초록 ▼

An optimal feedforward control design for controlling equipment performs intermittent re-initialization based on estimated state information. A model-based constrained optimization is explicitly performed during the feedforward calculation. During the course of operation of the equipment, state esti

An optimal feedforward control design for controlling equipment performs intermittent re-initialization based on estimated state information. A model-based constrained optimization is explicitly performed during the feedforward calculation. During the course of operation of the equipment, state estimation is continuously performed. When a load target change is detected, the estimated state may serve as the new signal baseline.

대표청구항 ▼

1. A method of controlling a power generating unit, the method comprising: receiving, at a computing device, a signal indicative of a first load demand set point, the first load demand set point indicative of a first desired output of the power generating unit;determining, via the computing device,

1. A method of controlling a power generating unit, the method comprising: receiving, at a computing device, a signal indicative of a first load demand set point, the first load demand set point indicative of a first desired output of the power generating unit;determining, via the computing device, a control signal to be used to drive the power generating unit to operate to generate power according to an initial load ramp process, the initial load ramp process based on a particular model and the first load demand set point;periodically or continuously performing a state estimation while the power generating unit is ramping, according to the initial load ramp process, towards the first load demand set point to obtain a current state estimation of the power generating unit; andwhile the power generating unit is ramping, according to the initial load ramp process, towards the first load demand set point: receiving a signal indicative of a second load demand set point, the second load demand set point indicative of a second desired output of the power generating unit;using the current state estimation to calculate, via the computing device, a modified control signal for use in driving the power generating unit to operate to generate power according to a modified load ramp process, the modified load ramp process based on the particular model and the second load demand set point; andusing the modified control signal to modify operation of the power generating unit to operate according to the modified load ramp process to drive the power generating unit towards the second desired output specified by the second load demand set point signal. 2. The method of claim 1, wherein performing the state estimation comprises a performing a linear state estimation process. 3. The method of claim 1, wherein performing the state estimation comprises performing a nonlinear state estimation process. 4. The method of claim 1, wherein using the modified control signal to modify the operation of the power generating unit to operate according to the modified load ramp process comprises calculating an optimal feedforward trajectory according to an optimization formulation. 5. The method of claim 4, wherein using the modified control signal to modify the operation of the power generating unit to operate according to the modified load ramp process comprises using the modified control signal to generate a predicted process output. 6. The method of claim 5, wherein using the modified control signal to modify the operation of the power generating unit to operate according to the modified load ramp process further comprises combining the predicted process output with a measured process value to generate a combined process output and sending the combined process output to a feedback controller to produce a control signal to operate the power generating unit. 7. The method of claim 1, further comprising using the modified control signal to compute an error value between the second load demand set point and an actual operational power output of the power generating unit and creating a new modified model which minimizes the error value. 8. A control system for controlling a power generating unit comprising: a calculation unit adapted to: receive a signal indicative of a first load demand set point, the first load demand set point indicative of a first desired output of the power generating unit; andreceive, while the power generating unit is ramping according to an initial load ramp process towards the first load demand set point, a signal indicative of a second load demand set point, the second load demand set point indicative of a second desired output of the power generating unit;a state estimation unit coupled to the calculation unit, the state estimation unit adapted to measure at least one characteristic associated with a current operational state of the power generating unit while the power generating unit is ramping, according to the initial load ramp process, towards the first load demand set point, the state estimation further adapted to generate a current state calculation based on the at least one characteristic;wherein the calculation unit is further adapted to calculate a first set of operational parameters based on the first load demand set point, and the calculation unit is further adapted to calculate a second set of operational parameters based on the second load demand set point and the current state calculation; anda control signal generator adapted to: generate an initial control signal to drive the power generating unit towards the first desired output according to the initial load ramp process, the initial load ramp process based on a particular model and the first set of operational parameters; andgenerate, while the power generating unit is ramping towards the first load demand set point according to the initial load ramp process, a modified control signal to drive the power generating unit towards the second desired output according to a modified load ramp process, the modified load ramp process based on the particular model and the second set of operational parameters. 9. The control system of claim 8, wherein the calculation unit is further adapted to calculate an optimal feedforward trajectory according to an optimization formulation. 10. The control system of claim 9, wherein the control signal generator is further adapted to generate a predicted process output using the modified control signal. 11. The control system of claim 10, wherein the control signal generator is further adapted to combine the predicted process output with a measured process value to generate a combined process output and send the combined process output to a feedback controller to produce a further control signal to operate the power generating unit. 12. The control system of claim 8, wherein the at least one characteristic associated with the current operational state includes at least one of an operational pressure of the power generating unit or an operational power generation of the power generating unit. 13. The control system of claim 8, wherein the calculation unit is further adapted to calculate an error value representing a difference between actual operational parameters at a given time and the second set of operational parameters at the given time, wherein upon the difference exceeding a threshold value, the state estimation unit is adapted to measure at least one subsequent characteristic, the calculation unit is adapted to calculate a subsequent set of operational parameters based on the at least one subsequent characteristic, and the control signal generator is further adapted to control the power generating unit based on the particular model and the subsequent set of operational parameters. 