Abstract Major equipment plays a vital role in national infrastructure and industrial development, and accurate fatigue load prediction is essential for ensuring its operational safety and service life. This paper proposes a deep learning-based framework for advanced fatigue load prediction by integrating a convolutional neural network (CNN), an Efficient channel attention for deep convolutional neural networks (ECA-net), and a long short-term memory network (LSTM). Using actual strain data collected from a tunnel boring machine (TBM) cutter head—sampled at 200 Hz—the study first preprocesses the raw signals through comparative filtering, with median filtering selected for optimal noise reduction. The preprocessed data are converted into supervised learning samples via sliding windows and standardized before model training. The proposed CNN–ECA–LSTM model extracts high-dimensional temporal features through CNN layers, enhances feature relevance via ECA-net, and captures long-term dependencies using LSTM. A multicondition recognition module is also incorporated to adapt to varying operational states. Hyperparameter tuning is performed systematically, and Smooth L1 loss is adopted to improve robustness against outliers. Experimental results demonstrate that the model achieves a root-mean-square error (RMSE) of 33.2626 and a mean absolute percentage error (MAPE) of 15.37677 in predicting 200 future time steps, with 44.5% of predictions reaching over 90% accuracy. Ablation studies confirm the contributions of each component, showing that the integrated model outperforms standalone CNN, LSTM, and CNN–LSTM variants. The study confirms that the proposed approach effectively predicts strain-based load trends, providing a reliable basis for fatigue load spectrum construction and remaining useful life estimation of major mechanical equipment.
Abstract The unique operational mode of coiled tubing (CT) inevitably subjects it to bending and damage, making its collapse strength difficult to predict and making it prone to collapse failure. In this study, the tensile mechanical properties of CT material (CT130 steel) under different prestrains were tested, and a method for calculating the yield strength of CT130 steel after bending was established. The mechanical response of CT during the bending–straightening process was simulated using nonlinear finite element analysis (NLFEA). Calculation models for ovality, eccentricity, and residual stress induced by bending were established and introduced into the collapse strength calculation model as the defect influence factor. Based on the full-wall yield failure principle and the Unified Strength Theory, combined with the methods for calculating the yield strength and the defect influence factors of CT130 steel, a collapse strength calculation model for CT based on four typical yield criteria was constructed. A comparison of the model's calculation results with field data and simulation results shows that the model based on the geometric midline (GM) yield criteria is the most accurate model. This model improves the predictive accuracy of collapse strength and helps reduce the probability of collapse failure in CT.
Abstract Due to different material parameters between the cladding layer and the base metal, significant tensile residual stresses still remain within the internal cladding layer after traditional heat treatment, seriously affecting the service safety of pressure vessels. Local rapid cooling heat treatment is an effective method to reduce the tensile residual stress in the cladding layer until a compressive stress state is achieved. In this paper, the influence of local rapid cooling heat treatment process parameters on the effectiveness of residual stress reduction was analyzed. Local rapid cooling heat treatment is more effective in reducing residual stress than conventional heat treatment. As the rapid cooling start temperature increases, the cooling rate increases, and the rapid cooling end temperature decreases, the effectiveness in reducing residual stress within the welding layer of strip clad welding structure improves. A mechanical model clarified the basic mechanism that local rapid cooling heat treatment reduces residual stress in the cladding layer through the contraction effect of the base metal during the cooling process. Theoretical analysis indicated that the control effect is related to the temperature difference of the base metal. And the influence of process parameter changes on the temperature difference magnitude is consistent with the change in control effect, which confirms the validity of the conclusions derived from finite element analysis.
Abstract This study examined the challenges and necessary steps for performing a fitness-for-service (FFS) assessment of cracked clad pressure vessels in sour service through a level 3 FFS assessment of a degraded Inconel 625–clad A516 Gr 70 pressure vessel containing crack-like flaws, in accordance with API 579-1/ASME FFS-1. The vessel, exposed to well fluids containing H2S, CO2, and chlorides, exhibited cracks due to weld defects in both the clad and the base metal. Advanced nondestructive testing (NDT) inspections identified critical defects, including a 3105-mm-long circumferential crack resulting from the welding process. Advanced elastic–plastic finite element analysis (FEA) was used to derive J-integral and reference stress solutions. Fracture and material properties, such as Master Curve toughness, true stress–strain curves, and residual stress distributions, obtained from FEA, were included. Results from the J-based failure assessment diagram (FAD) indicated that the vessel fails the level 3 FFS criteria, mainly due to the low fracture toughness of the A516 Grade 70 base metal under sour conditions, the presence of tensile residual stresses, and the presence of a long crack. The study highlights the challenges and the necessary steps for FFS assessment of clad vessels and presents an advanced FEA method for evaluating such vessels.
