System and method for nonlinear dynamic control based on soft computing with discrete constraints
원문보기
IPC분류정보
국가/구분
United States(US) Patent
등록
국제특허분류(IPC7판)
G06F-009/00
G05B-013/80
출원번호
US-0209671
(2002-07-30)
발명자
/ 주소
Ulyanov, Sergei V.
Panfilov, Sergei
Takahashi, Kazuki
출원인 / 주소
Yamaha Hatsudoki Kabushiki Kaisha
대리인 / 주소
Knobbe Martens Olson &
인용정보
피인용 횟수 :
24인용 특허 :
17
초록▼
A control system using a genetic analyzer based on discrete constraints is described. In one embodiment, a genetic algorithm with step-coded chromosomes is used to develop a teaching signal that provides good control qualities for a controller with discrete constraints, such as, for example, a step-
A control system using a genetic analyzer based on discrete constraints is described. In one embodiment, a genetic algorithm with step-coded chromosomes is used to develop a teaching signal that provides good control qualities for a controller with discrete constraints, such as, for example, a step-constrained controller. In one embodiment, the control system uses a fitness (performance) function that is based on the physical laws of minimum entropy. In one embodiment, the genetic analyzer is used in an off-line mode to develop a teaching signal for a fuzzy logic classifier system that develops a knowledge base. The teaching signal can be approximated online by a fuzzy controller that operates using knowledge from the knowledge base. The control system can be used to control complex plants described by nonlinear, unstable, dissipative models. In one embodiment, the step-constrained control system is configured to control stepping motors.
대표청구항▼
1. A self-organizing control system for optimization of a knowledge base, comprising:a fuzzy neural network configured to develop a knowledge base for a fuzzy controller; a genetic analyzer configured to develop a teaching signal for said fuzzy-logic classifier, said teaching signal configured to pr
1. A self-organizing control system for optimization of a knowledge base, comprising:a fuzzy neural network configured to develop a knowledge base for a fuzzy controller; a genetic analyzer configured to develop a teaching signal for said fuzzy-logic classifier, said teaching signal configured to provide a desired set of control qualities, said genetic analyzer using chromosomes, a portion of said chromosomes being step coded; and a PID controller with discrete constraints, said PID controller configured to receive a gain schedule from said fuzzy controller. 2. A self-organizing control system of claim 1, further comprising a feedback module for simulation of look-up tables for said fuzzy-logic suspension controller.3. The self-organizing control system of claim 1, wherein said genetic analyzer module uses a fitness function that reduces entropy production in a plant controlled by said PID controller.4. The self-organizing control system of claim 1, where said genetic analyzer module comprises a fitness function that is based on physical laws of minimum entropy.5. The self-organizing control system of claim 1, wherein said genetic analyzer is used in an off-line mode to develop said training signal.6. The self-organizing control system of claim 1, wherein said step-coded chromosomes include an alphabet of step up, step down, and hold.7. The self-organizing control system of claim 1, further comprising an evaluation model to provide inputs to an entropy-based fitness function.8. The self-organizing control system of claim 7, further comprising a fuzzy controller that approximates the teaching signal using knowledge from the knowledge base.9. A control system for a plant comprising:a neural network configured to control a fuzzy controller, said fuzzy controller configured to control a linear controller with discrete constraints; a genetic analyzer configured to train said neural network, said genetic analyzer that uses step-coded chromosomes. 10. The control system of claim 9, wherein said genetic analyzer uses a difference between a time derivative of entropy in a control signal from a learning control unit and a time derivative of an entropy inside the plant as a measure of control performance.11. The control system of claim 10, wherein entropy calculation of an entropy inside said plant is based on a thermodynamic model of an equation of motion for said plant that is treated as an open dynamic system.12. The control system of claim 9, wherein said genetic analyzer generates a teaching signal.13. The control system of claim 9, wherein said linear control system produces a control signal based on data obtained from one or more sensors that measure said plant.14. The control system of claim 13, wherein said plant comprises a suspension system and said cone or more sensors comprise angle and position sensors that measure angle and position of elements of the suspension system.15. The control system of claim 9, wherein fuzzy rules used by said fuzzy controller are evolved using a kinetic model of the plant in an offline learning mode.16. The control system of claim 15, wherein data from said kinetic model are provided to an entropy calculator that calculates input entropy production and output entropy production of the plant.17. The control system of claim 16, wherein said input entropy production and said output entropy production are provided to a fitness function calculator that calculates a fitness function as a difference in entropy production rates constrained by one or more weights.18. The control system of claim 17, wherein said