Study On Pipeline Magnetism-Leakage Detection Real-time Data Collection

Safety inspection of ferromagnetic oil vapor pipelines has been emphasized much more due to the development of online pipeline detection technology, whilst pipeline magnetism-leakage detection being such an effective technology. However, the huge amount of original signal data generated during detection, with higher system resolution coexisting, contribute to the exponential data size growth, which obstructs the signal storage and process. Thus for the sake of higher detecting precision as well as the applicable expansibility of the system, also the lesser system pressure, appropriate compression arithmetic, stylized by the data feature of pipeline magnetism-leakage, is indispensable.It is the integral exaltation wavelet coding for signal compression of the pipeline magnetism-leakage detection system that put forward on the basis of a good many summary of data compression scheme both at home and abroad. Wavelet signal decomposition is an adept and effective conveyance, meanwhile the integral exaltation wavelet, being called the 2nd generation wavelet, possesses the identity of high compression efficiency. Pipeline magnetism-leakage data will be turned into integral wavelet coefficients by way of integral exaltation wavelet, and then the decomposed wavelet coefficients are compressed via self-adaptation arithmetic coding. It is the betterment of Huffman coding and approach to the lossless compression coding limit. Self-adaptation updates probability estimate according to current and coded symbols one at a time, consequently probability statistics goes per signal, and algorithm effects, and real-time meets the system needs.Dual buffer system structure is designed for real-time collection, compression and storage of pipeline magnetism-leakage data. Based on that structure, the compression algorithm, of integral exaltation wavelet variation and self-adaptation arithmetic coding combined, processes the pipeline magnetism-leakage data lossless and real-time, with ratio of 9.166:1. And the feasibility is experimentally confirmed.