基于BP与SA-GA的飞机发动机风扇叶片清洗参数优化  

OPTIMIZATION OF AIRCRAFT ENGINE FAN BLADE CLEANING PARAMETERS BASED ON BP AND SA-GA

作　　者：牛国臣 朱通 Niu Guochen;Zhu Tong(College of Electronic Information and Automation,Civil Aviation University of China,Tianjin 300300,China)

机构地区：[1]中国民航大学电子信息与自动化学院,天津300300

出　　处：《计算机应用与软件》2021年第1期76-81,110,共7页Computer Applications and Software

基　　金：第八届波音基金项目(20180159204)。

摘　　要：针对目前飞机发动机风扇叶片清洗存在工人劳动强度大等问题,提出采用超声波清洗代替人工清洗,进而对风扇叶片清洗参数进行优化。通过飞机发动机风扇叶片清洗效果影响因素分析建立BP神经网络模型;对模型进行训练和测试;采用模拟退火算法与改进遗传算法相结合的方式进行全局寻优。对比优化分析得到的清洗优化参数进行清洗实验,使用图像检测方式,提出一种飞机发动机风扇叶片清洁度计算算法。实验数据充分表明,采用混合算法优化后清洁度能够达到92%以上,有效提高了清洗效果,为飞机发动机风扇片叶的自动化清洗提供了指导。Ultrasonic cleaning is proposed to optimize the fan blade cleaning parameters regarding to the problems such as labor intensity of workers in the current labor engine fan blade cleaning,and we can optimize the fan blade cleaning parameters.The BP neural network model was established by analyzing the factors affecting the cleaning effect of aircraft engine fan blades;the model was trained and tested;through the combination of simulated annealing algorithm and improved genetic algorithm,the global optimization was obtained.Using image detection to compare the cleaning optimization parameters for cleaning experiments,an algorithm for calculating the cleanliness of aircraft engine fan blades was proposed.The experimental data fully shows that the cleanliness can be more than 92%after optimization by the hybrid algorithm,which effectively improves the cleaning effect and provides guidance for the automatic cleaning of the aircraft engine fan blades.

关 键 词：超声波清洗 清洁度 正交试验 BP神经网络 改进遗传算法 模拟退火算法 

分 类 号：TP18[自动化与计算机技术—控制理论与控制工程]