Controller Setting Selection for Technical Oscillation Reduction
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Complex technical systems like gas turbines, wind turbines, and power grids often experience oscillations due to complex dynamical interactions, which reduce efficiency and increase wear, and current machine learning controllers struggle to effectively detect and reduce these oscillations.
Innovation Solution
A method that processes operational data signals into the frequency domain to determine entropy values, allowing for the selection of optimal controller settings to reduce oscillations, using techniques such as autocorrelation and Fast Fourier Transformation, and employing machine learning methods like artificial neural networks to identify and mitigate oscillatory patterns.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Adaptability or versatility
If machine learning methods are used to control complex technical systems, then the ability to model complex dynamical behavior is improved, but complex oscillation patterns are generated that are difficult to detect and reduce
Solution Approach 1:
The patent introduces an intermediary evaluation system that uses entropy calculation as a mediator between the controller and the technical system. This intermediary layer transforms the complex oscillation patterns into a scalar entropy value that quantifies oscillation severity, making them detectable and reducible without modifying the underlying machine learning controller
Solution Approach 2:
The patent replaces direct mechanical or control-theoretic oscillation detection methods with an information-theoretic approach using entropy calculation. This substitution transforms the detection problem from analyzing complex time-domain signals to evaluating a single entropy parameter, significantly simplifying the detection process
2Reliability
If sophisticated control strategies are used to operate technical systems, then productive and stable operation is achieved, but oscillations are induced that impair efficiency and increase wear
Solution Approach 1:
The patent implements a feedback mechanism where the entropy value calculated from system signals is used to evaluate controller settings and select optimal configurations. This feedback loop continuously monitors oscillation levels and adjusts controller parameters to minimize entropy, thereby reducing harmful oscillations while maintaining stable operation
Solution Approach 2:
The patent changes the evaluation parameter from traditional control metrics to entropy-based oscillation measurement. By selecting controller settings that optimize entropy values rather than traditional performance metrics, the system achieves stable operation with reduced oscillations and improved efficiency
Data Source
Figure 1~2
AI summary
For reducing oscillations in a technical system (GT) a plurality of different controller settings (CP) for the technical system (GT) is received. For a respective controller setting (CP) a signal (SG) representing a time series of operational data of the technical system controlled by the respective controller setting (CP) is received, the signal (SG) is processed, whereby the processing comprises a transformation into a frequency domain, and an entropy value (S) of the processed signal (PSG) is determined. Depending on the determined entropy values (S) a controller setting (SCP) from the plurality of controller settings (CP) is selected, and the selected controller setting (SCP) is output for configuring the technical system (GT).