is described in which the oven temperature of about 50° C is maintained within ±1 millideg.Discussion on “Electrical control of dangerous machinery and processes. Part 2: Electrically interlocked guards on machines” before the East Midland Centre, at Nottingham, 6th February, 1951
http://dl-live.theiet.org/content/journals/10.1049/pi-2.1953.0070
Discussion on “The dimming of low-pressure discharge lamps” before the North-Eastern Centre, at Newcastle upon Tyne, 26th November, 1951
http://dl-live.theiet.org/content/journals/10.1049/pi-2.1953.0156
Discussion on “The dimming of low-pressure discharge lamps” before the Utilization Section, 15th February, 1951
http://dl-live.theiet.org/content/journals/10.1049/pi-2.1951.0138
Discussion on “Electrical control of dangerous machinery and processes. Part 2: Electrically interlocked guards on machines” before the Mersey and North Wales Centre, at Chester, 22nd October, 1951
http://dl-live.theiet.org/content/journals/10.1049/pi-2.1953.0071
The author's reply to the discussions on “Electrical control of dangerous machinery and processes. Part 2: Electrically interlocked guards on machines”
http://dl-live.theiet.org/content/journals/10.1049/pi-2.1953.0073
Application of crosscorrelating equipment to linear-system identification
http://dl-live.theiet.org/content/journals/10.1049/piee.1965.0394
It is well known that the impulse response of a linear system subject to a ‘white-noise’ input can be obtained by crosscorrelating input and output. In recent years, workers have, however, become aware of the benefits of employing pseudorandom test signals with idealised autocorrelation functions. The paper includes a description of equipment developed to generate and crosscorrelate such signals. Experience is then described and discussed in applying the methods to (<i xmlns="http://pub2web.metastore.ingenta.com/ns/">a</i>) analogue simulated systems, (<i xmlns="http://pub2web.metastore.ingenta.com/ns/">b</i>) a small distillation column used for teaching purposes and (<i xmlns="http://pub2web.metastore.ingenta.com/ns/">c</i>) an industrial evaporator plant.Electrical control of dangerous machinery and processes. Part 2: Electrically interlocked guards on machines
http://dl-live.theiet.org/content/journals/10.1049/pi-2.1951.0102
In 1946, the author read a paper before this Institution on the “Electrical Control of Dangerous Machinery and Processes.” A little earlier, Mr. H.A. Hepburn had read a paper on “The Fencing of Dangerous Parts of Machinery” to The Institution of Mechanical Engineers, and he suggested to the author that the interlocking of such guards and fencing would prove an appropriate subject for a useful continuation of both papers.The present critical study, which is a development of one Section of the 1946 paper, arises from that suggestion; its subject matter is sufficiently indicated by its title, and it is only necessary to add that, in it, an attempt has been made to deduce a number of guiding principles of design and not to describe particular mechanisms in detail.Design of model reference control systems using a Lyapunov synthesis technique
http://dl-live.theiet.org/content/journals/10.1049/piee.1967.0062
The <i xmlns="http://pub2web.metastore.ingenta.com/ns/">V</i>function synthesis technique for designing adaptive loops, so that a model reference system will be controlled in a stable manner, is extended so that each parameter of a high-order linear plant is adjustable. The plant need not be the same order as the model, and both could incorporate zeros as well as poles. Use of either of the two control forms that are derived leads to a stable adaptive step response; a comparison between these loops and others that may alternatively be used is made. The effect of parameter variation due to environmental changes is considered, and simulator studies verify that the behaviour of a simple system may still be adequately regulated.Discussion on “Electrical control of dangerous machinery and processes. Part 2: Electrically interlocked guards on machines” before the Southern Centre, at Southampton, 7th November, 1951
http://dl-live.theiet.org/content/journals/10.1049/pi-2.1953.0072
Discussion on “Electrical control of dangerous machinery and processes. Part 2: Electrically interlocked guards on machines” before the Utilization Section, 18th January, 1950
http://dl-live.theiet.org/content/journals/10.1049/pi-2.1951.0103
Some aspects of transductor design, with particular reference to discharge-lamp circuits
http://dl-live.theiet.org/content/journals/10.1049/pi-2.1953.0185
The design of transductors for dimming discharge lamps gives rise to certain problems whose solution is of interest in a wider field of application. The combination of these non-sinusoidal devices raises in an acute form the question of the general validity of transductor characteristic curves. It is shown that within certain limitations the use of these characteristics is justified. The wide range of load current required in this application is discussed in connection with the magnetic properties of the core material. Some of the effects of varying the curvature of the knee of the saturation curve are mentioned. Discrepancies between theory and practice are discussed with reference to a specific design.Dynamic analysis of nonlinear control systems
http://dl-live.theiet.org/content/journals/10.1049/piee.1967.0021
A method of transient analysis which applies to autonomous and nonautonomous nonlinear systems of any order or configuration is presented. After noting that the transients which occur in such systems may be represented by an exponentially damped sinewave with time-varying damping and frequency, it is shown, by assuming the damping and frequency to be piecewise constant, that the nonlinearity in the system may be adequately characterised by an equivalent gain for transient phenomena. The equation governing the damping and frequency is then derived and shown to be nonlinear. An approximate solution of this equation is obtained, and a method is then developed for calculating the error in the approximation. Application of the theory to a bang-bang position-control system and a position-control system with resetter backlash illustrates the technique for systems with single-valued and phase-shifting nonlinearities. It is shown that the accuracy of prediction is similar to that obtained when the describing-function technique is used to predict limit cycles. Finally the results from this method are compared with those based on the describing-function characterisation, and it is found that appreciable improvement in the accuracy of prediction is obtained.The author's reply to the discussion on “Electrical control of dangerous machinery and processes. Part 2: Electrically interlocked guards on machines”
http://dl-live.theiet.org/content/journals/10.1049/pi-2.1951.0104
On the construction of Macfarlane's generalised functional matrix
http://dl-live.theiet.org/content/journals/10.1049/piee.1967.0026
Present-day control engineering—is it latin to the undergraduate?
