Modern aspects of informatization of agricultural production based on modeling and forecasting the production process of lentils under different conditions of moisture supply
Abstract
The article presents the results of the application of modern systems for modeling and forecasting the production process of lentils in the Southern Steppe of Ukraine. The correlation-regression analysis shows the high reliability and practical value of the obtained mathematical models of growing lentils for grain depending on the conventional tillage, fertilizer rate, and plant density under different moisture conditions; that is confirmed by the curves based on the experimental data and calculations. Mathematical models of lentils grain yield under different moisture conditions were compiled according to the obtained regression coefficients and free members: without irrigation - Y=1,5896+0,0032×Х1+0,0007×Х2-0,2561×Х3, and when applying irrigation Y=1,0200+0,0051×Х1+0,0022×Х2+0,2656×Х3.
The following results were obtained for the dependent variable for different conditions of moisture supply after conducting a regression-normalized analysis of the researched factors in view of yield of lentils, where: in variant without irrigation R = 0.7059; R2 = 0.4983; adjusted R2 = 0.4682; F (3,50) = 16,551 p <0,00000 and standard estimation error was 0,1232; in variant with irrigation R = 0,6131; R2 = 0.3759; adjusted R2 = 0.3385; F (3.50) = 10.04 p <0.00003 and the standard estimation error was 0,2591.
Nonlinear multilayer artificial neuron models have been developed for the first time to predict lentils grain yields. Generalized regression artificial neural network GRNN (4-12-7-1) with 12 neurons in the first hidden layer and seven ones in the second hidden layer; learning productivity was 0.215; control productivity was 0.290; test productivity was 0.362; learning error was 0.136; control error was 0.049; test error was 0.066. Taking into account nonlinear patterns of factor effect on lentils grain yield the multiple correlation was 0.96. Based on the results of ranking the researched factors' effect on the dynamics of formation and yield of lentils, it was found that moisture conditions (water consumption, m3/ha) with an impact factor of 4.21 which exceeds other researched factors by almost 2.2 times, are in the first place. Plant density (million/ha) was in second place with a factor of 1.62. The rate of mineral fertilizers (kg/ha of active substance) was in third place, which was slightly inferior to the density of standing plants, resulting in a total of 1.61. The depth of tillage (cm) was in the last fourth place with an impact factor of 1.01.
References
2. Extreme Trading. (2019). Nejronnye seti dlya trejderov. Extreme Trading. Retrieved from: https://medium.com/@ExtremeTrading/nejronnye-seti-dlya-trejderov-395d6e04b024 [in Russian]
3. Ushkarenko, V.A., Lazarev, N.N., Goloborodko, S.P., & Kokovihin, S.V. (2011). Dispersionnyj i korrelyacionnyj analiz v rastenievodstve i lugovodstve: Monografiya. Moskva: Izd. RGAU–MSHA im. K.A. Timiryazeva. 336 s. [in Russian]
4. Ushkarenko, V.O., Kovalenko, V.P., Plotkin, S.Ya., & Polyakov, M.G. (2001). Vikoristannya personalnih komp’yuteriv dlya virishennya zadach optimizaciyi silskogospodarskogo virobnictva: Navchalnij posibnik. Herson: Ajlant. [in Ukrainian]
5. Ushkarenko, V.O., Najdenova, V.O., Lazer, P.N., Sviridov, O.V., Lavrenko, S.O., & Lavrenko, N.M.(2016). Naukovi doslidzhennya v agronomiyi: Navchalnij posibnik. Herson: Grin D.S. [in Ukrainian]
6. Milyutin, I. (2018). Arhitektur nejronnyh setej dlya resheniya zadach NLP. Retrieved from: https://neurohive.io/ru/osnovy-data-science/7-arhitektur-nejronnyh-setej-nlp/ [in Russian]
7. Alyoshin, S., Borodina, E., Hafiiak, A., & Nosach, O. (2018). Neural Network Technology of the Financial and Economic Model Synthesis of Production as the Fragment of the Economy Digitalization. International Journal of Engineering & Technology, Vol. 7 (4.8), 355-363.
8. Govindaraju, R.S. (2000). Artificial neural networks in hydrology. I: Preliminary concepts. J. Hydrol. Eng., Vol. 5, 115-123.
9. Hansen, L.K., & Salamon, P. (1990). Neural network ensembles. IEEE T. Pattern Anal. Mach. Intell., Vol. 10, 993-1001.
10. Hebb, D.O. (1949). The organization of behavior. New York: Wiley.
11. Irmak, A., & Kamble, B. (2009). Evapotranspiration data assimilation with genetic algorithms and SWAP model for on-demand irrigation. Irrig. Sci., Vol. 28, 101-112.
12. Jarratano, D., & Riley, G. (2006). Expert systems. Principles of programming development. Ed. Williams
13. Khaikin, S. (2006). Neural networks: Full course. 2nd edition. Williams.
14. Kumar, M., Raghuwanshi, N.S., & Singh, R. (2011). Artificial neural networks approach in evapotranspiration modeling: A review. Irrig. Sci., Vol. 29, 11-25.
15. Lavrenko, S., Lavrenko, N., & Pichura, V. (2015). Neural network modeling of chickpea grain yield on ameliorated soils. Scientific Journal of Russian Scientific Research Institute of Land Improvement Problems, Vol. 2 (18), 16-30.
16. McCulloch, W.S., & Pitts, W.H. (1943). A logical calculus of ideas immanent in nervousactivity. Bull. Math. Biophys., Vol. 5, 115-133.
17. Michie, D., Spiegelhalter, D.J., Taylor, C.C. (1994). Machine Learning, Neural and Statistical Classification. Chichester: Ellis Horwood.
18. Minsky, M., & Papert, S.A. (1969). Perceptrons: An Introduction to Computational Geometry. Cambridge, Mass.: MIT Press.
19. Schumann, J.M.P., & Liu ,Y. (2010). Applications of Neural Networks in High Assurance Systems. Springer: Berlin, Germany,Vol. 268. 345 r.
20. Turan, N.G., Mesci, B., & Ozgonenel, O. (2011). Artificial neural network (ANN) approach for modeling Zn(II) adsorption from leachate using a new biosorbent. Chemical Engineering Journal, Vol. 173, 1, 98-105.
