摘要
为了克服传统BP网络收敛速度慢,隐含层节点数不确定等缺点,将免疫算法与BP网络理论相结合,提出了应用免疫神经网络建立固井质量预测模型。在免疫神经网络算法实现中,增添了抗体浓度进行免疫调节,提高了群体的多样性。仿真结果表明,方法比BP网络建立的模型具有更短的训练时间和更高的预测精度,能够提高固井质量,实现固井质量的预测和跟踪分析,对固井中各种未知信息的预测有着较好的适用性,为固井质量预测提供了一种新方法。
In order to overcome the shortcomings of traditional BP network, such as slow convergence, uncertain node number in the hidden layer and so on, an immune algorithm is integrated with BP network theory, and it is put forward that the prediction model of cementing quality is built by using immune neural network. In the immune neural network algorithm, antibody thickness is added to carry out immune adjustment and increase the group diversity. The simulation results show that this method has shorter training time and higher prediction accuracy than the BP network model, and it can improve cementing quality and realize prediction and tracking analysis of cementing quality. It has good serviceability for predicting all kinds of information not known in cementing. It has provided a new method for cementing quality prediction.
出处
《计算机仿真》
CSCD
北大核心
2009年第7期267-269,324,共4页
Computer Simulation
关键词
网络
免疫算法
免疫神经网络
固井质量
Network
Immune algorithm
Immune neural network
Cementing quality