Abstract Aiming at the problem of packer sealing failure caused by casing deformation in ultradeep horizontal wells and intervals with complex crustal stress, this study reveals the characteristics of casing elliptical deformation under the combined action of nonuniform crustal stress and formation creep. Based on the theory of large deformation of rubber and elastic mechanics, the three-dimensional numerical model of the packer sealing mechanism considering casing deformation is established. Using sealing contact strength as the evaluation index, the evolution law of rubber cylinder deformation, stress distribution, and sealing performance indices of the packer under different elliptical aspect ratios and pressure loads are analyzed. The research shows that elliptical deformation results in an “elliptical” distribution of contact stress between the rubber cylinder and the casing, with contact stress in the minor axis region significantly higher than that in the major axis region. When the aspect ratio is ≤1.02, the packer can maintain effective sealing through adaptive deformation of the rubber cylinder. When the aspect ratio is ≥1.04, the circumferential stress nonuniformity intensifies, and the contact stress in the major axis region is insufficient, resulting in a significant increase in the risk of sealing failure. The research results provide a theoretical basis and engineering guidance for the optimal design and sealing technology of packers under complex well conditions.
Abstract Thermal autofrettage is a promising approach for generating beneficial residual stress patterns in thick cylinders subjected to internal pressure. However, its effectiveness is constrained by the allowable thermal gradient, as excessive temperature differences can negatively impact the mechanical and microstructural properties of the material. To overcome these limitations, this study proposes a combined approach that integrates thermal autofrettage with wire-winding prestressing. Analytical and numerical approaches, supplemented by ASME code provisions, are applied to investigate the impact of the integrated procedure on the cylinders' resistance to yielding and fatigue life. The results indicate that the combined technique significantly increases the yield onset pressure and enhances the fatigue life of the vessel compared to stand-alone thermal autofrettage and an untreated “monobloc cylinder” (A monoblock cylinder is a pressure-retaining cylindrical component manufactured from a single, solid piece of material.).
Abstract The aim of this work is to examine the effect of TiO2 nanoparticles-coated E7018-H4R electrode on microstructure and mechanical characteristics of A-1011 steel welded joints. The shielded metal arc welding (SMAW) process is used to join the base metal, and output responses including ultimate tensile strength, hardness, percentage elongation, and residual stresses of joints have been evaluated. The effects of SMAW parameters such as welding current, welding speed and dipping time of electrodes in TiO2 solution on mechanical responses and microstructure have been analyzed through Response surface methodology (RSM) with central composite design. Mathematical models have also been developed to predict the output responses. Results revealed that tensile strength has been primarily influenced by welding current followed by dipping time and welding speed. The maximum tensile strength of 509 MPa, hardness of 230 HV and percentage elongation of 23% has been achieved by using 60 min dipping time of electrodes in TiO2 solution. The optimum combination of parameters, i.e., welding current of level 2 (130 A), welding speed of level 2 (35 cm/min), and dipping time of electrodes in TiO2 solution of level 2 (60 min) has been attained. The residual stress results showed that the specimen welded with coated electrodes have less residual stress value 142.32 MPa as compared to specimen with uncoated electrode (179 MPa). The optical and scanning electron microscopic (SEM) analysis confirmed the presence of TiO2 nanoparticles in the welded zone resulted in a smaller grains boundary of ferrite and presence of acicular ferrite, which ultimately led to increase in tensile strength and hardness of a specimen. The EDX analysis confirmed the presence of Ti peaks, along with Mn and Si, suggest inclusion formation favorable for acicular ferrite nucleation, leading to grain refinement and enhanced mechanical properties.
