Distributed Consensus Control Under Asymmetrical Network Delays
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
In distributed control systems, existing consensus control protocols fail to achieve stable convergence of state index values to a consensus value due to asymmetrical communication delays, leading to deviations and deteriorated convergence performance.
Innovation Solution
A new control protocol that compensates for asymmetrical communication delays by using the difference between the latest state index values of adjacent devices and the self-device, with a control gain adjustment based on communication delay times, ensuring the state index values converge to the average of their initial values.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If conventional consensus control protocols are used in distributed control systems, then the system can operate with multiple devices connected via network, but the state index values fail to converge to a stable consensus value due to asymmetrical communication delays
Solution Approach 1:
The patent applies preliminary action by having each control unit preemptively compensate for communication delays before they affect consensus accuracy. Specifically, control units calculate and apply delay compensation values based on their local clock time and predetermined delay characteristics, allowing the system to proactively counteract the effects of asymmetrical delays before they degrade convergence stability
Solution Approach 2:
The patent implements feedback mechanisms where control units continuously monitor their local clock time, compare it with received state index values from neighboring units, and adjust their state index values based on detected time differences. This closed-loop feedback ensures that asymmetrical communication delays are dynamically compensated, maintaining reliable convergence to consensus values
2Adaptability or versatility
If asymmetrical communication delays occur in the network, then devices can be located at spaced apart positions, but the consensus value deviates from the average of initial state index values and convergence performance deteriorates
Solution Approach 1:
The patent applies local quality by allowing each control unit to independently determine and apply its own delay compensation value based on its local clock time and specific communication characteristics. Instead of requiring a centralized timing mechanism, each unit customizes its compensation strategy to its local conditions, enabling devices to be placed flexibly at various positions while maintaining consensus accuracy through locally-adapted delay compensation
3Ease of operation
If state index values are transmitted between control units via communication network, then devices can be remotely controlled and monitored, but time delays in transmission cause the latest state index value to be used incorrectly in consensus calculations
Solution Approach 1:
The patent applies preliminary action by having control units prepare and apply delay compensation values before performing consensus calculations. Each control unit calculates its local clock time difference and determines the appropriate compensation value in advance, ensuring that when state index values are received from remote devices, the timing inaccuracies have already been corrected, preserving the integrity of the consensus calculation
Solution Approach 2:
The patent introduces the local clock time difference as an intermediary element that mediates between the transmitted state index values and the consensus calculation process. By using the clock time difference as a compensating factor, the system bridges the gap caused by transmission delays, allowing remote control capabilities to function accurately despite the time lag in information transmission
Data Source
Figure 1A~1D
Figure 2A~2C
Figure 3~4
AI summary
A distributed control system includes a plurality of control units that controls respective units of a plurality of devices, and a communication network that includes a communication line that connects the control units. A control unit of each of the devices is configured to determine a state target value, according to consensus control of a multi-agent system, by using a current state index value of a self-device, and a difference between a latest state index value of an adjacent device received from a control unit of the adjacent device, and a latest state index value of the self-device received from the control unit of the adjacent device.