System and method for materials process control

• US 5,805,452 A
• Filed: 08/01/1996
• Issued: 09/08/1998
• Est. Priority Date: 08/01/1996
• Status: Expired due to Term

• Current Assignee: Agricutlture United States of Americas The As Represented By The Secretary of
• Original Assignee: United States Department Of Agriculture
• Inventors: Byler, Richard K., Anthony, William S.
• Primary Examiner(s): RUGGIERO, JOSEPH F
• Application Number:
  US08/691,069
• Time in Patent Office: 768 Days
• Field of Search: 364/470.01, 364/470.13, 364/470.14, 364/470.15, 364/468.05-468.1, 364/148, 364/156, 364/470.1, 395/207, 395/208, 19/66 R, 19/66 CC, 19/200, 19/300, 57/264, 57/265, 705/7, 705/8
• US Class Current: 700/142
• CPC Class Codes: D01G 31/006 On-line measurement and recording of process and product parameters
• CPC Class Codes: D01H 13/32 Counting, measuring, recording or registering devices
• CPC Class Codes: G05B 13/026 using a predictor

Abstract

System and method for optimizing the processing of materials, particularly the processing of cotton in a cotton gin or a textile mill. Cotton moisture, color, and foreign matter measurements are made with electronic sensors at three stations. For a gin, a dynamic programming model uses input information relative to the moisture and trash content, color, lint turnout, staple length, cotton price structure, and energy costs to select an optimum drying and cleaning sequence for the cotton. The dynamic programming model optimizes cotton producer profits by selecting the amount of gin machinery necessary to achieve the most beneficial market value while minimizing the machinery used. Automated directional valves are used to route the cotton through the selected optimum combination of seed cotton cleaners, multi-path driers, and lint cleaners. For a mill, algorithms are used to select an optimum mill machine sequence that maximizes lint quality, and automated directional values are used to route lint through the selected sequence of cleaners.

Claims

1. A computer program product comprising a computer useable medium having computer program logic recorded thereon for enabling a processor in a computer system to control the processing of cotton through a gin to produce lint, said computer program logic comprising:

measuring means for enabling the processor to measure sensor data that correspond to color, moisture content, and trash content of the cotton as it enters the gin;

gin decision matrix generating means for enabling the processor to generate a gin decision matrix that includes optimum process control decisions that maximize net return from the lint as a function of input state of the cotton entering the gin;

gin process control decision means for enabling the processor to select an optimum process control decision from said gin decision matrix for an input state corresponding to said sensor data, wherein said optimum process control decision corresponds to an optimum gin machine sequence for cotton having said input state corresponding to said sensor data, said optimum gin machine sequence having a first portion and a second portion;

predicting means for enabling the processor to compute and store predicted values for color, moisture content, and trash content for the cotton after processing through said first portion of said optimum gin machine sequence; and

gin process control implementation means for enabling the processor to implement said optimum process control decision by routing cotton through said optimum gin machine sequence.

2. The computer program product of claim 1, further comprising:

second measuring means for enabling the processor to measure second sensor data that correspond to color, moisture content, and trash content of the cotton after it has been processed through said first portion of said optimum gin machine sequence.

3. The computer program product of claim 2, further comprising:

error feedback means for enabling the processor to compare said second sensor data to said predicted values to generate an error; and

error adjusting means for enabling the processor to adjust said optimum gin machine sequence and to adjust said predicted values to compensate for said error.

4. The computer program product of claim 3, further comprising:

third measuring means for enabling the processor to measure lint sensor data that correspond to color, moisture content, and trash content of the cotton after it has been processed through said second portion of said optimum gin machine sequence.

5. The computer program product of claim 4, further comprising:

second predicting means for enabling the processor to compute and store second predicted values for color, moisture content, and trash content for the cotton after processing through said second portion of said optimum gin machine sequence.

6. The computer program product of claim 5, further comprising:

second error feedback means for enabling the processor to compare said lint sensor data to said second predicted values to generate a second error; and

second error adjusting means for enabling the processor to adjust said second portion of said optimum gin machine sequence and to adjust said second predicted values to compensate for said second error.

