Control systems

Abstract

A closed loop control system for controlling a plant comprises a controller includes an input arranged to receive a feedback signal from the plant. The controller is arranged to compare a value of the feedback signal to a set point value x(s) and to produce an error signal ε(s) that is at least partially dependent on a difference between the value of the feedback signal and the set point value. The controller also includes an output arranged to supply the error signal ε(s) to the plant. A monitor is arranged to compare a value of the error signal ε(s) produced by the controller to a threshold value and to produce a warning signal when the value of the error signal ε(s) exceeds the threshold value for a period of time greater than a predetermined period.

Claims

1. A closed loop control system for controlling a plant, the control system comprising: a controller comprising an input arranged to receive a feedback signal from the plant, said controller being arranged to compare a value of said feedback signal to a set point value and to produce an error signal that is at least partially dependent on a difference between the value of the feedback signal and the set point value, said controller further comprising an output arranged to supply said error signal to the plant; a monitor arranged to compare a value of the error signal produced by the controller to a threshold value; wherein the monitor comprises a timer arranged to measure a period in which the value of the error signal exceeds the threshold value and to compare the measured period to a predetermined period; wherein the monitor is further arranged to produce a warning signal when said value of the error signal exceeds the threshold value for a period of time greater than the predetermined period.

2. The closed loop control system as claimed in claim 1, comprising a comparator arranged to perform the comparison of the value of the feedback signal to the set point value.

3. The closed loop control system as claimed in claim 2, wherein the comparator comprises a sum block arranged to determine the difference between the value of the feedback signal and the set point value.

4. The closed loop control system as claimed in claim 3, comprising an inverter arranged to invert the feedback signal and a sum block arranged to add the value of the inverted feedback signal to the set point value to determine the difference between the value of the feedback signal and the set point value.

5. A closed loop controlled plant system comprising: a plant arranged to receive an input signal and to provide an output signal at least partially dependent on the input signal; a controller comprising an input arranged to receive a feedback signal derived from the output signal produced by the plant, said controller being arranged to compare a value of said feedback signal to a set point value and to produce an error signal that is at least partially dependent on a difference between the value of the feedback signal and the set point value, said controller further comprising an output arranged to supply said error signal to the plant; a monitor arranged to compare a value of the error signal produced by the controller to a threshold value; wherein the monitor comprises a timer arranged to measure a period in which the value of the error signal exceeds the threshold value and to compare the measured period to a predetermined period; wherein the monitor is further arranged to produce a warning signal when said value of the error signal exceeds the threshold value for a period of time greater than a predetermined period.

6. The closed loop controlled plant system as claimed in claim 5, further comprising: one or more further plants each arranged to receive a respective input signal and to provide a respective output signal at least partially dependent on the respective input signal; one or more further controllers each comprising a respective input arranged to receive a respective feedback signal derived from the output signal produced by the respective plant connected to said controller, each controller being arranged to compare a value of the respective feedback signal to a respective set point value and to produce a respective error signal that is at least partially dependent on a difference between the value of the respective feedback signal and the set point value of that controller, each controller further comprising an output arranged to supply said error signal to the respective plant; one or more further monitors each arranged to compare a value of the error signal produced by a respective controller to a respective threshold value; wherein each monitor is further arranged produce a respective warning signal when said value of the respective error signal exceeds the corresponding threshold value for a period of time greater than a respective predetermined period.

7. The closed loop controlled plant system as claimed in claim 5, comprising a comparator arranged to perform the comparison of the value of the feedback signal to the set point value.

8. The closed loop controlled plant system as claimed in claim 7, wherein the comparator comprises a sum block arranged to determine the difference between the value of the feedback signal and the set point value.

9. The closed loop controlled plant system as claimed in claim 8, comprising an inverter arranged to invert the feedback signal and a sum block arranged to add the value of the inverted feedback signal to the set point value to determine the difference between the value of the feedback signal and the set point value.

10. A method of controlling a plant in closed loop, the method comprising: receiving a feedback signal from the plant; comparing a value of said feedback signal to a set point value; producing an error signal that is at least partially dependent on a difference between the value of the feedback signal and the set point value; supplying said error signal to the plant; comparing a value of the error signal to a threshold value; measuring a period in which the value of the error signal exceeds the threshold value; comparing the measured period to a predetermined period; and producing a warning signal when the value of the error signal exceeds the threshold value for a period of time greater than a predetermined period.

