Abstract
The greenhouse environment is a complex multi-scale integrated nonlinear system. This agro-ecosystem is composed of crop and greenhouse climate which are based on the existence of two different timescales. The effect of temperature plays a vital role in the variation of phenotypic data in crop growth. Thus, two different models, one for capturing the climate dynamics inside the greenhouse and other for crop dynamics, are essential. First, the neural network is used to predict the inside environment, given the outside conditions and the operation of the control equipment. The inputs of the network are meteorological parameters, whose measurements are costly and time consuming to acquire. So, instead of measuring all the parameters used in the physical modeling, the most significant relevant input parameters which give same modeling efficiency are identified. To avoid overfitting of the data and to realize the best prediction results with the simplest structure, an enhanced pruning algorithm is implemented for topology optimization of the artificial neural network. The pruning algorithm is discussed and exemplified via simulations. By plotting the training error, test error, and final prediction error (FPE) estimates over the course of pruning the network weights, it is inferred that a network with minimum of 15 weights is reliable to model greenhouse environment dynamics. With the Optimal Brain Surgeon (OBS) algorithm, a reduction of the number of weights from 141 to 76 (46%) in the first step and finally to 15 (89%) was achieved and the percentage prediction error is reduced from 13.13% for the complete network structure to 4.35% for final pruned network. Secondly, in order to study the progress of crop ontogeny, bootstrap resampling-based artificial neural network is developed with limited destructive measurements. The notion of prediction performance and the efficiency of the bootstrapped crop phenotypic neural network model are evaluated by root mean square error (RMSE), mean square error (MSE) and Nash and Sutcliffe efficiency (NSE) criteria. The net assimilation rate which determines the growth rate of the plant inside the greenhouse environment is calculated. The resulting model can be used for growth assessment, understanding crop physiology and yield prediction.
Similar content being viewed by others
Abbreviations
- \(S_{i}\) :
-
Intercepted solar radiant energy (W)
- \(T_{\text{out}}\) :
-
Outside temperature (°C)
- \(W_{o}\) :
-
Exterior absolute humidity (\({\text{g}}\,{\text{H}}_{2} {\text{O}}/{\text{m}}^{3} )\)
- \({\dot{\text{V}}}\) :
-
Ventilation rate (\({\text{m}}^{3} /{\text{s}}\))
- \(Q_{p}\) :
-
Water capacity of the evaporative cooling system (\({\text{g}}\,{\text{H}}_{2} {\text{O}}/{\text{s}})\)
- J :
-
Julian day
- H :
-
Julian hour
- \(T_{\text{in}}\) :
-
Inside temperature (°C)
- \(\rho\) :
-
Air density (\({\text{kg/m}}^{3}\))
- \(C_{p}\) :
-
Specific heat of air (\({\text{J}}\,\left( {\text{KgK}} \right)^{ - 1}\))
- \(V\) :
-
Greenhouse volume (\({\text{m}}^{3}\))
- \(q_{\text{heater}}\) :
-
Heat provided by the greenhouse heater (W)
- \(\lambda\) :
-
Latent heat of vaporization (J/g)
- \(U\) :
-
Heat transfer coefficient
- \(A\) :
-
Greenhouse area \(({\text{m}}^{2} )\)
- \(\omega_{\text{in}}\) :
-
Interior absolute humidity (\({\text{g}}\,{\text{H}}_{2} {\text{O}}/{\text{m}}^{3} )\)
- \(E\) :
-
Evapotranspiration rate of the plants \(( {\text{g}}\,{\text{H}}_{2} {\text{O}}/{\text{s}})\)
- \(\beta_{\text{T}}\) :
-
Thermodynamic coefficient
- LA:
-
Leaf area (m3)
- W :
-
Dry weight of the plant (g)
- E :
-
Net assimilation rate (\({\text{g}}\, {\text{m}}^{ - 2} \,{\text{day}}^{ - 1}\))
References
Clark OG, Kok R (1998) Engineering of highly autonomous biosystems: review of the relevant literature. Int J Intell Syst 13(8):749–783
Revathi S, Radhakrishnan TK, Sivakumaran N (2017) Climate control in greenhouse using intelligent control algorithms. In: Proceedings of the American control conference. https://doi.org/10.23919/ACC.2017.7963065
