Fault diagnosis has traditionally been employed to some safety-critical mechanical systems or to those for which downtime constitutes considerable financial loss. Typically, sensors monitor environmental and loading conditions in the application environment. Algorithms are then designed to extract information from the sensor readings, and compare them against a baseline to determine whether abnormal conditions exist and, if yes, what the root cause might be. While this technology has been matured to some degree for mechanical (and also structural) systems, what has been ignored until recently is that most of today's complex systems contain significant amount of electronics. Indeed, there is a priori reliability evidence that electronics may fail earlier than mechanical components.
In the aerospace domain, flight and ground crews require health state awareness and prediction technologies across all systems (including structures, propulsion, and various subsystems) that can accurately diagnose faults, anticipate failures, and predict the remaining life. This includes those from avionics. Indeed, electronic components have an increasingly critical role in on-board, autonomous functions for vehicle controls, communications, navigation, radar systems, etc. Future aircraft systems, such as the more electric aircraft or the Next Generation Air Transportation Systems (NGATS) will certainly rely on more electric and electronic components. The assumption of new functionality will also increase the number of electronics faults with perhaps unanticipated fault modes. In addition, the move toward lead-free electronics and Micro-Electro-Mechanical Systems (MEMS) will further result in unknown behavior. To improve aircraft reliability, assure in-flight performance, and reduce maintenance costs, it is therefore imperative to provide system health awareness for digital electronics. To that end, an understanding of the behavior of deteriorated components is needed as well as the capability to anticipate failures and predict the remaining life of embedded electronics.
The development and advancement of this capability is also relevant to multiple Exploration Systems Mission Directorate (ESMD) vehicles, including the Orion, Ares, and future vehicles such as Lunar Surface Access Module (LSAM). In addition, there is relevance to long-endurance robotic space missions from ESMD and SMD.
Generally, an understanding of intrinsic and extrinsic degradation mechanisms of component level devices is crucial for the adoption and application of health management to systems. Within the field of electronics, knowledge of semiconductor degradation under various system and environmental scenarios may be coupled with prognostic algorithms to predict future state and time–to–failure of semiconductor components.
The existence of measurable extrinsic degradation precursors, pertaining to device packaging, has been well established in literature for power transistor devices. In recent literature intrinsic degradation precursors, related to the physical properties of the semiconductor, have also been observed. However, it is not widely known how degradation mechanisms propagate as a function of environmental conditions and various stressors. The attainment of such knowledge is critical for advancements in the field of power electronics health management and prognostics. Therefore, the ability to perform large scale experiments on semiconductor devices for characterization of degradation precursors under various scenarios is of great interest. Additionally, the first phase of system implementation and its initial application to Insulated Gate Bipolar Transistors (IGBTs) in a thermal overstress scenario is presented.