Build on greenhouse roses disease prediction model based on genetic BP neural network

摘要

Through the application of genetic algorithms (genetic algorithm, simplified as GA) and BP(Back Propation) neural network, I built a prediction model of roses diseases, in which I choose six indicators as the input of network, they are the minimum temperature, maximum temperature, average te-mperataure, minimum humidity, maximum humidity, average humidity in the greenhoue, then I choose three diseases index as the output of the network, they are rose powdery mildew, do-wny mildew, gray mold. Using Matlab platform to complete the model, the results show that: based on genetic algorithm and BP neural network model the diseases of roses can be predicted. The average relative error is about 6.37%, so it has good effect, and compared with the traditional BP network model, the results show that the model based on genetic algorithms and BP neural network is superior to the traditional BP network model, which have higher precision and stablity.