Using a Simplified Canny Edge Detector to Detect Shock Arrival in Expansion Tubes and Shock Tunnels

The accurate detection of shock arrival in impulse facilities is important because it is a necessary parameter for the calculation of shock speeds, which are used to evaluate the state of the generated test flow. Inaccurate or incorrect calculation of shock speeds can cause either large errors or large uncertainties in calculated or inferred values based on these shock speed measurements. Traditionally, shock speeds have been calculated using the time-of-flight method between wall mounted instrumentation, using user selected shock arrival times. More complex techniques have been proposed, but they are computationally expensive and require user input for each calculation due to their computational expense. This paper proposes using a simplified Canny edge detector to automatically calculate shock arrival. This method is able to run on the whole dataset for each wall pressure signal, while other more complex methods have required cut down data sets, which necessitated user intervention. It runs in much less than a second, and is shown to be able to accurately find shock arrival for a representative set of sample wall pressure sensor data from a high enthalpy expansion tube facility. The Canny edge detector based method is also compared to user selected shock arrival times and shock arrival times calculated automatically using a Bayesian changepoint analysis method with good agreement seen.