11. A non-transitory computer-readable medium comprising instructions that, when executed by a suitable processor, carry out a method of controlling a plant in closed loop, the method comprising: receiving a feedback signal from the plant; comparing a value of said feedback signal to a set point value; producing an error signal that is at least partially dependent on a difference between the value of the feedback signal and the set point value; supplying said error signal to the plant; comparing a value of the error signal to a threshold value; measuring a period in which the value of the error signal exceeds the threshold value; comparing the measured period to a predetermined period; and producing a warning signal when the value of the error signal exceeds the threshold value for a period of time greater than a predetermined period.

Description

BRIEF DESCRIPTION OF THE DRAWINGS

(1) Certain examples of the present disclosure will now be described with reference to the accompanying drawings, in which:

(2) FIG. 1 is a block diagram of a typical closed loop controlled system;

(3) FIG. 2 is a control model of a typical current loop;

(4) FIG. 3 is a graph showing the response of the current loops of FIG. 2;

(5) FIG. 4 is a control model of a closed loop control system in accordance with an example of the present disclosure;

(6) FIG. 5 is a control model of a particular closed loop control system in accordance with an example of FIG. 4;

(7) FIG. 6 is a graph showing the monitor output for design-case operation of the system of FIG. 5;

(8) FIG. 7 is a graph showing the monitor output of operation of the system of FIG. 5 when the plant is compromised but under control;

(9) FIG. 8 is a block diagram of a multi-loop system in accordance with a further example of the present disclosure; and

(10) FIG. 9 is a graph showing the monitor output of operation of a system that uses persistence rather than a filter, in accordance with a further example of the present disclosure.

DETAILED DESCRIPTION

(11) FIG. 1 is a block diagram of a typical closed loop controlled system 2. The system 2 comprises a plant 4, a controller 6 including a sum block 8, and an inverter block 10. In practice, the controller 6 and inverter block 10 will generally all form part of the ‘controller’, however for the ease of description and illustration, these components are treated separately here. It will be appreciated that, as is standard notation in control theory, the functions shown in the accompanying drawings and described herein are in the Laplace domain (or ‘s-domain’), operating on the complex variable ‘s’.

(12) An input—or ‘demand’—x(s) is supplied to the sum block 8, together with the output of the inverter block 10, which is an inverted copy of the output y(s) of the plant 4. The sum block 8 produces an error signal ε(s) which is dependent on the difference between the input signal x(s) and the output signal y(s). The input signal x(s) may be a desired set point to which the system 2 is aiming to set the output y(s) of the plant 4. In this example, the input x(s), output y(s), and error signal ε(s) are fed into a scope 5, which is used to trace these signals. In practice, the system may not include such a scope.

(13) The error signal ε(s) is input to the controller 6, which subjects the error signal ε(s) to its transfer function C(s). The output of the controller 6, which is dependent on error signal ε(s), is supplied to the input of the plant 4, which has a transfer function P(s) of its own. The output of the plant y(s) is fed back to the input of the controller 6 via the inverter block 10 and sum block 8 as outlined above.

(14) An exemplary prior art system 2 that uses this scheme is shown in FIG. 2, which is a control model of a typical current loop, using the principles described above with reference to FIG. 1.

(15) In this system, the controller 6 can be seen in more detail, and comprises a proportional component 12, an integral component 14, a constant 16, and a sum block 18. The error signal ε(s), produced by the sum block 8 as outlined previously, is fed to both the proportional and integral components 12, 14 of the controller 6.

(16) By way of example, the system 2 may comprise a simple current loop, i.e. the innermost loop of a motor drive acting on the resistor-inductor (R-L) of the stator. In such an example, the transfer function P(s) of the plant 4 may be as per Equation 1 below:

(17) $P\left(s\right)=\frac{k}{\mathrm{Ls}+R}=\frac{\frac{1}{R}}{\frac{L}{R}.Math.s+1}$

(18) Equation 1: Exemplary Transfer Function P(s) of a Plant

(19) where P(s) is the transfer function of the plant 4, k is the admittance of the motor, L is the inductance of the stator, R is the resistance of the stator, and s is the complex Laplace variable.

