张欣雨　李　茹,2　高俊杰　王　浩　王晋宇　赵红燕

1(山西大学计算机与信息技术学院　山西 太原 030006)

2(山西大学计算智能与中文信息处理教育部重点实验室　山西 太原 030006)

3(山西煤炭进出口集团科技发展有限公司技术管理处　山西 太原 030006)

4(太原科技大学计算机科学与技术学院　山西 太原 030024)



改进最大Lyapunov指数的多变量瓦斯浓度预测研究

张欣雨1李茹1,2高俊杰1王浩1王晋宇3赵红燕4

1(山西大学计算机与信息技术学院山西 太原 030006)

2(山西大学计算智能与中文信息处理教育部重点实验室山西 太原 030006)

3(山西煤炭进出口集团科技发展有限公司技术管理处山西 太原 030006)

4(太原科技大学计算机科学与技术学院山西 太原 030024)

摘要煤矿井下瓦斯浓度受到多个环境参数的影响。首先对煤矿井下同一时间段的瓦斯浓度、风速、压力、温度、CO2、O2的时间序列数据进行统计学相关性分析，选择出对瓦斯浓度影响大的几种因素，并作为基于最大Lyapunov指数改进预测模型的输入参数来预测未来一段时间的瓦斯浓度。改进算法为：在用C-C方法重构多变量时间序列的相空间和Wolf方法计算最大Lyapunov指数的基础上，同时考虑夹角余弦和欧氏距离求取预测中心点的相近点。结果表明，改进预测算法提高了预测精度，平均绝对误差和标准差分别为2.11%和2.15%。

关键词相关性分析最大Lyapunov指数C-C方法Wolf方法夹角余弦

STUDY ON IMPROVED MULTIVARIATE GAS CONCENTRATION PREDICTION BASED ON LARGEST LYAPUNOV EXPONENT

Zhang Xinyu1Li Ru1,2Gao Junjie1Wang Hao1Wang Jinyu3Zhao Hongyan4

1(School of Computer and Information Technology, Shanxi University, Taiyuan 030006,Shanxi,China)2(Key Laboratory of Computational Intelligence and Chinese Information Processing of Ministry of Education, Shanxi University, Taiyuan 030006,Shanxi,China)3(Shanxi Coal Import and Export Group Co,Ltd, Technology Management Office, Taiyuan 030006,Shanxi,China)4(School of Computer Science and Technology, Taiyuan University of Science and Technology, Taiyuan 030024,Shanxi,China)

AbstractThe gas concentration of coal mine is affected by lots of environmental parameters. First, we made the statistics correlation analysis on gas concentration, wind velocity, pressure, temperature, and the time series data of CO2and O2during the same period in coal mine, and selected some factors having big impacts on gas concentration as the input parameters of the improved prediction model, which was based on the largest Lyapunov exponent, to predict the gas concentration in coming period. The improved algorithm is that based on using C-C method to reconstruct the phase space of multivariate time series and using Wolf method to calculate the largest Lyapunov exponent, the cosine of angle and the Euclidean distance are taken into account at the same time to obtain the close points to the prediction centre point. Results showed that the improved prediction algorithm improved the prediction accuracy, its’ average absolute error and standard deviation were 2.11% and 2.15% respectively.

KeywordsCorrelation analysisLargest Lyapunov exponentC-C methodWolf methodCosine of angle

0引言

我国煤矿事故发生频率相对较高，其中瓦斯灾害是威胁我国煤矿井下安全生产的主要灾害之一。因此对瓦斯有效准确的预测是预防瓦斯灾害的关键。……