Back to results
US 12,607,974 B2Granted
Adaptive engine with pre-processed control laws
AssigneeAdvanced Energy Industries, Inc.
InventorChad S. Samuels
Life of this patent
Estimating the term…
Abstract
This disclosure describes systems, methods, and apparatus for an adaptive engine with a bifurcated nonlinear model partially represented by a mapping function. The adaptive engine uses a nonlinear model able to produce a possible control signal via a mapping function. The model also has estimation portion using a time-varying linear system to approximate nonlinear behavior of a power system. A boot up phase can be used wherein the mapping function provides possible control signals based on measured system outputs rather than an estimated model parameter tensor.
Claims18
An adaptive engine configured to receive a series of reference signals and in response provide a control to one or more actuators controlling parameters of a power system, the adaptive engine comprising:
- a processor and a memory storing instructions that, when executed by the processor, cause the adaptive engine to implement:
- a plurality of estimation law modules each configured to implement at least one estimation law, and each configured to estimate estimated model parameter tensors, θ, for a nonlinear model of the one or more actuators and/or the power system controlled by the one or more actuators, based at least on an estimation error, e, or cost function, J, calculated from two or more of (1) measured system outputs, (2) estimated system outputs, and (3) a reference signal, the nonlinear model having a structure of a time-varying linear system, Wm, that changes from frame to frame to approximate large nonlinear behaviors in the power system, but remains constant within each frame, such that adaptation of the estimated model parameter tensor, θ, is based on a linear system within each frame;
- a first mapping function configured to generate possible control signals, Use, based on respective ones of the estimated model parameter tensors, θ;
- a selector module configured to generate a control, uout, and provide the control, uout, to the one or more actuators, wherein the generating is based on either (1) selection of one or more of the possible control signals, use, as the control, uout, or (2) blending two or more of the possible control signals, use, as the control, uout, and provide the control, uout, to the one or more actuators controlling parameters of the power system.
1 more element
A method of adaptive control comprising:
- estimating an estimated model parameter tensor, θ, for every control sample, wherein the estimating is based at least on an estimation law that is a function of an estimation error, ê, or cost function, J, wherein the estimated model parameter tensor, θ, comprises estimated parameters of a nonlinear model;
- generating a possible control signal, use, using a mapping function related to a nonlinear model;
- generating an estimated system output, yest_se, using an estimation portion of the nonlinear model or another mapping function;
- wherein both the estimation portion of the nonlinear model and the mapping function are functions of at least the estimated model parameter tensor, θ;
- wherein the estimation portion is also a function of a structure of a time-varying linear system, Wm, and wherein the structure of the time-varying linear system changes from frame to frame, but remains constant within each frame;
- and
- selecting (1) a best possible control signal, use, from a set comprising the possible control signal, Use, and additional possible control signals, or (2) a best combination of possible control signals, use, blended from two or more of the set.
4 more elements
A method of adaptive control, the method comprising:
- analyzing a reference signal, r, to determine an optimal frame length;
- accessing a model of a power system's nonlinear behavior comprising an estimation portion, the estimation portion comprising a time-varying linear system approximation, Wm, comprising a system of equations having matrix coefficients, the time-varying linear system approximation, Wm, being a function of:
- a possible control signal, use;
- and
- an estimated model parameter tensor, θ, of the model;
- for each of the multiple frames, providing a matrix structure for the matrix coefficients, wherein the matrix structure changes from frame to frame to approximate large nonlinear behaviors in the power system, but remains constant within each of the multiple frames;
- and
- for control samples of each of the multiple frames:
- estimating the estimated model parameter tensor, θ;
- determining a possible control signal, use, via a mapping function that takes the estimated model parameter tensor, θ, as an input;
- determining an estimated system output, yest_se, by applying the possible control signal, use, and the estimated model parameter tensor, θ, to the estimation portion of the model or another mapping function;
- and
- selecting a best possible control signal or best combination of possible control signals from a set including the possible control signal, use, and other possible control signals, as a control output, uout, the selecting based on an error or cost function dependent on at least the estimated system output, yest_se.
10 more elements
Dependencies are inferred from the claim text.