System and method for controlling a dynamic system

• US 10,247,170 B2
• Filed: 06/07/2016
• Issued: 04/02/2019
• Est. Priority Date: 06/07/2016
• Status: Active Grant

• Current Assignee: GE Infrastructure Technology, LLC (General Electric Company)
• Original Assignee: General Electric Company
• Inventors: Evans, Scott Charles, Delport, Sara Simonne Louisa, Davoust, Samuel, Xu, Yunwen
• Primary Examiner(s): Fennema, Robert E
• Assistant Examiner(s): Monty, Marzia T
• Application Number:
  US15/175,771
• Publication Number:
  US 20170350369A1
• Time in Patent Office: 1,029 Days
• Field of Search: 700287, 700290, 700170, 700219, 700292
• CPC Class Codes: F03D 1/06 Rotors
• CPC Class Codes: F03D 13/20 Arrangements for mounting or supporting wind motors; Masts or towers for wind motors
• CPC Class Codes: F03D 17/00 Monitoring or testing of wind motors, e.g. diagnostics testing during commissioning of wind motors F03D13/30
• CPC Class Codes: F03D 7/0224 Adjusting blade pitch
• CPC Class Codes: F03D 7/045 with model-based controls
• CPC Class Codes: F03D 7/046 with learning or adaptive control, e.g. self-tuning, fuzzy logic or neural network
• CPC Class Codes: F03D 7/047 characterised by the controller architecture, e.g. multiple processors or data communications
• CPC Class Codes: F03D 7/048 controlling wind farms
• CPC Class Codes: F05B 2270/32 Wind speeds
• CPC Class Codes: G05B 13/027 using neural networks only
• CPC Class Codes: G05B 19/04 Programme control other than numerical control, i.e. in sequence controllers or logic controllers G05B19/418 takes precedence
• CPC Class Codes: G05B 2219/2619 Wind turbines
• CPC Class Codes: Y02E 10/72 Wind turbines with rotation axis in wind direction
• CPC Class Codes: Y02E 10/728 Onshore wind turbines

Abstract

A control system for a dynamic system including at least one measurement sensor. The system includes at least one computing device configured to generate and transmit at least one regulation device command signal to at least one regulation device to regulate operation of the dynamic system based upon at least one inferred characteristic.

Claims

1. A control system for a dynamic system, said control system comprising:

at least one measurement sensor coupled to the dynamic system;

at least one regulation device configured to regulate operation of the dynamic system as a function of at least one dynamic system characteristic; and

at least one computing device coupled to said at least one measurement sensor and said at least one regulation device, said at least one computing device configured to;

receive at least one measurement signal from said at least one measurement sensor, wherein the at least one measurement signal includes measurement data;

classify, using the measurement data, a wind shear profile of the dynamic system from which to infer the at least one dynamic system characteristic;

infer the at least one dynamic system characteristic based, at least in part, on the wind shear profile;

generate at least one regulation device command signal based, at least in part, on the inferred at least one dynamic system characteristic;

wherein the dynamic system is a wind turbine;

wherein said at least one computing device is further configured to classify the wind shear profile using a random forests algorithm on the measurement data;

wherein the at least one regulation device command signal is utilized to change the pitch of one or more rotor blades of the wind turbine to enhance a wind turbine torque;

wherein the said at least one computing device is further configured to;

classify the wind shear profile and to determine whether the measurement data is consistent with a wind shear power law profile;

determine that the measurement data is not consistent with the wind shear power law profile, wherein the at least one computing device uses a fitted quadratic model to infer a wind shear characteristic as the at least one dynamic system characteristic; and

wherein the fitted quadratic model is given by;

V(h)=V(HH)[a(h−

HH)²+b(h−

HH)+1],

where V=wind velocity, HH=lower height, h=upper height, a=first quadratic fit coefficient, and b=second quadratic fit coefficient.

2. The control system in accordance with claim 1, wherein the wind shear power law profile includes a shear exponent α

, said at least one computing device further configured to determine that the measurement data is consistent with the wind shear power law profile, wherein said at least one computing device uses the shear exponent α

(alpha) as the at least one dynamic system characteristic.

3. The control system in accordance with claim 1, wherein said at least one computing device is further configured to infer the at least one dynamic system characteristic for at least one of directional variations, wind speeds at one or more heights, veer, density, turbulence intensity, yaw misalignment, up-flow, vertical shear, horizontal shear, waked state, and non-waked state.

4. The control system in accordance with claim 1, wherein said at least one computing device uses a single characteristic model to infer the at least one dynamic system characteristic.