7. The computer program product of claim 3, wherein said error adjusting means comprises weighting means for enabling the processor to weight said error.

8. The computer program product of claim 6, wherein said second error adjusting means comprises second weighting means for enabling the processor to weight said second error.

9. The computer program product of claim 1, wherein said gin decision matrix generating means comprises a plurality of tabular transition functions.

10. The computer program product of claim 1, wherein said gin decision matrix generating means comprises an algorithm for each gin machine in said optimum gin machine sequence, wherein said algorithm quantifies the effect of that gin machine on leaf trash (LF), moisture content (M), reflectance (RD), turnout (TO), length (LN), and +b (PB).

11. The computer program product of claim 10, wherein said gin decision matrix generating means comprises a cylinder cleaner algorithm as follows:

SLFL=-0.1305-0.0619*SM+1.0599*SLF

SML=SM

SRDL=3.4151-0.0008*SM+0.9606*SRD

STOL=-0.6071-0.0057*SM+1.0127*STO

SLNL=0.2557-0.0012*SM+0.7764*SLN

SPBL=2.8104-0.0137*SM+0.6457*SPB

wherein SLFL, SML, SRDL, STOL, SLNL, SPBL represent the values of leaf trash (LF), moisture content (M), reflectance (RD), turnout (TO), length (LN), and +b (PB), respectively, after processing by a cylinder cleaner, and SLF, SM, SRD, STO, SLN, SPB represent the values of leaf trash (LF), moisture content (M), reflectance (RD), turnout (TO), length (LN), and +b (PB), respectively, input to the cylinder cleaner.

12. The computer program product of claim 10, wherein said gin decision matrix generating means comprises a ginstand algorithm as follows:

SLFL=SLF

SML=SM

SRDL=19.52-0.12*SM+0.75*SRD

STOL=9.22+0.14*SM+0.72*STO

SLNL=0.31+0.002*SM+0.72*SLN

SPBL=1.50+0.02*SM+0.81*SPB

wherein SLFL, SML, SRDL, STOL, SLNL, SPBL represent the values of leaf trash (LF), moisture content (M), reflectance (RD), turnout (TO), length (LN), and +b (PB), respectively, after processing by a ginstand, and SLF, SM, SRD, STO, SLN, SPB represent the values of leaf trash (LF), moisture content (M), reflectance (RD), turnout (TO), length (LN), and +b (PB), respectively, input to the ginstand.

13. The computer program product of claim 10, wherein said gin decision matrix generating means comprises an impact cleaner algorithm as follows:

SLFL=-0.1305-0.0619*SM+1.0599*SLF

SML=SM

SRDL=3.4151-0.0008*SM+0.9606*SRD

STOL=-0.6071-0.0057*SM+1.0127*STO

SLNL=0.2557-0.0012*SM+0.7764*SLN

SPBL=2.8104-0.0137*SM+0.6457*SPB

wherein SLFL, SML, SRDL, STOL, SLNL, SPBL represent the values of leaf trash (LF), moisture content (M), reflectance (RD), turnout (TO), length (LN), and +b (PB), respectively, after processing by an impact cleaner, and SLF, SM, SRD, STO, SLN, SPB represent the values of leaf trash (LF), moisture content (M), reflectance (RD), turnout (TO), length (LN), and +b (PB), respectively, input to the impact cleaner.

14. The computer program product of claim 10, wherein said gin decision matrix generating means comprises a one lint cleaner algorithm as follows:

SLFL=-0.6665+0.0937*SM+0.8296*SLF

SML=SM

SRDL=25.9279-0.0834*SM+0.6757*SRD

STOL=-8.0604-0.1694*SM+1.2043*STO

SLNL=0.1742+0.0003*SM+0.8288*SLN

SPBL=3.3751-0.0222*SM+0.5970*SPB

wherein SLFL, SML, SRDL, STOL, SLNL, SPBL represent the values of leaf trash (LF), moisture content (M), reflectance (RD), turnout (TO), length (LN), and +b (PB), respectively, after processing by one lint cleaner, and SLF, SM, SRD, STO, SLN, SPB represent the values of leaf trash (LF), moisture content (M), reflectance (RD), turnout (TO), length (LN), and +b (PB), respectively, input to the one lint cleaner.

