Modeling And Control Of Water Jet For Forest Fire Truck

Fire spreads quickly and has strong destructive power.Especially,it threats the life the fire-fighters' safety.It is of great significance to realize the automatic control of fire suppression to improve the efficiency and accuracy of fire-fighting,and to ensure the safety of fire fighters.In this paper,the water jet of fire-fighting lance is taken as the research object.The landing position prediction model and control method of fire-fighting lance which are based on genetic BP neural network are proposed.The main research content and conclusions are as follows:1.A water jet experimental system based on forest fire truck was established.The experimental platform was mainly used to collect the data of the working condition,the external environment parameters and the data of the water landing point in order to provide the basis for the model establishment.The experimental platform mainly included:fire-fighting lance,holder,lifting platform,several sensors and computer.In the experimental platform,the construction of the equipment needed to complete the design of the pipeline,the connection of the circuit and the arrangement of the sensors.The software part needed to solve the problems of the data transmission,data processing and the control of the main equipment.Experimental scheme was proposed based on the experimental platform.2.The influence factors of water jet were analyzed by statistical method at first.Then,the appropriate input variables of the prediction model of water jet were selected.BP neural network prediction model and genetic algorithm optimization BP neural network prediction model(GA-BP neural network prediction model)were constructed,and the two models were compared through experiments.The experimental results showed that the genetic algorithm can effectively optimize the BP neural network,and the prediction accuracy of GA-BP neural network prediction model was improved.In GA-BP prediction model.the average error distance of each group was about 0.43m.the average relative error was about 6.5%and the best average relative error can reach 5.69%.The proposed GA-BP prediction model was able to predict the landing point of water jet effectively.3.The attitude control method of the fire-fighting lance based on the landing position prediction model of the water jet was studied,and the PID controller is designed.Based on the established model,the predicted coordinate placement and fire positon were compared.The variation of the fire-fighting lance attitude was calculated by using the deviation,and the working condition parameters of the fire-fighting equipment were adjusted.After several adjustments,the predicted landing point of water jet was close to the fire position,so as to accomplish the automatic orientating fire extinguishing.The simulation results showed that the proposed control method can adjust the attitude of the water gun in 10s to achieve the state that the deviation between the predicted point of water jet and the fire position was within 0.1m.