摘要
为提高粮食产量预测的精度,针对LSSVM模型的预测精度受惩罚参数C和核函数参数g选择的影响,将非线性惯性权重引入萤火虫算法,提出一种基于自适应权重的萤火虫算法(Self-Adaptive Firefly Algorithm,SAFA),并将SAFA应用于惩罚参数C和核函数参数g优化,提出一种基于SAFA-LSSVM的粮食产量预测算法。选择1978—2017年我国粮食产量数据为研究对象,与FA-LSSVM、PSO-LSSVM和LSSVM相比,研究结果表明本文提出的算法SAFA-LSSVM可以有效提高粮食产量预测的精度,相关系数R达0.9893,为粮食产量预测提供新的方法和途径。
In order to improve the precision of grain yield prediction, the nonlinear inertial weight is introduced into the Firefly algorithm for the influence of selecting the parameter C and the kernel function parameter g of the LSSVM model, and an adaptive weight based Firefly algorithm is proposed was applied to optimize the penalty parameter C and the kernel function parameter g. A grain yield prediction algorithm based on SAFA-LSSVM was proposed. China's grain yield data from 1978 to 2017 were selected as the research object, Compared with FA-LSSVM, PSO-LSSVM and LSSVM, the results show that the proposed algorithm SAFA-LSSVM can effectively improve the accuracy of grain yield prediction, R was 0.989 3, and provide new methods and approaches for grain yield prediction.
引文
[1] 孙东升, 梁仕莹. 我国粮食产量预测的时间序列模型与应用研究[J]. 农业技术经济, 2010(3): 97-106.Sun Dongsheng, Liang Shiying. Research on China's grain output forecast time series model and application [J]. Journal of Agrotechnical Economics, 2010(3): 97-106.
[2] 韩书成, 李丹, 熊建华, 等. 广州市耕地资源数量变化及其对粮食安全的影响[J]. 农林经济管理学报, 2016, 15(6): 648-654.Han Shucheng, Li Dan, Xiong Jianhua, et al. Changes in cultivated land amount and their impacts on food security in Guangzhou [J]. Journal of Agro-Forestry Economics and Management, 2016, 15(6): 648-654.
[3] 朱迎春, 周志刚, 王大鹏. 江苏省农机化水平与影响因素灰色关联分析[J]. 中国农机化学报, 2009, 30(4): 12-14.Zhu Yingchun, Zhou Zhigang, Wang Dapeng. Correlative degree analysis between the level of agricultural mechanization and influencing factors in Jiangsu Province [J]. Journal of Chinese Agricultural Mechanization, 2009, 30(4): 12-14.
[4] Fei S W, Miao Y B, Liu C L. Chinese grain production forecasting method based on particle swarm optimization-based support vector machine [J]. Recent Patents on Engineering, 2009, 3(1): 8-12.
[5] 张宇青, 易中懿, 周应恒. 一种线性ARIMA基础上的非线性BP神经网络修正组合方法在粮食产量预测中的运用[J]. 数学的实践与认识, 2013, 43(22): 135-142.Zhang Yuqing, Yi Zhongyi, Zhou Yingheng. Forecast of grain output using a method of nonlinear BP neural network based on linear ARIMA modified [J]. Mathematics in Practice and Theory, 2013, 43(22): 135-142.
[6] 苏博, 刘鲁, 杨方廷. GM(1,N)灰色系统与BP神经网络方法的粮食产量预测比较研究[J]. 中国农业大学学报, 2006, 11(4): 99-104.Su Bo, Liu Lu, Yang Fangting. Comparison and research of grain production forecasting with methods of GM(1,N) gray system and BPNN [J]. Journal of China Agricultural University, 2006, 11(4): 99-104.
[7] 罗利娟, 丁宏飞. 基于PSO-SVR的粮食产量预测模型[J]. 统计与决策, 2010(22): 37-38.Luo Lijuan, Ding Hongfei. Grain production forecasting model based on PSO-SVR [J]. Statistics and Decision, 2010(22): 37-38.
[8] Suykens J A K, Gestel T V, Brabanter J D, et al. Least squares support vector machines [J]. International Journal of Circuit Theory & Applications, 2002, 27(6): 605-615.
[9] Suykens J A K, Vandewalle J. Least squares support vector machine classifiers [J]. Neural Processing Letters, 1999, 9(3): 293-300.
[10] Baesens B, Viaene S, Gestel T V, et al. Least squares support vector machine classifiers: an empirical evaluation [J]. Access & Download Statistics, 2000: 1-16.
[11] 冯艳红, 刘建芹, 贺毅朝. 基于混沌理论的动态种群萤火虫算法[J]. 计算机应用, 2013, 33(3): 796-799.Feng Yanhong, LiuJianqin, He Yichao. Chaos-based dynamic population firefly algorithm [J]. Journal of Computer Applications, 2013, 33(3): 796-799.
[12] 刘佳昆, 周永权. 一种最大最小萤光素值人工萤火虫算法[J]. 计算机应用研究, 2011, 28(10): 3662-3664.Liu Jiakun, Zhou Yongquan. Glowworm swarm optimization algorithm based on max-min luciferin [J]. Application Research of Computers, 2011, 28(10): 3662-3664.
[13] 秦嵩, 赵道致, 王震蕾. 影响粮食产量因素的组合预测和经济分析[J]. 中国农机化学报, 2009, 30(2): 58-61.Qin Song, Zhao Daozhi, Wang Zhenlei. Economical analysis on factors related to the output of rice [J]. Journal of Chinese Agricultural Mechanization, 2009, 30(2): 58-61.
[14] 张浩, 王国伟, 苑超, 等. 基于AIGA—BP神经网络的粮食产量预测研究[J]. 中国农机化学报, 2016, 37(6): 22-27.Zhang Hao, Wang Guowei, Yuan Chao, et al. Research on forecast of grain production based on AIGA—BP neural network [J]. Journal of Chinese Agricultural Mechanization, 2016, 37(6): 22-27.
[15] 潘永明, 米冠旭. 基于主成分分析法的天津现代物流发展综合评价模型构建[J]. 中国农机化学报, 2009, 30(4): 45-47.Pan Yongming, Mi Guanxu. Model of the modern logistics developmental of Tianjin integrated evaluation based on PCA [J]. Journal of Chinese Agricultural Mechanization, 2009, 30(4): 45-47.
[16] Yuefeng Y, Xuerong X. Principal component analysis and GM(1,1) forecast of grain output influencing factors in Fujian Province [J]. Journal of Southern Agriculture, 2014, 45(4): 697-703.
[17] 魏津瑜, 陈锐, 刘曰波. 影响我国粮食产量的因素分析及对策研究[J]. 中国农机化学报, 2008, 29(5): 56-59.Wei Jinyu, Chen Rui, Liu Yuebo. Analysis of effecting factors on China's grain yield and countermeasure [J]. Journal of Chinese Agricultural Mechanization, 2008, 29(5): 56-59.