5. A method for controlling a dynamic system, the dynamic system including at least one measurement sensor, at least one regulation device, and at least one computing device coupled to the at least one measurement sensor and the at least one regulation device, said method comprising:

receiving, by the at least one computing device, at least one measurement signal from the at least one measurement sensor, wherein the at least one measurement signal includes measurement data;

classifying, by the at least one computing device using a random forests algorithm on the measurement data, a wind shear profile of the dynamic system from which to infer at least one dynamic system characteristic;

inferring, by the at least one computing device, the at least one dynamic system characteristic based, at least in part, on the wind shear profile;

generating, by the at least one computing device, at least one regulation device command signal based, at least in part, on the inferred at least one dynamic system characteristic;

wherein the dynamic system is a wind turbine;

wherein the at least one regulation device command signal is utilized to change the pitch of one or more rotor blades of the wind turbine to enhance a wind turbine torque;

wherein the method further comprises;

classifying, by the at least one computing device, the wind shear profile and to determine whether the measurement data is consistent with a wind shear power law profile;

determining, by the at least one computing device, that the measurement data is not consistent with the wind shear power law profile, wherein the at least one computing device uses a fitted quadratic model to infer a wind shear characteristic as the at least one dynamic system characteristic; and

wherein the fitted quadratic model is given by;

V(h)=V(HH)[a(h−

HH)²+b(h−

HH)+1],

where V=wind velocity, HH=lower height, h=upper height, a=first quadratic fit coefficient, and b=second quadratic fit coefficient.

6. The method in accordance with Claim 5, wherein the wind shear power law profile includes a shear exponent α

, the method further comprising determining, by the at least one computing device, that the measurement data is consistent with the wind shear power law profile, wherein the at least one computing device uses the shear exponent α

(alpha) as the at least one dynamic system characteristic.

7. The method in accordance with claim 5 further comprising inferring, by the at least one computing device, the at least one dynamic system characteristic for at least one of directional variations, wind speeds at one or more heights, veer, density, turbulence intensity, yaw misalignment, up-flow, vertical shear, horizontal shear, waked state, and non-waked state.

8. The method in accordance with claim 5 further comprising using, by the at least one computing device, a single characteristic model to infer the at least one dynamic system characteristic.

9. A wind turbine park comprising:

a plurality of wind turbines;

at least one measurement sensor coupled to at least one wind turbine of said plurality of wind turbines;

at least one wind turbine regulation device configured to regulate operation of said at least one wind turbine as a function of at least one wind characteristic; and

at least one computing device coupled to said at least one measurement sensor and said at least one wind turbine regulation device, said at least one computing device configured to;

receive at least one measurement signal from said at least one measurement sensor, wherein the at least one measurement signal includes measurement data;

classify, using the measurement data, a wind shear profile from which to infer the at least one wind characteristic;

infer the at least one wind characteristic based, at least in part, on the wind shear profile;

generate at least one wind turbine regulation device command signal based, at least in part, on the at least one wind characteristic;

wherein said at least one computing device is further configured to classify the wind shear profile using a random forests algorithm;

wherein the at least one wind turbine regulation device command signal is utilized to change the pitch of one or more rotor blades of the at least one wind turbine to enhance a wind turbine torque;

wherein the said at least one computing device is further configured to;

classify the wind shear profile and to determine whether the measurement data is consistent with a wind shear power law profile;

determine that the measurement data is not consistent with the wind shear power law profile, wherein the at least one computing device uses a fitted quadratic model to infer a wind shear characteristic as the at least one wind characteristic; and

wherein the fitted quadratic model is given by;

V(h)=V(HH)[a(h−

HH)²+b(h−

HH)+1],

where V=wind velocity, HH=lower height, h=upper height, a=first quadratic fit coefficient, and b=second quadratic fit coefficient.

10. The wind turbine park in accordance with claim 9, wherein the wind shear power law profile includes a shear exponent α

, said at least one computing device further configured to determine that the measurement data is consistent with the wind shear power law profile, wherein said at least one computing device uses the shear exponent α

(alpha) as the at least one wind characteristic.

11. The wind turbine park in accordance with claim 9, wherein said at least one computing device is further configured to infer the at least one wind characteristic for at least one of directional variations, wind speeds at one or more heights, veer, density, turbulence intensity, yaw misalignment, up-flow, vertical shear, horizontal shear, waked state, and non-waked state.

12. The wind turbine park in accordance with claim 9, wherein said at least one computing device uses a single characteristic model to infer the at least one wind characteristic.