Because thermowells are prone to fatigue damage in petroleum cracking gas pipelines, in this paper, the LES method is used to simulate the flow around thermowells through two‐way thermal‐fluid‐solid coupling, the internal causes of thermowell damage are explored, and measures for improving the thermowell safety are proposed. According to this study, when high‐speed, high‐temperature gas passes the thermocouple bushing, the main factors affecting the structural safety of the thermocouple bushing are the alternating stress caused by the vortex falling off, the thermal stress cycle due to the temperature gradient, and the pressure gradient impacted by the gas. Furthermore, this paper proposes improving the thermowell safety by installing the interference devices and optimizing the installation angle. The improvement measures were studied by conducting a two‐way thermal fluid‐structure coupling simulation. The results of this study show that after installing the interference device and optimizing the installation angle the displacement deformation of the thermowell and the equivalent stress is reduced by 57.2% and 72.1%, respectively, which indicates the safety improvement of the thermowell structure and the effectiveness of the method. The research contents of this paper can provide guidance for the installation and use of thermocouple bushing.

In view of the nonlinear and nonstationary characteristics of the weak fault signal of rolling bearings and the characteristics that are easy to be masked by strong background noise, a weak fault diagnosis method of rolling bearings combining Variational Mode Decomposition (VMD) and Maximum Correlated Kurtosis Deconvolution (MCKD) is proposed. In order to realize the adaptive parameter selection of VMD and MCKD, the particle swarm optimization algorithm was used to optimize the parameters in the two algorithms. Firstly, the Particle Swarm Optimization (PSO) was used to optimize the α and K in the VMD algorithm, and then the optimal mode components were selected based on the results of VMD decomposition of weak fault signals. Secondly, the PSO is used to optimize the sum in the MCKD algorithm, and then the fault shock component in the optimal component signal is strengthened based on the MCKD algorithm. Finally, the weak fault characteristics of the bearing were extracted by the envelope spectrum. The experimental results show that this method can adaptively enhance the impact component in the weak fault of the bearing, and effectively extract the weak fault characteristics of the bearing submerged by the strong noise.