Comprehensive Casing Corrosion Inspection Using Multi-Frequency Array EM Technology

Abstract

Abstract Historically, only total metal thickness measurement was possible using frequency domain electromagnetic (EM) logging tools. With advances in technology, it is critical to develop a frequency domain alternative answer using a multi-frequency array EM pipe inspection tool to accurately estimate the individual wall thicknesses of as many as five concentric pipes. Results from yard testing in a special design mockup, as well as field logs, are demonstrated as part of the technology assessment process. The new multi-frequency array EM tool uses the eddy current principle and includes two transmitters and eight receivers. It operates in continuous wave mode at multiple frequencies. Optimized transmitter-receiver spacing configurations and multi-frequency operation provide sufficiently diverse information to help assess the metal loss in each individual pipe for a wide range of configurations. The tool uses a sophisticated workflow of data processing and inversion algorithms to decouple individual thickness information from the measured data. The capabilities of the tool are demonstrated using two 400-ft long pipe mockups, each having 18 different combinations of overlapping and non-overlapping defects in five-, four-, and three-pipe sections. The configurations of the pipes used in the mockups were chosen to cover typical well completions commonly used in the Middle East. Data from the mockups are validated using synthetic data generated using two-dimensional (2D) computer models. The tool has delivered unprecedented accurate assessments of the fourth and fifth pipes, as well as an accurate assessment of the commonly evaluated first, second, and third pipes. The sensitivity of the inversion to model mismatches, such as those introduced by decentralized pipes, is studied by deliberately decentralizing one of the mockup pipes over a length of the log. An algorithm designed to correct for pipe eccentricity is also demonstrated. The results from the surface testing are discussed along with the performance of the tool in a test well with pipe configurations similar to the mockups. In the studied test well, the tool was able to identify defects in the outermost strings. This solution utilizes a novel inversion algorithm of big data from multi-frequency array sensors to derive individual pipes corrosion. This technology can significantly improve proactive decision making for mature well operations, especially in areas with high corrosion rates and shallow outer casing corrosion.