Research on Fire-Engine Pressure Balance Control System Based Upon Neural Networks

摘要

The pressure produced by the water coming out of the fire-engine pump outlet is controlled by the rotate speed of the fire pump. However, this RS is controlled through fire-engine accelerator voltage which is controlled by the ECU. In order to control and keep the fire-engine pressure balanced, it is necessary to take pressure, rotate speed and current rate as input parameters and control voltage as output parameter through BP neural network control system. Related researches indicate that BP neural network is appropriate for building the system whose target is to keep the pressure balanced. And, some modification can be done to the standard BP neural network algorithm. These modified BP neural network algorithms are BP neural network with momentum factors and self-adapting learning speed which can improve the response speed and performance of this control system dramatically.