Multi-Step Rolling Prediction Model of Greenhouse Microclimate Based on R-BP Neural Network

doi:10.3969/j.issn.1000-6362.2018.05.003

Abstract

Abstract: High greenhouse efficiency depends on the proper greenhouse environment. So it is of great significance to establish an accurate multi-step greenhouse microclimate prediction model to optimize the greenhouse environment control. A rolling back propagation (R-BP) neural network group model was proposed in this paper. The R-BP model mainly includes two stages: (1) Establish an initial BP neural network model, which adopts an auto-encoder (AE) to learn initial network factors and optimize the network factors by improved particle swarm algorithm; (2) Establish a rolling prediction model to realize multi step prediction of greenhouse microclimate. It used the output of the previous network as partial input of the next network. To prove the effectiveness of the R-BP model, several experiments were implemented in the Abu Dhabi auto-controlled green house and non-controlled Suzhou greenhouse. The experiments in Abu Dhabi greenhouse proved that the R-BP model achieved an average of 69.9% error decrease in 6h temperature prediction in the greenhouse and an average of 47% error decrease in relative humidity prediction, compared with the traditional BP neural network. In Suzhou greenhouse, the average prediction error of temperature was reduced by 43.3% and the average prediction error of humidity was reduced by 55.6%. The experimental results prove that the R-BP model can accurately predict the change of the greenhouse microclimate for future 6 hours, to provide the basis for greenhouse microclimate control optimization.

Key words: Greenhouse microclimate, Prediction model, Automatic encoder, BP neural network, Particle swarm optimization, Rolling prediction

REN Shou-gang, LIU Xin, GU Xing-jian,WANG Hao-yun, YUAN Pei-sen,XU Huan-liang. Multi-Step Rolling Prediction Model of Greenhouse Microclimate Based on R-BP Neural Network[J]. Chinese Journal of Agrometeorology, 2018, 39(05): 314-324.

