GB2162972A - Control of system parameters - Google Patents

Abstract

A feed back control system incorporates a controller which provides a controlling output signal which is derived from a desired value (x) and a measured value (y). The measured value (y) will vary with time. At each stage (3,4,5,6,7,8) in the controller an output signal (w,v,u,r,q,z,) is produced which is a function of time if that output signal lies within predetermined limits for each stage and which is constant if the output signal lies outside those limits. <IMAGE>

Description

SPECIFICATION Control of system parameters This invention relates to control of system parameters.

Control of system or process parameters such as temperature, flow, etc., can- be attempted using a variety of techniques. In many applications, it is extremely important that the parameters are controlled to set values with a minimum of steady-state and/or oscillatory error. Traditionally, control is achieved using a feed-back or "closed loop" method, whereby the parameter value is-automatically compared with its set value, and adjusted accordingly.

The value of a system or process parameter can be particularly difficult to control, especially when the system is slow to react to changes in input conditions. Most control methods are affected to some degree by one or more of-the following factors; (a) oscillation around the set value due to system hysteresis; (b) oscillation around the set value duced through instability in the control action; (c) iong start-up time caused by slow approach to the set value to avoid (b); (d) overshoot on start-up causing undesirably extreme conditions in system; and (e) steady-state error introduced by controller.

According to one aspect of the present invention there is provided a method of controlling a parameter of a system using a controller, the method comprising: inputting a desired value of the parameter and a measured value of the parameter to an amplifier to obtain a first output signal which is proportional to the difference between the desired value and the measured value if said difference lies between a first two predetermined values and which is constant if said difference is above or below respective ones of the predetermined values;; inputting the first output signal of the amplifier both directly' and via a differentiating device to a first summing device, the differentiating device providing a second output signal which is proportional to the change per unit time in the output of the amplifier if said change per unit time lies between a second two predetermined values and which is constant if said change per unit time is above or below respective ones of the second two predetermined values, and the summing device providing a third output which is proportional to the sum of the first and second output signals if said sum lies between a third two predetermined values and which is constant if said sum is above or below respective ones of the third predetermined values; inputting the third output signal to (a) a second summing device which receives as its other input signal a constant value and which provides a fourth otput which is proportional to the sum of the third output signal and the constant value if said sum lies between a fourth two predetermined- values and which is constant if said sum is above or below respective ones of the fourth predetermined values and to (b) an integrating device which provides a fifth output signal which is the sum of a value proportional to the integral between first and second limits of the third output signal with respect to time-and the value of the fifth output signal at the first limit if the last-mentioned sum lies between a fifth two predetermined values and which is constant if said sum is above or below respective ones of the fifth predetermined values; and inputting the fourth signal and the fifth signal to a multiplication means which provides a controller output for controlling the system which is proportional to the product of the fourth and fifth signals if said product lies between a sixth two predetermined values and which is constant if said product is above or below respective ones of the sixth predetermined values.

According to another aspect of the present invention there is- provided an apparatus for controlling a parameter of a system using a controller, the apparatus comprising: an amplifier which receives, as its input signals, a desired value of the parameter and a measured value of the parameter and which provides a first output signal which is proportional to the difference between the desired value and the measured value if said difference lies between a first two predetermined values and which is constant if said difference is above or below respective ones of the predetermined values;; a differentiating device which receives as its input signal the first output signal of the amplifier and which provides a second output signal which is proportional to the change per unit time in the output of the amplifier if said change per unit time lies between a second two predetermined values and which is contant if said change per unit time is above or below respective ones of the second two predetermined values; a first summing device which receives the first and second output signals and provides a third output which is proportional to the sum of the first and second output signals if said sum lies between a third two predetermined values and which is constant if said sum is above or below respective ones-of the third predetermined values; a second summing device which receives the third output signal and a constant value and which provides a fourth output signal which is proportional to the sum of the third output signal and the constant value if said sum lies between a fourth two predetermined values and which is constant if said sum is above or below respective ones of the fourth predetermined values; an integrating device which receives the third output signal and provides a fifth output signal which is the sum of a value proportional to the integral between first and second limits of the third output signal with respect to time and the value of the fifth output signal at the first limit if the last-mentioned sum lies between a fifth two predetermined values and which is constant if said sum is above or below respective ones of the fifth predetermined values; and a multiplication means which receives the fourth signal and the fifth signal and which provides a controller output for controlling the system which is proportional to the product of the fourth and fifth signals if said product lies between a sixth- two predetermined values and which is constant if said product is above or below respective ones of the sixth predetermined values.

