摘要
油漆沉积率模型是自动编程工艺参数选取的重要依据,为了建立符合实际工况的漆膜模型,采用喷涂机器人喷涂时椭圆型雾锥的实验数据,将贝叶斯归一化神经网络法和遗传算法分别用于漆膜模型的拟合。经过对比分析,采用2种算法得出模型都具有较高的精度,但遗传算法收敛速度更快,并可得出油漆沉积率方程的具体表达式,更适合油漆沉积率建模。
Deposition rate model of air spray painting is important for determining technological parameters in automatic trajectory programming.Based on the experimental datas of elliptical paint pattern,and in order to build the film model of realistic operating conditions,the deposition rate model is fitted by using the Bayes normalization algorithm and genetic algorithm respectively.The result shows that all the two models have the high precision.However,compared with Bayes normalization algorithm,the genetic algorithm converges faster and can obtain a definite function expression of the paint deposition rate model.So genetic algorithm is better than Bayes normalization algorithm in modeling deposition rate.
出处
《表面技术》
EI
CAS
CSCD
北大核心
2010年第3期29-33,共5页
Surface Technology
基金
国家自然科学基金(50765001)
广西教育厅科研基金(200708MS028)
广西科技厅科技创新能力基金(桂科能0842006_012_Z)
关键词
遗传算法
贝叶斯归一化法
油漆沉积率模型
优化建模
genetic algorithm
Bayes normalization algorithm
deposition rate model
optimization modeling
