A Prediction Method for Hydrate Formation Conditions Based on Wavelet Neural Network
Gao Hai-kang · Computer and Modernization · 2013
There are a variety of shortcomings in forecast methods of traditional hydrate formation conditions.Wavelet neural network 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.