Servo Control Parameter Search Range Adjustment for Vibration Reduction
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
The search range for machine learning in servo control devices is often not appropriately set, leading to suboptimal parameter selection for servo motors in machine tools, robots, and industrial machines, which can result in inefficient operation and increased vibration.
Innovation Solution
A machine learning device that includes a search solution detection unit, an evaluation function expression estimation unit, and a search range changing unit to dynamically adjust the search range based on evaluation function values, ensuring that the search solution is at the edge or within a predetermined range, allowing for optimal parameter selection and minimization of position errors.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If the search range is set in advance with a fixed width, then the machine learning device can perform parameter search efficiently, but the search may not cover the appropriate range leading to suboptimal parameter selection
Solution Approach 1:
The patent implements dynamic search range adjustment by detecting when the search solution is near the edge of the current search range and automatically expanding the range in that direction. This transforms the static search range into a dynamic one that adapts during the machine learning process, allowing the system to maintain both search efficiency and parameter selection accuracy.
Solution Approach 2:
The patent employs feedback mechanisms where the search solution detection unit monitors the position of the optimal parameter within the search range, and this information feeds back to the search range changing unit which adjusts the range accordingly. This closed-loop feedback ensures that the search range always encompasses the appropriate parameter values while maintaining efficient search performance.
2Measurement precision
If the search range is expanded to ensure optimal parameter detection, then the accuracy of parameter selection improves, but the search time and computational resources increase
Solution Approach 1:
Rather than using a uniformly large search range from the beginning, the patent dynamically adjusts the search range width based on the detected search solution position. The range expands only when necessary (when solution is near edge) and maintains a reasonable width otherwise, thus balancing accuracy requirements with time efficiency.
Solution Approach 2:
The patent changes the parameter of search range width dynamically during the machine learning process. By adjusting this parameter based on the detected solution position, the system achieves high accuracy when needed while minimizing unnecessary search time when the solution is well within the range.
3Measurement precision
If the search range is adjusted dynamically based on search solution position, then the parameter selection accuracy improves, but the device complexity increases
Solution Approach 1:
The patent divides the search range adjustment function into distinct modular units: a search solution detection unit that monitors solution position, and a search range changing unit that adjusts the range. This segmentation makes the complex control logic more manageable and implementable while achieving high precision parameter selection.
Solution Approach 2:
The patent introduces an intermediary evaluation function that bridges the search process and the search range adjustment. The evaluation function values are used to detect the search solution position and trigger appropriate range adjustments, serving as a mediator that simplifies the overall control complexity while maintaining high accuracy.
Data Source
AI summary
A machine learning device that searches for a first parameter of a component of a servo control device that controls a servo motor includes: a solution detection unit that acquires a set of evaluation function values used during machine learning or after machine learning, plots the set of evaluation function values in a search range of the first parameter or a second parameter used for searching for the first parameter, and detects whether a search solution is at an edge of the search range or is in a predetermined range from the edge; and a range changing unit that changes the search range to a new search range of the first parameter or the second parameter based on the estimation made on evaluation function values of an evaluation function expression when the search solution is at the edge of the search range or is in the predetermined range.