34. Data Based Modeling of Aero Engine Vibration Responses

Data based modeling has garnered increased interests in the last decade in vibration response classification, particularly so if the relationships between the response variables and the forcing functions are complex and dependent on multiple factors. Aero engines are one of the most heavily instrumented parts of an aircraft, and the data from various types of instrumentation across these engines are continuously monitored both offline and online for potential anomalies. Measured aero engine responses vary widely in character and amplitude depending on the operating conditions and prevailing environmental conditions. The majority of the vibration assessment is done via monitoring engine vibration levels to fundamental shaft rotational orders. However, focus on shaft orders in isolation may not expose the full picture when a range of other factors are also known to be in effect during operation of a complex machine such as an aero engine. Various complex relationships exist between different parameters such as vibration, temperature, pressure, etc., which are all captured via different instrumentation and vary from engine to engine. A global model to establish the association among the sensitivities of various parameters via their existing instrumentation is highly desirable to establish accurate engine behavior.

In the present work, a data-driven approach towards modeling aero engine vibration responses will be presented. This will utilize instrumentation sensor data of various types, which has been generated from a series of test bed runs. Further, empirical relationships and correlation among different types of sensor measurements with the aero engine responses will also be presented.