Research Of Intelligent Inspection System For The Fire Water Supply Equipment In Changing Weather Conditions

The fire fighting equipment is directly related to people’s lives and property. Whether the fire can be extinguished in the beginning, mainly depends on the normal fire water supply equipment. Actually, the fire water supply equipment in standby state for a long time, like prone to rust pumps, equipment moisture, action not normal are often found, so inspection function is needed when fire water supply equipment put into use. To ensure that the equipment can put into operation normally when needed, the fire water supply equipment will be started regular or irregular low-frequency, performance and key parameters will be tested all the time.Fire water supply equipment often placed in pump house, ambient temperature, relative humidity, and leakage current value are the main factors for the equipment in changing weather conditions, the impact of the results and sampling methods of these three factors are discussed in this paper, and these three parameters are selected as the primary control variables. Markov theory and BP neural network theory are introduced in this paper. The establishment of methods and Markov Model transition probability calculation method and adaptive control are also discussed in this paper, the use of BP neural network theory to complete the single failure rate is proposed, the calculation method of BP neural network and back-propagation of error is analyzed. The fire water supply equipment failure rate data which obtained under constant hot and humid environment is used, by multiple linear regression and residual correction combined with the time factor to obtain the theoretical failure rate, with the theory of fuzzy control theory, divide the failure rate and the rate of change, and fuzzy rule is designed to guide the completion of inspection control. MATLAB program is written to simulate Markov- three variable control parameters of fuzzy neural network fault prediction model for the control parameters, for the use of the intelligent inspection system for the fire water supply equipment in changing weather conditions, guide the design of the system. The frequency, inspection results and the impact of fire water supply equipment which controlled by the periodic inspection, direct fuzzy control model and Markov- three variable control parameters of neural network fault prediction model are compared. By way of example, based on temperature, relative humidity, leakage current’s Markov- three variable control parameters of fuzzy neural network fault prediction model are validated. Finally, the increase in non-critical input variables, optimize prediction algorithm to further improve the prediction accuracy is mentioned.In this paper, the intelligent inspection system is designed for the fire water supply equipment in changing weather conditions, also for the use of parameters acquisition, normalization, fuzzy and neural network for the multivariable predictive control system is very useful.