A Theory for Multiresolution Signal Decomposition: The Wavelet Representation
IEEE Transactions on Pattern Analysis and Machine Intelligence
Integrated Computer-Aided Engineering
| Hi-index | 0.00 |
Corrosion occurs frequently at the interior or exterior of steel casing pipes used in underground gas storage or oil fields. It causes not only production loss but also environmental pollution. Hence, corrosion monitoring and early detection of corrosion pits and wall thinning on casing pipes are considerably important to the gas and petroleum industry. We introduce in this paper a new version of the direct current electromagnetic induction system which is capable of detecting small, isolated corrosion pits and holes on the casing wall. A sensor system based on the direct current electromagnetic induction instrument is coupled to an updated data acquisition system. Information represented by enormous data sensed by the downhole instruments is to be transmitted to the computer logging system on surface, and a wavelet data compression technique is incorporated. A complete internal/external circumferential survey of corrosion extent on casing pipes can be obtained. Unlike the conventional direct current induction instrument, this system provides all the measured flux leakage signals and eddy current signals, and hence a full signature logging response. It allows an accurate differentiation of the casing pipe hardware from corrosion, and much easier interpretation of corrosions on casing pipes. To achieve the real time mode during the logging operation, a VLSI integrated circuit realizing the wavelet transform has been designed. Simulations have been conducted, and the integrated circuit has been fabricated. The test results show that the design is advantageous and the circuit performance is satisfying.