Maize yield combined prediction system and method

Abstract

The invention discloses a maize yield combined prediction system and method. The prediction system comprises a data input module, a model base and intelligent combined module, a combined prediction module, a parameter management module and a result check and output module. The prediction method includes the steps that yield prediction is divided into sown area prediction and per-unit-yield prediction, models are selected to be combined according to predicted targets, the weights of sub-models of the combined model are determined according to the basic data change amplitude, the combined prediction result is acquired according to weight distribution of the combined model and checked, and finally the combined prediction result is output through diagrams and characters. Different prediction models are correspondingly combined according to different prediction durations, prediction variable factors and prediction basic data, maize yield prediction is pertinently conducted, the prediction result is higher in accuracy, the application range is wider, short-period and medium-and-long-term dynamic prediction can be achieved at the same time, and agricultural research requirements are met.

Description

technical field [0001] The invention relates to the technical field of intelligent agriculture, in particular to a corn yield combination prediction system and method. Background technique [0002] Taking scientific methods to predict grain output is an important basis for grain production decision-making. Corn has the highest yield and consumption among the major food crops in my country, and it plays a vital role in socio-economic development and improvement of residents' living standards. It is estimated that the growth of my country's food demand in the future will mainly focus on corn. By the end of the "Twelfth Five-Year Plan", my country's corn consumption will exceed 220 million tons, and the annual increase in demand will reach 10 million tons. According to the US Department of Agriculture's forecast report, China's corn imports in 2024 / 25 will more than double the current level. The pattern of tight balance between supply and demand of corn may be broken. Accurat...

AI Technical Summary

Problems solved by technology

[0008] Judging from the various existing forecasting methods, there are various types of forecasting models, lack of comprehensive consideration of various influencing factors, the forecasting time of various methods is inconsistent, and the consistency of forecasting results is poor, which leads to inaccurate forecasting results and is difficult to obtain convincing results. conclusion

Images