Servo Predictive Control With Integral Action for Overshoot Control
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Model predictive control in servo systems often results in overshoot and deteriorated transient responses due to steady-state deviations, and the use of disturbance observers increases computational load and design complexity.
Innovation Solution
Incorporating an integral term into the prediction model of the control device, which multiplies a predetermined integral gain with the deviation between the command and output, allowing for effective elimination of steady-state deviations without deteriorating transient responses and reducing computational load by avoiding the use of disturbance observers.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If an integrator is connected to the compensator in series to eliminate steady-state deviation, then steady-state deviation is eliminated, but the output overshoots and transient response deteriorates
Solution Approach 1:
The patent segments the integral action from the main control loop by implementing it as a separate feedforward path. The integrator processes the deviation between command and output independently, and its output is added to the model predictive control output rather than being connected in series. This segmentation allows the integral action to eliminate steady-state deviation without directly affecting the transient response characteristics of the main control loop.
2Measurement precision
If a disturbance observer is constructed to cancel steady-state deviation, then steady-state deviation is eliminated, but the number of state variables increases and computational load increases
Solution Approach 1:
The patent extracts the integral action from the complex disturbance observer framework and implements it as a simple, standalone feedforward integrator. This extracted integral component achieves steady-state deviation elimination through basic integration of the deviation signal, avoiding the need for additional state variables and complex computational algorithms associated with traditional disturbance observers.
3Manufacturing precision
If model predictive control is used to suppress trajectory deviation, then trajectory tracking is improved, but computational load increases
Solution Approach 1:
The patent merges the feedforward integral action with the model predictive control output through simple addition. This combination allows the system to benefit from both the optimal trajectory tracking of model predictive control and the steady-state accuracy of integral action, while avoiding the computational burden of more complex integrated control algorithms.
Data Source
AI summary
The present invention provides a control device that executes model predictive control related to a predetermined control target corresponding to an actual target device which is an actual target of servo control in order to cause an output of the actual target device to track a predetermined command. The control device includes: an integrator which receives input of a deviation between the predetermined command and an output of the predetermined control target; and a model predictive control unit which has a prediction model that defines a correlation between a predetermined state variable and an input to the predetermined control target in a form of a predetermined state equation, and which performs model predictive control based on the prediction model according to a predetermined evaluation function in a prediction section having a predetermined time width using the output of the integrator as an input.