Adaptive control for a cooling system

• US 20030094004A1
• Filed: 11/27/2002
• Published: 05/22/2003
• Est. Priority Date: 06/07/1995
• Status: Active Grant

• Current Assignee: Emerson Climate Technologies Incorporated (Emerson Electric Company)
• Original Assignee: Abtar Singh, Hung M. Pham, Jean-Luc M. Caillat, Mark Bass
• Inventors: Pham, Hung M., Caillat, Jean-Luc M., Singh, Abtar, Bass, Mark
• Granted Patent: US 6,662,583 B2
• Time in Patent Office: Days
• US Class Current: 62/126
• CPC Class Codes: A47F 3/04 air-conditioned, refrigerated refrigerators F25D; heating arrangements specially adapted for transparent or reflecting areas H05B3/84
• CPC Class Codes: F04C 18/0215 where only one member is moving
• CPC Class Codes: F04C 2270/015 Controlled or regulated
• CPC Class Codes: F04C 2270/86 Detection
• CPC Class Codes: F04C 23/008 Hermetic pumps
• CPC Class Codes: F04C 27/005 Axial sealings for working fluid
• CPC Class Codes: F04C 28/00 Control of, monitoring of, or safety arrangements for, pumps or pumping installations specially adapted for elastic fluids
• CPC Class Codes: F04C 28/02 specially adapted for several pumps connected in series or in parallel
• CPC Class Codes: F04C 28/06 specially adapted for stopping, starting, idling or no-load operation
• CPC Class Codes: F04C 28/08 characterised by varying the rotational speed
• CPC Class Codes: F04C 28/22 by changing the eccentricity between cooperating members
• CPC Class Codes: F04C 28/265 being obtained by displacing a lateral sealing face
• CPC Class Codes: F04C 28/28 Safety arrangements; Monitoring
• CPC Class Codes: F25B 1/04 with compressor of rotary type F25B1/005, F25B1/10 take precedence
• CPC Class Codes: F25B 2400/22 Refrigeration systems for supermarkets
• CPC Class Codes: F25B 2600/0261 external to the compressor
• CPC Class Codes: F25B 2700/193 of the compressor
• CPC Class Codes: F25B 2700/1933 Suction pressures
• CPC Class Codes: F25B 2700/2106 of fresh outdoor air
• CPC Class Codes: F25B 2700/2117 of an evaporator
• CPC Class Codes: F25B 2700/21174 : of the refrigerant at the inlet of the evaporator
• CPC Class Codes: F25B 2700/21175 : of the refrigerant at the outlet of the evaporator
• CPC Class Codes: F25B 41/22 : between evaporator and compressor
• CPC Class Codes: F25B 41/35 : by rotary motors, e.g. by stepping motors
• CPC Class Codes: F25B 49/005 : of safety devices F25B49/02 and F25B49/04 take precedence
• CPC Class Codes: F25B 49/022 : Compressor control arrangements
• CPC Class Codes: F25B 5/02 : arranged in parallel
• CPC Class Codes: G05D 23/1909 : whose output amplitude can only take two discrete values
• CPC Class Codes: Y02B 30/70 : Efficient control or regulation technologies, e.g. for control of refrigerant flow, motor or heating

Abstract

A method and controller for a refrigerant pressure effecting component of a cooling system includes providing closed loop control over a pressure effecting component based on an error signal, performing a control algorithm that includes at least one programmably adjustable gain factor, and adaptably changing the programmably adjustable gain factor by periodically generating a new gain factor. More specifically, the system and method includes monitoring fluctuation in the error signal and generating a numeric value indicative of the percent fluctuation in the error signal over a predetermined period of time, then fuzzifying a numeric value to generate a set of fuzzy input values. A predetermined set of rules are applied to the fuzzy input values to generate a set of fuzzy output values, which are then defuzzified by a combining operation to yield the new gain factor.