Albright LD, Gates RS, Arvanitis KG, Drysdale AE (2001) Environmental control for plants on earth and in space. IEEE Control Syst Mag. https://doi.org/10.1109/37.954518
Vanthoor BHE, van Henten EJ, Stanghellini C, de Visser PHB (2011) A methodology for model-based greenhouse design: part 3, sensitivity analysis of a combined greenhouse climate-crop yield model. Biosys Eng 110(4):396–412. https://doi.org/10.1016/j.biosystemseng.2011.08.006
Fuchs M, Dayan E, Presnov E (2006) Evaporative cooling of a ventilated greenhouse rose crop. Agric For Meteorol 138:203–215. https://doi.org/10.1016/j.agrformet.2006.05.002
Kittas C, Karamanis M, Katsoulas N (2005) Air temperature regime in a forced ventilated greenhouse with rose crop. Energy Build 37:807–812. https://doi.org/10.1016/j.enbuild.2004.10.009
Linker R, Seginer I (2004) Greenhouse temperature modeling: a comparison between sigmoid neural networks and hybrid models. Math Comput Simul 65:19–29. https://doi.org/10.1016/j.matcom.2003.09.004
Seginer I, Boulard T, Bailey BJ (1994) Neural network models of the greenhouse climate. J Agric Eng Res 59(3):203–216. https://doi.org/10.1006/jaer.1994.1078
Potdar MV, Pawar KR (1991) Non-destructive leaf area estimation in banana. Sci Hortic 45(3–4):251–254. https://doi.org/10.1016/0304-4238(91)90070-F
Mokhtarpour H, Christopher BS, Saleh G, Selamat AB, Asadi ME, Kamkar B (2010) Non-destructive estimation of maize leaf area, fresh weight, and dry weight using leaf length and leaf width. Commun Biom Crop Sci 5(1):19–26
Jia Y, Culver TB (2006) Bootstrapped artificial neural networks for synthetic flow generation with a small data sample. J Hydrol 331(3–4):580–590. https://doi.org/10.1016/j.jhydrol.2006.06.005
Efron B, Tibshirani RJ (1993) An introduction to the bootstrap. Refrig Air Cond 57(57):436. https://doi.org/10.1111/1467-9639.00050
Yu H, Wilamowski BM (2011) Levenberg–Marquardt training. Industrial electronics handbook. Intell Syst 5:12-1-16. https://doi.org/10.1201/b10604-15
Nagy ZK (2007) Model based control of a yeast fermentation bioreactor using optimally designed artificial neural networks. Chem Eng J 127(1–3):95–109. https://doi.org/10.1016/j.cej.2006.10.015
Efron B (1979) Bootstrap methods: another look at the jackknife. Ann Stat 7(1):1–26. https://doi.org/10.1214/aos/1176344552
Ronga D, Zaccardelli M, Lovelli S, Perrone D, Francia E, Milc J, Pecchioni N (2017) Biomass production and dry matter partitioning of processing tomato under organic vs conventional cropping systems in a Mediterranean environment. Sci Hortic 224:163–170. https://doi.org/10.1016/j.scienta.2017.05.037
Chernick MR, Labudde RA (2009) Revisiting qualms about bootstrap confidence intervals. Am J Math Manag Sci 29(3–4):437–456. https://doi.org/10.1080/01966324.2009.10737767
Nash JE, Sutcliffe JV (1970) River flow forecasting through conceptual models part I—a discussion of principles. J Hydrol 10(3):282–290. https://doi.org/10.1016/0022-1694(70)90255-6
Acknowledgements
The authors thank ICAR-National Research Centre for Banana (NRCB) for providing the greenhouse structural facility for the research reported in this paper. We are also thankful to Dr. Uma, the Director and Dr. Ramajayam Devarajan, Senior Scientist of NRCB for their support to pursue the research studies.
Author information
Authors and Affiliations
Corresponding author
Ethics declarations
Conflict of interest
The authors declare that they have no conflict of interest.
Rights and permissions
About this article
Cite this article
Soundiran, R., Radhakrishnan, T.K. & Natarajan, S. Modeling of greenhouse agro-ecosystem using optimally designed bootstrapping artificial neural network. Neural Comput & Applic 31, 7821–7836 (2019). https://doi.org/10.1007/s00521-018-3598-7
Received:
Accepted:
Published:
Issue Date:
DOI: https://doi.org/10.1007/s00521-018-3598-7