Multi-agent system distributed optimization control method and storage medium
A multi-agent system and optimal control technology, applied in complex mathematical operations and other directions, can solve the problems of poor overall performance of network systems and large communication loss in multi-agent network systems, and achieve excellent iteration speed, strong robustness and universality. The effect of adaptability and optimization consistency
Pending Publication Date: 2021-11-30
CHINA UNIV OF GEOSCIENCES (WUHAN)
View PDF0 Cites 0 Cited by
- Summary
- Abstract
- Description
- Claims
- Application Information
AI Technical Summary
Problems solved by technology
[0004]
One of the main problems solved by the present invention is the large communication loss of the multi-agent network system, which leads to poor overall performance of the network system
Method used
the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
View moreImage
Smart Image Click on the blue labels to locate them in the text.
Smart ImageViewing Examples
Examples
Experimental program
Comparison scheme
Effect test
Embodiment 1
[0047] Embodiment one, such as figure 2 As shown, it is a schematic structural diagram of a fully connected multi-intelligence system in this embodiment. The number of agents included in the multi-agent system (hereinafter referred to as the system) is denoted as n, and each independent agent has its own state information. It is stipulated that the i-th The state value of the agent is The state value function is requires that the local state function f i (x) is a convex function and there is an optimal solution to ensure that the algorithm can finally find the optimal value and there is only one optimal value.
[0048] The system communicates through an undirected connected network, which is defined as G={υ,ε}, υ is a non-empty node set, each node represents an agent; ε represents a directed edge set, if two Nodes i and j can communicate, then there is (i, j)∈ε, node j represents the neighbor of i, and the set of neighbors of node i is denoted as N i , where the number ...
the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to View More PUM
Login to View More Abstract
The present invention provides a multi-agent system distributed optimization control method and a storage medium. According to the distributed optimization control method for the multi-agent system provided by the invention, a framework is optimized by using a distributed alternating direction multiplier method, the algorithm has excellent iteration speed and also has strong robustness and universality, and through the design of a communication threshold function, the intelligent agent can communicate only when the local information update is effective enough, so that the high efficiency of communication is ensured, the communication resources consumed by system optimization are greatly reduced, and the setting of the number of iterations also has an influence on the capability of reducing the communication loss of the system. According to the design of an adaptive penalty function, a penalty term of an intelligent agent in each iteration is specifically designed according to different state values of the intelligent agent, so that the convergence speed of the system can be ensured, the system has enough effectiveness in each iteration, and the setting of the maximum number of iterations kmax2 can influence the convergence speed of the system.
Description
technical field [0001] The invention relates to the technical field, in particular to a multi-agent system distributed optimization control method and a storage medium. Background technique [0002] In the existing technology, the distributed alternating direction multiplier algorithm to improve the performance of the multi-agent network system integrates many classic optimization ideas, such as the gradient descent method, the original dual algorithm, etc., and combines the problems encountered in modern statistical learning. A relatively easy-to-implement distributed computing framework, starting from the perspective of the original dual operator, the core is the augmented Lagrangian method of the original dual algorithm, which can solve specific forms of constrained convex optimization problems. The framework can solve many optimization problems and has a good convergence rate, but the communication problem cannot be solved, and the conditions that limit communication are...
Claims
the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to View More Application Information
Patent Timeline
Login to View More IPC IPC(8): G06F17/10G06F17/16G06F17/15
CPCG06F17/10G06F17/16G06F17/15Y02P90/02
Inventor 姜晓伟曹爽张斌李刚
Owner CHINA UNIV OF GEOSCIENCES (WUHAN)