Aiming at the local stress concentration problem at the large‑section rectangular inlet of the double‑inlet decoking drums with a diameter of 3 400 mm used in a 1.2 million tons per year ethylene cracking unit，the finite element calculation model of the decoking tank cylinder and the large‑sized double tangential inlet nozzle was established by using the Ansys software.The stress distribution of the large‑size tangential rectangular inlet pipe was obtained under various scenarios of rib thickness and spacing.The analysis primarily investigates the effect of rib thickness and spacing on the stress at the connection between the decoking drums and the tangential inlet pipe.The results indicate that the optimized rib thickness is 32 mm，with a spacing of 280 mm.Upon applying the actual combined loads encountered in engineering applications to the dual tangential inlet drum and assessing its strength according to pressure vessel design standards under both design and operational conditions，it is confirmed that the strength of the optimized drum meets the required standards.The research results provide a reliable basis for the engineering application of dual tangential inlet coke cleaning tanks.

Addressing the critical issue that defects such as pits and cracks significantly reduce the load‑bearing capacity of pressure vessels and lead to safety accidents under harsh operating conditions involving high temperature，high pressure，and corrosive media，this study employs the linear matching method（LMM） to systematically analyze the influence mechanism of geometric characteristics of pitting defects on the fatigue behavior of high‑temperature pressure‑bearing structures.The results indicate that local thinned pits significantly reduce the reverse plastic limit，with the axial local thinned pit having the most significant impact，while the limit load of elliptical pits is largely unaffected by pit size.Under high‑temperature environments，the axial local thinned pit increases the total strain range by 53.8%，whereas the circumferential local thinned pit increases it by only 18.1%.The total strain range of elliptical pits is significantly influenced by pit size，increasing with the short axis（axial dimension） and decreasing with the long axis（circumferential dimension）.This study provides a theoretical reference for extending equipment service life and improving production efficiency.

For the mechanical analysis of the D‑shaped flanged convex head structure，three analytical methods are adopted in this paper：GB 150/ASME Ⅷ-1（Method 1），the method from ASME Ⅷ-2 based on the improved approximate solution theory for spherical shells approximated as cylindrical shells（Method 2），and the proposed method based on the approximate solution theory for spherical shells approximated as shallow shells（Method 3）.A comparative study is carried out among the three analytical solutions and the finite element analysis（FEA） results.The results show that for flange stress，the errors between Method 1 and FEA range from 82.32% to 130.73%，those between Method 2 and FEA range from 3.99% to 5.19%，and those between Method 3 and FEA range from 7.09% to 9.21%.For the edge stress of the spherical head（not calculated by Method 1），the errors relative to FEA range from 3.16% to 7.73% for Method 2 and from 0.4% to 6.24% for Method 3.

To address the problem of premature aging failure of some domestic P91 pipes in process industry due to excessive aging rates under high‑temperature and high‑pressure environments，a scheme is proposed to identify abnormal aging rates based on the distribution state of grain boundary misorientation angles from electron backscatter diffraction（EBSD） testing.The results indicate that high‑mobility high‑angle grain boundaries generated when the high‑temperature tempering temperature exceeds the actual heating phase equilibrium temperature Ac_{\mathrm{1}} are one of the causes of abnormal aging rates in P91.When such abnormal aging rates occur，abnormal “spikes” appear near 30° and 45° on the EBSD grain boundary misorientation angle distribution map.Based on the above characteristics，the problem of abnormal aging rates in P91 caused by excessively high high‑temperature tempering temperatures can be identified.This study provides a reference for improving the manufacturing quality inspection of P91 pipes and avoiding premature aging failure.

To address the issues of insufficient generalization ability and low leakage detection accuracy of models under small sample conditions for oil and gas pipelines，a retrospective prototype network（VRPN） based on learnable variational mode decomposition was proposed and applied to the leakage detection of oil and gas pipelines.The end‑to‑end optimization of mode decomposition and dynamic weight allocation were achieved by using learnable convolution kernels and adaptive parameters.Introducing a double‑layer Inception multi‑scale convolutional branch to capture multi‑dimensional signals and extract more precise feature information.Introducing retrospective prototype classification into the historical prototype buffer，the integration of current and historical prototype network to suppress vibration，improve the identification accuracy.A joint optimization objective function of cross‑entropy loss and prototype separation loss was designed to enhance training stability and efficiency.Experiments show that the proposed VRPN network framework performs excellently under limited sample and cross condition scenarios，achieving an average accuracy of 98.50%，which is 0.67% higher than that of ProtoNet.The research can provide a new effective solution for oil and gas pipeline leakage detection.