Research on Food Security Risk Assessment and Early Warning in China Based on BP Neural Network Model

China has always attached great importance to food security issues; especially in today’s changeable world, it is particularly important build a feasible and accurate early warning system. According the influencing factors security, this paper uses PCA method AHP construct index system that includes 4 secondary indicators 13 tertiary of total trade ecological security. There are four levels no warning, light moderate heavy finally comprehensive evaluation from 2000 2019 specific various obtained. This constructs perspective data, breaks through limitations existing research, improves completeness indicators. Because BP neural network multilayer feedforward with strong adaptability, one most widely used successful models at present. Finally, simulate China’s design standardized risk prevention control processes classified response strategies—routine monitoring, control, emergency response—to provide signal guidance reference for respond risks early.