Soft computing optimizer of intelligent control system structures

• US 20050119986A1
• Filed: 07/23/2004
• Published: 06/02/2005
• Est. Priority Date: 07/25/2003
• Status: Active Grant

• Current Assignee: Yamaha Hatsudoki Kabushiki Kaisha (Yamaha Corporation)
• Original Assignee: Yamaha Hatsudoki Kabushiki Kaisha (Yamaha Corporation)
• Inventors: Takahashi, Kazuki, Panfilov, Sergey A., Ulyanov, Sergey V., Litvintseva, Ludmila, Ulyanov, Viktor S.
• Granted Patent: US 7,219,087 B2
• Time in Patent Office: Days
• US Class Current: 706/59
• CPC Class Codes: G06N 5/022 Knowledge engineering; Knowledge acquisition
• CPC Class Codes: G06N 5/048 Fuzzy inferencing

Abstract

The present invention involves a Soft Computing (SC) optimizer for designing a Knowledge Base (KB) to be used in a control system for controlling a plant such as, for example, an internal combustion engine or an automobile suspension system. The SC optimizer includes a fuzzy inference engine based on a Fuzzy Neural Network (FNN). The SC Optimizer provides Fuzzy Inference System (FIS) structure selection, FIS structure optimization method selection, and teaching signal selection and generation. The user selects a fuzzy model, including one or more of: the number of input and/or output variables; the type of fuzzy inference model (e.g., Mamdani, Sugeno, Tsukamoto, etc.); and the preliminary type of membership functions. A Genetic Algorithm (GA) is used to optimize linguistic variable parameters and the input-output training patterns. A GA is also used to optimize the rule base, using the fuzzy model, optimal linguistic variable parameters, and a teaching signal. The GA produces a near-optimal FNN. The near-optimal FNN can be improved using classical derivative-based optimization procedures. The FIS structure found by the GA is optimized with a fitness function based on a response of the actual plant model of the controlled plant. The SC optimizer produces a robust KB that is typically smaller that the KB produced by prior art methods.

Claims

1. A soft computing optimizer for designing a knowledge base to be used in a soft computing control, comprising:

a fuzzy inference engine;

a user input module configured to allow a user to select at least one optimization parameter, said optimization parameter comprising at least one of, a number of input variables of said knowledge base, a number of output variables of said knowledge base, a type of fuzzy inference model used by said fuzzy inference engine, and a preliminary type of membership function;

a genetic algorithm configured to optimize said knowledge base using said fuzzy inference engine according to said at least one optimization parameter.

2. The soft computing optimizer of claim 1, wherein said fuzzy inference engine comprises a Fuzzy Neural Network.

3. The soft computing optimizer of claim 1, wherein said fuzzy inference model comprises a Mamdani model.

4. The soft computing optimizer of claim 1, wherein said fuzzy inference model comprises a Sugeno model.

5. The soft computing optimizer of claim 1, wherein said fuzzy inference model comprises a Tsukamoto model.

6. The soft computing optimizer of claim 1, wherein said fuzzy genetic algorithm is configured to optimize said knowledge base according to a teaching signal.

7. The soft computing optimizer of claim 1, further comprising a classical derivative-based optimizer to further optimize an optimized knowledge base produced by said genetic algorithm.

8. The soft computing optimizer of claim 1, where said genetic algorithm uses a fitness function based on a response of a plant model.

9. The soft computing optimizer of claim 1, where said genetic algorithm uses a fitness function based on a response of a physical plant.

10. The soft computing optimizer of claim 1, where said genetic algorithm uses a fitness function based on minimizing entropy production.

11. A method for optimizing a knowledge base in a soft computing controller, comprising:

selecting a fuzzy model by selecting one or more parameters, said one or more parameters comprising at least one of a number of input variables, a number of output variables, a type of fuzzy inference model, and a teaching signal;

optimizing linguistic variable parameters of a knowledge base according to said one or more parameters to produce optimized linguistic variables;

ranking rules in said rule base according to firing strength;

eliminating rules with relatively weak firing strength leaving selected rules from said rules in said rule base;

optimizing said selected rules, using said fuzzy model, said linguistic variable parameters and said optimized linguistic variables, to produce optimized selected rules.

12. The method of claim 11, further comprising optimizing said selected rules using a derivative-based optimization procedure.

13. The method of claim 11, further comprising optimizing parameters of membership functions of said optimized selected rules to reduce approximation errors.

14. A soft computing optimizer, comprising:

a first genetic optimizer configured to optimize linguistic variable parameters for a fuzzy model in a fuzzy inference system;

a first knowledge base trained by a use of a training signal;

a rule evaluator configured to rank rules in said first knowledge base according to firing strength and eliminating rules with a relatively low firing strength to create a second knowledge base; and

a second genetic analyzer configured to optimize said second knowledge base using said fuzzy model.

15. The soft computing optimizer of claim 14, further comprising an optimizer configured to optimize said fuzzy inference model using classical derivative-based optimization.

16. The soft computing optimizer of claim 14, further comprising a third genetic optimizer configured to optimize a structure of said linguistic variables using said second knowledge base.

