基于动态无偏灰色马尔科夫模型的煤炭需求量预测
|
摘要
煤炭消费系统是一个复杂的非线性系统,具有随机性、非线性、动态性等特点,科学地预测煤炭消费量及其结构对于优化配置能源有重要意义.传统的单一预测方法预测精度较低.在对陕西省煤炭消费历史数据分析的基础上,构建了煤炭消费总量的GM(1,1)模型、灰色GM(1,1)预测模型和动态无偏的马尔科夫结构预测模型.拟合结果表明,动态无偏灰色马尔科夫模型既能消除传统灰色GM(1,1)模型的固有偏差,又能提高预测精度,其平均相对误差为2.10%,分别低于传统灰色GM(1,1)模型和一般灰色马尔科夫模型的17.37%和6.37%,可用于煤炭需求的预测,在此基础上对陕西省2017-2025年煤炭消费进行了预测,为未来能源消费发展规划提供依据.
Coal consumption system is a complex nonlinear system with characteristics of randomness,non-linearity and dynamics.It is important to predict coal consumption and its structure for scientifically optimizing energy allocation.The traditional single prediction method has low prediction accuracy.Based on the analysis of the historical data of coal con sumption in Sha nxi province,this paper constructs the GM(1,1)model,the grey GM(1,1)model and the dynamic unbiased Markov structure prediction model.The results show that th e dyna mic unbiased grey Markov model can not only eliminate the inherent deviation of the traditional grey GM(1,1)model,but also improve the prediction accuracy.The average relative error is 2.10%,which is lower than 17.37% and 6.37% of the traditional grey GM(1,1)model and the general grey Markov model respectively.It can be used to forecast the coal dem and,On this basis,the coal consumption in Shanxi province from 2017 to 2025 is forecasted,which providing a basis for future energy consumption development planning.
Coal consumption system is a complex nonlinear system with characteristics of randomness,non-linearity and dynamics.It is important to predict coal consumption and its structure for scientifically optimizing energy allocation.The traditional single prediction method has low prediction accuracy.Based on the analysis of the historical data of coal con sumption in Sha nxi province,this paper constructs the GM(1,1)model,the grey GM(1,1)model and the dynamic unbiased Markov structure prediction model.The results show that th e dyna mic unbiased grey Markov model can not only eliminate the inherent deviation of the traditional grey GM(1,1)model,but also improve the prediction accuracy.The average relative error is 2.10%,which is lower than 17.37% and 6.37% of the traditional grey GM(1,1)model and the general grey Markov model respectively.It can be used to forecast the coal dem and,On this basis,the coal consumption in Shanxi province from 2017 to 2025 is forecasted,which providing a basis for future energy consumption development planning.
引文
[1]张金锁,冯雪,邹绍辉.基于趋势组合的我国煤炭需求预测模型研究[J].商业研究,2014(6):51-56.
[2]杨文清.我国煤炭需求组合预测模型及应用[J].西安科技大学学报,2014,34(5):553-557.
[3]索瑞霞,王福林.组合预测模型在能源消费预测中的应用[J].数学的实践与认识,2010,40(18):80-85.
[4]朱瑞祥,黄玉祥,杨晓辉.用灰色神经网络组合模型预测农机总动力发展[J].农业工程学报,2006,22(2):107-110.
[5]卢奇,顾培亮,邱世明.组合预测模型在我国能源消费系统的构建及应用[J].系统工程理论与实践,2003(3):24-30.
[6]陈华友.组合预测权系数确定的一种合作对策方法[J].预测,2003,22(1):75-77.
[7]文炳洲,索瑞霞.基于组合模型的我国能源需求预测[J],数学的实践与认识,2016,46(20)45:53.
[8]杨珊,陈建宏,郭宏斌.基于无偏灰色模型的煤矿百万吨死亡率预测[J].中国安全科学学报,2011,21(9):22-27.
[9]Nai ming Xie,Chao qing Yuan,Ying jie Yang.Forecasting China's energy demand and self-sufficiency rate by grey forecasting model and Markov model[J].International Journal of Electrical Power&Energy Systems,2015(3):1-8.
[10]杨锦伟,孙宝磊.基于灰色马尔科夫模型的平顶山市空气污染物浓度预测[J].数学的实践与认识,2014,44(2):64-70.
[11]邓聚龙.灰色理论基础[M].武汉:华中科技大学出版社,2002.
[12]刘思峰.灰色系统理论及其应用[M].北京:科学出版社,2008.
[2]杨文清.我国煤炭需求组合预测模型及应用[J].西安科技大学学报,2014,34(5):553-557.
[3]索瑞霞,王福林.组合预测模型在能源消费预测中的应用[J].数学的实践与认识,2010,40(18):80-85.
[4]朱瑞祥,黄玉祥,杨晓辉.用灰色神经网络组合模型预测农机总动力发展[J].农业工程学报,2006,22(2):107-110.
[5]卢奇,顾培亮,邱世明.组合预测模型在我国能源消费系统的构建及应用[J].系统工程理论与实践,2003(3):24-30.
[6]陈华友.组合预测权系数确定的一种合作对策方法[J].预测,2003,22(1):75-77.
[7]文炳洲,索瑞霞.基于组合模型的我国能源需求预测[J],数学的实践与认识,2016,46(20)45:53.
[8]杨珊,陈建宏,郭宏斌.基于无偏灰色模型的煤矿百万吨死亡率预测[J].中国安全科学学报,2011,21(9):22-27.
[9]Nai ming Xie,Chao qing Yuan,Ying jie Yang.Forecasting China's energy demand and self-sufficiency rate by grey forecasting model and Markov model[J].International Journal of Electrical Power&Energy Systems,2015(3):1-8.
[10]杨锦伟,孙宝磊.基于灰色马尔科夫模型的平顶山市空气污染物浓度预测[J].数学的实践与认识,2014,44(2):64-70.
[11]邓聚龙.灰色理论基础[M].武汉:华中科技大学出版社,2002.
[12]刘思峰.灰色系统理论及其应用[M].北京:科学出版社,2008.