Although the present invention is particularly useful in controlling temperature, it may also be used to control other process and system parameters such as flow etc.

For a better understanding of the present invention, and to show how the same may be carried into effect, reference will now be made, by way of example, to the accompanying drawings, in which: Figures la and ib are block diagrams of analogue and sampled-data systems respectively; Figures 2a and 2b show algorithms for use in the system of Figs. 1 a and 1 b respectively; and Figure 3 is a block diagram of the controller in Figs. 1a and 1 b.

Fig. 1 a shows a controller 1 having as its input x(t), y(t) being analogue functions of time as follows: x is an input signal representing a set value for the system 2 to be controlled; and y is a signal representing the fed-back measured value outputted from the system 2.

The controller output z (t), an analogue function of time, is used to modulate a system parameter for example to switch a heater on and off in response to temperature changes of the system 2. For a temperature control system, the value x is a voltage representing the set value temperature and y is a voltage corresponding to the temperature sensed by a temperature sensor in the system 2, and amplified as necessary. The output z is used to adjust heating (or cooling) power (or rate of heat transfer) applied to the system by means of phase control, burst control (zero-voltage switch), or other similar device. The output z (t) is obtained from the inputs x, y using the algorithm shown in Fig. 2a.In Fig. 2a, the capital letters A to H, J to N, P to X represent arbitrary constants which may take any value between - cc and + m. This algorithm is implemented in the controller as shown in Fig.

3.

The inputs x, y are fed to a differential amplifier 3. The output w from the amplifier 3 is a constant value B or C if A (x-y) > B or A (x-y) < C respectively, and is a function of time if A(x-y) lies between B and C.

The signal w is fed (i) directly to an adder 4 and-(ii) via a non-inverting differential amplifier 5 to the adder 4. The output v from the amplifier 5 is constant if the differentiated signal d (w (t)) dt is above or below predetermined values E/D and F/D respectively, and is proportional to the differentiated signal d (w (t)) dt if the differentiated signal lies between the constant values E/D and F/D.

The signal v and w are summed in the adder 4 to produce an output u which is such that it is constant if the sum v + w lies above or below predetermined values H/G, J/G respectively and is proportional to the sum of v + w if that sum lies between these values.

The adder 4 output u is fed to a further adder 6, where it is summed with an offset constant value K to provide an output r which varies in relation to u (t) and K in the same manner as the output u of the adder 4 varies in relation to v (t) and w (t), but with predetermined values M/L, N/L which may be different.

The adder 5 output u is also fed to a noninverting integrator 7 which limits its input to values greater than or equal to zero and then integrates to provide an output q which is constant if the sum of the instantaneous output q (to), and a value proportional to the integral of u (t) with respect to time between t and t, P.

u (t) dt is above or below predetermined constant values Q, R respectively, and which is the function q(t,) + P.

u (t) dt if that function lies between q and r.

Finally, the output z (t) is derived from an analogue multiplier 8 from signals q and r so that it is constant if the product of q and r is above or below predetermined values W/V, X/V respectively and is proportional to that product if that product lies between those predetermined values.

Figs. 1 b and 2b show a system and algorithm respectively for sampled data, with numerals 1' and 2' representing the controller and system respectively. The operation is substantially as described above with reference to Figs. 1 a, 2a and 3 except that x,, y,is the value of parameter x, y at an instant T, and xmi, YT-1 is the value of the parameter x, y at the previously sampled time T-1. The differentiation and integration may be carried out according to known sampled data methods or may be approximated as indicated in Fig. 2b.

In particular the differentiator may be replaced by a sampled data device whose output is proportional to the difference between adjacents amples and whose effective time constant may be varied by varying the sampling frequency.

