遗传算法与小波神经网络在ET_0预测中的应用
|
摘要
ET_0是计算作物需水信息的重要依据,对于发展精准灌溉、制定中长期的农业用水策略具有重要的指导意义。选择将P-M公式的计算值为ET_0标准值,利用SPSS软件对多种气象数据进行相关性分析并以最高气温、最低气温、日照时数和相对湿度作为输入因子,采取小波神经网络将归一化的历史ET_0标准序列分别进行高低频带的预测,建立结合遗传算法优化小波神经网络的ET_0预测模型。通过仿真对比,基于遗传算法优化的小波神经网络与未采用遗传算法优化的小波神经网络相比具有更高的精度和更快的收敛速度,其中的小波算法能有效处理ET_0序列在预测过程中的周期非平稳特性。经实验样机平台验证遗传算法优化小波神经网络在作物需水预测中具有良好的性能,模型能够为精准灌溉、农业中长期需水决策提供指导作用。
ET_0 is an important basis for the calculation of crop water requirement information,which has vital guiding significance for the precision irrigation development and the formulation of agricultural water strategy for medium and long term. The values of P-M formulation are selected as ET_0 standard values. SPSS is utilized to analyze the relevance between multiple meteorological factors and ET_0 values. Then the highest temperature, lowest temperature, sunshine hours and relative humidity factors are chosen as input factors. Wavelet neural network is adopted to predict the different frequency band of the normalized history ET_0 standard value sequence respectively. The prediction model of wavelet neural network was established based on genetic algorithm to predict the ET_0.By comparison, the wavelet neural network optimized with genetic algorithm has higher prediction accuracy and faster convergence speed than wavelet neural network model which is not optimized by genetic algorithm,the problem of the periodic non-stationary characteristics of ET_0 sequence in the prediction process is solved effectively.Tested in the experimental prototype platform, the wavelet neural network model optimized with genetic algorithm has good performance in water demand prediction, which will provide the guidance in precision irrigation and agricultural water strategy of medium and long term.
ET_0 is an important basis for the calculation of crop water requirement information,which has vital guiding significance for the precision irrigation development and the formulation of agricultural water strategy for medium and long term. The values of P-M formulation are selected as ET_0 standard values. SPSS is utilized to analyze the relevance between multiple meteorological factors and ET_0 values. Then the highest temperature, lowest temperature, sunshine hours and relative humidity factors are chosen as input factors. Wavelet neural network is adopted to predict the different frequency band of the normalized history ET_0 standard value sequence respectively. The prediction model of wavelet neural network was established based on genetic algorithm to predict the ET_0.By comparison, the wavelet neural network optimized with genetic algorithm has higher prediction accuracy and faster convergence speed than wavelet neural network model which is not optimized by genetic algorithm,the problem of the periodic non-stationary characteristics of ET_0 sequence in the prediction process is solved effectively.Tested in the experimental prototype platform, the wavelet neural network model optimized with genetic algorithm has good performance in water demand prediction, which will provide the guidance in precision irrigation and agricultural water strategy of medium and long term.
引文
[1] 张伟,何勇,裘正军,等.基于无线传感网络与模糊控制的精细灌溉系统设计[J].农业工程学报,2009,25(S2):7-12.ZHANG W,HE Y,QIU Z J,et al.Design of precision irrigation system based on wireless sensor network and fuzzy control[J].Transaction of the CASE,2009,25(S2):7-12.
[2] 罗金耀.节水灌溉理论与技术[M].武汉:武汉大学出版社,2003:41-48.LUO J Y.Water-saving irrigation theory and technology[M].Wuhan:Wuhan University Press,2003:41-48.
[3] 樊军,邵明安,王全九.黄土区参考作物蒸散量多种计算方法的比较研究[J].农业工程学报,2008,24(3):98-102.FAN J,SHAO M A,WANG Q J.Comparison of many equations for calculating reference evapotranspiration in the Loess Plateau of China[J].Transaction of the CASE,2008,24(3):98-102.
