Minggu, 02 Februari 2014

Magnetic Flux Leakage Device for Offshore Oil Pipeline Defect Inspection

To ensure the safety of oil transmission, routine inspections of offshore oil pipelines are essential to locate defects or corrosion. This article describes a test device for offshore oil pipeline defect inspection, based on magnetic flux leakage (MFL). The hardware and signal-processing system are described. The inspection device includes an MFL detector, driver robot, data-acquisition system, signal processing unit, and power unit. The device can be used for multi-radius pipelines and various work conditions.

Pipeline safety evaluation is an important issue in the world today. Corrosion, stresses, and mechanical damage of oil and gas pipelines can cause catastrophic failures. To ensure their integrity, the pipelines are periodically inspected using an inspection robot or "pig." The magnetic flux leakage (MFL) method can be used to measure and locate cracks in both circumferential and axial directions, and is a popular method of pipeline inspection.1 This article describes a test device for offshore oil pipeline defect inspection based on MFL.
Description of the Test Device
MFL nondestructive testing is an electromagnetic technique that can be used for testing conducting material; it originates from Michael Faraday's discovery of electromagnetic induction in 1831. MFL utilizes permanent magnets to magnetize a sample to saturation; the magnetic flux distributes equally in the sample when there is no defect. In the regions of reduced thickness, such as a corrosion defect or crack, the magnetic flux leaks into the air.2 This leakage flux, which is correlated with the size and location of the defect, can be detected by a magneto sensor. One can analyze pipeline defects from the leakage flux signals.
Figure 1 shows the structure schematic diagram of offshore pipeline MFL inspection. The MFL inspection device has six parts: a driver robot, a system controller, power supply, MFL sensors, an MFL data processor, and a position tracer.3 Figure 2 shows a simplified structure of the MFL inspection device. It includes a pressure tractor, MFL inspection unit, MFL signal processing unit, and power unit. In this simplified device, the driving force is fluid pressure, but usually the device is propelled by a driver robot (Figure 3). The three groups of driving wheels are oriented around the pipeline at 120 degrees.
For a 297-mm-diameter offshore oil pipeline, the MFL detector comprises 16 testing units placed equally around the pipeline circumference (Figure 4). Each unit includes a magnetic path, seven sensors, and a sensor box. The magnetic path includes a permanent magnet, a magnetizer, and a steel brush. Sensors fixed to the printed cicuit board are enclosed in sensor boxes. These sensor boxes are waterproof, oilproof, and can withstand high pressure (4 MPa). The magnetizer material is ingot iron and the permanent magnet is made from NdFeB material. In the inspection process, the permanent magnet provides the magnetic flux. Finite element analysis indicates that the main influence factor is the permanent magnet length when the permanent magnet has limited volume. In this device, the length and width of the permanent magnet are 48.1 and 23.2 mm for a 297-mm inner diameter pipeline. The dimensions are 44.5 and 20.4 mm for a 247-mm diameter pipeline.
Operation of the Test Device
Many researchers indicate that the air gap lift-off value of an MFL test device is an important factor in its sensitivity. Since the lift-off value has an uncertain relationship to the system output, one cannot use a mathematical equation to express it. Based on finite element analysis, one can obtain the best air gap lift-off value for a specific test device. Considering the movement of the test device in the pipeline and inspection sensitivity, 3 mm was selected as the test device air gap lift-off value.
Figure 5(a) shows the MFL testing circuit. The Hall[dagger] sensors record the remote field signal. Because of the low signal level expected, a high gain amplifier is required. Each Hall sensor inspection signal is coupled to the input of an instrument amplifier (AD623). To improve the system precision and eliminate the influence of temperature, a temperature measurement circuit is included (Figure 5[b]). The major circuit is a two-terminal temperature transducer (AD590), and its output current is proportional to the absolute temperature. The output current of AD590 is converted to a voltage signal by resistance R5 and then is enlarged 50 times in the amplifier in-phase port.

Figure 6 shows the MFL inspection data-acquisition function. The peripheral component interconnect (PCI) board mainly includes analog-to-digital (A/D) circuit, 32-channel analog input channels, digital signal processing (DSP), field programmable gate array (FPGA), a PCI controller with configuration chip, and a three-line SyncBus[dagger] for synchronized multiboard operation. Input signals include MFL signals, temperature data, and location data.
Because these devices are digital systems operating with process-specific software, all analog signals must be converted to digital numbers before a computer can read them. A/D conversion in a control system is performed by boards or boxes called "analog peripherals."4 They connect to the central processing unit via the system's back-plane bus. Each inspection channel can be programmed independently for one of the standard input ranges (0 to 5 V, 10 V, or 20 V) while operating from a single 5 V supply. MFL inspection signals are filtered in DSP module, and then compressed with other data in FPGA.5 A central computer controls the data-acquisition process through PCI9054 and stores pipeline inspection data in the storage device. The A/D converter sampling precision and sampling frequency are, respectively, 12 bit and 1.25 M/s.
The noise signal that is produced by seamless steel pipe is similar in shape and amplitude to a defect signal. One can adopt an adaptive filtering algorithm, however. Wavelet theory and Huffman coding are utilized to compress MFL inspection data.6 The MFL inspection data are then stored and analyzed in the computer. Figure 7 shows the signal approaching curve and depth approaching curve from a 15 by 12 by 37.1 mm^sup 2^ defect with noise through 2,500 steps of operation.
The pipeline inspection device speed is -0.5 Km/h, the operating temperature is from 5 to 80°C, and the maximum pressure is 400 MPa. The permanent magnet remanence density is 1.07 T, the intrinsic coercive force is 1,592 kA/m, and the maximum magnetic energy product is 239 KJ/m^sup 3^. The maximum permeability of the magnetizer is 12,000 μm or greater. The Hall sensor nonlinearity and temperature coefficient are separately 0.1 to 1.0% and -0.06%/°C.
Refinements to be made include analyzing the effects of the magnetic circuit configuration, improving the data sampling speed, and characterizing the MFL signals associated with common types of defects.

Sumber : Jin, TaoQue PeiwenZhengsu, Tao. "Magnetic Flux Leakage Device for Offshore Oil Pipeline Defect Inspection". 28 Januari 2014. http://search.proquest.com/docview/222951766?accountid=31562

Tidak ada komentar:

Posting Komentar