Identification Of Bearing Faults In Rail Vehicles Based On Chaos Theory

The railway is an important means of transportation in China.When mechanical equipment fails,it should promptly discover the fault and issue an early warning and repair or replace the faulty equipment to avoid serious safety accidents and economic losses.However,at the initial stage of the fault,the fault signal is very weak,and the traditional signal processing method cannot detect the weak fault signal.This paper designs the Duffing oscillator detection system based on chaos theory.The fault detection of rolling bearings is taken as an example to realize the detection of weak signals.First,we study the characteristics of the Holmes-Duffing oscillator.The chaotic system is very sensitive to parameters,and the system is turned into a critical state,which makes the system sensitive to weak signals with the same frequency in phase.By studying the influence of noise on the chaotic system,it is found that the Duffing oscillator has strong anti-interference ability to Gaussian white noise.At the same time,the intermittent chaos phenomenon of the system and the phase blind area of the detection system are found.Second,the Duffing oscillator is improved by combining the characteristics of the actual signal.The Duffing equation is transformed on the time scale to achieve weak signal detection at any frequency,and the detection efficiency is improved.It is proposed to use inverse phase change principle for signal detection.This method can more accurately determine whether the system has undergone phase change and improve the accuracy of detection.A phase shift method is used to eliminate the detection phase blind zone.This method increases the detection range and improves the detection efficiency.An improved simulation model of Duffing oscillator detection system is established.The simulation results show that the detection system can effectively detect weak signals and is immune to noise and other frequency signals.Finally,the detection system will be applied to the fault signal detection of rolling bearings.The fault signals of the outer ring,inner ring and rolling element of the rolling bearing are respectively tested.The test results show that the improved Duffing vibrator detection system can effectively detect the fault signal.