摘要
为了提高固体推进剂燃速预示精度,将遗传算法(Genetic Algorithm)与误差反传(Back Propagation)网络结构模型相结合,设计了用遗传算法优化神经网络权重的新方法。以固体推进剂燃速数据库为基础,对推进剂的燃速进行了预估,并与BP算法进行了比较。结果显示,预估值与实际值接近,误差小于BP算法模型,具有良好的预示效果,为推进剂燃速预估提供了新方法。
In order to improve predicting precision of burning-rate of solid propellant. Combine Genetic Algorithm and Back-Propagation neural network, A fresh method on optimizing biases and weights of neural networks by Genetic Algorithm was demonstrated. Based on the database of solid propellants. Burning rate of solid propellant was predicted and contrasted to BP. The result showed that calculations were corresponding to practice and the errors were smaller than BP. So it supplies a new method for predicting burning rate of solid propellant.
出处
《计算机与应用化学》
CAS
CSCD
北大核心
2006年第7期639-642,共4页
Computers and Applied Chemistry
关键词
推进剂燃速预估
遗传算法
神经网络
predicting of burning rate of solid propellant, genetic algorithm, neural network
