Research On Optimization Of Partial Discharge Acoustic Signal Demodulation Algorithm Based On DFB-FL

Electrical equipment insulation faults are often accompanied by partial discharge,and the detection of partial discharge can realize the monitoring of the insulation state.The fiber optic acoustic emission(AE)detection method has attracted much attention from researchers due to its advantages such as high sensitivity and strong anti-interference ability.However,the fiber optic partial discharge detection method still has problems such as low signal-to-noise ratio and difficulty in separating the sensing unit and the demodulation unit.Distributed feedback fiber laser(DFB-FL)has the advantages of small size of sensing unit and stable output of narrow linewidth lasers,which is suitable for fiber optic AE detection.The acoustic signal generated by the partial discharge will cause a change in the center wavelength of the DFB-FL emission laser,and this change needs to be demodulated by the optical fiber interferometer to realize detection.Based on the above background,the following research is conducted on the application of DFB-FL in partial discharge detection:The mathematical model for DFB-FL acoustic sensing is derived.Based on the analysis of the principle of optical fiber interferometer demodulation,the limitations of the 2×2 coupler are described,and the reason for selecting the 3×3 coupler for the signal demodulation system in this study is explained.On the basis of explaining the principle of the signal demodulation algorithm matched with the 3×3 coupler,the limitations of traditional Naval Postgraduate School(NPS)demodulation algorithm are explained,and an optimization method is proposed.The traditional NPS algorithm and the improved NPS algorithm are compared and analyzed by simulation.The simulation results show that non-ideal working state will cause distortion of the demodulated waveform of the traditional NPS algorithm,while the improved algorithm can overcome the influence of non-strict symmetry of the coupler and optical power fluctuations,and obtain consistent demodulation results with the waveform of the signal to be demodulated.In order to verify the difference in the demodulation ability of the actual acoustic signal to be measured before and after algorithm optimization,the results obtained by the two algorithms in each step of the actual acoustic signal demodulation are compared.The experimental results show that for continuous and instantaneous acoustic signals,the improved NPS algorithm can demodulate the measured signal,while the demodulation results of the traditional NPS algorithm will be severely distorted.In order to clarify the advantages of the improved NPS algorithm on the detection of partial discharge acoustic signals,a partial discharge detection experimental platform is built,and the detection results of partial discharge acoustic signals are compared by three methods: improved NPS algorithm,traditional NPS algorithm,and intensity demodulation.The experimental results show that the improved NPS algorithm has higher detection sensitivity and higher signal-to-noise ratio.