(20) In such an example, the transfer function C(s) of the controller 6 may as per Equation 2 below:

(21) $C\left(s\right)=K_p+\frac{K_i}{s}=\frac{K_{p}s+K_i}{s}$

(22) Equation 2: Exemplary Transfer Function C(s) of a Controller

(23) where C(s) is the transfer function of the controller 6, K.sub.p is the coefficient of the proportional component of the controller 6, K.sub.i is the coefficient of the integral component of the controller 6, and s is the complex Laplace variable.

(24) The generalised transfer function TF(s) of a controlled system is the ratio of the output y(s) to the input x(s), which is defined by Equation 3 below:

(25) $\mathrm{TF}\left(s\right)=\frac{y\left(s\right)}{x\left(s\right)}=\frac{P\left(s\right).Math.C\left(s\right)}{P\left(s\right).Math.C\left(s\right)+1}$

(26) Equation 3: Exemplary Generalised Transfer Function TF(s) of a Controlled System

(27) where y(s) is the system output, x(s) is the system input, P(s) is the transfer function of the plant, and C(s) is the transfer function of the controller.

(28) Thus, when Equation 3 is applied to the motor drive system 2 example given above, the transfer function TF(s) of the motor drive system 2 as a whole is derived as per Equation 4 below:

(29) $\mathrm{TF}\left(s\right)=\frac{y\left(s\right)}{x\left(s\right)}=\frac{K_i+K_{p}s}{{\mathrm{LRs}}^2+\left(R^2+K_p\right)s+K_i}$

(30) Equation 4: Exemplary Transfer Function TF(s) of a Controlled Motor Drive System

(31) FIG. 3 is a graph showing the response of the current loops of FIG. 2. Specifically, FIG. 3 shows the output signal y(t), and error signal ε(t) in the time domain, where the error signal ε(t) is equal to the difference between the input demand signal x(t) and the output signal y(t).

(32) In the graph of FIG. 3, the plant has a resistance R of 1Ω and an inductance L of 1 H. The controller 6 is tuned to have a natural frequency of √2 radians per second and has a Butterworth filter characteristic P=1 V/A, and I=2 Vs/A.

(33) Initially, the input signal x(t) undergoes a step change from 0 to 1, while the output signal y(t) remains at 0 because it cannot change instantaneously. This results in the error signal ε(t) also taking the value 1 at the initial time. This error signal ε(t) is applied to the plant 4, which drives an increase in the output y(t) of the plant, thereby reducing the error signal ε(t).

(34) As can be seen from the plot of the output signal y(t), the output of the plant 4 ‘overshoots’ the value set by the demand, i.e. the input signal x(t), and thus the error signal ε(t) undergoes a negative overshoot. Subsequently, the output signal y(t) settles at the value set by the input signal x(t) and the error signal ε(t) tends to zero.

(35) FIG. 4 is a control model of a closed loop control system in accordance with an example of the present disclosure. The system 102 of FIG. 4 comprises a plant 104, a controller 106 which includes a sum block 108, and an inverter block 110. However, in addition, the system 102 of FIG. 4 further comprises a monitor 103, which is described in further detail below.

(36) In this example, the warning signal 120, output signal y(t), error signal ε(t), and filtered error signal f(t) are fed into a scope 105, which is used to trace these signals (as shown in the graphs of FIGS. 6 and 7). In practice, the system 102 may not include such a scope, however this is shown for ease of illustration.

(37) The function of the control system of FIG. 4 may be understood with reference to a worked example, where FIG. 5 is a control model providing a specific implementation of the general model of FIG. 4 described above. As with FIG. 4, the system 102 of FIG. 5 comprises a plant 104, a controller 106, a sum block 108, an inverter block 110, and a monitor 103.

(38) Functionally, the controller 106, sum block 108, and inverter block 110 will generally all form part of the ‘controller’, however one or more of these components could be separate from the other components. Similarly, the monitor 103 may be integral with the control 106 or may be partially or wholly separate.

(39) It will be appreciated that, as is standard notation in control theory, the functions shown in the accompanying drawings and described herein are in the Laplace domain (or ‘s-domain’), operating on the complex variable ‘s’. Where time-domain inputs and outputs are connected to Laplace domain blocks, suitable domain conversions can be carried out as necessary.

(40) The monitor 103 is arranged to compare a value of the error signal ε(t) to a threshold value. The monitor produces a warning signal 120 when said value of the error signal ε(t) exceeds the threshold value for a period of time greater than a predetermined period, as outlined below.

