Adaptive Engine Control Law Selection for Stable Plasma Actuation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current adaptive controllers in plasma processing systems lack stability, scalability, and adaptability, struggling with nonlinear systems, modeling uncertainties, and unbounded control values, particularly in handling asynchronous actuators with different response times.
Innovation Solution
An adaptive engine that generates control signals based on an input regressor, using a combination of estimation laws and control laws, with a selector module to choose the best control signals minimizing an error or cost function, and an adaptation law generator producing estimated model parameter tensors for precise actuator control.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If existing adaptive controllers are used, then control signals can be generated, but the controllers lack stability and struggle with unbounded control values
Solution Approach 1:
The patent transforms unbounded control values into bounded control signals by applying saturation functions and scaling parameters. The adaptation law generator modifies control parameters dynamically to ensure outputs remain within acceptable bounds while maintaining system adaptability to changing conditions.
Solution Approach 2:
The controller implements continuous feedback mechanisms where the output is monitored and fed back to adjust subsequent control actions. This feedback loop ensures stability by correcting deviations and preventing unbounded behavior while maintaining adaptability through dynamic parameter adjustment based on system response.
2Adaptability or versatility
If a single control law is used, then the controller is simple, but it lacks adaptability to various situations within a given recipe
Solution Approach 1:
The controller transitions from a static single control law to a dynamic multi-law system. The adaptation law generator dynamically selects and switches between multiple control laws based on real-time system conditions, allowing the controller to adapt to various situations while maintaining a manageable structure through systematic selection criteria.
Solution Approach 2:
The control law is segmented into multiple specialized control laws, each optimized for specific operating conditions. The adaptation law generator acts as a selector that divides the overall control task into segments handled by different control laws, improving adaptability while organizing complexity through structured segmentation.
3Productivity
If the rail voltage is held at a high level for much of a pulse cycle, then the power amplifier can provide the desired pulsed waveform, but this leads to overheating of components and premature system failure
Solution Approach 1:
The controller implements periodic pulsing of the rail voltage rather than maintaining it continuously at high levels. By applying voltage in controlled pulses synchronized with the plasma processing cycle, the system maintains productivity during active phases while allowing cooling periods during idle phases, preventing overheating and extending component life.
Data Source
AI summary
This disclosure describes systems, methods, and apparatus for a selector and combiner of an adaptive engine. The adaptive engine can combinations of estimation and control laws to produce a multitude of possible control signals based on inputs such as a reference signal. A nonlinear model of the system can generate estimated system outputs for each of the possible controls signals. The selector and combiner can use the estimated system outputs to determine a best of the possible control signals, or a best combination of the possible control signals, such that the control approaches a desired measured output of the system.