Feedback Control with Thinned ML Sampling for Low-Frequency Vibration
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Feedback control devices that do not use machine learning struggle to effectively suppress low-frequency disturbance vibrations due to short sampling times, which limit the input data available for machine learning-based control systems, preventing adequate suppression of these vibrations.
Innovation Solution
A feedback control device that incorporates both a classical PID control system and a machine learning-based control system, where the machine learning system processes control deviation data thinned out at a predetermined period, allowing for longer input time lengths and improved suppression of low-frequency disturbances without increasing calculation complexity.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Speed
If the sampling frequency is increased to improve responsiveness, then the sampling time becomes shorter and responsiveness is improved, but the time length of control deviation data input to the neural network becomes shorter, preventing effective suppression of low-frequency disturbance vibration
Solution Approach 1:
The control system is divided into two independent control units: a first control unit that processes control deviation data at a first sampling frequency, and a second control unit that processes control deviation data at a second sampling frequency. This segmentation allows each control unit to operate with optimized sampling parameters for its specific function, resolving the contradiction between responsiveness and low-frequency vibration suppression.
Solution Approach 2:
The system dynamically adjusts sampling frequencies for different control functions. The first control unit uses a higher sampling frequency for responsive control, while the second control unit uses a lower sampling frequency for analyzing low-frequency disturbance vibrations. This dynamic parameter adjustment enables both responsiveness and effective vibration suppression.
2Loss of time
If the sampling time is shortened to improve responsiveness, then the time length of control deviation data input to the neural network becomes shorter, but machine learning does not proceed well and the control system cannot suppress low-frequency disturbance vibration
Solution Approach 1:
The control system is divided into two independent control units: a first control unit that processes control deviation data at a first sampling frequency, and a second control unit that processes control deviation data at a second sampling frequency. This segmentation allows each control unit to operate with optimized sampling parameters for its specific function, resolving the contradiction between responsiveness and low-frequency vibration suppression.
Solution Approach 2:
Different sampling frequencies are applied to different control functions based on their specific requirements. The first control unit uses a higher sampling frequency suitable for responsive control, while the second control unit uses a lower sampling frequency that provides sufficient data length for machine learning to effectively suppress low-frequency disturbance vibrations.
Data Source
AI summary
The feedback control device takes information regarding a control deviation between a measured value and a target value of a controlled object as input, and outputs a control amount for the controlled object; comprising: a first control unit that takes information regarding the control deviation as input, and outputs a first control amount for the controlled object; a second control unit that takes information regarding the control deviation as input and outputs a second control amount for the controlled object, and in which a parameter for calculating the second control amount is determined by machine learning; an operation unit that operates the controlled object using the first control amount output from the first control unit and the second control amount output from the second control unit; and a sampling unit for thinning out at a predetermined period information regarding the control deviation input to the second control unit.