Yunxia Chen

Reliability Prediction Using Physics–Statistics-Based Degradation Model

The reliability and life predictions of products using small samples represent a major challenge to reliability engineers. In this paper, we develop a physics-statistics (P-S)-based model and an adaptive Kalman filter approach for reliability and life predictions. The P-S-based model combines the physics-of-failure models with the statistics models to consider the randomness among identical products. The degradation path is modeled with a time-scale transformation Brownian motion with drift, which is updated by the Kalman filter. Time-scale transformation is used to adapt the linearly increasing drift for modeling a nonlinear degradation process. The degradation of units over time is used to obtain the parameters of the proposed model. The parameters in the model are estimated using the maximum-likelihood estimation and particle swarm optimization methods with the accelerated degradation data, so that it provides more prior information and the empirical model for reliability prediction. The validity of the proposed approach is demonstrated with an illustrative example using the data collected from an accelerated degradation test of accelerometers. The proposed method is compared with the basic one in terms of their accuracy of reliability and life predictions.

Competing failure with degradation and random shocks based on probabilistic failure threshold

As the products become more and more complicated, and the demands of users vary continuously, failure threshold of product may be probabilistic rather than determined. This paper develops an extension research based on probabilistic failure threshold of competing failure with degradation and random shocks. Firstly, general competing failure model is proposed based on the assumption that the degradation process and the cumulative shock-damage process are independent, and then reliability formula based on different distribution is elaborated. Secondly, another competing failure model based on probabilistic failure threshold is proposed with the consideration of dependent competing of degradation and random shocks based on the former model. Random shocks could affect degradation process with a damage factor, which adds to the probability of degradation failure to the product. At last, those two models are applied to two cases of accelerator with degradation data and vibrating-damage data, and the comparison reveals the differences between those two models as well as the traditional model with pre-determined threshold. It tests the reasonability and accuracy of the probabilistic threshold competing failure models. The result of this paper provides an extensional method for the product failure prediction and healthy management based on the competing failure modes of probabilistic threshold degradation and random shocks.

Coupling damage and reliability modeling for creep and fatigue of solder joint

Abstract Solder joint often plays a crucial role in the normal operation of electronic equipment due to its unique material properties and harsh working condition, making it very important to carry out the accurate reliability analysis of solder joint. The low-cycle fatigue due to temperature cycling and the creep brought by continuous high temperature are two dominant failure modes of the solder joint. Current modeling methods for these two mechanisms mainly focus on failure process of each mechanism separately, with little consideration of the coupling relationship in the material properties. This paper introduces a coupling damage model considering both low-cycle fatigue and creep. The coupling relationship between these two failure mechanisms is investigated with the effects of creep strain rate on the ductility and the effects of damage on mechanical properties of solder joint. The analysis of the former mechanism concerns the fatigue parameter of Coffin-Manson model, while the latter one focuses on the applied stress increasing with the accumulation of damage. Further, considering that creep degradation rate would increase once the cumulative damage reaches a trigger threshold, a generalized cumulative damage model is developed. Based on this assumption, a reliability model for solder joint considering the uncertainty of model parameters is then proposed. Finally, a case study of a lead-free solder joint is given to validate this method.

Outlier test and analysis method of degradation data under unequal error variances

For degradation data, its usually difficult to judge and elimination outlier data, especially when error variances are unequal. So, in the paper, under the assumption that the degradation data obey the same path model form, a degradation outlier test method is presented based on path model when the error variances are unequal. Since the path model gives the trend term of degradation data, its parameters can be considered as the degradation characterization. If the path model parameters are equal for two groups of degradation data, it shows that the two group data have the same degradation trend term. Otherwise, they have different trend terms, and the data of small size can be considered as outlier. For abnormal degradation data, an outlier analysis method is also proposed. It can test the differences in intercepts and slopes between two degradation path models, and determine whether the abnormity is caused by the differences in intercepts, slopes, or both. At the end, an example is given.

Multi-Stress Equivalent Optimum Design for Ramp-Stress Accelerated Life Test Plans Based on D-Efficiency

Due to time and cost limitations, multi-stress accelerated life tests (ALTs) have been gradually applied in the fields of experimental design and reliability estimation for highly reliable and long-life products. To develop more efficient and concise test plans, it is necessary to study an equivalent experimental design method to create an equivalent conversion approach for different ALT plans. This paper presents the equivalent optimum design method combined with D-efficiency and asymptotic variance. In this method, the Fisher information matrix determinant is proposed as the equivalent feature variable. D-efficiency is a new equivalent criterion applied in the ALT design method and is measured using the ratio of Fisher information matrix determinants. The objective is to minimize the test time at a given efficiency and under asymptotic variance constraints without loss of test estimation accuracy. Based on the proposed equivalent method, a constant-stress ALT plan was successfully converted into its equivalent step-stress or ramp-stress ALT plan. Finally, an example is presented to demonstrate the equivalent ALT design of an electrical product with two stresses. The constant-stress ALT is equivalently designed for the ramp-stress ALT plan, and the sensitivity of the D-efficiency is analyzed. The results show that the proposed equivalent test scheme can shorten test time while achieving the same estimation accuracy.

