System and method for monitoring and control

• US 7,245,101 B2
• Filed: 04/17/2002
• Issued: 07/17/2007
• Est. Priority Date: 04/19/2001
• Status: Expired due to Fees

• Current Assignee: Oxford University Innovation Limited (University of Oxford)
• Original Assignee: Isis Innovation Limited
• Inventors: Dadd, Michael William
• Primary Examiner(s): Vincent; David
• Assistant Examiner(s): Norton; Jennifer L.
• Application Number:
  US10/474,905
• Publication Number:
  US 20040119434A1
• Time in Patent Office: 1,917 Days
• Field of Search: 700 28- 33, 700/35, 700/47, 700/45, 700/52, 700/73, 700/104, 702/104, 318/565, 318/568.24, 318/561
• US Class Current: 318/561
• CPC Class Codes: G05B 17/02 electric
• CPC Class Codes: G05B 23/0254 based on a quantitative model, e.g. mathematical relationships between inputs and outputs; functions: observer, Kalman filter, residual calculation, Neural Networks
• CPC Class Codes: G05B 23/0297 Reconfiguration of monitoring system, e.g. use of virtual sensors; change monitoring method as a response to monitoring results

Abstract

The invention provides for the monitoring and control of a system, such as a linear electro-mechanical transducer, e.g. a linear electric motor, particularly when driven in a reciprocating manner. A dynamic model of the linear motor and its load is used to calculate from the input drive to the motor a predicted response of the motor. The predicted response is compared to a measured response and the model parameters are varied by global optimisation until the two match as closely as desired. The drive may be the voltage input to the electrical motor and the response may be the electrical current in the motor coils. The parameters and variables of the model, after optimisation, can be taken as good estimates of the actual operating parameters and variables of the linear motor. This gives a measurement of such quantities as the stoke and offset of the linear motor. The invention is particularly applicable to the use of linear motors to drive compressors, for instance in Stirling cycle coolers. The system can be applied to balanced pairs of compressors and to compressor systems which use an active balancer to cancel vibration in the system. In this case a balancer model is created and used in an analogous way to the motor model.

Claims

1. A method of monitoring the state of a linearly reciprocating machine comprising an electromechanical transducer in response to an input to the machine by providing an estimate of a value of at least one of a plurality of machine properties including machine variables and machine parameters, the method comprising:

(a) providing a parametrised model of the machine in which respective model parameters and variables correspond to the plurality of machine properties,

(b) providing an input to the model corresponding to a value of an input to the machine,

(c) measuring one of the machine variables to provide a measurement thereof,

(d) comparing the measurement of one of the machine variables with the corresponding model variable and outputting a measure of a difference between them,

(e) optimising the model in an iterative optimisation process by repeatedly varying a plurality of the model parameters corresponding to machine properties while using the same input to the model and comparing the same measurement of one of the machine variables with the corresponding model variable as the plurality of model parameters are varied to reduce the difference,

(f) generating a signal representative of an estimate of one of the machine properties other than the measured one of the machine variables based on the value of the model parameter or variable corresponding thereto, and

(g) periodically repeating steps (b) to (f) thereby to continuously provide as an output a real-time optimised estimate of the one of the machine properties other than the measured one of the machine variables.

2. A method according to claim 1 wherein the optimising of the model is continued iteratively for a preset number of iterations.

3. A method according to claim 1 wherein the optimising of the model is continued until the measured machine variable has a predetermined relationship with the corresponding model variable.

4. A method according to claim 3 wherein the predetermined relationship is that the difference between the measured machine variable and the corresponding model variable is less than a predetermined value.

5. A method according to claim 1 wherein the comparison between the measured machine variable and corresponding model variable is performed at a plurality of sampling points.

6. A method according to claim 5 wherein the comparison between the measured machine variable and corresponding model variable comprises calculating an RMS difference between them.

7. A method according to claim 5 wherein the comparison between the measured machine variable and corresponding model variable comprises calculating their cross-correlation.

