Soil testing and formulated fertilization method based on deep learning and weighting type multi-factor evaluation

Abstract

The invention provides a soil testing and formulated fertilization method based on deep learning and weighting type multi-factor evaluation, which comprises the following steps: sorting and archiving soil data and plot yield data, and carrying out normalization processing and label adding operation to obtain sample data; setting corresponding weights according to differences of sensitivities of different crops to different chemical elements, and grading fertility conditions and heavy metal pollution conditions of soil by utilizing a fertility coefficient standard formula and an improved Nemero index method to obtain a comprehensive fertility index; improving a fitness function in a classical genetic algorithm through weighting type multi-factor evaluation, selecting proper filial generation data through the fitness function to perform data enhancement, and expanding a data sample; and training the neural network model by using the expanded data sample to obtain a crop yield prediction and fertilizer preparation and supplement strategy. According to the method, the problem that a neural network requires a large number of training samples is solved, and abundance and deficiency degree evaluation and pollution index evaluation of soil are given.

Description

technical field [0001] The invention relates to planting technology, in particular to a soil testing and formula fertilization method based on deep learning and weighted multi-factor evaluation. Background technique [0002] The method of soil testing and formula fertilization is based on the results of soil testing and fertilizer field tests, and according to the law of crop fertilizer demand, soil fertilizer supply performance and fertilizer effect, it is a technology to reasonably propose the application amount, application period and method of fertilizer. This technology can be divided into For soil testing and formula fertilization, both soil testing and formula fertilization must be tested for soil nutrients. The latter still needs to draw conclusions based on a large number of field experiments, but because of the same purpose, it is collectively referred to as soil testing and formula fertilization technology. [0003] With the rapid development of network informatio...

AI Technical Summary

Problems solved by technology

[0009] To sum up, if the existing methods of soil testing and formula fertilization are based on agronomic formulas and mathematical methods, there will be large analysis errors for some uncommon crop objects or crop data with large sampling point deviations; if they are based on expert experience And probability theory, for some crops grown under complex conditions, there will be excessively complex high-dimensional situations in probability theory, which will greatly affect the accuracy of algorithm analysis and prediction; if it is based on a single neural network, it requires a large number of original samples, which is obviously not in line with Data characteristics of some uncommon crops

Images