Papers by Keyword: Lifetime Prediction

Abstract: This study investigated the reliability challenges of SiC MOSFETs under continuous dual-bias stress conditions in electric vehicle (EV) applications. Accelerated dual-bias time-dependent dielectric breakdown (DB-TDDB) tests were performed on 1.2kV SiC MOSFETs by applying a negative gate-source voltage and a high drain-source voltage simultaneously. Experimental analysis and TCAD simulations revealed a nonlinear coupling effect between gate and drain biases, leading to a spatial relocation of the maximum gate oxide electric field under dual-bias conditions. An improved dual-acceleration-factor E-model was proposed to characterize this behavior. Based on this model, the projected lifetime at 1 ppm failure rate is 5.11×10¹³ hours, exceeding the 20-year automotive standard (AEC-Q101) by several orders of magnitude. Furthermore, failure analysis identified edge-thinning and JFET-region effects as two primary degradation mechanisms affecting gate oxide reliability, and corresponding process and design optimizations are proposed to mitigate this vulnerability. The findings offer critical insights for improving the reliability of SiC MOSFETs in EV applications under prolonged dual-bias stress.

Abstract: While mixed-mode loading conditions became hot topic recently, it is still quite unexplored area, even in case of metals. In case of polymers, this is even more complex problem, thanks to their higher sensitivity on higher temperatures connected with friction. Since polymer materials are very popular, this topic is of high interest. One of the best experimental specimens for achieving mixed mode loading conditions is the CRB specimen, as it provides clear mixed-mode I + III conditions under tension + torsion. Furthermore, for lifetime predictions it is necessary to calculate some material constants based on measured crack growth rates. However, these crack growth rates are not as easily and accurately measured on CRB specimens, as on CT specimens. Therefore, the main focus of this study is to observe difference between lifetime predictions based on CT and CRB data and possible application of CT data for CRB specimens.

Abstract: Composite materials in structural applications that are subjected to static loads for several decades tend to change material performance over their lifetime. Classical creep tests with constant static loading are quite simple tests with low demands on the test equipment. Unfortunately, these tests require uneconomically long test times, which is why a shortening of the test times with various accelerated approaches is being researched. Within this work two approaches for reduction of the testing time were investigated. On the one hand a fatigue test with the variation of R-ratio and following extrapolation to an R-ratio of 1 was done. On the other hand a Stress Rate Accelerated Creep Rupture Test (SRCR) was developed, where a defined initial stress σ_i is applied at the beginning of the loading process, followed by an increase load with a constant rate instead of the static stress segment of the classic creep rupture tests. Changing the load rate in several individual tests leads to stress rate-dependent fracture strengths with associated fracture times, which allows extrapolation to a fracture time at a load rate of zero. In particular, the approach of the SRCR offers great potential for greatly reducing test times with an acceptable prediction quality.

Abstract: Reinforced concrete structures are generally affected by degradation phenomena, which results in a time variability in strength and stiffness beyond the baseline conditions which are assumed in structural design, in particular when the concrete is exposed to an aggressive environment. Therefore, structural safety should realistically be considered time-variant. This paper provides a probabilistic approach to predict the time-evolution of the mechanical and geometrical properties of a reinforced concrete structural element (i.e., bridge pier) subjected to corrosion-induced deterioration, due to diffusive attack of chlorides, in order to evaluate its service life. The proposed model is based on Monte Carlo simulations in order to evaluate time variant axial force-bending moment resistance domains, with the aim to estimate the time-variant reliability index. Finally, an application to estimate the expected lifetime of a deteriorating reinforced concrete bridge pile is proposed.

Abstract: The article is focused on complex analysis of the principles of degradation of composites based on alkali-activated materials exposed to extreme temperatures (max. temperature 1200 degree of Celsius). Thanks to specific properties these materials show a remarkable potential for the special application, for example for construction with high risk of fire etc. The effect of high temperatures shall be tested not only at such extreme levels but also at gradually increasing temperatures. The temperature effects shall be evaluated not only in of changes physic-mechanical parameters but also the physic-chemical determinations.

Abstract: A damage model for thin adhesively bonded joints is presented, which predicts the time to creep-fatigue failure of the joint subjected to combined static and cyclic sustained loadings with constant or variable amplitudes. The influences of particular model parameters on the predicted lifetime are elaborated suggesting the proposed stepwise parameter identification strategy by means of creep and Wöhler fatigue tests until rupture. The parameters are identified and computationally optimized. As a conclusion, the model prediction is verified and validated.

Abstract: Most of the existing methods for bearing real-time reliability evaluation employ real-time transformation of traditional reliability indices, performance degradation trajectory analysis, and performance degradation distribution, which are usually limited in terms of accuracy and applicability. A method for real-time reliability evaluation and life prediction for bearings based on normalized individual state deviation is proposed in this study. First, a self-organizing map neural network is utilized to obtain the individual state deviation of a running rolling bearing. Second, individual state deviation is normalized into a state deviation degree, which is used to formulate a modified real-time reliability model for the realization of real-time reliability evaluation and residual life prediction. The proposed method combines population information with real-time monitoring information of individual bearings, and thus avoids the negligence of the real-time transformation of the monitored individual. The errors caused by the randomness of the individual bearing operational process are also reduced. Finally, the feasibility and efficiency of the proposed method is validated by performing run-to-failure experiments on bearings.

Abstract: Experience obtained from often tragic road accidents has proven that in a fire the tunnel lining is destructed very intensively by high temperatures. Concrete, usually applied using the sprayed concrete technology (so-called shotcreting) is used for building the linings of road tunnels. The topic of the present article is the development of a concrete mix determined by the sprayed concrete technology so that this material is to the maximum extent resistant to extreme temperature effects occurred in a fire. It is apparent that using this type of concrete is one of the way of increasing the fire safety of road tunnels.

Abstract: In the paper the polymer pipes loaded by internal pressure with additional point load on the external surface are studied using numerical methods. It is shown that the additional external loading can significantly influence residual lifetime of the pressured pipe. In this study the effect of the pipe geometry is considered. The shape of the propagating crack is estimated based on a special algorithm and the stress intensity factor is evaluated by direct method. The relation between crack size and a geometric function YasYas for specific material and geometrical properties is found. The results of this paper should contribute to better understanding of the real pipe behavior and to prevent unexpected failure of the pressured pipe system due to non-homogenous distribution of the soil load.

Abstract: As a part of power transfer process, cylinder plays an important role in pneumatic system. Its failure can cause mechanical equipment downtime suddenly and gas leak, so that production and personnel security will be in danger. Cylinder lifetime prediction has been an important topic. In this paper, an adaptive method based on Kernel Density Estimation is put forward for predicting the cylinder lifetime and getting the reliability function of cylinders. Kernel Density Estimation is a nonparametric estimation method of statistics. It can make full use of samples data without assuming distribution model. In the end, a comparison is made on the cylinder experiment between the proposed method and the common used parameter estimation method, Weibull distribution, and the results show that the proposed method has a more satisfactory performance.