Development of a Random Loading Test Bench for UD E-Glass/Resin Rotating Shafts with Integrated Monitoring of Failure Prediction

In the today high-cost and complex drilling environment, the importance of drillstring failure issue has dramatically reappeared, in spite of many manufacturing and materials improvements. Most drillstring failures are due to fatigue, resulting from repeated cyclic bending loads and stresses in tensile or buckled drill pipes. Fatigue prediction is usually based on the cumulative fatigue damage model from Hansford and Lubinski as defined in API RP7G. This model, based on S-N curves, a failure criterion and a damage accumulation rule, initially requires a calculation of the drillpipe stress caused by bending when rotated in a dog leg. This bending stress calculation, key point of the cumulative fatigue damage model, is usually made by assuming that the curvature of the drill pipe is the same as the dog leg. However, this paper shows that this strong hypothesis may lead to major under-estimation of the cumulative fatigue damage. Moreover, the stress distribution within a drill pipe may be completely different depending mainly on the position of the drill pipes along the drillstring and the wellbore architecture and tortuosity. The cumulative fatigue damage model as defined in API RP7G has been implemented in an advanced torque and drag model, which enables to track any given point of the drill pipe while drilling, such as the transition zone, the tool joints and the drill pipe body. For the first time, it has also been possible to fully track variation of stresses at a given point in the drill pipe. Based on drill pipes S-N curve available in the literature and actual drilling data, this paper shows and compares results of fatigue damage calculations as obtained from the conventional way (strong hypothesis on the contact) with results obtained from advanced torque and drag model that make no assumptions about the contact. This extensive study as presented in this paper has never been done in the past. This advancement should probably lead to minimize the risk of drillstring failures in complex wells by a better monitoring of stresses in drill pipe. Introduction Drill pipe failures are still responsible for rising costs in drilling industry. In spite of many research studies carried out to mitigate this issue, it continues to occur with high frequency. Many drill pipe failures analyses have shown that fatigue accounts for the majority of drillstring failures. Fatigue is a cumulative and non-reversible phenomenon, resulting from repeated cyclic bending loads and stresses in tensile or buckled drill pipe. Fatigue occurs even if the cyclic stress is much lower than the static strength limit of the drillpipe material. Fatigue prediction is based usually on the Cumulative Fatigue Damage model (CFD), introduced by Hansford and Lubinski in API RP7G1. This model allows determining the life duration of the drill pipe based on S-N curves, damage accumulation rules and a failure criterion. However, it requires calculation of the drill pipe stress in the wellbore, which is the key point in fatigue prediction. Bending stress calculation has first been performed by Lubinski in case of rotated drill pipe subjected to tension, and Paslay suggested improvement by studying the case of compression. Lubinski/Paslay model is based on strong hypotheses on the contact between drill pipe and wellbore, and assumes the axial load to be known. The aim of this paper is to address a new approach to assess the bending stress in the drillpipe. This methodology allows implementing the Cumulative Fatigue Damage model as defined in API RP7G, in an advanced torque/drag and buckling model 2,3,4, which allows tracking any given drill pipe while drilling.

Solution of the ratchet-shakedown Bree problem with an extra orthogonal primary load

A Novel Predictive Selective Maintenance Strategy Using Deep Learning and Mathematical Programming

For more than 25 years sandwich connectors made from GFRP are known in different markets. In different parts of the world, national committees or authorities have defined different testing programs to assess the performance of such elements. Now in Europe an internationally counselled assessment document (EAD) is available and the first producers have submitted their evaluation reports. In this paper, the background, as well as some results of this EAD, are presented and discussed.The focus of this paper is on the evaluation of the thermal effects on the load-bearing capacity of sandwich connectors. In particular, the temperature effect causes cyclic shear and bending stress in the connector due to the temperature difference between the facing layer and the load-bearing layer. The effects of the cyclic temperature-induced stress are investigated by functional tests with a subsequent test of the residual load-bearing capacity.In the context of this article experimental investigations for the evaluation of the temperature-induced cyclic shear and bending stress on the load-carrying capacity of sandwich connectors made of glass fibre reinforced polymer are presented and subsequently evaluated.KeywordsEADETAPoint anchorSandwich connectorCertificationApprovalShear test

This study investigated the effects of cyclic bending stress levels and testing in simulated physiological solutions or air on the integrity of plasma-sprayed hydroxylapatite (HA) coatings of two different crystallinities. Hydroxylapatite-coated commercially pure (CP) Ti rods were evaluated by immersion testing in Hank's Balanced Salt Solution (HBSS) and by rotating bending in air and HBSS. Static immersion testing of nonstressed specimens resulted in significant microcracking of coating surfaces after 42 days. Specimens cyclically tested at bending stresses above the yield strength of Ti experienced low cycle fatigue failure of the Ti substrates prior to spallation of the HA coatings. Coatings tested at 1 x 10(6) cycles with interface bending stresses of 180 MPa displayed increased surface microcracking, but no bulk coating spallation. Coatings cycled in HBSS displayed greater amounts of microcracking and surface alteration than samples cycled in air. There was no apparent relation between HA crystallinity and mechanical integrity under cyclic bending stresses.

