Reliability Analysis of Dynamic Load-Sharing Systems with Constrained and Changing Component Performances
Heping Jia,Liudong Xing,Yi Ding,Yanbin Li,Dunnan Liu
DOI: https://doi.org/10.1109/tsmc.2023.3277497
2023-01-01
IEEE Transactions on Systems Man and Cybernetics Systems
Abstract:Considerable research efforts have been expended in modeling load-sharing systems. The existing models, however, have various limitations, such as being limited to the exponential time-to-failure distribution, constant component performances, or performances without constraints. In this article, we make contributions by modeling a dynamic load-sharing system (DLSS), where the performance of each component is dynamic according to prespecified load-sharing principles and is limited by its capacity constraint. Moreover, the capacity constraint of a component can reduce due to degradations. In the proposed model, increasing failure rates are also involved since the surviving components must share the load of the failed component and continue working with increasing stresses. When the desired performance for a component exceeds the limitation, the entire system fails. An extended Markov process (EMP) method is proposed for evaluating the reliability of the considered DLSS with nonrepairable components. The proposed analytical method is flexible in handling arbitrary component time-to-failure distributions and in handling diverse load allocation mechanisms. Numerical studies of a power transmission system and a water transmission system are provided to validate the proposed method and its advantages. Effects of several model parameters are also investigated through case studies.