(41) The monitor comprises a filter 122, characterised by a transfer function as per Equation 5:

(42) $F\left(s\right)=\frac{1}{T.Math.s+1}$

(43) Equation 5: Exemplary Transfer Function F(s) of a Filter

(44) where F(s) is the transfer function of the filter 122 and T is a time delay parameter.

(45) The filter 122 acts to ‘smooth’ the error signal ε(t) by averaging it over a time period determined through proper selection of the time delay parameter T. The output f(t) of the filter 122 is compared to a threshold by a comparator 124. If the output f(t) of the filter 122, i.e. the time-averaged value of the error signal ε(t), is greater than a threshold, a warning signal 120 is asserted by the monitor 103. The output f(t) of the filter 122 is supplied as an output of the monitor 103, together with the warning signal 120.

(46) As it takes some time for the average f(t) of the error signal ε(t) to ‘build up’ to the threshold value, this avoids triggering of the warning signal 120 in response to a momentary spike in the error signal ε(t). This effect may be readily understood with reference to FIG. 6.

(47) FIG. 6 is a graph showing the monitor output for design-case operation of the system 102 of FIG. 5. This shows operation of the system 102 when there are no defects in the system 102. Initially, a sudden change in the demand x(t) gives rise to a step change in the error signal ε(t). However, this is smoothed by the filter 122 and so the filtered signal f(s) at the output of the filter 122 does not cross the threshold, which in this instance is set to 0.5. Thus while the error signal ε(t) itself spikes to a value greater than the threshold, the filter provides a built in time delay such that the error signal ε(t) must be greater than the threshold for a sufficient amount of time before the warning signal 120 would be triggered. As such, the warning signal 120 is not triggered and remains at the value ‘0’, where the value ‘0’ indicates that no defects have been detected.

(48) Conversely, FIG. 7 is a graph showing the monitor output of operation of the system 102 of FIG. 5 when the plant 104 is compromised but under control.

(49) In this example, a change in the demand x(t) gives rise to a step change in the error signal ε(t). However, unlike in the case described above with reference to FIG. 6, in this example the error signal ε(t) sustains a high value for an abnormally long period of time. Thus, when the error signal ε(t) is smoothed by the filter 122, the filtered signal f(s) at the output of the filter 122 crosses the threshold for a period of time t.sub.warning, where the threshold is set to 0.5. During this time t.sub.warning, the warning signal 120 is triggered and changes from ‘0’ to ‘1’, where the value ‘1’ indicates a warning of a defect. It will of course be appreciated that the roles of ‘0’ and ‘1’ could be reversed, depending on the implementation.

(50) Thus in this instance, the error signal ε(t) itself has a value greater than the threshold for sufficient time that the such that the warning signal 120 is triggered. Once the warning signal 120 is triggered, an event may be logged, e.g. in the controller 106 or in an external device or system (such as a remote diagnostics system) that details the error. The logged event may include, for example, the date and time the error occurred, the location or devices associated with the error, the magnitude of the error, the sign of the error, the duration the error lasted, etc. Thus even if the plant 104 returns to normal, the occurrence of the error may be logged for later inspection.

(51) Hysteresis may be used such that the error signal ε(t) must fall below a different value lower than the initial threshold for asserting the warning signal 120 before the warning signal 120 is de-asserted. Of course, with catastrophic damage to the plant 104 (e.g. due to a locked rotor in a motor system), the warning signal 120 may become permanently asserted until the cause of the issue is resolved.

(52) FIG. 8 is a block diagram of a multi-loop system 202 in accordance with a further example of the present disclosure. In this example, the multi-loop system 202 is a controlled motor drive system, which has a ‘velocity loop’ 201a and a ‘current loop’ 201b. The components in each loop 201a, 201b are similar to those described in reference to FIG. 5 above, where elements having reference numerals starting with a ‘2’ in FIG. 8 correspond to like elements having reference numerals starting with a ‘1’ in FIG. 5. For the ease of reference and illustration, the sum blocks 208a, 208b are shown separately to the rest of the respective controllers 206a, 206b, however it will be appreciated that in practice the sum blocks may be combined within their respective controllers.

