基于小波神经网络的水合物形成条件预测方法  

A Prediction Method for Hydrate Formation Conditions Based on Wavelet Neural Network

作　　者：唐永红[1] 王超[1] 龚安[1] 徐加放[1] 高海康[1] 

机构地区：[1]中国石油大学(华东),山东青岛266580

出　　处：《计算机与现代化》2013年第6期5-8,共4页Computer and Modernization

基　　金：山东省自然科学基金资助项目(ZR2012EEM020)

摘　　要：传统的水合物形成条件预测方法都存在各种缺点,而小波神经网络预测水合物形成条件的精度比较高,利于推广。针对水合物形成条件预测值之间相对差距较大,本文提出群体最大误差比率代表机制来改进小波神经网络的学习方式。实验结果表明,该算法有效可行,预测准确度高。There are a variety of shortcomings in forecast methods of traditional hydrate formation conditions. Wavelet neural net-work prediction of hydrate formation conditions is of high precision and is conducive to the promotion. For the relative gap between the predictive values of hydrate formation conditions being larger, the groups maximum error ratio representation mechanism is presented to improve wavelet neural network learning style. The experimental results show that the algorithm is feasible and effective to predict with high accuracy.

关 键 词：预测 水合物形成条件 小波神经网络 学习方式 

分 类 号：TP301.6[自动化与计算机技术—计算机系统结构]