http://dl-live.theiet.org/content/journals/10.1049/ree.1975.0115
The development of control engineering design techniques over the last two decades is traced and their application is examined to show how they may be reflected back into the undergraduate teaching of engineers. Control engineering can be regarded as a thought-provoking orderly discipline fulfilling the same functions for science students as does latin to the arts student.Study of the sensitivity of systems to time-delay variations
http://dl-live.theiet.org/content/journals/10.1049/el_19670133
The purpose of this letter is to extend the methods of sensitivity analysis to systems described by differential difference equations, with major emphasis on the determination of the effects of time-delay perturbations. The sensitivity coefficients to the variations of a time delay are defined and a method for the derivation of a sensitivity differential difference equation is presented.Graphical analysis of a continuous-control hill-climbing system fitted for a process with an inertia after the extremal characteristic
http://dl-live.theiet.org/content/journals/10.1049/el_19670007
An extension of the graphical method used for the analysis of a pulse-control hill-climbing system fitted for a process with inertia after the extremal characteristic allows the behaviour of a hill-climbing system to be studied for continuous control.Simulation of dynamic physical systems using a digital process-control computer
http://dl-live.theiet.org/content/journals/10.1049/el_19680291
A digital computer designed for process control can be used for the simulation of complex continuous systems. The program is to contain facilities for the inclusion of items of operating equipment within the simulation study.On the derivative of the trace of a matrix quadratic form
http://dl-live.theiet.org/content/journals/10.1049/el_19670089
In the optimisation of linear systems with a random plant gain, the problem of computing the derivative of the trace of a matrix quadratic form is encountered. A derivation of this derivative is presented.Further comments on the Lyapunov matrix equation
http://dl-live.theiet.org/content/journals/10.1049/el_19670120
A matrix-iterative scheme for solution of the Lyapunov matrix equation <i xmlns="http://pub2web.metastore.ingenta.com/ns/">S</i>′<i xmlns="http://pub2web.metastore.ingenta.com/ns/">D</i>+<i xmlns="http://pub2web.metastore.ingenta.com/ns/">DS</i>=−<i xmlns="http://pub2web.metastore.ingenta.com/ns/">F</i>, with <i xmlns="http://pub2web.metastore.ingenta.com/ns/">F</i> symmetrical and <i xmlns="http://pub2web.metastore.ingenta.com/ns/">S</i> in Schwarz canonical form, is described. The largest matrix to be inverted is a bidiagonal <i xmlns="http://pub2web.metastore.ingenta.com/ns/">n</i>×<i xmlns="http://pub2web.metastore.ingenta.com/ns/">n</i>.Erratum: On the derivative of the trace of a matrix quadratic form
http://dl-live.theiet.org/content/journals/10.1049/el_19670175
Adaptive compensation of mechanical resonances
http://dl-live.theiet.org/content/journals/10.1049/el_19670433
Mechanical oscillations occurring in a servoloop put severe limitations on system performance. In this letter, an adaptive antiresonant compensating system is described which is capable of cancelling the complex mechanical poles.Pull-in range of phase-lock circuits with arbitrary feedback filter
http://dl-live.theiet.org/content/journals/10.1049/ree.1968.0074
Existing methods for the theoretical determination of the capture range of selected automatic phase control loops, using a sin (ϕ-phase comparator, are analysed. From the results of this analysis, an equivalent method to the one used by Goldstein in his treatment of loops with sawtooth comparator is developed that can be applied to a wider variety of interesting loop filters. The method is based on the experimentally suggested hypothesis that all non-synchronized states of loops of most interest are asymptotically periodic. A function is derived which allows quick determination of the capture range of most stable a.p.c. loops containing one large integration time-constant in their filter branch. The method is directly applied to a few technically interesting cases, and very good agreement with experimental results is found.Exact evaluation of responses of control systems with time delay
http://dl-live.theiet.org/content/journals/10.1049/el_19660173
A method is formulated for the rapid evaluation of the response of a feedback control system with time delay to a deterministic input. The solution is obtained analytically from that of an auxiliary system without time delay. The method is exact and is generally applicable to all linear systems with constant coefficients.Sensitivity of an eigenvalue of a multivariable control system
http://dl-live.theiet.org/content/journals/10.1049/el_19660374
A formula for the evaluation of the sensitivity function of an eigenvalue is proposed. The result is arrived at by consideration of the problem of sensitivity of an eigenvalue of a linear multivariable time-invariant control system subjected to parameter variations.