In majority of engineering structures and mechanical components/systems, the operating load is usually varying randomly. The material property also suffers degradation with the loading conditions. This leads to continuously changing load-strength relationship and results in a variation in the failure rate. In the load amplitude domain, the situation can be described as one in which the structure or component is subjected to multiple actions of stochastic load, and the material property degrades continuously with loading pattern. Since a failure is caused when the load applied becomes higher than the relevant strength, a closed-form failure rate model is developed for a mechanical component subjected to a random load process. Consequently, the variation in failure rate is interpreted in both the load uncertainty and the component property uncertainty. Based on such a failure rate model, the effects of component strength degradation, as well as load dispersion and component strength dispersion on the failure rate are highlighted. The present paper analyzes failure rate evolvement from the competing mechanisms of statistically-increasing load and degrading strength and interprets the three stages uniformly as the manifestation of ever-changing load-strength interaction.
Received from the Guest Editor on August 25, 2009
References: 11

A method to model component reliability is presented, where the life is measured by the number of load applications. With static strength failure and fatigue failure as the scenario, the dynamic models for calculating the reliability and failure rate of components under random repeated loads both without strength degradation and those with strength degradation are derived. The relationship between reliability and the number of load applications, and that between failure rate and the number of load applications are discussed in different cases. The result shows that both the reliability and the failure rate of components decrease as the number of load applications increases. This is even though the strength does not degenerate, and the failure rate curve has the partial characteristics of a bathtub curve, with an early failure period and a random failure period. When strength degenerates with the number of load applications, the reliability of components decreases more obviously, and the failure rate curve of components is bathtub-shaped.
Received from the Guest Editor on August 25, 2009
References: 15

A method of analyzing probability distribution of fatigue life of notched specimens is proposed in this paper based on the average stress model for fatigue life calculation. The factors which have effect on the probability distribution of fatigue life of notched specimens were separated into two parts, the non-homogeneity of microstructure and the dispersion of local stress at the notch root. The effect of the former part could be obtained by means of test data of fatigue life of the smooth specimens, and that of the later part was achieved by the Monte Carlo stochastic finite element method (SFEM). Lastly the two portions were combined organically, and thus the probability distribution of fatigue life of notched specimens was got. Examples at two different stress levels were accomplished, and the predicted distribution was in good agreement with the experimental result, which demonstrates that this method is feasible and effective.
Submitted on May 19, 2009. Received from the Guest Editor on August 25, 2009
References: 13

The movement principium of rotation joint of umbrella antenna was studied and the motion reliability model of rotation joint based on unascertained theory was built. Synthetically considering the undulatory property of the output moment of motor, non-homogeneity of material parameter and dimension errors, this paper treats the mechanism motion function as one with some unascertained variables. With the application of computational theory of unascertained rational numbers, the reliability calculation formulas are built corresponding to the performance functions according to moment and work respectively and the motion reliability of rotation moment in the whole spreading process is predicted. Compared to the existing random method, the proposed method is easy to use. Reliability can be predicted with higher confidence even with inadequate data or insufficient information. Finally, the possible values and confidences of unascertained performance function of antenna rotation joint are given by an example, which confirms the rationality and feasibility of the presented method.
Submitted on March 31, 2009. Received from the Guest Editor on August 25, 2009
References: 13

As an important tool for performance assessment in multi-state systems, the universal generating function approach is introduced and developed aiming at strength reliability analysis of structural systems. Static reliability models of statically determinate and statically indeterminate systems are built by constructing special generating functions, which are used to describe probability distributions of the resistance and the load. Composition operators and performance structure functions thereof are defined to composite different universal generating functions. When composition operators are executed, computational costs can be reduced by means of collecting like terms. The results of theoretical analysis and numerical simulation show that the generating function approach can be widely used for probability modeling of large complex systems with the failure dependence.
Submitted on Dec.19, 2008, received from the Guest Editor on August 25, 2009
References: 07

Common cause failure is an important phenomenon for a system with failure dependent parts. In this paper, several common cause failure models are analyzed and compared. A new common cause failure model for system reliability estimation is presented based on Bayesian Networks. Examples of series system, parallel system and series-parallel system are given to explain how to use the model to evaluate the reliability of system, through which the weak parts of the system can be identified. Also, Monte-Carlo simulation method is used to estimate the reliability based on Bayesian Networks, the results of which are compared with the system reliability under failure independence assumptions. The simulation results show that the reliability model based on Bayesian Network is consistent with the traditional qualitative analysis which proves that the Bayesian Networks model is accurate and valid.
Submitted on October 15, 2008 and made available by Guest Editor on August 25, 2009
References: 19

Petri Nets can represent static structure and dynamic change in a system, and reflect the state of a system and the development of event. It can express spread relationship of faults and identify the minimal cut sets (MCS) according to the relationship between inputs and outputs implied in the Incidence Matrix of Petri Nets. Petri Nets can also be used for qualitative analysis according to the program of the method and shows its advantage when dealing with models containing repeatable events. The proposed method differs from other cut sets-finding methods in fault tree analysis, and utilizes the property of Petri nets. It is easy for programming. In addition, failure detection using reachability concepts and state equation of Petri nets according to the number of tokens or initial states can predict the next state and final state of a system. A sample system is analyzed to show the process of the proposed method.
Received from the Guest Editor on August 25, 2009
References: 14

Complex systems are usually repairable and subject to a few types of different preventive maintenance (including overhaul), which can be imperfect. This paper reanalyzes the bus-motor-failure data of [1]. Four different competing risk models, each of which has three parameters, are fitted to the data and the best model is identified. Based on an examination of the failure rate of the fitted model, two basic aging characteristics are identified and analyzed. The results are useful for modeling and analyzing similar systems.
Received from the Guest Editor on August 25, 2009
References: 10

Hierarchical Bayesian method for estimating the failure probability under Data Only One Failure by taking the s-quasi-Beta distribution as the prior distribution is proposed in this paper. The weighted Least Squares Estimate method was used to obtain the formula for computing reliability distribution parameters under Data Only One Failure. Taking one type of electrical connectors as an example, the estimator of reliability characteristic values was obtained using the above method. Finally the feasibility of the above mentioned method was verified.
Submitted on January 02, 2009.Received from the Guest Editor on August 25, 2009
References: 07

Reliability and security are two major criteria for survivable communications in error-prone wireless environments. To ensure reliable communications, Forward Error Correcting (FEC) codes such as Reed-Solomon (RS) codes are employed for error detection and correction by adding redundancies into the original data to form code words. Secure data communications based on FEC are achieved in many traditional approaches by encrypting the whole code words, which is not computationally or energy efficient. In this paper, we propose a new selective encryption approach based on FEC code words to effectively sustain both green and survivable communications in wireless networking systems.
Received on November 10, 2009; revised March 09, 2010
References: 06