The analysis and design of square cross-section vessels have always been difficult and complicated due to the lack of theoretical solutions. Current design codes rely on empirical formulas for stress calculations developed based on analysis results of prismatic cross-section frames. The accuracy and applicability of empirical formulas are not satisfied in most engineering applications. With the recently proposed closed-form solutions for square shells, it has become possible to optimize the analysis and design of square cross-section containers. In this paper, two dimensionless parameters are introduced into the closed-form solutions. The stress in a square shell is a function of those two parameters. Parametric analysis was performed to investigate the stress characteristics in the square shell. Empirical expressions are proposed to simplify stress analysis. By using the conception of virtual rolled square, the theoretical solution of rolled square shell is extended to welded square shell. By performing regression analysis on the parameter research results, the corresponding empirical expression for design stress calculation of welded square shells was established. The accuracy and applicability of the empirical formulas proposed in this paper are confirmed.
When flaws are detected in power plants, they are evaluated to determine their impact on component integrity. Three conditions are imposed on the components in question to ensure they can operate safely. Firstly, the applied stress must be less than the allowable stress. The second is that the maximum allowable flaw depth should be set to prevent coolant leakage from the pressurized pipes. According to the ASME Code Section XI, the allowable flaw depth should be less than 75 % of the pipe wall thickness, even if the first condition is met. The third condition is the maximum allowable flaw length, which is intended to prevent a guillotine break in the case of a circumferential flaw or a split fracture in the case of an axial flaw. The current maximum allowable flaw length is defined as the length at which through-wall flawed piping fails due to applied stress. Therefore, current maximum allowable lengths are irrespective of flaw depth. However, the failure stress for a shallow flaw is higher than for a through-wall flaw, and the elongation for a shallow flaw is greater than for a through-wall flaw in a plate subjected to tensile loading. Furthermore, if the length of a shallow flaw exceeds the maximum allowable length determined by a through-wall flaw, the shallow flaw is not acceptable. This paper uses a flat plate model with surface flaws to examine the characteristics of flaw lengths and proposes a new methodology for determining the maximum allowable flaw lengths.
This study systematically investigated spot heating local PWHT for pressure vessel nozzle to cylinder butt welds through integrated experimental and numerical analysis. A dual-stage local heat treatment methodology, i.e., the primary plus secondary local PWHT (PS-PWHT), was developed specifically for the nozzle to cylinder butt welds, with a systematic investigation of secondary heating parameters affecting residual stress reduction. The results demonstrate that the PS-PWHT method achieves 50-70% residual stress reduction at critical inner surface locations. Two optimized secondary heating configurations, i.e. elliptical saddle-shaped secondary heating (E-SH) and rectangular saddle-shaped secondary heating (R-SH), were established based on reverse bending moment analysis of axial/hoop heating bands. Geometric parameters were quantified with recommended secondary heating distances (WDC=2Rt and WDS=4Rt) and (2-3)t width range (where WDC and WDS are the primary and secondary heating distances for the hoop and axial heating bands, respectively. t represents wall thickness). A heating temperature range of 300-450°C was identified for secondary heating, balancing effective stress relief without material degradation and damage. The developed process control strategy demonstrates significant improvements in mitigating welding-induced residual stresses through the PS-PWHT method.
This research presents a novel analytical model for above ground liquid storage tanks. Using elasticity theory, the tank is modeled as an axisymmetric structure with a multi-course cylindrical wall welded to a flat bottom floor under hydrostatic pressure loads. While all courses satisfy compatibility and equilibrium conditions when the tank is filled with liquid, a unique flexible constraint is introduced at the bottom connection with the floor at the corner joint using equivalent torsional and radial springs. The spring parameters are obtained by solving the moving boundary equations for the floor critical zone and its vicinity region on foundation under hydrostatic pressure. The constraint equations are transformed into matrix form for computational implementation. The analytical model provides a comprehensive and rigorous solution for the entire tank structure under various liquid fill heights. Deformation, stresses, and strains are calculated through iterations. Results from the analytical model are validated by finite element analysis on different tanks and liquid load conditions. The innovative method enables the relation of minimum required component wall thicknesses to hydrostatic load. Especially, the method provides an analytical procedure for the lower course and bottom floor critical zone as part of the solution for the first time, which allows the two components to be reclassified from Type C to Type B components in API 579. The method is intended to be used as a procedure for design analysis or fitness-for-service assessment of liquid storage tanks in relevant industrial codes and standards.
