Grain quality multi-index prediction method and device

Abstract

The invention relates to a grain quality multi-index prediction method and apparatus. The method comprises the steps of obtaining N kinds of grain quality index time series data; inputting the N typesof grain quality index time series data into a trained LSTM-GAN prediction model to obtain a prediction value of each index; wherein the LSTM-GAN prediction model comprises an LSTM model and a GAN model; wherein the LSTM model at least comprises N LSTM units; wherein N is the type number of the indexes; inputting the grain quality index time series data of each category in the training set into the corresponding LSTM unit for prediction; a prediction result is sent to the GAN model; and determining corresponding state information by integrating the prediction result of each index through theGAN model, and adjusting the parameters of each LSTM unit according to the state information corresponding to the real index data in the training set, so that the prediction error meets the requirement, and the training of the LSTM-GAN prediction model is realized. According to the method, the relevance and interaction of various grain quality indexes are considered, so that the accuracy of a prediction result is improved.

Description

technical field [0001] The invention relates to a grain quality multi-index prediction method and device, belonging to the technical field of grain quality evaluation and analysis. Background technique [0002] The quality of wheat grains gradually deteriorates with the prolongation of storage time, and it is reflected in the numerical changes of multiple indexes at different degrees. Therefore, the degree of deterioration of the storage quality can be understood by studying the change trend of wheat multi-index time-series data, and can be adjusted accordingly. The storage process of wheat. Due to the differences in the wheat storage environment and the different interactions between multiple indicators, there are certain errors in the prediction of the multi-index time series data, and the error increases with the prolongation of the storage time, which may also affect the accuracy of wheat quality evaluation. sex. Therefore, it is necessary to explore a more effective m...

AI Technical Summary

Problems solved by technology

[0024] The purpose of the present invention is to provide a grain quality multi-indicator prediction method and device, which is used to solve the problem of inaccurate prediction of the existing grain multi-indicator data

Images