Safety early warning method based on multiple physical parameters for intelligent early warning security and protection of energy storage system

The invention belongs to the technical field of safety early warning, and particularly relates to a multi-physical-parameter-based safety early warning method for intelligent early warning security and protection of an energy storage system. The multi-physical-parameter-based safety early warning method for intelligent early warning security and protection of the energy storage system belongs to the technical field of safety early warning. The invention provides a security early warning method based on multiple physical parameters, which is used for intelligent early warning security and protection of an energy storage system. The method comprises a thermal runaway early warning step; a thermal runaway early warning step: the central controller determines an input layer neuron vector x = {x1, x2, x3, x4, x5} of the five-layer BP neural network in two continuous sampling periods; wherein x1 is the air pressure value in the battery module, and x2 is the concentration of the VOC gas in the battery module; x3 is