Study On The Method Based On Wavelet Transform Theory To Inhibit Geomagnetic Disturbance Imposed By Railway

Under the remarkable supporting and funding of our country,the network of geomagnetic observatory is developing day to day, as a result of which the observed geomanetic materials are getting richer and richer.However,as China is experiencing the period of rapid economic development,the construction and operation of electrical facilities have caused an increasingly serious disturbance to the geomagnetic observation.As a serious consequence,some geomagnetic stations are forced to be relocated and reconstructed.Terefore,it is an urgent task that we should identify the characteriscs of various typical disturbing sources and seek effective methods to inhibit these geomagnetic disturbance.We need not only the traditional analytical methods but also more sophisticated methods to identify and inhibit the man-made disturbance in order to improve the data quality of the observation. Moreover the improved data would provide more reliable material to be applied for further understanding the geophysical phenomena related to geomagnetic changes.The characteriscs of various typical sources that have disturbing effects on the geomagnetic observation stations were studied systematically.Combined with the theory of electromagnetic induction law (i.e. Biot-Savart Law), the mechanic model was formed to descript the disturbing effect on the geomagnetic observing imposed by the railway disturbance. Based on the modern signal processing theory, the power spectra estimation theory were elaborated in detail and then made some computer simulations. Simulation of the power spectrum estimation analysis showed that the classical power spectra estimation and the parametric model power spectra estimation both have their own advantages and disadvantages each. Thus the two spectral estimation methods have their own scope of application, of which the former one is more suitable for harmonic analysis, the latter one is more suitable for stationary signal analysis. The observed geomagnetic data is regarded as the superposition of the geomagnetic diurnal variation and the man-made disturbance. The quiet daily variation field is relatively stable, which can be fitted out with four harmonic waves whose periods are 24 hours, 12 hours, 8 hours and 6 hours respectively. As it's a non-stationary harmonic signal, it's more suitable to be analyzed with the classical power spectrum estimation method.The wavelet transform theory was elaboratedly discussed in detail and then made some computer simulations.The simulation analysis of the wavelet transform showed that the wavelet has unparalleled advantages in the non-stationary denoising where there exists sudden mutation and peak. Therefore the wavelet transform has more broad applicating value in the practical engineering applications.In this study, based on the data selected from Jinghai geomagnetic station, Nanjing geomagnetic station and Shanghai geomagnetic station which were all imposed with typical railway disturbances and Quanzhou geomagnetic station which was introduced as reference station in May 2008,the classical power spectrum estimation and wavelet transform method were brought forward and combined together. firstly, the observed data from stations with railway disturbance and reference station were executed independantly.The comparative analysis of the spectral characteristics was made to reveal the dominant range of frequency bands of the railway disturbance. Then the wavelet transform method was used to process the corresponding frequency bands so as to achieve a better inhibiting effect of the railway disturbance.The Welch approach of power spectrum estimation is used to make comparative analysis. It's revealed that the railway disturbance mainly impacted the Z-component and D-component of the geomagnetic field in the High-frequency band unanimously, especially in the bandwidth of 0.004～0.03Hz. On this basis, the wavelet transform was applied to inhibit the railway disturbance imposed on the Z-component and D-component. Then the comparative anaysis was made to study the differece of the waveform, the power spectra and the correlation coefficient between the original data and the processed data.In this thesis,the preliminary results of the work suggested that our method is effective to inhibit the railway disturbance.