Power generation system control through adaptive learning

A method, comprising:

• receiving, via one or more processors, a first set of inputs associated with a power generation system, wherein the first set of inputs comprise one or more operating parameters related to the power generation system;
• generating, via the one or more processors, a first set of modeled outputs of the power generation system based on a model of the power generation system and the first set of inputs;
• receiving, via the one or more processors, a first set of measured outputs of the power generation system via one or more sensors of the power generation system, wherein the first set of measured outputs correspond to the first set of modeled outputs;
• determining, via the one or more processors, a filter component that, when applied to the model of the power generation system, approximately matches the first set of modeled outputs to the first set of measured outputs;
• determining, via the one or more processors, an adaptive component based on an algorithm that represents a previous correction factor as a function of the first set of measured outputs, wherein the algorithm is based on least squares regression, auto-regression moving average, subspace identification, or linear quadratic estimator;
• determining, via the one or more processors, a current correction factor comprising a product of the filter component and the adaptive component;
• receiving, via the one or more processors, a second set of inputs associated with the power generation system, wherein the second set of inputs comprise the one or more operating parameters related to the power generation system;
• generating, via the one or more processors, a second set of modeled outputs of the power generation system based on the model of the power generation system, the second set of inputs, and the current correction factor;
• and
• a second set of modeled outputs of the power generation system controlling, via the one or more processors, an operation of the power generation system based on the second set of modeled outputs.

One or more tangible, non-transitory, machine-readable media, comprising machine-readable instructions to cause one or more processors to:

• receive a first set of inputs associated with a power generation system, wherein the first set of inputs comprise one or more input operating parameters of the power generation system;
• generate a first set of modeled outputs of the power generation system based on a model of the power generation system and the first set of inputs;
• receive a first set of measured outputs of the power generation system via one or more sensors of the power generation system, wherein the first set of measured outputs correspond to the first set of modeled outputs;
• determine a filter component that, when applied to the model of the power generation system, approximately matches the first set of modeled outputs to the first set of measured outputs;
• determine an adaptive component based on an algorithm that represents a previous correction factor as a function of the first set of measured outputs, wherein the algorithm is based on least squares regression, auto-regression moving average, subspace identification, or linear quadratic estimator;
• determine a first correction factor comprising a product of the filter component and the adaptive component;
• receive a second set of inputs associated with the power generation system, wherein the second set of inputs comprise the one or more input operating parameters of the power generation system;
• generate a second set of modeled outputs of the power generation system based on the model of the power generation system, the second set of inputs, and the first correction factor;
• and
• control an operation of the power generation system based on the second set of modeled outputs.