| With the development of industry,systems in aerospace,nuclear energy engineering,weaponry and other engineering fields are becoming more and more complex.And systems in these domains also have the characteristic of multi-state and multi-phases.Once these systems fail,it will cause huge casualties or economic losses.As technology improves,these systems often use high-reliability components or redundant designs to improve the reliability.The application of these technologies leads to common cause failure becoming the main reason for system failure and researches show that CCF has a great impact on the reliability of the system.Therefore,in order to make the reliability assessment of complex systems more accurate,CCF must be considered.Based on the existing literature,this paper proposes reliability assessment method considering CCF for complex systems under the characteristics of complex systems in multi-stage,phased-mission,and phased-mission multi-state.First of all,for a phased-mission system(PMS),since there are dependence between the phases and the interaction among components,there will be dynamic logic such as backup and functional dependency in the system.We can only use a dynamic fault tree to model the system.Dynamic Bayesian Network can not only express the dependence between the phases of a PMS,but also an effective tool for analyzing DFT.Therefore,this chapter first proposes the corresponding multi-stage system dynamic Bayesian network model based on the DFT of the PMS.Then consider the PMS may be affected by CCF,adopts the Efficient Decompose Aggregation method to simplify the CCF.Then obtain the simplified DFT and then convert the simplified DFT into the corresponding PMS-DBN.Moreover use the conditional probability formula to calculate the reliability of the system.Finally,an example is used to illustrate the method,and the correctness of the method is proved by comparing with the result of the traditional method.Then for multi-state systems,due to lack data or fuzzy judgments of experts,it is difficult to use certain values to describe the state performance of the components and the corresponding performance level probability.They can only be expressed by fuzzy values.Therefore,combine fuzzy theory with universal generating function(UGF)to establish a fuzzy universal generating function for MSS.Then,CCF can also occur in MSS,consider the impact of CCF,divide system failures into independent failures and CCF,according to the implicit method.For CCF,use α -factor model to quantitatively description and presenting a method for estimating the α -factor when the data is insufficient.Then,combine the fuzzy UGF and the α -factor model estimating method to conduct the reliability evaluation of the MSS with CCF.At last,the proposed method is illustrated by an example,which proves the effectiveness of the method.Finally,for the multi-state phased mission system(MS-PMS).Considering that the MSPMS system in engineering often has the characteristic of performance sharing,but due to the limitation of the transmission capacity of the public bus during the energy transmission process,the remaining energy of the system cannot be completely transferred to other sub-systems.Moreover,since the subsystems that make up a multi-state and multi-stage system are generally redundant designs such as voting systems or parallel systems,this type of system can be affected by CCF.Therefore,it is considered to extend the general generating function.The establishment of a binary general generating function can indicate the performance level of the system or whether the system affected by CCF.Finally,an example is used to illustrate the method. |