The integration of the Industrial Internet of Things (IoT) and machine learning techniques in condition-based maintenance (CBM) systems represents a significant advancement in the monitoring and maintenance of electromotors, particularly those used in elevator systems. This study explores the effectiveness of a Condition Based Maintenance (CBM) framework utilizing Arduino microcontrollers and Microelectromechanical Systems (MEMS) sensors for real-time vibration data collection, coupled with the analysis through a Deep Variational Auto-Encoder (D-VAE) model for anomaly detection in electromotor operation particularly in the elevator system. Traditional methods such as Isolation Forest and K-means clustering, enhanced by Principal Component Analysis (PCA), are also implemented for comparative analysis. The performance of these models is evaluated using unsupervised learning metrics, including the silhouette score, Davies-Bouldin Index, and Calinski-Harabasz score, to assess their efficacy in detecting anomalies within vibration data indicative of potential electromotor malfunctions. Our findings suggest the superior capability of the D-VAE model in discerning between normal and faulty operational states through detailed clustering, despite some limitations under problematic conditions. This study underscores the potential of data-driven approaches and IoT technologies in enhancing operational efficiency, reliability, and safety in elevator systems through predictive maintenance strategies.

Effective predictive maintenance (PdM) strategy is needed in microelectronic manufacturing to reduce cost and loss of cycle time associated with unplanned maintenance events in fab. Adoption of recent advances in big data analytics methods will help in developing an effective PdM strategy. This paper discusses advanced machine learning based PdM strategy. Various approaches to formulate the PdM problem and solve it are discussed. The performance of respective machine learning algorithms on equipment data is evaluated. To benefit from the strengths of various algorithms a super learning based approach with deep neural networks is proposed. This approach is robust compared to single method selection and handles changing production environment well. This cross validation based stacked model ensemble predictive maintenance strategy is also easy to automate.

A method for analysis of stress-strain state of the tubing column in the spatially curved well section with an arbitrary intensity of changes as the zenith and azimuth angle is developed in the article.Stresses and strains are calculated using the rod mechanics fundamentals. The system of vector differential equilibrium equations in the projections on the axes of the moving coordinate system is used for the analysis of elastic equilibrium of the tubing axis.The analysis is carried out by solving the direct problem of the rod bending by means of iterative numerical method implemented in software-oriented mathematical environment.A method to account for the limiting effect of the well walls to move of the elastic tubing axis is provided based on the processing of its geometrical parameters, as well as data and directional survey and profilometric research of the well bore.Using the designed calculation schemes, it is found that local bend of the well axis substantially increases the bending stress, which can reach the yield strength of the tubing material. For example that discussed in the article, the bending stress is approximately 300 MPa, while the minimum yield strength of steel, which are made tubing, is equal to 379 MPa.Moreover, the axial force change caused by the reciprocating motion of a rod suspension may initiate the occurrence of cyclic bending moments and stresses in the tubing. So, for the considered operating conditions, we received the following: the minimum stress of the cycle σmin = 303 MPa; the maximum stress of the cycle σmax = 346 MPa; stress amplitude of the cycle σa = 43 MPa; loading cycle asymmetry factor Rσ = 0,88. This load pattern allows to speak about the cyclic fatigue of the tubing.

Failure analysis and experimental stress analysis of a threaded rotating shaft

FAILURE ANALYSIS AND EXPERIMENTAL STRESS ANALYSIS OF A THREADED ROTATING SHAFT

The damage experiments of reinforced concrete (RC) samples under synergistic effects of cyclic freeze-thaw, deicing-salt attack, rebar corrosion and bending stress were investigated using a comprehensive experimental method. Synergistic effects of these factors on the damage evolvement of RC were studied by measuring the change of dynamic modulus of concrete, rebar strain and concrete strain. Experimental results showed similar rule in damage evolvement but different damage rate between the normal-strength concrete (C45, the 28-day compressive strength of 52 MPa) and the high-strength concrete (C70, the 28-day compressive strength of 77 MPa). The dynamic modulus of the reinforced concrete degraded with increasing effecting factors.

Fatigue Strength of Thin-Walled Tube-to-Plate T-Joints Under In-Plane Bending

Machine learning-based diagnosis in wave power plants for cost reduction using real measured experimental data: Mutriku Wave Power Plant

Three types of structural ceramics, Al2O3/SiC, mullite/SiC and Si3N4/SiC having excellent crack-healing ability were sintered. A semi-elliptical surface crack of 75-110μm in surface length was made on each specimen with a Vickers indenter. The basic crack-healing behavior was studied systematically on Al2O3/SiC as a function of crack-healing temperature and time. Three point bending tests were carried out to investigate the strength of crack-healed samples. The specimens containing the semi-elliptical surface crack were healed under static or cyclic bending stress on mullite/SiC and Si3N4/SiC. It was shown that a crack could be healed completely either under a static or a cyclic stress. The competition between self crack-healing rate (SCHR) and crack growth rate (CGR=da/dt) determines the crack-healing behavior under stress: (i) In case of SCHR>CGR, a crack can be healed completely even under stress and recover of strength is complete. (ii) In case of SCHR≅CGR, a crack cannot be healed completely and recover of strength is insufficient. (iii) In case of SCHR<CGR, the specimen fails due to crack growth under stress.

The essential aim of Industry 4.0 is to enable industries to be more productive, efficient, and flexible. A predictive maintenance strategy can make a positive contribution to all these things, as it uses industrial IoT technologies to monitor asset health, optimise maintenance schedules, provide real-time alerts about operational risks, and maximise uptime, and can provide digital services to customers based on data from its machines. It improves productivity, improves customer satisfaction, and therefore gives the company a competitive advantage. Nevertheless, decision making in relation to a predictive maintenance strategy is not systematised, and this may lead to some inappropriate decisions, which do not achieve the goal sought. This chapter describes a multicriteria model, designed with the analytic hierarchy process, to systematise decision making with respect to a predictive maintenance strategy.