Multi-dimensional time series forecasting method of total grain yield based on LSTM neural network

Abstract

The invention discloses a multi-dimensional time series forecasting method of total grain yield based on an LSTM neural network. Based on LSTM neural network, the output data of many kinds of agricultural products are used as the input variables of LSTM neural network, and a multidimensional time series forecasting model based on LSTM neural network is established. Because the output data of manykinds of agricultural products are considered and the superposed LSTM is used to process the sequence data information, the problem of gradient disappearance is avoided, the forecasting effect is good, the precision is high, and the applicability is good.

Description

technical field [0001] The invention relates to a method for predicting total grain output, specifically, a multidimensional time series prediction method for total grain output based on an LSTM neural network, belonging to the field of grain output forecasting. Background technique [0002] At present, a variety of forecasting methods have been formed in the field of grain production forecasting, such as exponential smoothing model, gray forecasting method, ARIMA model, remote sensing technology forecasting model, statistical dynamic growth model, meteorological yield forecasting model, neural network model, etc. After investigation, most of these methods only use the historical data of grain production to predict the grain production in the future, and the total grain production is often affected by other agricultural products, such as the increase in the area of ​​corn field under a certain planting area. At the same time, the area of ​​farmland for planting other crops w...

AI Technical Summary

Problems solved by technology

However, its memory function is limited, and it cannot remember the results of a long time ago, so it cannot solve the problem of long-distance dependence.

Images