genetic analyzer uses said fitness function to develop a training signal for an off-line control system, the training signal corresponding to an operational environment.19. The control system of claim 18, wherein said step-coded chromosome includes codes for step up, step down, and hold.20. The control system of claim 9, wherein control parameters in the form of a knowledge base from an off-line control system are provided to an online control system that, using information from said knowledge base.21. A method for controlling a nonlinear plant by obtaining an entropy production difference between a time derivative dSu/dt of an entropy of the plant and a time derivative dSc/dt of an entropy provided to the plant from a controller; using a genetic algorithm that uses the entropy production difference as a performance function to evolve a control rule in an off-line controller with discrete constraints; and providing filtered control rules to an online controller with discrete constraints to control the plant.22. The method of claim 21, further comprising using said online controller to control a stepping motor to change a damping factor of one or more shock absorbers in the vehicle suspension system.23. The method of claim 21, further comprising evolving a control rule relative to a variable of the controller by using a genetic algorithm, said genetic algorithm using a fitness function based on said entropy production difference.24. A self-organizing control system, comprising: a simulator configured to use a thermodynamic model of a nonlinear equation of motion for a plant, a fitness function module that calculates a fitness function based on an entropy production difference between a time differentiation of an entropy of said plant dSu/dt and a time differentiation dSc/dt of an entropy provided to the plant by a step-constrained linear controller that controls the plant; a genetic analyzer that uses said fitness function to provide a teaching signal; a fuzzy logic classifier that determines one or more fuzzy rules by using a learning process and said teaching signal; and a fuzzy logic controller that uses said fuzzy rules to set a step control variable of the step-constrained linear controller.25. The self-organizing control system of claim 24, wherein said genetic analyzer uses chromosomes, at least a portion of said chromosomes being step-coded.26. A control system comprising: a step-coded genetic algorithm that computes a teaching signal using a fitness function that provides a measure of control quality based on reducing production entropy; a local entropy feedback loop that provides control by relating stability of a plant to controllability of the plant.27. The control system of claim 26, wherein said step-coded genetic algorithm comprises step-coded chromosomes to limit the search space of the genetic algorithm.28. The control system of claim 27, wherein a fuzzy neural network is used to create a database from said teaching signal.29. The control system of claim 28, wherein said plant is a vehicle suspension system.30. The control system of claim 29, wherein a knowledge base corresponds to stochastic characteristics of a selected stochastic excitation signal used to develop the teaching signal.31. An optimization control method for a shock absorber comprising the steps of:obtaining a difference between a time differential of entropy inside a shock absorber and a time differential of entropy given to said shock absorber from a control unit that controls said shock absorber; and optimizing at least one control parameter of said control unit by using a discrete-constrained genetic algorithm, said discrete-constrained genetic algorithm using said difference as a fitness function, said genetic algorithm constrained by at least one step constraint of a chromosome. 32. The optimization control method of claim 31, wherein said time differential of said step of optimizing reduces an entropy provided to said shock absorber from said control unit.33. The optimization control method of claim 31, wherein said control unit comprises a fuzzy neural network, and wherein a value of a coupling coefficient for a fuzzy rule is optimized by using said genetic algorithm.34. The optimization control method of claim 31, wherein said control unit comprises an offline module and an online control module, said method further including the steps of optimizing a control parameter based on said discrete-constrained genetic algorithm by using said performance function, determining said control parameter of said online control module based on said control parameter and controlling said shock absorber using said online control module.35. The optimization control method of claim 34, wherein said offline module provides optimization using a simulation model, said simulation model based on a kinetic model of a vehicle suspension system.36. The optimization control method of claim 34, wherein said shock absorber is arranged to alter a damping force by altering a cross-sectional area of an oil passage controlled by a stepping motor, and said control unit controls said stepping motor to adjust said cross-sectional area of said oil passage.37. A method for control of a plant comprising the steps of: calculating a first entropy production rate corresponding to an entropy production rate of a control signal provided to a model of said plant; calculating a second entropy production rate corresponding to an entropy production rate of said model of said plant; determining