投影寻踪在瓦斯涌出量预测中的应用
发布时间:2018-08-09 13:44
【摘要】:针对瓦斯涌出量预测时常用的瓦斯监测数据降维方法会不同程度地造成数据信息丢失、导致预测精度降低的问题,利用投影寻踪原理并结合差分进化算法将高维样本数据转化成1维投影数据,运用Matlab曲线拟合工具箱建立了一种正弦和函数瓦斯涌出量预测模型。实验结果表明,该模型具有较高的预测精度和可操作性。
[Abstract]:Aiming at the problem of reducing the dimension of gas monitoring data, which is commonly used in gas emission prediction, it will cause the loss of data information to some extent and lead to the decrease of prediction accuracy. Based on projection pursuit principle and differential evolution algorithm, a prediction model of sinusoidal and functional gas emission is established by using Matlab curve fitting toolbox to transform high-dimensional sample data into 1-dimensional projection data. The experimental results show that the model has high prediction accuracy and maneuverability.
【作者单位】: 兰州石化职业技术学院信息处理与控制工程系;兰州石化职业技术学院石油化学工程系;
【基金】:甘肃省科技厅资助项目(1204GKCA004) 甘肃省财政厅专项资金资助项目(甘财教〔2013〕116号)
【分类号】:TD712.5
[Abstract]:Aiming at the problem of reducing the dimension of gas monitoring data, which is commonly used in gas emission prediction, it will cause the loss of data information to some extent and lead to the decrease of prediction accuracy. Based on projection pursuit principle and differential evolution algorithm, a prediction model of sinusoidal and functional gas emission is established by using Matlab curve fitting toolbox to transform high-dimensional sample data into 1-dimensional projection data. The experimental results show that the model has high prediction accuracy and maneuverability.
【作者单位】: 兰州石化职业技术学院信息处理与控制工程系;兰州石化职业技术学院石油化学工程系;
【基金】:甘肃省科技厅资助项目(1204GKCA004) 甘肃省财政厅专项资金资助项目(甘财教〔2013〕116号)
【分类号】:TD712.5
【参考文献】
相关期刊论文 前8条
1 王磊;周玉文;;基于投影寻踪的城市排水系统洪涝风险评估模型[J];中国给水排水;2011年23期
2 章立清;秦玉金;姜文忠;井庆贺;赵光明;;我国矿井瓦斯涌出量预测方法研究现状及展望[J];煤矿安全;2007年08期
3 蔡武;袁莎莎;戚洋;;基于多元线性回归分析法预测矿井瓦斯涌出量[J];煤矿安全;2010年01期
4 朱红青;常文杰;张彬;;回采工作面瓦斯涌出BP神经网络分源预测模型及应用[J];煤炭学报;2007年05期
5 范作鹏;李启月;徐敏;王卫华;;基于投影寻踪自回归模型的一次爆破成井效果预测研究[J];矿冶工程;2013年02期
6 王江荣;;基于遗传算法模糊多元线性回归分析的瓦斯涌出量预测模型[J];工矿自动化;2013年12期
7 王江荣;王s,
本文编号:2174254
本文链接:https://www.wllwen.com/kejilunwen/anquangongcheng/2174254.html