Microwave filter debugging method based on multi-population particle swarm optimization method

Abstract

The invention provides a microwave filter debugging method based on a multi-population particle swarm optimization method, and the method comprises the steps: initializing the speed and position of each particle when the number of initial sub-populations is 1, and calculating the adaptive value of each particle; grading the particles on the basis of the particle fitness values; selecting a speed updating strategy of the particles according to the distribution characteristics of the particle grades, updating the speed and the position of each particle, and recalculating the adaptive value of each particle; determining the number of sub-populations after division and the particle composition of each sub-population through a particle division method and a k-means method; and for each dividedsub-population, repeating particle grading, particle speed and position updating and sub-population division operations until an optimal solution is found. The method has the advantages that the problem that the population number needs to be set manually is solved, a new particle speed updating framework is provided, the calculation speed, the population information utilization rate, the optimization speed and precision are improved, and practicability and applicability are achieved.

Description

technical field [0001] The invention relates to the field of microwave filter debugging, in particular to a microwave filter debugging method based on a multi-swarm particle swarm optimization method. Background technique [0002] Optimization problems are ubiquitous in practical applications. It can be said that where there is a choice, there will be optimization. A general single-objective optimization problem can be described as Among them, x is the n-dimensional optimization variable, f is the optimization objective function, also known as the fitness value function, and x* is the optimal value of the optimization variable. The optimization process is to search for the optimal value x* that minimizes the optimization objective function f in the n-dimensional optimization space. [0003] Microwave filter is a passive frequency selection device, which is widely used in the filter unit of base station and relay station. However, due to the existence of design errors and...

AI Technical Summary

Problems solved by technology

[0004] Most of the optimization methods choose the particle swarm optimization method. At present, there are many methods to improve the particle swarm optimization method. Most of them are to add some correction items or restriction items to the basic update formula of gbest or lbest, and combine the advantages of these two update strategies. There are fewer ways to combine

For the particle swarm optimization algorithm (MS-PSO) with multiple population structures, the number of populations needs to be manually set, and the algorithm is highly subjective and poor in generality.

[0005] The Chinese patent with the publication number CN107153837A and the publication date of September 12, 2017 does not solve the problem that k-means clustering needs to manually set the number of clusters; in addition, the speed update strategy of the invention only weights the two search strategies , and did not maximize the advantages of fast search speed of gbest and high search accuracy of lbest

[0006] The overall architecture of the Chinese patent with the publication number CN109858606A and the publication date of June 07, 2019 is consistent with the gbest architecture, and still faces the problem of being prone to falling into local optimum

Images