Abstract Triple-offset butterfly valves are manufactured under standard temperature conditions. During cryogenic service, critical sealing structures undergo uneven deformation due to thermal contraction. This leads to insufficient sealing pressure, which may result in seal failure. To enhance the cryogenic sealing performance of triple-offset butterfly valves, this study introduces an iterative method for optimizing the valve seat's outer contour. A coupled thermal-structural finite element analysis is first performed to evaluate the sealing behavior under cryogenic conditions. Based on the contact stress distribution and deformation of key components, targeted contour modifications are applied. Finally, the sealing performance of the optimized valve is evaluated through cryogenic temperature testing. After the optimization of the valve seat structure, the minimum contact stress in the sealing weak area increased from 4.57 MPa to 8.08 MPa, resulting in a substantial improvement in the overall sealing performance. Cryogenic testing results indicated that the optimized butterfly valve reduced leakage by 44%, meeting the practical sealing requirements of butterfly valves under cryogenic temperature conditions and further validating the effectiveness and applicability of the proposed optimization approach in improving sealing performance. This study proposes an optimization method for seal structures to enhance the sealing performance of cryogenic butterfly valves, offering valuable insights for the design of other cryogenic sealing components.
A selective cell-based smoothed finite element method (CSFEM) is incorporated into a partitioned semi-implicit coupling scheme for fluid–structure interaction (FSI) based on four-node quadrilateral element. The selective CSFEM invokes full and reduced integrations at different steps of the characteristic-based split (CBS) scheme that solves the Navier–Stokes equations in a fractional-step manner. Because the semi-implicit coupling framework is also underpinned by the CBS scheme, an hourglass control is adopted to stabilize both the partitioned algorithm and selective CSFEM over under-integrated smoothing cells (SCs). The elastodynamics equations are computed via the conventional CSFEM. The developed technique is validated against available data for different FSI examples. In addition to satisfactory results, it can save more computer time than the original semi-implicit coupling algorithm using the standard CSFEM.
Irradiation assisted stress corrosion cracking (IASCC) growth rate equations based on the Hashimoto-Koshiishi model, a mechanistic model that is in turn based on the slip oxidation mechanism, have been proposed and studied. This study optimized the crack growth rate (CGR) equation by revising the strain rate equation at the crack tip according to the irradiation dose. The Gao-Hwang equation was used for the crack tip strain rate when strain hardening occurred, and the Rice-Drugan-Sham (RDS) equation was used when strain hardening was negligible. It was confirmed that the optimized CGR equation could predict experimental data of L-grade stainless steels more accurately than the previously developed CGR equation. In addition, a comparison was made between the CGRs obtained with some model calculations. Among the three models, the optimized equation was the best and the PLEDGE (Plant Life Extension Diagnosis by GE) model was second in accuracy. The Eason and Pathania model adopted by the ASME (American Society of Mechanical Engineers) was third. It was thought that one of the reasons why the calculation results using the Eason and Pathania model were less in agreement with the experimental data was that the K dependence of the CGR was treated as constant regardless of the irradiation fluence.
A finite element (FE) model is developed to predict the burst pressure of internally corroded elbows with high accuracy and reliability. This model considers various parameters, including the length, depth, and width of the corrosion defect, as well as the pipe diameter and elbow bending radius. It is found that increases in the depth and length of the corrosion defect have a significant impact on the reduction of the corroded elbow burst pressure. Meanwhile, the corrosion defect width has a minimal effect. Corroded pipeline elbows with a smaller diameter or bending radius are more prone to burst than those with larger dimensions. A novel burst pressure prediction model for corroded pipeline elbows is proposed, and its predictions are compared with the FE simulations and several existing models. The mean value and coefficient of variation (COV) of the ratio between the FE results and the proposed model's predictions are 0.994 and 4.42%, respectively. These results demonstrate that the proposed model provides more accurate predictions than other models. This study offers a reliable foundation for predicting burst pressure and evaluating the integrity of pipeline elbows with internal corrosion defects.
In the pressure vessels and heat exchanger manufacturing, some codes and specifications require the final acceptance test (hydrotest) to be performed at temperature higher than the ambient temperature. This may also be a code requirement depending on the conditions. This is done to reduce the risks of embrittlement under stress which may happen on certain high-strength materials (especially Cr-Mo steels). The execution of a hydrotest in warm conditions poses some challenges including the pressure variations induced by the temperature drop during the hydrotest duration. This essay provides a simple practical methodology for calculating such pressure variations with the aim of properly preparing the system to perform the hydrotest.
