Sensing Study Of Thermo-Optic Oscillation In Optical Microcavities

The optical cavity of the whispering gallery mode has an ultra-high quality factor and a small mode volume.Therefore,the optical biochemical sensor based on such optical microcavity has an ultra-high detection sensitivity,which makes the sensor have an important application prospect.In this paper,the nature of the resonance state of the microcavity changes when the substance to be tested changes,and a sensing mechanism using the heat dissipation rate as the sensing index is proposed.Combined with the advanced parameter measurement method of artificial neural network,it can effectively extract sensor information.The parameter measurement accuracy is high and the acquisition of the entire transmission process can be avoided,enabling low-cost detection.When the CaF2 optical cavity is swept by the probe laser,a significant optically-thermal oscillation phenomenon is observed on the transmitted waveform.The reason for this oscillation is the interaction between the thermal expansion effect,the photothermal effect,and the Kerr effect.In this paper,the theoretical model of the optically-thermal oscillation effect is established and theoretical analysis shows that the oscillation period value on the optically-thermal oscillation is related to the heat dissipation rate.Heat dissipation rate is defined as the rate at which heat is transferred from the optical microcavity to the environment surrounding the cavity.As the nature of the substance to be tested changes,the heat dissipation rate changes accordingly,which manifests itself as a change in the oscillation period during the transmission.When the value of the heat dissipation rate increases,the number of oscillations gradually decreases,and finally no oscillation occurs.By using this feature,the value of the heat dissipation rate at this time can be determined according to the transmission process,thereby achieving the purpose of high performance sensing.However,multiple oscillation periods have a nonlinear relationship with the heat dissipation rate,and the heat dissipation rate cannot be effectively measured by a certain oscillation period value.Therefore,this paper proposes an artificial neural network,which is an advanced nonlinear parameter measurement method.In this paper,two artificial neural networks are used,which are BP neural network and GRNN neural network.The transmission data measurement modelsbased on these two neural networks are built.The corresponding thermal dissipation rate is measured by inputting the optically-thermal oscillation period value.The numerical simulation results show that the neural network-based measurement model can effectively reflect the trend of heat dissipation rate and the measurement accuracy is high.At the same time,by comparing the performance of these two neural networks,it is found that the GRNN neural network not only has a simpler optimization process,but also has better measurement performance.