Method and apparatus for real-time control

• US 4,975,865 A
• Filed: 05/31/1989
• Issued: 12/04/1990
• Est. Priority Date: 05/31/1989
• Status: Expired due to Fees

• Current Assignee: Mitech Company Limited
• Original Assignee: Mitech Company Limited
• Inventors: Fossheim, Gregory H., Clancy, James E., Carrette, George J.
• Primary Examiner(s): MacDonald, Allen
• Application Number:
  US07/359,871
• Time in Patent Office: 552 Days
• Field of Search: 364/500, 364/513, 364/550, 364/139, 364/496, 364/148, 364/300, 364/131-138
• US Class Current: 700/10
• CPC Class Codes: G05B 13/028 using expert systems only
• CPC Class Codes: Y10S 706/906 Process plant

Abstract

Digital processing methods and apparatus for monitoring, controlling, and simulating industrial processes operate on data signals representative of process parameter values and generate response values representative of desired or requested process parameter values. A knowledge base stores knowledge of the industrial process in the form of rules, and an inference engine applied the rules to calculate the response values. A time-stamp value is assigned to each data signal, representing the time of receipt of each signal. Currency evaluation elements, responsive to user-selected currency range values and the time-stamp signals, assign an expiration-time value to the data signals and disregard data signals having an expiration-time value outside a corresponding user-selected currency range value.

Claims

1. Digital processing apparatus for monitoring and controlling an industrial process, the apparatus comprising

processing means for processing a set of data signals representative of a corresponding set of periodically sensed process parameter values, and for generating in real-time, in response to said data signals, a set of calculated response values representative of requested process parameter values, said processing means including

knowledge base means for providing knowledge relating to the industrial process, to enable said processing means to process said data signals in accordance with said knowledge, said knowledge base means including rule base means for providing a set of rules representative of said knowledge,

inference engine means for calculating intermediate data values in response to said data signals and said rules, and for calculating, in real-time, said response values, in response to any of said data signals, said intermediate data values, and said rules,

said inference engine means including expression evaluation means for evaluating logical expressions representative of any of said data signals, said intermediate data values, and said rules, each said logical expression including a set of variables corresponding to said data signals,

time-stamp means, for determining and assigning a time-stamp value to each said data signal, said time-stamp values being representative of a time at which each said process parameter value is generated, and

currency evaluation means, responsive to said time-stamp signals and to user-selected currency range values, for assigning an expiration-time value for each data signal, intermediate data value, and variable, and for discarding data signals having an expiration-time value outside a corresponding user-selected currency range value.

2. Apparatus according to claim 1, wherein said currency evaluation means includes

means for assigning expiration-time values to said variables in accordance with a selected logical function of any of said variables corresponding to a given logical expression.

3. Apparatus according to claim 2, wherein said means for assigning expiration-time values includes

means for assigning expiration-time values to a given variable in accordance with a lowest expiration-time value corresponding to a given logical expression.

4. Apparatus according to claim 1, further comprising

sensor means, including a plurality of sensor elements, for periodically sensing said process parameter values and providing in response thereto said set of data signals representative of said process parameter values,

feedback control signal generating means, coupled to said processor means, for generating control signals in response to said calculated response values, and

feedback control elements, coupled to said feedback control signal generating means, for controlling the industrial process in response to said asserted control signals, to attain said requested process parameter values.

5. Apparatus according to claim 4, wherein said processing means includes

sensor accessing means, responsive to said currency evaluation means, for requesting data from said sensor elements only when a data signal previously generated in correspondence with a given sensor has associated therewith an expiration-time value outside a corresponding user-selected currency range value.

6. Apparatus according to claim 1, wherein said inference engine means further includes

forward chaining means for executing forward chaining of said rules in accordance with said data signals, and

backward chaining means for executing backward chaining of said rules in accordance with said data signals.

7. Apparatus according to claim 1, wherein said processing means further includes

alarm means, for calculating, in real-time, a set of alarm threshold values corresponding to said process parameter values, and generating an alarm signal when a sensed process parameter value exceeds a corresponding alarm threshold value.

8. Apparatus according to claim 7, further comprising

display means, coupled to said processing means, for displaying any of said data signals, said calculated response values, said requested process parameter values, and said alarm signals, to advise a user in real-time.

9. Apparatus according to claim 1, wherein said processing means further includes

simulation means, for simulating said industrial process in response to user-selected process parameter values and in accordance with said knowledge.

