Servo Filter Coefficient Tuning for Resonance and Phase Delay
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing servo control systems face challenges in setting optimal filter parameters, such as attenuation coefficients and central frequencies, which can lead to insufficient resonance suppression and increased phase delay, affecting the overall performance of the servo control unit.
Innovation Solution
A machine learning system that optimizes filter coefficients by initially setting values to attenuate specific frequency components, calculating input/output gain and phase delay, and removing initial filter characteristics to enable machine learning, thereby adjusting coefficients to minimize gain and phase delay through reinforcement learning.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Object-affected harmful factors
If filter parameters are set to attenuate specific frequency components, then resonance suppression is improved, but phase delay increases and servo control performance deteriorates
Solution Approach 1:
The patent applies parameter changes by using machine learning to automatically optimize filter parameters (attenuation coefficient, central frequency, bandwidth) based on measured frequency characteristics. The system learns optimal parameter combinations that balance resonance suppression with minimal phase delay impact, transforming manual parameter setting into an adaptive optimization process that resolves the contradiction between attenuation effectiveness and control performance.
Solution Approach 2:
The patent implements feedback by measuring the actual frequency characteristics (gain and phase delay) of the servo control device and using this information to train the machine learning model. The system continuously refines filter parameters based on feedback from performance measurements, creating a closed-loop optimization process that resolves the trade-off between resonance attenuation and phase delay by learning from actual system behavior.
2Object-affected harmful factors
If multiple filter parameters are adjusted to optimize performance, then resonance suppression improves, but device complexity increases
Solution Approach 1:
The patent applies self-service by implementing an automated machine learning system that performs parameter optimization without requiring manual intervention. The system automatically measures frequency characteristics, trains the learning model, and determines optimal filter parameters independently, eliminating the need for complex manual parameter adjustment procedures while achieving superior resonance suppression.
Solution Approach 2:
The patent applies preliminary action by pre-training the machine learning model with measured frequency characteristics before actual filter parameter optimization. The system performs preliminary data collection and model training offline, so that during operation, optimal parameters can be quickly determined without complex real-time adjustments, simplifying the overall process while maintaining high performance.
Data Source
AI summary
A machine learning system that optimizes coefficients of at least one filter provided in a servo control device that controls a motor includes: an initial setting unit that sets initial values of the coefficients of the filter so as to attenuate at least one specific frequency component; a frequency characteristic calculation unit that calculates at least one of an input/output gain and an input/output phase delay of the servo control device based on an input signal and an output signal of the servo control device, of which the frequencies change; and a filter characteristic removing unit that removes filter characteristics of an initial filter in which the initial values are set to the filter from at least one of the input/output gain and the input/output phase delay obtained based on the input signal and the output signal obtained using the initial filter at the start of machine learning.