| Soybean is one of the main food crops in China and is widely used in food,feeding and industry.Therefore,the demand for soybean is very large.China is the first soybean importer in the world,but its soybean output ranks only the fourth in the world.For soybean production,China is about ten times less than the United States.The main soybean importers of our country are mainly the United States,Brazil and Argentina,where soybean production is in the top three in the world.Because of the huge demand for soybean in the international soybean market,China has been in a passive state.How to improve soybean yield has become an urgent problem for soybean planting industry in China.The main factors affecting soybean yield are soybean varieties,planting methods,fertilization,soil,water and sunshine,etc.This research is carried out on the planting density and fertilization systems which are currently lacking in China.At present,in the process of soybean planting in China,soybean planting density and fertilizer amount are basically carried out according to the conventional experience of production workers.This method not only fails to make rational use of agricultural production resources,but also leads to the waste of agricultural resources and the increase of production costs.The excessive application of fertilizers will even cause serious damage to the ecological environment of farmland,such as soil consolidation,water pollution,etc.In this study,field experiment is conducted to investigate the practical problems faced by soybean production in China at the present stage.The purpose of the experiment is to study and popularize the soybean planting density and nitrogen,phosphorus and potassium fertilizer application models suitable for soybean production in northeast China.The field experiment was conducted from October 2017 to October 2018 at 93 Administration Bureau Da Xi Jiang Farm,which belongs to Heilongjiang Agricultural Reclamation General Administration.The representative experimental land was selected and plots were divided.The soybean variety "Hei he 52" was used as the experimental material.Through rigorous experimental design,the purpose,requirements,methods and steps of the experiment were clearly defined,and the planting density,nitrogen application rate and plot of soybean were determined.Phosphorus and potassium application were the experimental factors,and soybean yield was the index.The field experiment includes single factor experiment and multi-factor experiment.The multi-factor experiment is designed by quadratic rotation orthogonal regression.After obtaining the experimental data,regression analysis and BP neural network are used to establish and optimize the fertility effect model,and the optimum fertility effect model is obtained with the aim of maximizing soybean yield.Based on the above model,the optimum planting density of "Hei he 52" soybean in Northeast China is 235,000 plants/ha,N application rate 67.9 kg/ha,P application rate 104.6 kg/ha and K application rate 17.4 kg/ha.Under this condition,soybean yield is 3953.2 kg/ha.This model combines the planting density of soybean with the amount of fertilizer applied.This model carries out optimization analysis through BP neural network and proves that it has practical significance to popularize in Northeast China,especially in Heilongjiang soybean producing area by field experiments.The popularization of the model will greatly increase the yield of soybean,improve the deficiency of fertilizer and domestic system theory,and alleviate the serious shortage of domestic soybean supply and demand.At the same time,it is also the popularization and application of industrial engineering optimization idea in the field of agriculture. |
PDF Full Text Request