摘要
将人工神经网络技术引入曲面磨削加工领域,介绍了利用BP算法建立的曲面磨削表面粗糙度随磨削用量变化的进化神经网络预测模型.针对BP算法存在收敛速度慢、容易陷入局部极小值及全局搜索能力弱等缺陷,采用遗传算法训练BP神经网络,取代了一些传统的学习算法,设计了基于进化神经网络的学习算法.实验和仿真结果表明,基于进化计算的BP神经网络不仅可以克服单纯使用BP网络易陷入局部极小等问题,而且预测精度较高.
Artificial neural networks were introduced in the area of curve grinding. The prediction model of surface roughness in curve grinding based on back propagation (BP) algorithm was proposed. There are some disadvantages in BP algorithm, such as low rate of convergence, easily falling into local minimum point and weak global search capability. In order to settle these problems, a genetic algorithm was used to train BP neural network to replace classical learning algorithms. An evolutionary neural network learning algorithm was founded. The results of simulations and experiments show that the evolutionary neural network based genetic algorithm can effectively overcome the problem of falling into local minimum point. This method can get higher accuracy of predictions.
出处
《上海交通大学学报》
EI
CAS
CSCD
北大核心
2005年第3期373-376,共4页
Journal of Shanghai Jiaotong University
基金
上海市科委重点科技资助项目(021111125)
关键词
进化神经网络
遗传算法
曲面磨削
表面粗糙度
预测
Backpropagation
Evolutionary algorithms
Genetic algorithms
Learning algorithms
Neural networks
Predictive control systems
Surface roughness