a fitness function for a step-constrained genetic optimizer using said first entropy production rate and said second entropy production rate; providing said fitness function to said genetic optimizer; providing a teaching output from said step-constrained genetic optimizer to a fuzzy neural network configured to produce a knowledge base; providing said knowledge base to a fuzzy controller, said fuzzy controller using an error signal and said knowledge base to produce a coefficient gain schedule; and providing said coefficient gain schedule to a step-constrained linear controller.38. The method of claim 37, wherein said genetic optimizer minimizes entropy production under one or more constraints.39. The method of claim 38, wherein at least one of said constraints is related to a weight based on user-perceived evaluation of control performance.40. The method of claim 37, wherein said model of said plant comprises a model of a suspension system.41. The method of claim 37, wherein said second control system is configured to control a physical plant.42. The method of claim 37, wherein said second control system is configured to control a shock absorber.43. The method of claim 37, wherein said second control system is configured to control a damping rate of a shock absorber.44. The method of claim 37, wherein said linear controller receives sensor input data from one or more sensors that monitor a vehicle suspension system.45. The method of claim 44, wherein at least one of said sensors is a vertical motion sensor that measures a vehicle vertical movement.46. The method of claim 44, wherein at least one of said sensors is an accelerometer that measures a vehicle vertical acceleration.47. The method of claim 44, wherein at least one of said sensors is an accelerometer that measures at least one component of a vehicle acceleration.48. The method of claim 44, wherein at least one of said sensors is a length sensor that measures a change in length of at least a portion of said suspension system.49. The method of claim 44, wherein at least one of said sensors is an angle sensor that measures an angle of at least a portion of said suspension system with respect to said vehicle.50. The method of claim 44, wherein at least one of said sensors is an angle sensor that measures an angle of a first portion of said suspension system with respect to a second portion of said suspension system.51. The method of claim 44, wherein said fitness function comprises a weighted combination of one or more state variables of said suspension system.52. The method of claim 44, wherein said fitness function comprises a weighted combination of one or more variables related to passenger comfort.53. The method of claim 44, wherein said fitness function comprises a weighted combination of one or more variables related to passenger comfort, and wherein one or more weights of said weighted combination can be adjusted according to passenger desires.54. The method of claim 44, wherein said fitness function comprises a weighted combination of vehicle movements such that said control system will reduce one or more selected movements of said vehicle.55. The method of claim 37, wherein said second control system is configured to control a throttle valve in a shock absorber.56. The method of claim 37, wherein said fitness function comprises a weighted combination of one or more state variables.57. The method of claim 37, wherein said fitness function comprises a weighted combination of one or more system variables.58. A control apparatus comprising: off-line optimization means for determining a control parameter from an entropy production rate to produce a knowledge base from a teaching signal found by a step-constrained genetic analyzer; and online control means for using said knowledge base to develop a control parameter to control a plant.
연구과제 타임라인
LOADING...
LOADING...
LOADING...
LOADING...
LOADING...
이 특허에 인용된 특허 (17)
Harada, Hiroshi; Matsushita, Yukio; Kaji, Hirotaka; Yamaguchi, Masashi, Apparatus for controlling watercraft pilot control.
Frerichs Donald K. (Shaker Heights OH) Kaya Azmi (Akron OH) Keyes ; IV Marion A. (Chagrin Falls OH), Method and procedure for neural control of dynamic processes.
Kaji, Hirotaka; Yamaguchi, Masashi; Harada, Hiroshi; Matsushita, Yukio, Method of real-time optimization of control device equipped with fuzzy inference system.
Savaresi, Sergio M.; Spelta, Cristiano; Delvecchio, Diego; Bonaccorso, Gabriele; Ghirardo, Fabio; Campo, Sebastiano, Control of a suspension system of a vehicle provided with four semi-active suspensions.
Funes, Pablo; Popovici, Elena; Gaudiano, Paolo; Buchsbaum, Daphna; Garagic, Denis; Ecemis, M. Ihsan; Bingham, Chris; Bonabeau, Eric, Method and system for fast, generic, online and offline, multi-source text analysis and visualization.
Bonabeau, Eric; Anderson, Carl; Scott, John M.; Budynek, Julien; Malinchik, Sergey, Methods and systems for applying genetic operators to determine system conditions.
Bonabeau, Eric; Anderson, Carl; Scott, John M.; Budynek, Julien; Malinchik, Sergey, Methods and systems for applying genetic operators to determine system conditions.
Bonabeau,Eric; Anderson,Carl; Scott,John M.; Budynek,Julien; Malinchik,Sergey, Methods and systems for applying genetic operators to determine system conditions.
Bonabeau, Eric; Anderson, Carl; Scott, John M.; Budynek, Julien; Malinchik, Sergey, Methods and systems for applying genetic operators to determine systems conditions.
※ AI-Helper는 부적절한 답변을 할 수 있습니다.