摘要
农机总动力的预测可以为农业装备需求规划提供数据参考,以黑龙江省农机总动力数据作为研究对象,应用混沌理论中G-P算法计算农机总动力的关联维数,将求出的关联维数与灰色系统理论中的灰色关联分析法相结合研究其影响因素,建立了农机总动力灰色多变量预测模型GM(1,m).计算结果:应用Matlab软件编程计算得到了农机总动力7个具体的主导影响因素,并建立了农机总动力的灰色多变量GM(1,7)预测模型,模型的平均相对误差为2.14%,由结果对比可知该预测模型的预测结果高于GM(1,1),方法具有一定的预测精度与科学性.
In this paper take the Agricultural Machinery Total Power of Heilongjiang Province as researcher subject. It takes the advantage of chaos theory GP algorithm and gray correlation analysis of gray system theory to study the influencing factors. By using Matlab software programming to calculate correlation dimension and relational degree of total Power of Heilongjiang Province, the calculation results are seven specific factors. Then the multivariate grey model prediction GM(1,7) is constructed by Matlab software programming. The forecasting model relative error is -0.174. The results show that the result of multivariate grey model forecast is higher than the result of the prediction model GM(1,1).
引文
[1]刘静,朱达荣.考虑自变及因变影响的农机总动力组合预测模型[J].农机化研究,2015, 4(4):230-236.
[2]陈宝峰,白人朴,刘广利.影响山西省农机化水平的多因素逐步回归分析[J].中国农业大学学报,2005,10(4):115-118.
[3]鞠金艳,王金武·黑龙江省农业机械化发展的关键影响因素分析[J].农机化研究,2014, 02(2):60-67.
[4]张锋伟,赵武云等.甘肃省农用大中型拖拉机需求量预测模型及时效分析[J].中国农机化,2010(2):27-30.
[5] GRASSERGER P, PROCACCIA I, Measuring the strangeness of strange attractors[J]. Physcia D,1983, 9(1,2):189-208.
[6]李小霞,文晓涛.关联维在砂体储层预测上的应用[J].石油化工应用,2013, 12(12):39-42.
[7] Mehdi Askari, Hadi Askari. Times series grey system prediction—based models gold price forecasting[J]. Trends in Applied Science Research, 2011, 6(11):1287-1292.
[8]戴立勋,魏宏安.甘肃省农业机械化影响因素的量化分析[J].中国农机化,2010(4):24-32.
[9]国家统计局.中国统计年鉴(1994-2013)[M].北京,中国统计出版社,1994-2013.
[10]国家统计局.中国农村统计年鉴(1994-2013)[M].北京,中国统计出版社,1994-2013.
[11]黑龙江省统计局.黑龙江统计年鉴(1994-2013)[M].北京,中国统计出版社,1994-2013.