15. The computer program product of claim 10, wherein said gin decision matrix generating means comprises a two lint cleaner algorithm as follows:

SLFL=-1.0150+0.1483*SM+0.7278*SLF

SML=SM

SRDL=38.1983-0.2905*SM+0.5302*SRD

STOL=-12.6405-0.1969*SM+1.3191*STO

SLNL=-0.0520+0.0003*SM+1.0266*SLN

SPBL=2.3649-0.0183*SM+0.7439*SPB

wherein SLFL, SML, SRDL, STOL, SLNL, SPBL represent the values of leaf trash (LF), moisture content (M), reflectance (RD), turnout (TO), length (LN), and +b (PB), respectively, after processing by two lint cleaners, and SLF, SM, SRD, STO, SLN, SPB represent the values of leaf trash (LF), moisture content (M), reflectance (RD), turnout (TO), length (LN), and +b (PB), respectively, input to the two lint cleaners.

16. The computer program product of claim 10, wherein said gin decision matrix generating means comprises a stick machine algorithm as follows:

SLFL=0.1625-0.0710*SM+1.0326*SLF

SML=SM

SRDL=2.3777-0.0299*SM+0.9736*SRD

STOL=1.8168-0.0011*SM+0.9419*STO

SLNL=0.1996-0.0003*SM+0.8226*SLN

SPBL=1.8629-0.0025*SM+0.7587*SPB

wherein SLFL, SML, SRDL, STOL, SLNL, SPBL represent the values of leaf trash (LF), moisture content (M), reflectance (RD), turnout (TO), length (LN), and +b (PB), respectively, after processing by a stick machine, and SLF, SM, SRD, STO, SLN, SPB represent the values of leaf trash (LF), moisture content (M), reflectance (RD), turnout (TO), length (LN), and +b (PB), respectively, input to the stick machine.

17. The computer program product of claim 1, wherein said gin decision matrix generating means comprises a drier algorithm as follows:

SLFL=SLF

SRDL=SRD

STOL=STO*(1.0-(SM-SML)/100.0)

SLNL=SLN

SPBL=SPB

SML=4.2767+0.6397*SM-0.01713*TEMP

wherein SLFL, SRDL, STOL, SLNL, SPBL, SML represent the values of leaf trash (LF), reflectance (RD) , turnout (TO), length (LN), +b (PB), and moisture content (M), respectively, after processing by the drier, and SLF, SRD, STO, SLN, SPB, SM represent the values of leaf trash (LF), reflectance (RD), turnout (TO), length (LN), +b (PB), and moisture content (M), respectively, input to the drier, and TEMP represents the temperature of the drier.

18. The computer program product of claim 1, wherein said gin process control implementation means comprises:

valve control means for enabling the processor to change a position of valves that divert the flow of cotton through said optimum gin machine sequence; and

valve position determining means for enabling the processor to determine the position of the valves.

19. The computer program product of claim 1, further comprising:

display means for enabling the processor to display said sensor data and said optimum gin machine sequence.

26. The computer program product of claim 6, wherein said gin decision matrix generating means comprises an algorithm for each gin machine in said optimum gin machine sequence, wherein said algorithm quantifies the effect of that gin machine on leaf trash (LF), moisture content (M), reflectance (RD), turnout (TO), length (LN), and +b (PB), wherein coefficients for said algorithm are determined as a function of said error and said second error.

20. A control system for controlling the processing of cotton through a gin to produce lint, the gin having a plurality of auxiliary treatment units and a duct through which entrained cotton flows, said control system comprising:

measuring means for measuring sensor data that correspond to color, moisture content, and trash content of the cotton;

processing means coupled to said measuring means, said processing means processing said sensor data and selecting a set of said plurality of auxiliary treatment units through which the cotton is processed to produce lint, said set corresponding to an optimum gin machine sequence that maximizes net return from the lint; and

cotton diversion means coupled to each of said plurality of auxiliary treatment units and coupled to said processing means, wherein said processing means activates said cotton diversion means to divert cotton to flow through each of said plurality of auxiliary treatment units in said set and wherein said processing means deactivates said cotton diversion means to bypass each of said plurality of auxiliary treatment units not in said set.