8. A method according to claim 1 wherein the model is optimised by a global optimisation algorithm.

9. A method according to claim 1 wherein the electromechanical transducer is a linear motor.

10. A method according to claim 1 wherein the machine further comprises at least one of a load connected to the electromechanical transducer and a load connected to mounting components for the machine.

11. A method according to claim 1 wherein the input is an electrical drive signal in response to which the transducer produces a mechanical drive, and the measured one of the machine variables is an electrical response to the electrical drive signal.

12. A method according to claim 11 wherein the electrical drive signal is an electrical voltage and the electrical response is electrical current in the transducer.

13. A method according to claim 11 wherein the electrical drive signal is an electrical current and the electrical response is electrical voltage across the transducer.

14. A method according to claim 11 wherein the electrical drive signal is substantially sinusoidal.

15. A method according to claim 1 wherein a force-electrical characteristic of the electromechanical transducer is a function of its position.

16. A method according to claim 1 wherein the model parameters comprise parameters describing mechanical behaviour of the transducer and parameters describing electrical behaviour of the transducer.

17. A method according to claim 16 wherein the parameters describing the mechanical behaviour of the transducer comprise at least one of moving mass, a damping coefficient, a spring rate coefficient and the model variables include an offset representing a mean position of the transducer.

18. A method according to claim 16 wherein the parameters describing the electrical behaviour of the transducer comprise at least one of an effective circuit resistance, an effective circuit inductance, an effective circuit capacitance, and the model variables include an offset voltage.

19. A method according to claim 1 wherein the machine property whose value is estimated is an offset of the transducer.

20. A method according to claim 1 wherein the machine property whose value is estimated is a stroke of the transducer.

21. A method according to claim 1 wherein the estimate of the at least one machine property is calculated from the optimised model.

22. Apparatus for monitoring a state of a linearly reciprocating machine comprising electromechanical transducer in response to an in put to the machine by providing an estimate of a value of at least one of a plurality of machine properties including machine variables and machine parameters, comprising:

a parametrised model of the machine in which respective model parameters and variables correspond to the plurality of machine properties,

an input for providing to the model a value corresponding to a value of an input to the machine,

a sensor for measuring one of the machine variables to provide a measurement of the one of the machine variables,

a comparator for comparing the measurement of one of the machine variables with the model variable corresponding to that machine variable to output a measure of a difference between them,

an optimiser for optimising the model in an iterative optimisation process by repeatedly varying a plurality of the model parameters corresponding to machine properties while using the same input to the model and comparing the same measurement of one of the machine variables with the corresponding model variable as the plurality of model parameters are varied to reduce the difference, and

an output for providing a signal representative of an estimate of one of the machine properties other than the measured one of the machine variables based on the value of the model parameter or variable corresponding thereto,

wherein the apparatus is adapted to periodically repeat operation of the sensor, the comparator, the optimiser and the output, thereby to continuously provide as an output a real-time optimised estimate of the one of the machine properties other than the measured one of the machine variables.

23. Apparatus according to claim 22 wherein the optimiser is adapted to optimise the model iteratively for a preset number of iterations.

24. Apparatus according to claim 22 wherein the optimiser is adapted to optimise the model until the measured machine variable has a predetermined relationship with the corresponding model variable.

25. Apparatus according to claim 24 wherein the predetermined relationship is that the difference between the measured machine variable and the corresponding model variable is less than a predetermined value.

26. Apparatus according to claim 22 wherein the comparison between the measured machine variable and corresponding model variable is performed at a plurality of sampling points.

27. Apparatus according to claim 26 wherein the comparison between the measured machine variable and corresponding model variable comprises calculating an RMS difference between them.

28. Apparatus according to claim 26 wherein the comparison between the measured machine variable and corresponding model variable comprises calculating their cross-correlation.

29. Apparatus according to claim 22, wherein the optimiser is a global optimiser.

30. Apparatus according to claim 22 wherein the electromechanical transducer is a linear motor.

31. Apparatus according to claim 22 wherein the machine further comprises at least one of a load connected to the electromechanical transducer and a load connected to mounting components for the machine.