A strategy for fatigue life estimation under variable amplitude loading in consideration of load sequence based on damage mechanics

Most mechanical and structural components are typically subjected to variable amplitude loading during their entire life. For practical purpose, it is computationally advantageous to approximate the load sequence by blocks of cycles with constant amplitude and stress ratio. But the indispensable influence on life evaluation of the correlation characteristics between individual load processes can be ignored in many cases. The paper first reviews the historical achievements on the equivalence disposal and life-prediction models applied for variable amplitude loading. Based on the idea of Gaussian copula type model, the fatigue life process can be discretized into blocks. The dependence character is described with correlation length. Then an ameliorated model based on Corten-Dolan accumulated damage rule is introduced to draw to the equivalent stress range. Using the disposed stress sequence, a life estimation algorithm based on damage mechanics is presented, during which the coupling between damage field and stress field is considered to deduce the damage evolution equation. Finally, the strategy discussed above is applied examplarily to a practical case.

Fatigue life prediction for structure under fatigue and low-energy impact based on continuum damage mechanics

A product characterized by fatigue-induced failure is subjected to complex cyclic load which cause fatigue damage accumulated over time. At the same time, the product under cyclic load may suffer from low-energy impact during the whole service life. For its discontinuity, the low-energy impact does not lead to the product failure immediately. The low-energy impact not only cause structural damage and the residual strength degradation instantly but also influence the effect of cyclic load. And cyclic load has the same effect for structural in return. This article shows an approach for fatigue life prediction based on continuum damage mechanics with coupling relationship between fatigue damage and impact damage. First, the features and coupling relationships of fatigue damage and impact damage are discussed in this essay and both of them result in a drop in residual strength of structure. Second, the initial stress response under impact and fatigue loads were obtained. The existed evolution equations of fatigue and impact damage were respectively introduced. Third, based on these damage evolution equations, a new coupling damage model for fatigue damage and impact damage has been established and the damage and residual strength evolution of the structure is proposed. Finally, an engineering case is introduced to validate the approach.

Study of Preferred Profile of Accelerated Stability Test Method for Quartz Flexible Accelerometer

Manufactured quartz flexible accelerometers have problems of poor long-term stability, which can be attributed to drifting of key parameters (e.g., zero bias, scale factor) that is influenced by a combination of time and external environmental stress inducing factors that include temperature fluctuation, thermal cycling, and dynamic mechanical loading. This paper focused on the experimental design for such a device. Accelerated stability experiments are performed and are divided into three kinds: thermal cycling test, test of concurrent application of thermal cycling and vibrational loading, and test of thermal cycling with subsequent vibrational loading. Experimental data are analyzed using the mobile linear regression coefficient test and the mobile standard deviation, after which the steady-state cycle number is determined. Based on the above analysis, the rapid effectiveness of the profiles is evaluated. Finally, the preferred profile is determined by thoroughly considering the equipment capacity and engineering operational convenience. It is illustrated that the vibrational loading has a significant effect on accelerated stability for various parameters. Overall, the research on the accelerated stability test enables an accelerometer to quickly attain a steady-state and also provides initial technical support to improve the robustness of the accelerometers long-term stability.

Notice of Retraction Reliability model of missile with disturbance-dependence

In this paper, firstly, considering the mechanism of disturbance factors which have effect on missiles reliability, three types of correlations is proposed with definitions and descriptions. Then specific reliability model formissile with each type of the correlations is developed. Furthermore, the synthetical reliability model for missile with each type isformulated by synthesizing and modifying these three aforementioned reliability models. Detailed steps of reliability calculationis given. Finally, the correctness and feasibility of the classification of correlation and the reliability model proposed in this paper is verified by taking a missile as a typical case.

A descriminant method for consistency of accelerated degradation mechanism based on MC-ApEn

The method of moving cut data-approximate entropy (MC-ApEn) is used in this paper. With the data obtained by an enhancement testing of a certain kind of conductive film resistors, the discriminant method of consistency of accelerated degradation mechanism is presented. Finally, the new judgment method of consistency of accelerated degradation mechanism based on the data of enhancement testing is discussed.