Adaptive Turbine Control for Fast Transient Power Response
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing power generation systems, such as gas turbine systems, face challenges in accurately controlling power output during fast transient events due to delays in response time, leading to reduced accuracy of estimated parameters and poor controllability.
Innovation Solution
A system that uses a controller with processors to generate modeled outputs based on a power generation system model and inputs, and adjusts a correction factor using both a filter component and an adaptive component to match modeled outputs to measured outputs, thereby improving model accuracy and controllability.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If real-time tuning is applied to match modeled output to measured output, then model accuracy is improved, but response delay occurs during fast transient events
Solution Approach 1:
The correction factor is made dynamic by separating it into a filter component (for steady-state accuracy) and an adaptive component (for transient response). The adaptive component uses past correction factors and current measured outputs to predict future correction needs, allowing the system to adapt quickly to changing conditions without the delay inherent in traditional real-time tuning approaches.
Solution Approach 2:
The adaptive component performs preliminary action by using historical data and algorithms (least squares regression, auto-regression moving average, subspace identification, or linear quadratic estimator) to predict the required correction factor before the actual transient event occurs. This predictive approach eliminates the response delay by preparing the correction in advance based on patterns learned from past behavior.
2Stability of the object's composition
If traditional correction factor tuning is used, then model stability is maintained, but controllability deteriorates during fast transient events
Solution Approach 1:
The correction factor is segmented into two distinct components: a filter component that ensures model stability by smoothing out noise and variations, and an adaptive component that enhances controllability by providing predictive adjustments during transient events. This segmentation allows each component to specialize in its function without compromising the other.
Solution Approach 2:
The adaptive component implements feedback by continuously using past correction factors and current measured outputs to inform future corrections. This feedback mechanism allows the system to learn from past behavior and improve its predictive accuracy, enhancing controllability during transient events while the filter component maintains overall stability.
Data Source
Figure 1
Figure 2~3
Figure 4~5
AI summary
A system (40) includes a power generation system (10) and a controller (18). The controller (18) includes processors (19) that receive a first set of inputs (42). The processors (19) also generate a first set of modeled outputs (46) system based on a model (44) of the power generation system (10) and the first set of inputs (42). The processors (19) further receive a first set of measured outputs (56) corresponding to the first set of modeled outputs (46). The processors (19) determine a first correction factor (58) based on the first set of modeled outputs (46) and the first set of measured outputs (56). The first correction factor (58) includes differences between the first set of modeled outputs (46) and the first set of measured outputs (56). The processors (19) also generate a second set of modeled outputs (46) based on the model, a second set of inputs (42), and the first correction factor (58). The processors (19) further control an operation of the power generation system (10) based on the second set of modeled outputs (46).