The algorithms described herein enable the establishment of a degree of control over the system which is limited by the system and not by the controller.

The controller may be optimised by the adjustment of three parameters: i) the gain of the input differential amplifier; ii) the time constant (or sampling frequency) of the differentiator 5; and iii) the time contant of the integrator 7.

The possible signal values at each stage are -limited to values governed by supply rail voltages used for the controller.

As the block diagram is schematic, the stages shown may each be achieved, if necessary, by cascading more than one stage.

Further, it is possible to implement a controller having the same control function if the multiplier is eliminated and the outputs r, q from the second adder and the integrator 7 are used one to modulate a burst controller and the other to phase control the same mains output, thus achieving multiplication.

Claims (6)

1. A method of controlling a parameter of a system using a controller, the method comprising: providing a first signal which is proportional to the difference between a desired value of the parameter and a measured value of the parameter if said difference lies between a first two predetermined values and which is constant if said difference is above or below respective ones of the predetermined values; providing from said first signal a second signal which is proportional to the change per unit tjme in the first signal if said change per unit time lies between a second two predetermined values and which is constant if said change per unit time is above or below respective ones of the second two predetermined values;; providing from said first and second signals a third signal which is proportional to the sum of the first and second signals if said sum lies between a third two predetermined values and which is constant if said sum is above or below respective ones of the third predetermined values; providing from said third signal a fourth signal which is proportional to the sum of (a) the third signal and (b) a constant value if said sum lies between a fourth two predetermined values and which is constant if said sum is above or below respective ones of the fourth predetermined values;; providing from said third signal a fifth signal which is a function of time (t) which is the sum of a value proportional to the integral between a first limit (t,) and time (t) of the third signal with respect to time and a constant value which is the value of the function at the first limit (t,) if the last-mentioned sum lies between a fifth two predetermined values and which is constant if said sum is above or below respective ones of the fifth predetermined values; and providing a controller output for controlling the system which output is proportional to the product of the fourth and fifth signals if said product lies between a sixth two predetermined values and which is constant if said product is above or below respective ones of the sixth predetermined values.

2. A method of controlling a parameter of a system substantially as hereinbefore described with reference to the drawings.

3. An apparatus for controlling a parameter of a system using a controller, the apparatus comprising: first means arranged to receive a desired value of the parameter and a measured value of the parameter and to provide a first output signal which is proportional to the difference between the desired value and the measured value if said difference lies between a first two predetermined values and which is constant if said difference is above or below respective ones of the predetermined values;; differentiating means arranged to receive as its input signal the first output signal and to provide a second output signal which is proportional to the change per unit time in the first output signal if said change per unit time lies between a second two predetermined values and which is constant if said change per unit time is above or below respective ones of the second two predetermined values; a first summing device arranged to receive the first and second output signals and to provide a third output signal which is proportional to the sum of the first and second output signals if said sum lies between a third two predetermined values and which is constant if said sum is above or below respective ones of the third predetermined values;; a second summing device arranged to receive the third output signal and a constant value and to provide a fourth output signal which is proportional to the sum of the third output signal and the constant value if said sum lies between a fourth two predetermined values and which is constant if said sum is above or below respective ones of the fourth predetermined values; an integrating device arranged to receive the third output signal and to provide as a fifth output signal a function of time which is the sum of a value proportional to the integral between a first limit and time (t) of the third output signal with respect to time and a constant value which is the value of the function at the first limit if the last-mentioned sum lies between a fifth two predetermined values and which is constant if said sum is above or below respective ones of the fifth predetermined values; and means for providing a controller output for controlling the system which output is proportional to the product of the fourth and fifth signals if said product lies between a sixth two predetermined values and which is constant if said product is above or below respective ones of the sixth predetermined values.

4. An apparatus as claimed in claim 3, in which the first means comprises a differential amplifier.

5. An apparatus as claimed in claim 3 or 4, in which the controller output providing means comprises a multiplier.

6. An apparatus for controlling a parameter of a system substantially as hereinbefore described with reference to the accompanying drawings.