14. The control system of claim 8, wherein the calculation unit is adapted to calculate the second set of operational parameters only upon receiving a load demand set point signal. 15. A method of controlling equipment in a plant, the method comprising: configuring an initial optimal operational control of the equipment based on a particular model and a signal indicative of an initial load demand set point, the initial load demand set point indicative of an initial desired output of the equipment;controlling the equipment in the plant to operate according to an initial load ramp process, the initial load ramp process based on the initial optimal operational control;periodically or continuously running, while the equipment is ramping according to the initial load ramp process towards the initial load demand set point, a model-based state estimation to obtain a current state estimation, the model-based state estimation being based on a plurality of operational values of the equipment; andwhile the equipment is ramping according to the initial load ramp process towards the initial load demand set point: determining whether a signal indicative of a new load demand set point is received, the new load demand set point indicative of a new desired output of the equipment;upon receiving the signal indicative of the new load demand set point, computing a new optimal operational feedforward control trajectory using the current state estimation; andmodifying the controlling of the equipment in the plant so that the equipment operates in accordance with a modified load ramp process to drive the equipment towards the new load demand set point, the modified load ramp process based on the particular model and the new optimal operational feedforward control trajectory. 16. The method of claim 15, further comprising running a feedback control calculation in addition to the model-based state estimation, the feedback control calculation comprising one of a proportional-integral-derivative controller, a lead-lag controller, a model predictive controller, and a linear-quadratic-Gaussian controller. 17. The method of claim 16, wherein the modified load ramp process is further based on the feedback control calculation. 18. The method of claim 15, further comprising: determining whether or not the equipment has reached the new load demand set point; andupon determining the equipment has not reached the new load demand set point, computing a subsequent optimal operational feedforward and feedback control trajectory using a model-based state estimation representative of a current time. 19. The method of claim 15, wherein computing the new optimal operational feedforward control trajectory comprises performing a minimization calculation involving at least one of an initialization parameter, a state constraint, an input constraint, an input change constraint, or an output constraint. 20. A power plant, comprising: a turbine;a boiler coupled to the turbine that operates to create steam to drive the turbine; anda control unit communicatively connected to the boiler to control the operation of the boiler, the control unit including; a feedback controller that produces a feedback control signal;a feedforward controller that produces a feedforward control signal, the feedforward controller including: a calculation unit adapted to: receive a signal indicative of a first load demand set point, the first load demand set point indicative of a first desired output of the power plant; andreceive, while the plant is ramping according to an initial load ramp process towards the first load demand set point, a signal indicative of a second load demand set point, the initial load ramp process based on a particular model and the first load demand set point, and the second load demand set point indicative of a second desired output of the power plant;a state estimation unit adapted to, while the power plant is ramping according to the initial load ramp process towards the first load demand set point, measure at least one characteristic associated with a current operational state of the power plant and generate a current state calculation based on the at least one characteristic;wherein the calculation unit is further adapted to calculate a first set of operational parameters based on the first load demand set point, and the calculation unit is further adapted to calculate, while the plant is ramping according to the initial load ramp process towards the first load demand set point, a second set of operational parameters based on the second load demand set point and the current state calculation; anda feedforward control signal generator adapted to: produce, upon receiving the signal indicative of the first demand set point, the feedforward control signal, the feedforward control signal including a first response characteristic, the first response characteristic based on the first set of operational parameters, and the feedforward control signal used to drive the power plant towards the first desired output according to the initial load ramp process; andproduce, upon receiving the signal indicative of the second demand set point while the power plant is ramping according to the initial load ramp process towards the first load demand set point, a modified feedforward control signal, the modified feedforward control signal including a second and different response characteristic, the second and different response characteristic based on the second set of operational parameters, and the modified feedforward control signal used to modify the driving of the power plant towards the second desired output according to a modified load ramp process, the modified load ramp process based on the particular model and the second load demand set point; anda control signal combiner that combines the modified feedforward control signal and the feedback control signal to create a master control signal for controlling the power plant. 21. The boiler operated power plant of claim 20, wherein the feedback controller comprises one of a proportional-integral-derivative controller, a lead-lag controller, a model predictive controller, and a linear-quadratic-Gaussian controller. 22. The boiler operated power plant of claim 21, wherein the feedforward controller is adapted to produce control signals only upon receiving a load demand set point signal. 23. The boiler operated power plant of claim 20, wherein when the state estimation unit is described by a nonlinear dynamic process, the calculation unit is a nonlinear calculation unit. 24. The boiler operated power plant of claim 20, wherein when the state estimation unit is described by a linear dynamic process, the calculation unit is a linear calculation unit.