[4] JABLOUN M,SAHLI A.Evaluation of FAO-56 methodology for estimating reference evapotranspiration using limited climatic data:application to Tunisia[J].Agricultural Water Management,2008,95(6):707-715.
[5] SHIRI J,NAZEMI A H ,SADRADDINI A A ,et al.Comparison of heuristic and empirical approaches for estimating reference evapotranspiration from limited inputs in Iran[J].Computers and Electronics in Agriculture,2014,108:230-241.
[6] IRMAK S,IRMAK A,ALLEN R G,et al.Solar and net radiation-based equations to estimate reference evapotranspiration in humid climates[J].Journal of Irrigation and Drainage Engineering,2003,129(5):336-347.
[7] 张展羽,王声锋,段爱旺,等.基于天气预报的参考作物腾发量LS-SVM预测模型[J].水科学进展,2010,21(1):63-68.ZHANG Z Y,WANG S F,Duan A W,et al.Least squares support vector machines model for predicting reference evapotranspiration based on weather forecasts[J].Advance in Water Science,2010,21(1):63-68.
[8] 冯禹,崔宁博,龚道枝,等.基于极限学习机的参考作物蒸散量预测模型[J].农业工程学报,2015,31(S1):153-160.FENG Y,CUI N B,GONG D Z,et al.Prediction model of reference crop evapotranspiration based on extreme learning machine[J].Transaction of the CASE,2015,31(S1):153-160.
[9] 王升,陈洪松,聂云鹏,等.基于基因表达式编程算法的参考作物腾发量模拟计算[J].农业机械学报,2015,46(4):106-112.WANG S,CHEN H S,NIE Y P,et al.Simulation of reference evapotranspiration based on gene-expression programming method[J].Transactions of The Chinese Society of Agricultural Machinery,2015,46(4):106-112.
[10] 蔡甲冰,刘钰,雷廷武,等.应用自适应神经模糊推理系统(ANFIS)的ET0预测[J].农业工程学报,2004,20(4):13-16.CAI J B,LIU Y,LEI T W,et al.Prediction of daily reference evapotranspiration using adaptive neuro-fuzzy inference system[J].Transaction of the CASE,2004,20(4):13-16.
[11] 李志磊,魏正英,周建平,等.ET0影响因素及ANFIS预测模型研究[J].机械设计与制造,2016(10):22-26.LI Z L,WEI ZY ,ZHOU J P,et al.Research on ET0 influence factors and ANFIS forecast model[J].Machinery Design & Manufacture,2016(10):22-26.
[12] 吴宏霞.基于BP神经网络的参考作物蒸发蒸腾量预测研究[D].南京:河海大学,2006:35-51.WUHX.Prediction of reference crop evapotranspiration base on back-propagation network[D].Nanjing:Hohai University:2006:35-51
[13] 徐俊增,彭世彰,张瑞美,等.基于气象预报的参考作物蒸发蒸腾量的神经网络预测模型[J].水利学报,2006,37(3):376-379.XU J Z,PENG S Z ,ZHANG R M,et al.Neural network model for reference crop evapotranspiration prediction based on weather forecast[J].Journal of Hydraulic Engineering,2006,37(3):376-379.
[14] 王乙江,彭世彰,徐俊增,等.神经网络方法计算参考作物腾发量精度分析[J].节水灌溉,2012(9):37-40.WANG Y J,PENG S Z,XU J Z,et al.Precision analysis of reference crop evapotranspiration calculation by neural network method[J].Water Saving Irrigation,2012(9):37-40.
[15] 王堃,陈涛涛,李雪,等.基于Matlab的小波神经网络参考作物腾发量预测模型研究[J].沈阳农业大学学报,2013,44(4):457-460.WANG K,CHEN T T,LI X,et al.Wavelet neural network forecasting model for the reference crop evapotranspiration based on Matlab[J].Journal of Shenyang Agricultural University,2013,44(4):457-460.
