Study Of Fault Tolerance Estimation For Contaminants Parameters In Cabins Based On WSN

The cabin of the aircraft is so tightly closed and the air flow in cabin is poor that the inappropriate concentration of pollutants will affect the comfort and health of the passengers.The single wired sensor measurement method to monitor environments is commonly used in aircraft cabin,which sometimes caused false alarms and a large number of unnecessary losses to human and material resources.Therefore,it is necessary to carry out a new measurement method,which can still achieve an accurate estimation of the concentration of pollutants in some emergencies.To this end,a pollutant parameters monitoring model based on wireless sensor network was established,and the related parameters state estimation algorithm from the perspective of fault tolerance was studied,which purpose is to provide an estimation algorithm with practical engineering significance for the monitoring of cabin pollutant concentration.In order to solve the problem of large measurement error in single-sensor,the output-corrected primary and secondary sensor node are used to establish a combined sensing monitoring model based on wireless sensor network.For the packet loss and path loss in the wireless sensor network,a Kalman-consensus filter algorithm based on event-triggered mechanism for state estimation was proposed.The optimal gain of the algorithm is given theoretically,and the stability condition of the algorithm is proved by Lyapunov method.After that,considering the limitation of energy consumption in the actual application environment,the model of the algorithm energy consumption is established,and the impact of the triggered threshold on energy consumption is analyzed.Finally,in order to reduce the bandwidth of signal transmission bandwidth,the measured value and estimated value of the node are quantized,and the quantized state estimation algorithm is proposed.The effectiveness of the algorithm under quantization communication is verified by simulation.Simulation results verify the effectiveness and accuracy of the above algorithms,the simulation results show that compared with the traditional local Kalman filter,the proposed suboptimal event-triggered Kalman consensus filter has the higher level of fault tolerance,and can accurately estimate the state in the environment with packet loss.In addition,the consensus error between the sensors is low,which effectively solves the problem of false alarms;Compared with the traditional Kalman consensus filter,the proposed algorithm can achieve the same accuracy under the appropriate triggered threshold,and the traditional Kalman consensus filter will diverge while the proposed algorithm can stabilizely convergence under the case of packet loss.Compared with the time-triggered algorithm,the algorithm greatly reduces the energy consumption in transmission and reduces the communication burden.Then,considering the actual bandwidth requirement,the algorithm is quantized to reduce the bandwidth pressure.