Claims

1. A controller for a refrigerant pressure effecting component of a cooling system, comprising:

a control processor coupled to said pressure effecting component for providing closed loop control over said component based on an error signal;

said control processor including memory for storing first processing instructions for performing a control algorithm that includes at least one programmably adjustable gain factor;

said control processor including memory for storing second processing instructions for adaptively changing said programmably adjustable gain factor by periodically generating a new gain factor;

said second processing instructions causing said control processor to;

(a) monitor fluctuation in said error signal and generate a numeric value indicative of the percent fluctuation in said error signal over a predetermined time;

(b) fuzzify said numeric value by applying a membership function to generate a set of fuzzy input values indicative of degree of membership;

(c) apply a predetermined set of rules upon said fuzzy input values to generate a set of fuzzy output values, the predetermined set of rules reflecting the degree to which the gain factor associated with a given member of said membership function is to be changed; and

(d) defuzzifying said fuzzy output values by a combining operation yielding said new gain factor.

2. The controller of claim 1 wherein said first processing instructions cause said control processor to effect proportional closed loop control and wherein said gain factor is a proportioning constant.

3. The controller of claim 2 wherein said first processing instructions further cause said control processor to effect integral closed loop control.

4. The controller of claim 1 wherein said first processing instructions cause said control processor to effect integral closed loop control and wherein said gain factor is an integrating constant.

5. The controller of claim 1 wherein said membership function defines a plurality of linguistic variables representing different ranges of error values.

6. The controller of claim 5 wherein said linguistic variables include at least three linguistic variables representing predefined, LARGE, MEDIUM and SMALL error value ranges.

7. The controller of claim 1 wherein said membership function defines at least three overlapping ranges of membership function values such that at least some of said numeric values indicative of percent fluctuation map onto more than one membership function value.

8. The controller of claim 1 wherein said second processing instructions further cause said control processor to monitor direction of fluctuation in said error signal and wherein said predetermined set of rules includes a first subset of rules for positive error signal fluctuations and a second subset of rules for negative error signal fluctuations.

9. The controller of claim 1 wherein said second processing instructions cause said controller to defuzzify said fuzzy output values by calculating the centroid of said fuzzy output values to derive a multiplier that is applied to said programmably adjustable gain factor to yield said new gain factor.

10. The controller of claim 1 wherein said pressure effecting component is a compressor.

11. The controller of claim 1 wherein said pressure effecting component is a valve.

12. The controller of claim 1 wherein said pressure effecting component is a stepper regulator valve.

13. A method for controlling a refrigerant pressure effecting component of a cooling system, said steps comprising:

providing closed loop control over a pressure effecting component based on an error signal;

storing first processing instructions for performing a control algorithm that includes at least one programmably adjustable gain factor; and

storing second processing instructions for adaptably changing said programmably adjustable gain factor by periodically generating a new gain factor;

wherein said step of storing said second processing instructions includes monitoring fluctuation in said error signal;

generating a numeric value indicative of the percent fluctuation in said error signal over a predetermined time;

fuzzifying said numeric value by applying a membership function;

generating a set of fuzzy input values indicating degree of membership of said numeric values;

applying a predetermined set of rules upon said fuzzy input values, said predetermined set of rules reflecting the degree to which said gain factor associated with a given member of said membership function is to be changed;

generating a set of fuzzy output values; and

defuzzifying said fuzzy output values by a combining operation to yield said new gain factor.

14. The method of claim 13 wherein said step of storing first processing instructions includes effecting proportional closed loop control when said gain factor is a proportioning constant.

15. The method of claim 14 wherein said step of storing first processing instructions effects integral closed loop control.

16. The method of claim 13 wherein said step of storing first processing instructions includes effecting integral closed loop control when said gain factor is an integrating constant.

17. The method of claim 13 wherein said step of fuzzifying said numeric value by applying a membership function includes defining a plurality of linguistic variables representing different ranges of error values.

18. The method of claim 13 further comprising the step of monitoring a direction of fluctuation in said error signal, wherein said predetermined set of rules includes a first subset of rules for positive error signal fluctuations and a second subset of rules for negative error signal fluctuations.

19. The method of claim 13 further comprising the step of calculating the centroid of said fuzzy output value and deriving a multiplier that is applied to said programmably adjustable gain factor to yield said new gain factor.