Design and control of engineering systems utilizing component-level dynamic mathematical model with multiple-input multiple-output estimator

• US 8,131,384 B2
• Filed: 05/29/2009
• Issued: 03/06/2012
• Est. Priority Date: 11/03/2008
• Status: Active Grant

• Current Assignee: Rtx Corporation
• Original Assignee: United Technologies Corporation (Raytheon Technologies Corporation d/b/a RTX)
• Inventors: Karpman, Boris, Meisner, Richard P., Lacour, Mark E.
• Primary Examiner(s): Kasenge, Charles
• Application Number:
  US12/475,020
• Publication Number:
  US 20110230981A1
• Time in Patent Office: 1,012 Days
• Field of Search: 700/28, 700/29, 700/30, 700/32, 700/33, 700/34, 700/37, 700/44, 700/45, 700/52, 700/54, 700/282, 700/283, 60/392.81, 607/72, 607/90, 701/100, 701/101
• US Class Current: 700/45
• CPC Class Codes: G05B 13/04 involving the use of models or simulators

Abstract

A control system comprises an actuator, a control law and a processor. The actuator positions a control surface and the control law controls the actuator. The processor comprises an open loop module, a corrector, a comparator, and an estimator, and generates an output vector to direct the control law. The open loop module generates the output vector as a function of a state vector and an input vector. The corrector generates a corrector vector as a function of the output vector. The comparator generates an error vector by comparing the corrector vector to the input vector. The estimator generates the state vector as a function of the error vector, such that the error vector is minimized.

Claims

1. A control system comprising:

an actuator for positioning a control surface, the control surface comprising a vane or nozzle surface positioned to control a boundary condition of a working fluid flow;

a control law for controlling the actuator, and a processor for generating an output vector to direct the control law, the processor comprising;

an open loop module for generating the output vector as a function of a state vector and an input vector, wherein the input vector describes the boundary condition of the working fluid flow;

a corrector for generating a corrector vector as a function of the output vector;

a comparator for generating an error vector by comparing the corrector vector to the input vector; and

an estimator for generating the state vector as a function of a multiple-input, multiple-output gain tensor that operates to decouple cross-correlated elements of the error vector and the state vector, such that the error vector is minimized;

wherein the state vector describes a spool speed and the processor has a cycle time of 50 ms or less, such that the state vector has a response time slower than the cycle time.

2. The control system of claim 1, wherein the multiple-input, multiple-output gain tensor varies as a nonlinear function of the input vector.

3. The control system of claim 1, wherein the error vector is a function of a time derivative of the state vector.

4. The control system of claim 1, wherein a time derivative of a solution state element of the state vector is zero.

5. The control system of claim 1, wherein the open loop module generates the output vector as a further function of a continuity constraint on the state vector, and wherein the continuity constraint is based on the boundary condition.

6. The system of claim 1, wherein the multiple-input, multiple-output gain matrix has no substantially triangular form.

7. A method for controlling a rotational state of a turbine engine, the method comprising:

sensing a boundary state with a sensor, the boundary state describing flow at a boundary of the turbine engine;

controlling an actuator state with a control law as a function of a model feedback, wherein the actuator state describes a control surface positioned in the flow;

generating the model feedback with a closed-loop processor as a function of the boundary state and the actuator state, and as a further function of the rotational state of the turbine engine, wherein the rotational state is related to the flow at the boundary;

correcting the model feedback for errors with a corrector module of the closed-loop processor, based on the boundary state; and

estimating the rotational state by minimizing the errors with a model state estimator of the closed-loop processor, wherein the model state estimator decouples cross-correlated errors and states as a function of a multiple-input, multiple-output gain matrix, such that individual errors and individual states are decoupled; and

controlling the rotational state of the turbine engine, such that a time rate of change of the rotational state is minimized;

wherein the closed-loop processor has a cycle time of 50 ms or less, such that the rotational state has a response time slower than the cycle time.

8. The method of claim 7, wherein the boundary state describes the flow at one of an inlet flow boundary of the turbine engine or an outlet flow boundary of the turbine engine.

9. The method of claim 8, wherein the actuator state describes a control surface positioned in the flow between the inlet flow boundary and the outlet flow boundary.

10. The method of claim 9, wherein the rotational state is constrained by flow continuity between the inlet flow boundary and the outlet flow boundary.

11. The method of claim 7, wherein the multiple-input, multiple-output gain matrix has no substantially triangular form.

12. A control system for a gas turbine engine, the system comprising:

a sensor for sensing a boundary condition that constrains a spool speed;

an actuator for changing a control state in order to alter the boundary condition;

a module for generating output as a function of the boundary condition and the control state;

a comparator for generating errors by comparing the output to the boundary condition and the control state;

an estimator for estimating the boundary condition as a function of a multiple-input, multiple-output gain matrix that decouples cross-correlated errors, such that the errors are minimized; and

a controller for directing the actuator as a function of the output, such that the spool speed is controlled;

wherein the control system has a cycle time of 50 ms or less, such that the spool speed has a response time slower than the cycle time.

13. The system of claim 12, wherein the boundary condition describes flow related to the spool speed and constrains the spool speed based on flow continuity.

14. The system of claim 12, wherein the gain matrix is defined as a time rate of change.

15. The system of claim 12, wherein the gain matrix is a nonlinear function of the boundary condition.

16. The system of claim 12, wherein the multiple-input, multiple-output gain matrix has no substantially triangular form.