论文部分内容阅读
作者运用马尔柯夫过程理论,从动态上分析了多微机系统故障及其转移与修复状态的运动,建立了空间离散、时间连续的可维多机控制系统的可靠性数学模型,並独立地运用拉普拉斯变换理论,推导出多机系统的平均失效间隔时间和系统稳态有效度的简捷计算方法。根据多机控制系统的组成及其运行特点,建立了N/r/m(m≦N)系统(整个计算机系统共有N个子系统,其中r个冗余子系统、m个维修人员)在可靠度与有效度两种参量下的运动方程组,在这些方程组的基础上运用平均失效间隔和稳态有效度的计算方法,导出其求解方程组,分别归纳了系统的平均失效间隔和稳态有效度计算公式。接着,作者应用平均失效间隔计算公式求出双机可维系统与双机不可维系统的平均失效间隔,并分别与单机系统的平均失效间隔进行比较,分析了子系统的维修度对多机系统可靠性的重要影响。根据当前计算机系统实际运行情况,将子系统的平均失效间隔与平均修复时间代入,得到了具体的量化概念,并提出提高系统可靠性的具体措施。
Based on the theory of Markov process, the author dynamically analyzes the faults of multi-computer system and its movement in the state of transfer and repair, and establishes the reliability mathematical model of multi-machine control system which is discrete in space and continuous in time. Laplace transform theory, a simple method to calculate the average failure interval and the steady-state validity of the multi-machine system is deduced. According to the composition of the multi-machine control system and its operation characteristics, the reliability of the N / r / m (m≤N) system (N total subsystems, r redundant subsystems and m maintenance personnel) And the validity of the two parameters under the motion equations, the use of these equations based on the average failure interval and the steady-state validity of the calculation method, derived its solution equations, respectively, the system of the average failure interval and steady-state effective Degree calculation formula. Then, the author calculates the average failure interval of the two-dimensional and two-dimensional non-dimensional systems by using the average failure interval calculation formula and compares them with the average failure intervals of the single-machine systems respectively. The important influence of reliability. According to the actual operation of the current computer system, the average failure interval subsystem and the average repair time into, get a specific quantitative concept, and put forward specific measures to improve system reliability.