21. The control system of claim 20, further comprising:

second measuring means coupled to said processing means, said second measuring means measuring lint cleaner sensor data that correspond to color, moisture content, and trash content of the cotton as it enters a lint cleaning stage that contains a plurality of lint cleaners, wherein said processing means processes said lint cleaner sensor data and selects a set of lint cleaners through which the cotton is processed to produce lint, said set of lint cleaners corresponding to an optimum lint cleaning sequence that maximizes revenue from the lint, wherein said processing means activates said cotton diversion means to divert cotton to flow through each of said plurality of lint cleaners in said set and wherein said processing means deactivates said cotton diversion means to bypass each of said plurality of lint cleaners not in said set.

22. The control system of claim 21, wherein said cotton diversion means for each of said plurality of lint cleaners comprises:

a supply valve for controlling flow of cotton through a supply conduit connecting said lint cleaner to the duct at a supply connection location;

a return valve for controlling flow of cotton through a return conduit connecting said lint cleaner to the duct at a return connection location downstream from said supply connection location; and

a duct valve for controlling flow of cotton through the duct, said duct valve disposed between said supply connection location and said return connection location.

23. The control system of claim 22, wherein said processing means activates said cotton diversion means by closing said duct valve while substantially simultaneously opening said supply valve and said return valve.

24. The control system of claim 23, wherein said processing means deactivates said cotton diversion means by opening said duct valve while substantially simultaneously closing said supply valve, and then closing said return valve after a pre-determined time delay.

25. The control system of claim 21, further comprising:

third measuring means coupled to said processing means for measuring color, moisture content, and trash content of the lint after processing by said set of auxiliary treatment units.

27. A control system for controlling the processing of lint through a mill, the mill having a plurality of auxiliary treatment units and a duct through which lint flows, said control system comprising:

measuring means for measuring sensor data that correspond to color, moisture content, and trash content of the lint;

processing means coupled to said measuring means, said processing means processing said sensor data and selecting a set of said plurality of auxiliary treatment units through which the lint is processed to produce lint of a predetermined quality, said set corresponding to an optimum mill machine sequence; and

lint diversion means coupled to each of said plurality of auxiliary treatment units and coupled to said processing means, wherein said processing means activates said lint diversion means to divert lint to flow through each of said plurality of auxiliary treatment units in said set and wherein said processing means deactivates said lint diversion means to bypass each of said plurality of auxiliary treatment units not in said set.

28. The control system of claim 27, further comprising:

second measuring means coupled to said processing means, said second measuring means measuring sensor data that correspond to color, moisture content, and trash content of the lint as it enters a second cleaning stage, wherein said processing means processes said sensor data and selects a set of cleaners in said second cleaning stage through which the lint is processed, said set of cleaners corresponding to an optimum cleaning sequence that maximizes quality of the lint, wherein said processing means activates said lint diversion means to divert lint to flow through each of said set of cleaners and wherein said processing means deactivates said lint diversion means to bypass cleaners not in said set.

29. The control system of claim 27, wherein said lint diversion means for each of said plurality of auxiliary treatment units comprises:

a supply valve for controlling flow of lint through a supply conduit connecting said auxiliary treatment unit to the duct at a supply connection location;

a return valve for controlling flow of lint through a return conduit connecting said auxiliary treatment unit to the duct at a return connection location downstream from said supply connection location; and

a duct valve for controlling flow of lint through the duct, said duct valve disposed between said supply connection location and said return connection location.

30. The control system of claim 29, wherein said processing means activates said lint diversion means by closing said duct valve while substantially simultaneously opening said supply valve and said return valve.

31. The control system of claim 30, wherein said processing means deactivates said lint diversion means by opening said duct valve while substantially simultaneously closing said supply valve, and then closing said return valve after a pre-determined time delay.