| With the rapid development of computer science and software engineering technologies,they are widely used in all walks of life,and play a pivotal role in such safetycritical systems as modern avionics systems.For the modern avionics system,its scale and complexity are growing,and the accidental failure of the system or system failure can cause significant damage to property and endanger lives.Therefore,the reliability analysis of safety-critical systems such as avionics systems is essential,not only to ensure the normal function of the system,but also to fully consider and evaluate the reliability of the system in the process of requirements analysis and model design.The avionics system is a safety-critical system with high reliability requirements.It is an important part of modern civil and military aircraft,with information sensing and display functions,and is the information processing hub of the aircraft,in addition to providing essential functions such as navigation,human-machine interface and external communication for aircraft operations.With the increasingly complex and diverse needs of aircraft,there are higher requirements for the function,performance and reliability of the avionics system,especially the reliability of the avionics system,which directly affects the normal operation of the aircraft and the safety of the aircraft occupants.With the introduction of model-driven development method,researchers have combined system engineering and model-driven to propose model based system engineering(MBSE),in which modeling language is the core of MBSE idea.In this paper,we aim to support the formal modeling and verification of avionics software reliability requirements,and establish Safety_Sys ML Flow Model(S2MFM)based on the mainstream MBSE modeling language Sys ML,combined with AADL language,and on the basis of Safety_Sys ML Flow Model modeling Based on the Safety_Sys ML Flow Model,we analyze the reliability of avionics system using the dual flow graph error propagation theory to verify whether the system satisfies the reliability statute in the requirements,and our main contributions include:· Constructing Safety_SysML flow model,a safety-critical software modeling language for avionics software,defining the graphical elements needed for Safety_Sys ML flow model and the relevant attributes for system reliability verification,constructing the flow model reliability modeling elements,and completing the modeling for system reliability verification.· The Safety_SysML flow model error propagation analysis method based on dual flow diagrams for analyzing and verifying the software reliability of avionics system is proposed,enabling it to verify the reliability requirements of safety-critical systems.· Proposed a transformation method from the Safety_Sys ML flow model to the dual-flow graph error propagation model,enabling it to be automatically transformed to the dual-flow graph error propagation model after the Safety_Sys ML flow model modeling is completed.· A reliability verification tool for avionics system based on Safety_Sys ML flow model is designed and implemented,which supports Safety_Sys ML flow model language modeling and reliability verification.In order to demonstrate the feasibility as well as the effectiveness of the method,the specific application of Safety_Sys ML flow model reliability modeling,verification and analysis is illustrated in this paper with the flight management system as an example. |