Application Of HHT Method In GAT Maglev Gyroscope Data Processing

With the progress of science and technology, gyroscope technology is playing a more andmore important role in underground engineering surveying and inertial navigation. Extensiveuse demands the higher accuracy. Because it’s hard to make a breakthrough on hardware, sowe put the goal of improving precision of gyroscope in data processing.In view of predecessors have used a variety of kalman filtering method to data processing,and in order to try to use different method to analyze gyroscope data, the author used a newmethod–Hilbert Huang method to analyze the gyroscope data. The article research contentmainly includes the data denosing and the estimate of the rotor current cycle.The data denoising is based on the EMD bandpass filter denoising principle. Using theEMD method respectively decomposes the rotor current and the stator current. It’ll producethe IMFS from high frequency to low frequency. Filtering the IMF from high frequency tolow frequency one by one, and reconstructing the remain IMFS while calculating the meanvalue and mean variance of residual signal. Then we can draw the image after filtering,determine the distribution of noise’s IMFS and will know the EMD filtering result. Byclassified studying large of datas, the EMD denoising method achieved ideal result.Based on the IMF cycle to calculate the original data cycle is a new method of calculatingcycle put forward by author. It’s based on the cycle of useful components’ IMFs and variancevalues, the IMF’s cycle will be a useful composition of the value of variance as a right. Inaccordance with the rights, do a weighted average to the IMF cycle of useful components,making an estimate of the original data. Through extracting the extreme value point offiltered image and calculating the extreme value point cycle, this new IMF cycle estimationmethod has been proved to have certain reliability.