References

[1] 丁为民,汪小旵,李毅念,等.温室环境控制与温室模拟模型研究现状分析[J].农业机械学报,2009,40(5):162-168. Ding W M,Wang X C,Li Y N,et al.Review on environmental control and simulation models for greenhouses[J].Transactions of the Chinese Society for Agricultural Machinery,2009,40(5): 162-168.(in Chinese) [2] 徐立鸿,苏远平,梁毓明.面向控制的温室系统小气候环境模型要求与现状[J].农业工程学报,2013,29(19):1-15. Xu L H,Su Y P, Liang Y M.Requirement and current situation of control-oriented microclimate environmental model in greenhouse system[J]. Transactions of the CSAE,2013, 29(19):1-15 (in Chinese) [3] 张芳,方慧,杨其长,等.基于CFD模型的大跨度温室自然通风热环境模拟[J].中国农业气象,2017,38(4):221-229. Zhang F,Fang H,Yang Q C,et al.Ventilation simulation in a large-scale greenhouse based on CFD[J].Chinese Journal of Agrometeorology,2017,38(4):221-229.(in Chinese ) [4] 邹秋滢.温室小气候模型的建立及其控制策略研究[D].沈阳:沈阳农业大学,2010. Zou Q Y.Study on dynamic modelling and control strategy in greenhouse climate system[D].Shenyang:Shenyang Agricultural University,2010. [5] Seginer I,Boulard T,Bailey B J.Neural network models of the greenhouse climate[J].J.Agric.Engng.Res.,1994,59(3):203-216. [6] 汪小旵,丁为民,罗卫红,等.利用 BP 神经网络对江淮地区梅雨季节现代化温室小气候的模拟与分析[J].农业工程学报,2004,20(2):235-238. Wang X C,Ding W M,Luo W H,et al.Simulation and analysis of micro-climate of gutter connectedVenlo greenhouse during rainy season in Jianghuairegion of China using BP neural network[J].Transactions of the CSAE,2004,20(2):235-238.(in Chinese ) [7] 何芬,马承伟.华北地区冬季温室植物冠层温度建模[J].农业机械学报,2009,40(5):169-172. He F,Ma C W.Modeling greenhouse plant canopy temperature in North China during winter[J].Journal of Agricultural Machinery,2009,40(5):169-172.(in Chinese) [8] 李倩,申双和,曹雯,等.南方塑料大棚冬春季温湿度的神经网络模拟[J].中国农业气象,2012,33(2):190-196. Li Q,Shen S H,Cao W,et al.Neural network simulation on air temperature and relative humidity inside plastic greenhouse during winter and spring in southern China[J].Chinese Journal of Agrometeorology,2012,33(2):190-196.(in Chinese) [9] 金志凤,符国槐,黄海静,等.基于BP神经网络的杨梅大棚内气温预测模型研究[J].中国农业气象,2011,32(3): 362-367. Jin Z F,Fu G H,Hang H J,et al.Simulation and forecast of air temperature inside the greenhouse planted myica rubra based on BP neural network[J].Chinese Journal of Agrometeoro- logy,2011,32(3):362-367.(in Chinese ) [10] 程曼,袁洪波,蔡振江,等.基于全局变量预测模型的温室环境控制方法[J].农业工程学报,2013,29(S1):177-183. Cheng M,Yuan H B,Cai Z J,et al.Environment control method in greenhouse based on global variable prediction model[J].Transactions of the CSAE,2013,29(Supp.1):177-183. (in Chinese) [11] 柳芳,王铁,刘淑梅.天津市二代节能型日光温室内部温湿度预测模型:以西青为例[J].中国农业气象,2009,30(S1): 86-89. Liu F,Wang T,Liu S M. Research on temperature and humidity forecasting model for second energy saving solar greenhouse in Tianjin :taking Xiqing as a example[J]. Chinese Journal of Agrometeorology,2009,30(Supp.1): 86-89.(in Chinese ) [12] 李宁,申双和,黎贞发,等.基于主成分回归的日光温室内低温预测模型[J].中国农业气象,2013,34(3):306-311. Li N,Shen S H,Li Z F,et al.Forecast model of minimum temperature inside greenhouse based on principal component regression[J].Chinese Journal of Agrometeorology,2013, 34(3):306-311.(in Chinese ) [13] Bengio Y,Courville A,Vincent P.Representation learning:a review and new perspectives[J].IEEE Transactions on Pattern Analysis and Machine Intelligence,2013,35(8): 1798-1828. [14] Rumelhart D E,Hinton G E,Williams R J.Learning representations by back-propagating errors[J].Nature,1986, 323(6088): 533-536. [15] Bengio Y,Lamblin P,Popovici D,et al.Greedy layer-wise training of deep networks[C].Proc.of the 20th Annual Conference on Neural Information Processing System,2006, 19:153-160. [16] 曲建岭,杜辰飞,邸亚洲,等.深度自动编码器的研究与展望[J].计算机与现代化,2014,(8):128-134. Qu J L,Du C F,Di Y Z,et al.Research and prospect of deep auto-encoders[J].Computer and Modernization,2014,(8): 128-134.(in Chinese) [17] Yin H,Jiao X,Chai Y,et al.Scene classification based on single-layer SAE and SVM[J].Expert Systems with Applications,2014,42(7):3368-3380. [18] L?ngkvist M,Karlsson L,Loutfi A.A review of unsupervised feature learning and deep learning for time-series modeling [J].Pattern Recognition Letters,2014,42(1):11-24. [19] Romeu P,Zamora-Martínez F,Botella-Rocamora P,et al. Time-Series forecasting of indoor temperature using pre-trained deep neural networks[A].Artificial Neural Networks and Machine Learning: ICANN 2013[C].Springer Berlin Heidelberg,2013:451-458. [20] 陈啸,王红英,孔丹丹,等.基于粒子群参数优化和BP神经网络的颗粒饲料质量预测模型[J].农业工程学报,2016,32 (14):306-314. Chen X,Wang H Y,Kong D D,et al.Quality prediction model of pellet feed basing on BP network using PSO parameters optimization method[J].Transactions of the CSAE,2016, 32(14):306-314.(in Chinese) [21] 徐小力,刘秋爽,见浪護.光伏充气膜温室自跟踪发电系统发电量预测[J].农业机械学报,2012,43(S1):305-310. Xu X L,Liu Q S,Jian L H.Generating capacity prediction of automatic tracking power generation system on inflatable membrane greenhouse attached photovoltaic[J].Journal of Agricultural Machinery,2012,43(Supp.1):305-310.(in Chinese) [22] 武妍,徐敏.一种改进的粒子群优化算法[J].计算机工程与应用,2006,42(33):40-42. Wu Y,Xu M.Modified particle swarm optimization algorithm[J].Computer Engineering and Applications,2006, 42(33):40-42.(in Chinese) [23] 张文爱,刘丽芳,李孝荣.基于粒子进化的多粒子群优化算法[J].计算机工程与应用,2008,44(7):51-53. Zhang W A,Liu L F,Li X R.Particle swarm optimization based on particle evolution[J].Computer Engineering and Applications,2008,44(7):51- 53.(in Chinese) [24] 王存睿,段晓东,刘向东,等.改进的基本粒子群优化算法[J].计算机工程,2004,30(21):35-37. Wang C R,Duan X D,Liu X D,et al.A modified basic particle swarm optimization algorithm[J].Computer Engineering, 2004,30(21):35-37.(in Chinese)