Auto-Tuning Controller for Nonlinear Systems
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Feedback control systems with nonlinear elements pose challenges in optimizing control parameters, as existing auto-tuning methods struggle to determine appropriate PID parameters due to strong nonlinearity between manipulated and controlled values, leading to deteriorated control performance.
Innovation Solution
A controller with an auto-tuning function that alternately outputs manipulated values to produce limit cycles, successively changing their magnitude to determine PID parameters based on response characteristics, even in systems with strong nonlinearity, by correcting the magnitude of the manipulated value according to the ratio of output periods and evaluating linearity through change rates.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If conventional auto-tuning methods are used in systems with strong nonlinearity, then the control parameters cannot be optimized, but the control performance deteriorates
Solution Approach 1:
The patent applies dynamics by making the manipulated value magnitude adjustable and variable during the auto-tuning process. Instead of using a fixed magnitude, the system dynamically changes the manipulated value magnitude based on the detected nonlinearity degree, allowing the auto-tuning to adapt to nonlinear characteristics of the controlled object and achieve accurate parameter optimization even in strongly nonlinear systems
Solution Approach 2:
The patent changes the parameter of manipulated value magnitude during the auto-tuning process. By detecting the nonlinearity degree and adjusting the manipulated value magnitude accordingly, the system transforms the fixed-parameter approach into a variable-parameter approach, enabling effective auto-tuning in nonlinear systems where conventional fixed-magnitude methods fail
2Device complexity
If the manipulated value magnitude is kept constant during auto-tuning, then the tuning process is simple, but it cannot handle nonlinear controlled objects
Solution Approach 1:
The patent introduces feedback by detecting the nonlinearity degree during the auto-tuning process and using this information to adjust the manipulated value magnitude. The system continuously monitors the response characteristics and feeds back this information to modify the tuning parameters, creating a closed-loop adaptive tuning mechanism that handles nonlinearities effectively
Solution Approach 2:
The patent performs preliminary detection of nonlinearity degree before proceeding with the main auto-tuning process. By first assessing the nonlinear characteristics of the controlled object and then setting an appropriate manipulated value magnitude based on this preliminary information, the system prepares the optimal tuning conditions in advance, avoiding the need for complex real-time adjustments
Data Source
AI summary
A controller having an auto-tuning function has a manipulated value determination unit that selectively determines a first manipulated value for causing a first change in a controlled value of a controlled object or a second manipulated value for causing a second change opposite to the first change in the controlled value, in accordance with a preset parameter, so that a process value obtained from the controlled object follows a target value, and a tuning unit that alternately outputs the first manipulated value and the second manipulated value in accordance with the process value, and that determines the parameter from a response characteristic obtained through the alternate output. The tuning unit successively changes a magnitude of the first manipulated value each time output is switched during the alternate output of the first manipulated value and the second manipulated value.