Research On Detection Algorithms Of Weak Resonant Signal Of Resonator Based On Chaotic Oscillator

Resonant micro-cantilever sensors are widely used in environmental monitoring,quality control,food safety and other fields.Since its output signals are weak and susceptible to interference from noise and other factors,it is difficult to reliably extract the weak resonant signal output by the resonator.Whereas the Duffing chaotic oscillator is sensitive to weak signals and immune to noise,so it is often used for weak signals detection.In this paper,the weak resonant signal detection algorithms based on the Duffing oscillator is studied.The Duffing oscillator detection system and the Lyapunov exponent algorithm for state discrimination are studied.The system state is judged by the change of the maximum Lyapunov exponent sign,and then the threshold of the system is determined.The forward plus algorithm and the reverse minus algorithm for the weak resonant signal peak detection are studied.And the theoretical amplitude-frequency characteristic curve model is established.The accuracy of the two algorithms for detecting the weak resonant signal under different noise levels is tested.The amplitude-frequency characteristic curve actually collected is used as the detection model and the two algorithms are experimentally verified.It turns out that the minus algorithm is superior to the plus algorithm.The minus algorithm can detect signals with larger amplitude and its detection accuracy is higher than ±3Hz error of the open-loop detection.By modifying the coefficient of the nonlinear restoring force term of the Duffing equation,the signal-tonoise ratio of the system is improved,so that the system with noise variance 0.01 can effectively detect the resonant signal and improve the system detection performance.The theoretical phase frequency characteristic curve model is established.The thermal phase drift of the vibration signal and its influence on the detection system are analyzed.Under the condition of no frequency deviation,a phase oscillator array is constructed to detect the 100 k Hz resonant signal with thermal phase drift.By reducing the array range of phase oscillators to-90° ± 47°,the problem of detecting the phase shift of the resonant beam under 35 ?m thickness is solved.Under the condition of frequency deviation-40 Hz,the problem of detection failure caused by frequency deviation is improved by constructing the frequency offset oscillator array.Based on the detection model of the theoretical phase-frequency characteristic curve,simulation tests of the resonant signal of the phase and frequency offset are carried out and the system detection performance is improved by amplitude reduction method.The results show that the weak resonant signal detection algorithms based on the Duffing oscillator can effectively extract the weak resonant signal output by the resonator.In addition,because the system has good anti-noise performance and high detection accuracy,there is a good application value in the field of resonator signal detection.