Kravchenko O. Methods of forecasting reliability discrete devices by modeling the degradation of computer components

The thesis is devoted to solving important scientific challenges in developing a method of forecasting reliability discrete devices, taking into account the physical properties of the material in the study of the process of degradation. Built improved physical model of a discrete device through the use of models of composite materials and thin shells with inclusions consideration by interfacial layer that provides increased accuracy in the calculation of the fragility of the material discrete device. The improved model takes into account the material properties of the composite computer system, change their properties due to the influence of pressure and temperature during operation. Developed and refined theoretical method of forecasting technical condition of a discrete device, which is used for its proposed model of the composite computer system, which provided the opportunity to build algorithms and software to evaluate the reliability and forecasting technical condition of computer equipment The dependences of the function behaviors material discrete unit of interfacial layer thickness based on the model of discrete physical condition of the device by taking into account changes in the properties of interfacial layers, providing a practical assessment of the technical state of computer systems and components. The practical result of the research is the intelligent subsystem reliability prediction. The accuracy of predicting increased by 7-10%.