Configurable intelligent optimization method based on double-layer automatic generation strategy

An automatic generation and intelligent optimization technology, applied in artificial life, biological models, instruments, etc., can solve problems such as difficult to solve efficiently, mixed design of intelligent optimization methods is not flexible enough to solve, etc., to enhance dynamic and robustness, Ensure global optimization efficiency and efficient solution effect
CN113240159APending Publication Date: 2021-08-10BEIHANG UNIV +1
0 Cites 0 Cited by

Patent Information

Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
BEIHANG UNIV
Publication Date
2021-08-10

Smart Images

  • Figure 1
    Figure 1
  • Figure 2
    Figure 2
  • Figure 3
    Figure 3
Patent Text Reader

Abstract

A configurable intelligent optimization method based on a double-layer automatic generation strategy combines an operator configuration method in a serial intelligent optimization method and a topological structure configuration mode in a parallel intelligent optimization method, and divides sub-populations in an evolutionary iteration process through the double-layer automatic generation strategy. Self-adaptive communication topology selection is realized through automatic operator configuration and inter-group interaction, synchronous search of a plurality of processes or threads is formed, and diffusion and application of population overall information are realized in a periodic process interaction mode, so that the global optimization efficiency is ensured under the condition of sub-population dispersion independent solution, the dynamic nature and robustness of the intelligent optimization method in large-scale solution space search are enhanced, complex problem decomposition and optimization are more flexible, and more space is provided for hybrid design and solution of the intelligent optimization method.
Need to check novelty before this filing date? Find Prior Art

Description

technical field

[0001] The invention relates to a configurable intelligent optimization method based on a double-layer automatic generation strategy, which belongs to the field of parallel computing of complex optimization problems. Background technique

[0002] Since the early 1990s, in order to speed up the iterative process of intelligent optimization methods, researchers have designed a population-based parallel framework according to its characteristics of population-based iteration. Under this framework, the population of methods is divided into multiple sub-populations, and each sub-population is carried by different computing cores, and is optimized in parallel with the same evolutionary mechanism in the solution space. The intelligent optimization mode of the population can be divided into master-slave parallel mode, coarse-grained parallel mode and fine-grained parallel mode. In the fine-grained mode, the population is divided into several very small subgroups, wh...

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