煤矿瓦斯涌出量的非线性降维Elman动态预测模型  

Nonlinear dimension reduction and improved Elman dynamic prediction model of coal mine gas emission

作　　者：魏林[1] 付华[2] 尹玉萍[2] 

机构地区：[1]辽宁工程技术大学基础教学部,辽宁葫芦岛125105 [2]辽宁工程技术大学电气与控制工程学院,辽宁葫芦岛125105

出　　处：《辽宁工程技术大学学报（自然科学版）》2017年第4期359-365,共7页Journal of Liaoning Technical University (Natural Science)

基　　金：国家自然科学基金项目(51274118)

摘　　要：为对回采工作面绝对瓦斯涌出量进行有效预测,提出非线性降维的改进Elmand动态预测模型.模型采用非线性映射在特征空间内对数据进行有效降维,以此确定神经网络输入数目,并利用自适应蚁群微分进化算法对改进的Elman神经网络(IENN)的阈值、权值、自反馈因子和增益因子行全局寻优.将该预测模型用于矿井监测的历史数据进行检验.研究结果表明:模型能够有效地减少预测模型的输入变量个数,并且相比其他预测模型提高了预测的精度和效率.In order to achieve more effective predicted results for the absolute gas emission quantity, this paper put forward gas emission dynamic prediction model based on the nonlinear dimension reduction and the improved Elman. This model uses the nonlinear mapping in the feature space to reduce data dimension effectively and to determine the input numbers of neural network. For the purpose of achieving the optimal parameters of the improved Elman neural network（IENN）, this paper used the adaptive ant colony-differential evolution algorithm（ACDE）. With the historical data of mine actual monitoring to experiment and analysis, the results show that this model can effectively reduce the numbers of input variables, and compared with other prediction models this model improves the forecast accuracy and efficiency.

关 键 词：绝对瓦斯涌出量 非线性映射 蚁群算法 微分进化算法 Elman神经网络 

分 类 号：TD712[矿业工程—矿井通风与安全]