[16] 冯禹,王守光,崔宁博,等.基于遗传算法优化神经网络的参考作物蒸散量预测模型[J].资源科学,2014,36(12):2624-2630.FENG Y,WANG S G,CUI N B,et al.Artificial neural networks optimized by genetic algorithms formodeling reference evapotranspiration in hilly area in central Sichuan[J].Resources Science,2014,36(12):2624-2630.
[17] 王振午.基于数据挖掘和人工智能的参考作物腾发量预测[D].杭州:浙江大学,2016:45-59.WANG Z W.The estimation of reference crop evapotraspiration based on data mining and artificial intelligence [D].Hangzhou:Zhejiang University,2016:45-59.
[18] 胡庆芳.参考腾发量的计算和预测方法研究 [D].北京:清华大学,2005:48-78.HU Q F.Study on compuation and forcasting methods of reference evapotranspiration[D].Beijing:Tsinghua University,2005:48-78
[19] 刘丙军,邵东国,沈新平.基于小波变换的BP神经网络参考作物腾发量预测模型[J].武汉大学学报(工学版),2007,40(1):69-73.LIU B J,SHAO D G,SHEN X P.Reference crop evapotranspiration forecasting model for BP neural networks based on wavelet transform[J].Engineering Journal of Wuhan University (Engineering Science),2007,40(1):69-73.
[20] WU J H,WANG G L,LI X M,et al.Comparison of BP neural network model and logistic regression in the analysis of influencing factors of violence in hospitals[J].Applied Mechanics & Materials,2011,50/51:964-967.
[21] WU Z H,HUANG N E.Ensemble empirical mode decomposition:a noise-assisted data analysis method[J].Advances in Adaptive Data Analysis,2009,1(1):1-49.
[22] LI S T ,CHEN C C.Function approximation using robust wavelet neural networks[C]// IEEE International Conference on Tools with Artificial Intelligence,Washington,D.C.,2002.
[23] 许允斗,滕志程,郭江涛,等.基于遗传算法的锻造操作机缓冲机构尺度优化[J].燕山大学学报,2016,40(3):202-206.XU Y D,TENG Z C,GUO J T,et al.Dimension optimization of buffering mechanism of forgingmanipulator based on genetic algorithm[J].Journal of Yanshan University,2016,40(3):202-206.
[24] 莫珍丽.基于遗传算法优化小波神经网络的传染病发病率预测模型研究 [D].重庆:重庆医科大学,2015:20-23.MO Z L.Prediction model for infectious disease incidence based on wavelet neural network optimized by genetic algorithm [D].Chongqing:Chongqing Medical University,2015:20-23.
[25] 段春锋,缪启龙,曹雯.西北地区参考作物蒸散变化特征及其主要影响因素[J].农业工程学报,2011,27(8):77-83.DUAN C F,MIAO Q L,CAO W.Changing characteristics of reference crop evapotranspiration and main causes in the Northwest China[J].Transaction of the CASE,2011,27(8):77-83.
[26] 曹雯,申双和,段春锋.西北地区生长季参考作物蒸散变化成因的定量分析[J].地理学报,2011,66(3):407-415.CAO W,SHEN S H,DUAN C F.Quantification ofthe causes for reference crop evapotranspiration changes in growing season in Northwest China[J].Acta Geographica Sinica,2011,66(3):407-415.
[27] 任超,梁月吉,庞光锋,等.经验模态分解和遗传小波神经网络法用于边坡变形预测[J].测绘科学技术学报,2014,31(6):551-555.REN C,LIANG Y J,PANG G F,et al.The empirical mode decomposition and genetic algorithm-wavelet neural network for slope deformation prediction research[J].Journal of Geomatics Science and Technology,2014,31(6):551-555.
[2] 罗金耀.节水灌溉理论与技术[M].武汉:武汉大学出版社,2003:41-48.LUO J Y.Water-saving irrigation theory and technology[M].Wuhan:Wuhan University Press,2003:41-48.