10. Apparatus according to claim 1, further comprising

recovery means for providing recovery of processing of said data signals, without loss of said data signals, in the event of power failure.

11. Apparatus according to claim 1, wherein said inference engine means further includes

primary rule processing means for processing a first set of rules representative of a first set of knowledge relating to said industrial process, and

secondary rule processing means for processing a second set of rules representative of a second set of knowledge relating to said industrial process.

12. Apparatus according to claim 1, wherein said expression evaluation means includes

generic rule processing means for evaluating said logical expressions in response to generic rules, said generic rules being applicable to selected data signals, among a set of data signals, which satisfy a selected constraint,

said generic rule processing means including

pattern matching means for searching said set of data signals to identify said selected data signals which satisfy said selected constraint.

13. Apparatus according to claim 1, wherein said expression evaluation means includes

category processing means for assigning selected category values to said rules, and for controlling processing of said logical expressions in response to said selected category values.

14. A digital processing method for monitoring and controlling an industrial process, the method comprising the steps of

processing a set of data signals representative of a corresponding set of periodically sensed process parameter values, and

generating in real-time, in response to said data signals, a set of calculated response values representative of requested process parameter values, said processing step including

providing knowledge relating to the industrial process, to enable processing of said data signals in accordance with said knowledge,

said providing step including providing a set of rules representative of said knowledge,

calculating intermediate data values in response to said data signals and said rules,

calculating in real-time said response values, in response to any of said data signals, said intermediate data values, and said rules,

said calculating steps including evaluating logical expressions representative of any of said data signals, said intermediate data values, and said rules, each said logical expression including a set of variables corresponding to said data signals,

assigning a time-stamp value to each said data signal, said time-stamp values being representative of a time at which each said process parameter value is generated,

assigning, responsive to said time-stamp signals and to user-selected currency range values, an expiration-time value for each data signal, intermediate data value, and variable, and

discarding data signals having an expiration-time value outside a corresponding user-selected currency range value.

15. A method according to claim 14, further comprising the steps of

configuring a plurality of sensor elements for periodically sensing said process parameter values and providing in response thereto said set of data signals representative of said process parameter values,

generating control signals in response to said calculated response values, and

controlling the industrial process in response to said asserted control signals, to attain said requested process parameter values.

16. A method according to claim 15, wherein said processing step includes the step of

requesting data from said sensor elements only when a data signal previously generated in correspondence with a given sensor has associated therewith an expiration-time value outside a corresponding user-selected currency range value.

17. A method according to claim 14, wherein said calculating step further includes

executing forward chaining of said rules in accordance with said data signals, and

executing backward chaining of said rules in accordance with said data signals.

18. A method according to claim 14, wherein said processing step further includes the steps of

calculating, in real-time, a set of alarm threshold values corresponding to said process parameter values, and

generating an alarm signal when a sensed process parameter value exceeds a corresponding alarm threshold value.

19. A method according to claim 18, further comprising the step of

displaying any of said data signals, said calculated response values, said requested process parameter values, and said alarm signals, to advise a user in real-time.

20. A method according to claim 14, wherein said processing step further includes

simulating said industrial process in response to user-selected process parameter values and in accordance with said knowledge.

21. A method according to claim 14, further comprising the step of

providing recovery of processing of said data signals, without loss of said data signals, in the event of power failure.

22. A method according to claim 14, wherein said currency evaluation step includes

assigning expiration-time values to said variables in accordance with a selected logical function of any of said variables corresponding to a given logical expression.

23. A method according to claim 22, wherein said step of assigning expiration-time values includes

assigning expiration-time values to a given variable in accordance with a lowest expiration-time value corresponding to a given logical expressi,on.

24. A method according to claim 14, wherein said calculating step further includes

processing a first set of rules representative of a first set of knowledge relating to said industrial process, and

processing a second set of rules representative of a second set of knowledge relating to said industrial process.

25. A method according to claim 14, wherein said evaluating step includes the step of

responding to generic rules to evaluate said logical expressions, said generic rules being applicable to selected data signals, among a set of data signals, which satisfy a selected constraint,

said responding step including the step of

searching said set of data signals to identify said selected data signals which satisfy said selected constraint.

26. A method according to claim 14, wherein said evaluating step includes the steps of

assigning selected category values to said rules, and

controlling processing of said logical expressions in response to said selected category values.