17. The soft computing optimizer of claim 14, further comprising a third genetic optimizer configured to optimize a structure of membership functions in said fuzzy inference system.

18. The soft computing optimizer of claim 14, wherein said second genetic analyzer uses a fitness function based on measured plant responses.

19. The soft computing optimizer of claim 14, wherein said second genetic analyzer uses a fitness function based on modeled plant responses.

20. The soft computing optimizer of claim 14, wherein said second genetic analyzer uses a fitness function configured to reduce entropy production of a controlled plant.

21. The soft computing optimizer of claim 14, wherein said first genetic algorithm is configured to choose a number of membership functions for said first knowledge base.

22. The soft computing optimizer of claim 14, wherein said first genetic algorithm is configured to choose a type of membership functions for said first knowledge base.

23. The soft computing optimizer of claim 14, wherein said first genetic algorithm is configured to choose parameters of membership functions for said first knowledge base.

24. The soft computing optimizer of claim 14, wherein a fitness function used in said second genetic algorithm depends, at least in part, on a type of membership functions in said fuzzy inference system.

25. The soft computing optimizer of claim 14, further comprising a third genetic analyzer configured to optimize said second knowledge base according to a search space from the parameters of said linguistic variables.

26. The soft computing optimizer of claim 14, further comprising a third genetic analyzer configured to optimize said second knowledge base by minimizing a fuzzy inference error.

27. The soft computing optimizer of claim 14, wherein said second genetic optimizer uses an information-based fitness function.

28. The soft computing optimizer of claim 14, wherein said first genetic optimizer uses a first fitness function and said second genetic optimizer uses said first fitness function.

29. The soft computing optimizer of claim 14, wherein said second genetic optimizer uses a fitness function configured to optimize mechanical characteristics of a controlled plant.

30. The soft computing optimizer of claim 14, wherein said second genetic optimizer uses a fitness function configured to optimize entropy properties of a controlled plant.

31. The soft computing optimizer of claim 14, wherein said second genetic optimizer uses a fitness function configured to optimize based on user preferences.

32. The soft computing optimizer of claim 14, wherein said second genetic optimizer uses a nonlinear model of a controlled plant.

33. The soft computing optimizer of claim 14, wherein said second genetic optimizer uses a nonlinear model of an unstable plant.

34. The soft computing optimizer of claim 14, wherein said teaching signal is obtained from an optimal control signal.

35. The soft computing optimizer of claim 14, wherein said optimal control signal comprises a filtered measured control signal.

36. The soft computing optimizer of claim 14, wherein said optimal control signal comprises a lowpass filtered measured control signal.

37. The soft computing optimizer of claim 14, wherein said optimal control signal comprises a bandpass filtered measured control signal.

38. The soft computing optimizer of claim 14, wherein said optimal control signal comprises a highpass filtered measured control signal.

39. A method for creating a knowledge base for a fuzzy inference system, comprising:

selecting first linguistic variable parameters that describe membership functions in said fuzzy inference system;

varying said first linguistic variable parameters in a first genetic algorithm to create optimized linguistic variables describing optimized membership functions for said fuzzy inference system;

creating a teaching signal that describes a controlled plant;

creating a first knowledge base using at least said teaching signal and said optimized membership functions;

ranking rules in said first knowledge base according to a firing strength of each rule;

creating a second knowledge base from rules in said first knowledge base that have a relatively strong firing strength; and

using a second genetic algorithm to optimize said second knowledge base according to said optimized linguistic variables, said optimized membership functions, and said teaching signal to produce an optimized knowledge base.

40. The method of claim 39, further comprising using a derivative-based optimizer to further optimize said optimized knowledge base.

41. The method of claim 39, further comprising using a back-propagation algorithm to further optimize said optimized knowledge base.

42. The method of claim 39, further comprising using a derivative-based optimizer to further optimize said linguistic variables.

43. The method of claim 39, wherein said second genetic algorithm uses a fitness function based on measured plant response.

44. The method of claim 39, wherein said second genetic algorithm uses a fitness function based on predicted plant response.

45. The method of claim 39, wherein said first genetic algorithm optimizes the number of membership functions in said fuzzy inference system.

46. The method of claim 39, wherein said first genetic algorithm optimizes parameters of membership functions in said fuzzy inference system.

47. The method of claim 39, wherein said teaching signal comprises one or more input signals and one or more output signals.

48. The method of claim 39, wherein a history of activation of rules is associated with a history of activations of membership functions of output variables.

49. The method of claim 39, wherein said fuzzy inference system comprises a Sugeno system and a history of activation of rules is associated with intervals of output signals.

50. The method of claim 39, wherein said fuzzy inference system comprises a Sugeno system of order 0, and wherein intervals of the output signals are taken as a search space.

51. The method of claim 39, wherein said second genetic algorithm depends, at least in part, on a type of said fuzzy inference system.

52. The method of claim 39, wherein said fuzzy inference system comprises a Mamdani model.

53. The method of claim 39, wherein said fuzzy inference system comprises a Sugeno model.

54. The method of claim 39, wherein said fuzzy inference system comprises a Tsukamoto model.

55. The method of claim 39, wherein said second genetic algorithm uses an information-based fitness function.