[3] 樊军,邵明安,王全九.黄土区参考作物蒸散量多种计算方法的比较研究[J].农业工程学报,2008,24(3):98-102.FAN J,SHAO M A,WANG Q J.Comparison of many equations for calculating reference evapotranspiration in the Loess Plateau of China[J].Transaction of the CASE,2008,24(3):98-102.
[4] JABLOUN M,SAHLI A.Evaluation of FAO-56 methodology for estimating reference evapotranspiration using limited climatic data:application to Tunisia[J].Agricultural Water Management,2008,95(6):707-715.
[5] SHIRI J,NAZEMI A H ,SADRADDINI A A ,et al.Comparison of heuristic and empirical approaches for estimating reference evapotranspiration from limited inputs in Iran[J].Computers and Electronics in Agriculture,2014,108:230-241.
[6] IRMAK S,IRMAK A,ALLEN R G,et al.Solar and net radiation-based equations to estimate reference evapotranspiration in humid climates[J].Journal of Irrigation and Drainage Engineering,2003,129(5):336-347.
[7] 张展羽,王声锋,段爱旺,等.基于天气预报的参考作物腾发量LS-SVM预测模型[J].水科学进展,2010,21(1):63-68.ZHANG Z Y,WANG S F,Duan A W,et al.Least squares support vector machines model for predicting reference evapotranspiration based on weather forecasts[J].Advance in Water Science,2010,21(1):63-68.
[8] 冯禹,崔宁博,龚道枝,等.基于极限学习机的参考作物蒸散量预测模型[J].农业工程学报,2015,31(S1):153-160.FENG Y,CUI N B,GONG D Z,et al.Prediction model of reference crop evapotranspiration based on extreme learning machine[J].Transaction of the CASE,2015,31(S1):153-160.
[9] 王升,陈洪松,聂云鹏,等.基于基因表达式编程算法的参考作物腾发量模拟计算[J].农业机械学报,2015,46(4):106-112.WANG S,CHEN H S,NIE Y P,et al.Simulation of reference evapotranspiration based on gene-expression programming method[J].Transactions of The Chinese Society of Agricultural Machinery,2015,46(4):106-112.
[10] 蔡甲冰,刘钰,雷廷武,等.应用自适应神经模糊推理系统(ANFIS)的ET0预测[J].农业工程学报,2004,20(4):13-16.CAI J B,LIU Y,LEI T W,et al.Prediction of daily reference evapotranspiration using adaptive neuro-fuzzy inference system[J].Transaction of the CASE,2004,20(4):13-16.
[11] 李志磊,魏正英,周建平,等.ET0影响因素及ANFIS预测模型研究[J].机械设计与制造,2016(10):22-26.LI Z L,WEI ZY ,ZHOU J P,et al.Research on ET0 influence factors and ANFIS forecast model[J].Machinery Design & Manufacture,2016(10):22-26.
[12] 吴宏霞.基于BP神经网络的参考作物蒸发蒸腾量预测研究[D].南京:河海大学,2006:35-51.WUHX.Prediction of reference crop evapotranspiration base on back-propagation network[D].Nanjing:Hohai University:2006:35-51
[13] 徐俊增,彭世彰,张瑞美,等.基于气象预报的参考作物蒸发蒸腾量的神经网络预测模型[J].水利学报,2006,37(3):376-379.XU J Z,PENG S Z ,ZHANG R M,et al.Neural network model for reference crop evapotranspiration prediction based on weather forecast[J].Journal of Hydraulic Engineering,2006,37(3):376-379.
[14] 王乙江,彭世彰,徐俊增,等.神经网络方法计算参考作物腾发量精度分析[J].节水灌溉,2012(9):37-40.WANG Y J,PENG S Z,XU J Z,et al.Precision analysis of reference crop evapotranspiration calculation by neural network method[J].Water Saving Irrigation,2012(9):37-40.