32. Apparatus according to claim 22 wherein the input is an electrical drive signal in response to which the transducer produces a mechanical drive, and the measured one of the machine variables is an electrical response to the electrical drive signal.

33. Apparatus according to claim 32 wherein the electrical drive signal is an electrical voltage and the electrical response is electrical current in the transducer.

34. Apparatus according to claim 32 wherein the electrical drive signal is an electrical current and the electrical response is electrical voltage across the transducer.

35. Apparatus according to claim 32 wherein the electrical drive signal is substantially sinusoidal.

36. Apparatus according to claim 22 wherein a force-electrical characteristic of the electromechanical transducer is a function of its position.

37. Apparatus according to claim 22 wherein the model parameters comprise parameters describing mechanical behaviour of the transducer and parameters describing electrical behaviour of the transducer.

38. Apparatus according to claim 37 wherein the parameters describing the mechanical behaviour of the transducer comprise at least one of moving mass, a damping coefficient, a spring rate coefficient and the model variables include an offset representing a mean position of the transducer.

39. Apparatus according to claim 37 wherein the parameters describing the electrical behaviour of the transducer comprise at least one of an effective circuit resistance, an effective circuit inductance, an effective circuit capacitance, and the model variables include an offset voltage.

40. Apparatus according to claim 22 wherein said machine property whose value is to be estimated is an offset of the transducer.

41. Apparatus according to claim 22 wherein the machine property whose value is to be estimated is a stroke of the transducer.

42. Apparatus according to claim 22 wherein the estimate of the at least one machine property is calculated from the optimised model.

43. A program embodied on a computer-readable medium, the program, when executed, providing a method of monitoring the state of a linearly reciprocating machine comprising an electromechanical transducer in response to an input to the machine by providing an estimate of a value of at least one of a plurality of machine properties including machine variables and machine parameters, the method using a parametrised model of the machine in which respective model parameters and variables correspond to the plurality of machine properties and the method comprising:

(a) receiving an input to the model corresponding to a value of an input to the machine,

(b) receiving a measurement of one of the machine variables,

(c) comparing the measurement of one of the machine variables with the corresponding model variable and outputting a measure of a difference between them,

(d) optimising the model in an iterative optimisation process by repeatedly varying a plurality of the model parameters corresponding to machine properties while using the same input to the model and comparing the same measurement of one of the machine variables with the corresponding model variable as the plurality of model parameters are varied to reduce the difference,

(e) generating a signal representative of an estimate of one of the machine properties other than the measured one of the machine variables based on the value of the model parameter or variable corresponding thereto, and

(f) periodically repeating steps (a) to (e) thereby to continuously provide as an output a real-time optimised estimate of the one of the machine properties other than the measured one of the machine variables.

44. A method of monitoring the state of a linearly reciprocating machine comprising an electromechanical transducer, in response to an input to the machine, by providing an estimate of a value of at least one of a plurality of machine properties including machine variables and machine parameters, the method using a parametrised model of the machine in which respective model parameters and variables correspond to the plurality of machine properties and the method comprising:

(a) receiving an input to the model corresponding to a value of an input to the machine,

(b) receiving a measurement of one of the machine variables,

(c) comparing the measurement of one of the machine variables with the corresponding model variable and outputting a measure of a difference between them,

(d) optimising the model in an iterative optimization process by repeatedly varying one or more model parameters corresponding to machine properties while using the same input to the model and comparing the same measurement of one of the machine variables with the corresponding model variable as the plurality of model parameters are varied to reduce the difference,

(e) generating a signal representative of an estimate of one of the machine properties other than the measured one of the machine variables based on the value of the model parameter or variable corresponding thereto, and

(f) periodically repeating steps (a) to (e) thereby to continuously provide as an output a real-time optimised estimate of the one of the machine properties other than the measured one of the machine variables.