Shell and tube heat exchangers are widely used in various industries for effective heat transfer between fluids. However, these exchangers often encounter challenges such as corrosion and wear, which can greatly affect their performance and lifespan. To address these issues, utilizing stainless steel (SS) made heat exchangers can significantly improve the corrosion and wear resistance of heat exchangers and their components. In this study, welding procedure specifications (WPS) for welding tube-to-tube sheet (TTS) joint mock up is qualified as per ASME IX QW 193 and ASME Section VIII Div. 1 UW-20. Experimental studies were performed on a 50 mm thick SS tube sheet (SA240 Gr TP 316 L) welded with SS tubes that have an outside diameter (OD) 19.05 mm and thickness 1.65 mm (SA 312 Gr TP 316 L). Both the tube and tubesheet materials are SS which is widely used due to their excellent corrosion resistance, strength and versatility. The welding of 10 samples was carried out using a gas tungsten arc welding (GTAW). In order to evaluate the quality of TTS joint, visual testing (VT), penetrant testing (PT) and macro-examination for minimum leak path (MLP) was performed according to welding procedure which can also satisfy the inspection standard. Furthermore, hardness test, ferrite test, and chemical analysis were also conducted to confirm the quality and integrity of the welds. The successful WPS provides a good reference (process parameters) for the welding quality control during the manufacturing SS heat exchangers for corrosion resistance.
Water hammer in long-distance pumped water transmission pipelines can cause severe safety incidents, making effective protection for pipelines and pump stations essential. However, optimizing parameters of water hammer protection devices remains challenging, as engineers often rely on time-consuming, experience-based trial-and-error methods. To address this, a multi-objective optimization framework combining random forest (RF) and nondominated sorting genetic algorithm II (NSGA- II) is proposed. An RF model is trained to map relationships between device parameters and extreme water hammer pressures. A multi-objective optimization model is developed with unidirectional surge tower water level, maximum pressure, and minimum pressure as objectives. Furthermore, Shapley additive explanations (SHAP), an interpretable machine learning method, is employed to reveal the importance and interactions of parameters. Results show that the approach rapidly identifies optimal device settings, achieving a 79% increase in minimum pressure, a 25% reduction in surge tower water level, and negligible change in maximum pressure compared with the original design. SHAP analysis quantitatively verifies that connecting pipe diameter and local resistance coefficient of the downstream air vessel are the dominant parameters governing transient pressure behavior. The nonlinear interaction of the two parameters is quantitatively characterized, showing that the increase in positive-pressure peaks induced by a larger connecting pipe diameter can be counterbalanced by a corresponding rise in local resistance coefficient, reflecting a codependent mechanism between flow inertia and local head loss. This data-driven interpretation provides quantitative insight into parameter coupling and offers practical guidance for optimizing water-hammer protection device design.
In order to improve the safety of vehicle-mounted liquid hydrogen storage and transportation and reveal the leakage and diffusion characteristics of vehicle-mounted liquid hydrogen in open space, a numerical simulation model of liquid hydrogen leakage and diffusion was established. The leakage and diffusion behavior of liquid hydrogen during transportation was studied. The influence of wind speed, leakage rate, leakage time, wind temperature, ground temperature, and other factors on the diffusion behavior of hydrogen clouds was analyzed. The results show that the flammable hydrogen cloud formed after the leakage of liquid hydrogen diffuses from the near ground to the distant air, and the volume expands rapidly, resulting in a significant increase in the potential hazard area. The combustible hydrogen cloud has typical radial concentration gradient distribution characteristics, and the hydrogen concentration decreases from the center to the periphery. The spatial diffusion range and volume change are mainly affected by wind speed, leakage rate, and leakage duration, while the influence of wind temperature and ground temperature is relatively limited.
Nondestructive evaluation techniques are increasingly applied to estimate the buckling load of shell structures without inducing failure. This study employs the force-stiffness (F-s) method to evaluate the buckling load of oblate ellipsoidal shells subjected to external pressure. Experiments are conducted on five steel shells with comparable R/t ratios, recording pressure, strains, and crown deflections up to failure. Using deflection data up to 80% of the buckling load, the F-s technique predicts buckling with similar to 90% accuracy, while incorporating crown strain data improves accuracy to similar to 95%. All shells fail within a pressure range of 2.7-3.4 MPa. The method is further validated on 3D-printed aluminum shells, showing promising consistency, though additional testing is required to refine accuracy. Finally, the F-s approach is demonstrated for real-time prediction of a large-scale shell, nearly three times larger than the experimental models, highlighting its scalability and potential for structural applications.