[15] 王堃,陈涛涛,李雪,等.基于Matlab的小波神经网络参考作物腾发量预测模型研究[J].沈阳农业大学学报,2013,44(4):457-460.WANG K,CHEN T T,LI X,et al.Wavelet neural network forecasting model for the reference crop evapotranspiration based on Matlab[J].Journal of Shenyang Agricultural University,2013,44(4):457-460.
[16] 冯禹,王守光,崔宁博,等.基于遗传算法优化神经网络的参考作物蒸散量预测模型[J].资源科学,2014,36(12):2624-2630.FENG Y,WANG S G,CUI N B,et al.Artificial neural networks optimized by genetic algorithms formodeling reference evapotranspiration in hilly area in central Sichuan[J].Resources Science,2014,36(12):2624-2630.
[17] 王振午.基于数据挖掘和人工智能的参考作物腾发量预测[D].杭州:浙江大学,2016:45-59.WANG Z W.The estimation of reference crop evapotraspiration based on data mining and artificial intelligence [D].Hangzhou:Zhejiang University,2016:45-59.
[18] 胡庆芳.参考腾发量的计算和预测方法研究 [D].北京:清华大学,2005:48-78.HU Q F.Study on compuation and forcasting methods of reference evapotranspiration[D].Beijing:Tsinghua University,2005:48-78
[19] 刘丙军,邵东国,沈新平.基于小波变换的BP神经网络参考作物腾发量预测模型[J].武汉大学学报(工学版),2007,40(1):69-73.LIU B J,SHAO D G,SHEN X P.Reference crop evapotranspiration forecasting model for BP neural networks based on wavelet transform[J].Engineering Journal of Wuhan University (Engineering Science),2007,40(1):69-73.
[20] WU J H,WANG G L,LI X M,et al.Comparison of BP neural network model and logistic regression in the analysis of influencing factors of violence in hospitals[J].Applied Mechanics & Materials,2011,50/51:964-967.
[21] WU Z H,HUANG N E.Ensemble empirical mode decomposition:a noise-assisted data analysis method[J].Advances in Adaptive Data Analysis,2009,1(1):1-49.
[22] LI S T ,CHEN C C.Function approximation using robust wavelet neural networks[C]// IEEE International Conference on Tools with Artificial Intelligence,Washington,D.C.,2002.
[23] 许允斗,滕志程,郭江涛,等.基于遗传算法的锻造操作机缓冲机构尺度优化[J].燕山大学学报,2016,40(3):202-206.XU Y D,TENG Z C,GUO J T,et al.Dimension optimization of buffering mechanism of forgingmanipulator based on genetic algorithm[J].Journal of Yanshan University,2016,40(3):202-206.
[24] 莫珍丽.基于遗传算法优化小波神经网络的传染病发病率预测模型研究 [D].重庆:重庆医科大学,2015:20-23.MO Z L.Prediction model for infectious disease incidence based on wavelet neural network optimized by genetic algorithm [D].Chongqing:Chongqing Medical University,2015:20-23.
[25] 段春锋,缪启龙,曹雯.西北地区参考作物蒸散变化特征及其主要影响因素[J].农业工程学报,2011,27(8):77-83.DUAN C F,MIAO Q L,CAO W.Changing characteristics of reference crop evapotranspiration and main causes in the Northwest China[J].Transaction of the CASE,2011,27(8):77-83.
[26] 曹雯,申双和,段春锋.西北地区生长季参考作物蒸散变化成因的定量分析[J].地理学报,2011,66(3):407-415.CAO W,SHEN S H,DUAN C F.Quantification ofthe causes for reference crop evapotranspiration changes in growing season in Northwest China[J].Acta Geographica Sinica,2011,66(3):407-415.
[27] 任超,梁月吉,庞光锋,等.经验模态分解和遗传小波神经网络法用于边坡变形预测[J].测绘科学技术学报,2014,31(6):551-555.REN C,LIANG Y J,PANG G F,et al.The empirical mode decomposition and genetic algorithm-wavelet neural network for slope deformation prediction research[J].Journal of Geomatics Science and Technology,2014,31(6):551-555.