An Innovative ECA Approach For Reeled CRA Welds And Its Validation Programme

This paper presents an innovative defect growth ECA methodology for pipeline girth welds and its validation programme, applied specifically to reeling ECA of pipelines with under-matched strength welds. The ECA method is a tear-fatigue approach that accounts for the blunting limit in JR curves during pipe spooling and reel-lay. Fatigue crack growth may occur by low cycle high stress fatigue and by tearing, but the latter only if the crack tip opening displacement exceeds the blunting limit. Conventional ECA with BS7910 is limited because the weld's strength needs to be over-matched. Alternative industry methods for the application of FEA to under-matched strength welds are computationally more intensive than the presented innovative approach. Fatigue crack growth for low cycle high stress fatigue is calculated using Paris’ Law in the approach but, if the crack tip opening due to the tearing mechanism is less than the blunting limit then tearing growth is zero. With the innovative method, if the crack tip opening displacement exceeds the blunting limit then the tearing defect growth is included. Hence, the method is a combined tear-fatigue approach. Welded pipe strings were fabricated from pups composed of clad material; i.e. carbon backing steel pipe with a 3 mm layer of corrosion resistant alloy (CRA) on the inner circumference. Each test string was approximately 10.5m long and fabrication was from a mix of six 0.5m length pups in the central zone of each string and two longer end pups. Three girth welds included EDM notches for test purposes which simulated planar flaws. The notches were on the extreme tension fibre, as the test string gets pulled to the reel former in a reeling test rig. Full scale reeling simulations involved pulling the test strings up to 6 times to the reel former in a reeling test rig. Measurement of defect growth associated with the EDM notches was by scanning electron microscope (SEM), from specimen segments extracted from the test strings. Predictions of defect growth were by finite element models in combination with pipe-specific data that was the outcome of an associated small-scale test programme. Validation of the ECA-by-FEA approach is by a predictive best estimate study, for which there is excellent agreement between the measured values and the calculated defect growths. The ECA-by-FEA approach is conservative for project work, as shown by a high estimate study and an offset blunting limit study. Early development of the ECA approach was for small diameter CRA pipelines during the execution of the Guara-Lula project (Sriskandarajah et al, 2015). The presented full-scale tests, innovative defect growth measurement by scanning electron microscope and the FEA and defect growth calculations were full validation of the approach, with pipe strings that had outer diameter of 323.9mm.

Four Point Bend Test of 5LPP – Concrete Coated Pipe

The objective of this paper is to present a 4-point bend test of 5LPP (Five Layer Polypropylene) concrete coated pipe. This is the first of its kind of bend test for a complex coating combination of 5LPP and concrete layers. The bend tests have been carried out to simulate S-Lay installation loading conditions to assess the coating integrity of the pipeline during installation. This paper reports the test arrangements including instrumentation, load schedule, test procedure and the challenges involved. Finally, the preliminary results and conclusions of the tests are documented. Two separate full scale four-point bend tests are carried out to study the behavior of the 5LPP concrete coated pipe. The purpose of the first test is to understand the complex behavior of the 5LPP/CWC coated test pipe and validate previously made industry standard assumptions regarding the calculated coated joint stiffness. The purpose of the second test is to observe the coating integrity of the test joint and slippage behavior due to the simulated installation conditions (overbend and sagbend bending moments and/or corresponding curvatures). The nonlinear moment-curvature for the concrete coated pipe is estimated based on an analytical approach taking into consideration plane bending theory and slippage behavior of the coating layers. The moment-curvature is used to prepare the load schedule for the tests. The test string consists of a test joint (40ft) welded to half joints at the ends. The bend test is performed using industry established full scale 4-point bend test arrangements. A global finite element model is used to simulate the tests using the analytical moment-curvature of the concrete coated pipe. The stiffness of the test pipe is calculated using the first bend test and compared against the analytical stiffness. The second test is carried out by applying loads corresponding to an estimated maximum overbend bending moment and then the test string is unloaded and rebent in opposite direction by applying loads corresponding to an estimated maximum sagbend bending moment. The results of the second test are documented at each load step and the integrity of the coating is measured against specified concrete coating damage criteria for tension as well as compression. Finally, field observations from the actual installation operation are compared against the bend test results. Conclusions are presented to address various aspects of concrete coated pipe for S-Lay installations.

Chinese Connected Word Speech Recognition Based on Derivative Dynamic Time Warping

The algorithm of derivative dynamic time warping (DDTW) can overcome the shortcoming of dynamic time warping (DTW) and the computational complexity has not increased. In this paper, the algorithm of DDTW was applied to Chinese connected word speech recognition. For each isolated word, as an independent reference template and as basic recognition unit, there was an independent reference template to correspond; the matching between some word of the test string and a reference template was done by the DDTW, and the reference string which had the minimum cumulative distance was as output. The experimental results show that our method is obviously superior to all the methods based on DTW, and the recognition rate has reached 90%.