一种基于双空间知识转移与融合的动态多目标优化方法
By constructing a knowledge base for decision and objective spaces and using KD Tree and inverse distance weighted fusion, high-quality candidate solutions are generated, which solves the problem of slow response speed of dynamic multi-objective optimization methods in human-machine joint assembly systems, and enables rapid adaptation to environmental changes and efficient task allocation.
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- GUIZHOU UNIV
- Filing Date
- 2026-06-16
- Publication Date
- 2026-07-17
AI Technical Summary
Existing dynamic multi-objective optimization methods cannot fully utilize the complementary information between the decision space and the target space in human-machine collaborative assembly systems, resulting in slow algorithm response speed after environmental changes, and the convergence and diversity of solutions are difficult to meet actual production needs.
A knowledge base is constructed for the decision space and the goal space, storing the statistical characteristics of elite populations in the historical environment. A KD Tree is used to retrieve similar historical samples, and a weighted fusion is performed using inverse distance weights to generate candidate solutions for the decision and goal spaces. The optimal individual is selected by combining non-dominated ranking and crowding distance to form a new initial population.
It can quickly respond to environmental changes, generate high-quality initial populations, shorten convergence time, ensure solution diversity, improve the operational efficiency of human-machine collaborative assembly systems, rationally allocate worker workloads, and meet actual production needs.
Smart Images
Figure CN122413342A_ABST