一种多目标精细优化的光缆网络路由智能规划方法
By employing a multi-objective, fine-grained optimization intelligent planning method for fiber optic network routes, and utilizing reinforcement learning and collaborative decision-making, the problem of lack of local information in global optimization of submarine cable route planning is solved, thereby improving the safety and economy of submarine cable paths.
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- BEIJING UNIV OF POSTS & TELECOMM
- Filing Date
- 2023-10-31
- Publication Date
- 2026-07-17
AI Technical Summary
Existing submarine cable route planning methods lack local information during global optimization, which makes it impossible to achieve refined planning and decision-making, posing safety risks. Furthermore, methods based on numerical optimization algorithms cannot flexibly adjust the optimization target weights for different marine areas.
A multi-objective fine-optimization intelligent planning method for optical cable network routing is adopted. Through reinforcement learning and collaborative decision-making, multiple agents search and negotiate on their respective target map distributions, avoiding the transformation of multiple targets into a comprehensive distribution map and making optimization decisions directly at the local level.
It achieves fine-grained optimization of local multi-objective submarine cable routes, avoids potential hazards under extreme conditions, improves the safety and economy of submarine cable routes, and meets the specific needs of different marine areas.
Smart Images
Figure CN117610201B_ABST