Fast Processing Method Realization Of Weak Signal Of Optienl Current Transformer

With the development of smart power grid,the development prospect of Optical Current Transducer(OCT)in the field of power system measurement is getting wider and wider.However,a large number of random noises greatly affect the accuracy of OCT measurement,and smart power grid needs the real-time signal measurement.Therefore,rapid denoising of OCT weak signals is studied in this paper,aiming at improving the accuracy and real-time performance of the measurement.Firstly,on the basis of OCT principle,the source and composition of OCT noise are analyzed and studied.The results show that the noise of photodetector is mainly composed of granular noise,thermal noise,noise and generation-compound noise.The main noise source models of OCT were established,the factors affecting the noise were analyzed,and the output signals generated in the photoelectric conversion were collected.Then,from the perspectives of time domain and frequency domain,the problems such as complex noise distribution,overlapping frequency band with the measured signal,low signal-to-noise ratio,and difficulty in obtaining accurate components were analyzed.Aiming at the intrinsic DC component of OCT,an anti-shaking Kalman filtering algorithm is proposed based on signal-noise analysis.Firstly,the DC component state equation of OCT is established,and the noise recursive estimation and anti-oscillation function are introduced into the standard Kalman filtering algorithm to predict and correct the noise variance and overcome the defect of serious jitter in amplitude under low SNR.The stability of anti-oscillation Kalman filtering is analyzed by Lyapunov function.By writing a Python program of the proposed algorithm,the filtering effect of anti-shaking Kalman filtering algorithm and standard Kalman filtering algorithm is compared and analyzed.The calculation results show that the proposed filtering algorithm can effectively improve the measurement accuracy.A variational Kalman filter algorithm based on variational mode decomposition is proposed for weak AC signals.Firstly,a state model of AC signal measurement of OCT is established.Variational mode decomposition is combined with anti-shaking Kalman filtering algorithm to estimate the state parameters of signals from the perspective of time domain and frequency domain,and the real-time performance of variational Kalman filtering algorithm is analyzed.The Python program of the proposed algorithm is used to compare and analyze the filtering effects of the proposed algorithm and Central Difference Kalman filter(CDKF)on OCT AC signals.The calculation results show that the proposed algorithm can effectively improve the signal-to-noise ratio of OCT.Based on the compressed sensing theory,the parallelism of the above algorithms is optimized.The sampling strategy of variational Kalman filter is optimized.The filtering gain and error variance matrix are decoupled to improve the calculation speed,and the state estimates of the intrinsic DC component and weak signal are calculated in parallel.On this basis,we design the noise variance prediction correction module,anti-shaking module,frequency updating module,etc.,so that the algorithm can realize the parallel running of calculation under the requirement of accuracy.Based on Lab VIEW FPGA hardware,the proposed algorithm is implemented in parallel on the hardware platform.Finally,an experimental platform is built and the proposed algorithm is applied to OCT signal processing,and the effectiveness of the proposed algorithm is verified.The results show that the proposed signal processing algorithm improves the measurement accuracy of OCT and meets the real-time requirements.