(53) The multi-loop system 202 controls the operations of two plants 204a, 204b. The first plant 204a, controlled by the velocity loop 201a, is a motor. The second plant 204b, controlled by the current loop 201b, is a power FET. The output signal y1(t) of the motor is fed back to the sum block 208a of the velocity loop 201a via a first inverter 210a, while the output signal y2(t) of the power FET is fed back to the sum block 208b of the current loop 201b via a second inverter 210b.

(54) The controller 206a of the velocity loop 201a acts to steer the output of the motor (i.e. the first plant 204a). However, embedded within its path is the current loop, the controller 206b of which acts to steer the output of the power FET (i.e. the second plant 204b), the output of which is used to drive the motor. Thus the output of the controller 206a within the velocity loop 201a is provided as the input signal x2(t) to the current loop 201b.

(55) As outlined above, the output signal y1(t) from the motor (i.e. the first plant 204a) is fed back to the sum block 208a, which determines the difference between the output y1(t) from the motor and the demand set by the input signal x1(t) and outputs the difference as a first error signal ε1(t). The first error signal ε1(t) is used by the first controller 206a to set a signal which is supplied to the current loop 201b as its input signal x2(t).

(56) The controller 206b of the current loop 201b compares its input signal x2(t) to the feedback signal derived from the output signal y2(t) of the power FET, i.e. the second plant 204b, using the second sum block 208b. The second sum block 208b outputs a second error signal ε.sub.2(t) which is used by the second controller 206b to set a signal which is supplied to the power FET, i.e. to the second plant 204b. This sets the conductance of the power FET, thus varying the current supplied to the motor. Thus the current loop 201b ‘fine-tunes’ the control signal from the velocity loop 201a to account for operational changes of the switch (i.e. the power FET) itself.

(57) The first error signal ε1(t) produced by the first sum block 208a is input to a first monitor 203a which determines when the first error signal ε1(t) has been greater than a particular threshold for longer than a certain amount of time, in the same way described hereinabove in relation to the monitor 103 in the system 102 of FIG. 5. Similarly, the second error signal ε2(t) produced by the second sum block 208b is input to a second monitor 203b which determines when the second error signal ε2(t) has been greater than a particular threshold for longer than a certain amount of time. The thresholds and time periods used by the first and second monitors 203a, 203b may be (and generally are) different to one another.

(58) The monitor 203a of the velocity loop 201a may therefore detect ‘additional load’ faults such as degraded bearings, a degraded drive shaft, or additional shaft load from a higher mechanical drive chain. Meanwhile, the monitor 203b of the current loop 201b may detect ‘additional impedance’ faults such as degradation of an electrical machine stator or a compromised inverter output.

(59) FIG. 9 is a graph showing the monitor output of operation of a system that uses persistence rather than a filter, in accordance with a further example of the present disclosure. Unlike in the filter-based example, in this arrangement the warning signal 120 is asserted (i.e. set to ‘1’) when the error signal ε(t) is greater than a threshold—which, in this example, is set to 0.5—for longer than a predetermined amount of time tper_on.

(60) Initially, a change in the input demand x(t) gives rise to a step change in the error signal ε(t). This sudden rise in the error signal ε(t) does not cause the warning signal 120 to be asserted immediately. Instead, the monitor waits for a duration t.sub.per_on and if the error signal ε(t) still exceeds the threshold at that time, the warning signal 120 is asserted.

(61) After some time, the error signal ε(t) drops back below the threshold. However, this does not immediately cause the warning signal 120 to be de-asserted. Instead, the error signal ε(t) must remain below a further persistence time tper_off before the warning signal 120 is de-asserted.

(62) In either example, whether a filter or persistence is used, the system will generally log that the warning signal 120 has been asserted, e.g. in memory or by transmitting a suitable signal to an external maintenance centre. This log may contain various details relating to, e.g. the magnitude, sign, date, time, and operating conditions at the time of the error for further analysis.

(63) Thus it will be appreciated by those skilled in the art that examples of the present disclosure provide an improved closed loop control system, method, and associated controlled systems, in which the operational health of the controller as well as the controlled system can be monitored for defects proactively. This may advantageously avoid catastrophic failure. Examples of the present disclosure may also provide for more ‘intelligent’ maintenance operations based on actual maintenance requirements rather than relying on a set maintenance schedule that may not reflect actual need.

(64) While specific examples of the disclosure have been described in detail, it will be appreciated by those skilled in the art that the examples described in detail are not limiting on the scope of the disclosure.