摘要
考虑到冲击地压时序的特点,采用基于免疫进化规划的进化神经网络进行了冲击地压非线性系统的建模研究。并采用一个矿山实测得到的震级数列进行了进化神经网络方法的实用性验证,结果表明,进化神经网络不但模型拟合精度高,而且预测性能也较好。
Considering the dynamic character of rock burst, a time serial model of rock burst is constructed by evolutionary neural network based on immunized evolutionary program. The proposed model is tested by a real magnitude series from a mine, and results show that evolutionary neural network model has not only high precision, but also high prediction.
出处
《爆炸与冲击》
EI
CAS
CSCD
北大核心
2004年第6期524-528,共5页
Explosion and Shock Waves
基金
国家自然科学基金项目(50309041)
关键词
爆炸力学
进化神经网络
建模
冲击地压
时序
mechanics of explosion
evolutionary neural network
construct model
rock burst
time serial
