Modeling of raw data pattern classification of wind tunnel test data in ILST.

Wind tunnel testing plays an important role in determining the design of infrastructure and transportation media whose performance is influenced by aerodynamic effects. In a typical wind tunnel test, a set of raw data is generally produced that represents the results sensed by the measuring instrumentation. This data set can be patterned for various purposes, such as classifying wind tunnel test types and checking the readiness level of wind tunnel test systems. This research activity seeks to model the pattern of raw data from the Indonesian Low Speed Tunnel wind tunnel test, by compiling a data structure consisting of four raw data clusters, namely pressure, temperature, force/moment, and attitude of the test model, conditioning the data structure so that it forms a 16 different raw data patterns, and perform numerical tests to verify and validate the classification of raw data patterns using the K-Nearest Neighbor method. The results show that the developed model is capable of classifying according to the expected 16 patterns and can be projected to support decision making in examining the level of readiness of the wind tunnel test system. [ABSTRACT FROM AUTHOR]