基于神经网络和人工鱼群算法的惯性延时微通道优化  

Inertial Delay Microchannel Optimization Design Based on Neural Network and Artificial Fish Swarm Algorithm

作　　者：赵川霆 聂伟荣[1] 袁君鑫 席占稳[1] 曹云[1] ZHAO Chuanting;NIE Weirong;YUAN Junxin;XI Zhanwen;CAO Yun(School of Mechanical Engineering,Nanjing University of Science and Technology,Nanjing 210094,China)

机构地区：[1]南京理工大学机械工程学院,江苏南京210094

出　　处：《探测与控制学报》2024年第3期33-39,44,共8页Journal of Detection & Control

摘　　要：为了保证中、小口径弹引信的炮口安全性控制要求,设计了具有肋板阻尼结构的惯性延时微通道,同时为了保证其具有稳定的延时性能,研究了该微通道在高离心转速下的延时响应性能。采用了神经网络模型和人工鱼群算法对蛇形微通道内肋板的结构位置进行优化设计。用两相流水平集模型以微通道的延迟时间为研究对象进行了模拟仿真,得到180组样本数据,分析发现肋板结构在微通道内的不同位置与延迟时间呈现出高度的非线性关系。根据样本数据建立神经网络模型用以拟合设计变量与优化目标之间的映射函数,并采用人工鱼群算法对神经网络模型拟合的映射函数的参数进行优化。结果表明,经过结构优化之后,在1 000 g离心环境下微通道中流体的延迟时间从最短的11.446 ms提升到了25.054 ms,延时效果得到了显著提升。最后研究了优化后的结构在中、小口径弹引信使用环境下的延时特性,验证了其满足大部分中、小口径弹引信的延时控制要求。In order to ensure the safety control requirements for the use of medium and small caliber ammunition fuses,an inertial delay microchannel was designed with pin fin damping structure.At the same time,in order to ensure its stable delay performance,the delay performance of this microchannel under high centrifugation speed was studied.The neural network model and artificial fish swarm algorithm were adopted to optimize structural position of the pin fins in the serpentine microchannel.A two-phase flow level set model was applied to simulate the delay time of microchannels and 180 sets of sample data were obtained.Analysis showed that different positions of the pin fin structure in the microchannels exhibited a highly nonlinear relationship with the delay time.A neural network model was established based on sample data to fit the mapping function between design variables and optimization objectives,which used artificial fish swarm algorithm to optimize the parameters of mapping function fitted by neural network model.Experimental results showed that after structural optimization,the delay time of the fluid in the microchannel under high-speed centrifugal environment increased from 11.446 ms to 25.054 ms,and the delay effect improved a lot.Finally,delay characteristics of the optimized structure in the application environment of medium and small caliber ammunition fuzes were studied,which indicated that it met the delay control requirements of most medium and small caliber ammunition fuzes.

关 键 词：惯性 阻尼结构 肋板 延时微通道 人工鱼群算法 神经网络 

分 类 号：TJ43[兵器科学与技术—火炮、自动武器与弹药工程]