Parallel Device Control Using Averaged Integral Terms
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Control systems for parallel operating devices, such as mechanically coupled motors or electric power converters, face challenges in maintaining accurate control of a common operating quantity due to infinite gain at zero frequency in PI controllers, leading to excessive control force usage and inaccuracies, particularly in master-slave configurations and drooping methods.
Innovation Solution
Implementing a control system where each controller maintains a device-specific integral term and computes an arithmetic average of its own and other controllers' integral terms to correct deviations, allowing for dynamic role-shifting without human intervention and maintaining accuracy by avoiding unwanted drifts.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If a PI controller is used to control the operating quantity, then the controller can eliminate steady-state error, but the infinite gain at zero frequency causes excessive control force usage and instability in parallel operating devices
Solution Approach 1:
The patent modifies the integral term computation by introducing a stabilization factor that changes the integration parameter. Instead of using the standard integral term that accumulates indefinitely, the stabilized integral term applies a damping factor (e.g., 0.95) to reduce the accumulation rate, thereby preventing infinite gain while maintaining steady-state error elimination capability.
Solution Approach 2:
The patent implements a feedback mechanism where each controller receives and processes the stabilized integral terms from other controllers. This feedback loop allows controllers to adjust their control force based on the collective state of the parallel system, preventing any single controller from accumulating excessive control force that would cause instability.
2Device complexity
If a master-slave configuration is used to control parallel operating devices, then control stability can be improved, but the system complexity increases and dynamic role-shifting becomes difficult
Solution Approach 1:
The patent makes all controllers universal by giving each controller the same capability to compute and process stabilized integral terms. Unlike master-slave configurations where only the master computes integral terms, this approach enables any controller to perform the stabilization function, allowing dynamic role-shifting without reconfiguration when units are added or removed from the parallel system.
Solution Approach 2:
The patent introduces dynamic adaptability by allowing controllers to dynamically adjust their integral terms based on real-time feedback from the system state. The stabilization factor enables the system to dynamically balance control forces among parallel devices, and controllers can dynamically assume leadership roles without manual intervention when system conditions change.
3Device complexity
If drooping is applied to reduce control force usage, then stability in parallel systems improves, but control precision decreases due to the intentional error introduction
Solution Approach 1:
The patent changes the integral term parameter by applying a stabilization factor that reduces the accumulation rate without introducing the intentional error that drooping does. This parameter modification maintains control precision by accurately tracking the actual operating quantity while preventing excessive control force buildup, achieving stability without sacrificing precision.
Data Source
Figure 1a
Figure 1b
Figure 1c
AI summary
A device-system comprises parallel operating devices (105-107) for driving an operating quantity towards a target value, and a control system for controlling each device at least partly based on a device-specific integral term relating to a time integral of a device-specific error signal that is indicative of a deviation of the operating quantity from the target value. The control system comprises a stabilizing system that computes an arithmetic average of the device-specific integral terms and corrects the device-specific integral terms towards the computed arithmetic average. The correction of the device-specific integral terms makes it possible to avoid unwanted drifts in the device-specific integral terms in a situation where there are differences between the device-specific error signals. The devices can be peers to each other and thus redundancy is achieved because one device can be removed from or added to the device-system without actions from the other devices.