摘要
矿井瓦斯涌出预测是建立一个复杂的非线性变化系统的过程,传统的瓦斯涌出量预测方法存在一定的局限性。瓦斯涌出量的预测即是建立函数关系模型,预测模型建立的准确与否取决于各个影响因素之间的相互作用、相互耦合的特性。文中将灰色理论与遗传算法有机地结合起来,以灰色理论为基础,利用遗传算法优化隐含层神经元个数和网络中的连接权值,建立瓦斯涌出量预测模型。预测实验结果表明,该方法对瓦斯涌出量预测适应性强、方法快捷、预测结果精度高。
Mine gas emission system is a complex system and is also nonlinear change one.The traditional methods for the gas emission prediction have a certain limitations.Forecasting the gas emission depends on an establishment of a nonliner functional relation of many factors;the accuracy of the forecasting model for the gas emission is determined by the peculiarities of the in teraction and coupling between all the affecting factors.This paper combines neural networks with genetic algorithm;on the basis of grey model,and applying genetic neural network to optimize the construction and the power size of grey model,a forecasting model of gas emission is established.A predicting model with GM(GreyModel) and Genetic Neural Network was established by introducing grey theory into genetic neural network with high precision of forecast.Both of the subsequent training and the examination show that satisfactory forecasting results are obtained by using this method,which can meet with the requirement for an exact guidance to the practice.
出处
《微计算机信息》
2011年第6期192-194,共3页
Control & Automation
关键词
非线性特征
灰色理论
遗传神经网络
瓦斯涌出量
nonlinear characteristics
greymodel
genetic neural network
mine gas emission