Battaglia C., Bregant L., Zugna A.

To know the performance and characteristics of competition vehicles has always been a common aim in the motorsport world. This paper focuses on the development of a procedure to extract the straight-line performance of a racing motorcycle by measurement of the exhaust noise signal. The method is divided into three phases. In the first one, experimental measurements are carried out by placing a single microphone on the pit-wall, and acquiring an acoustic signal from the motorcycle as it passes by. The second phase consists of processing the acquired sound signal. Engine order tracking is performed by means of an autoregressive (AR) spectrogram, based on an adaptive Kalman filter; this proved to be an excellent algorithm to calculate the time-varying coefficients of the model. In the third phase, the signal associated with the extracted order is fitted by a mathematical model of the motorcycle, which provides as a final result the evolution of speed, acceleration and engine RPM on the straight. In this procedure, Doppler effect, time delay and microphone distance from the path are taken into account. Finally, an approximation of engine torque and power curves, as well as mass and drag coefficients, is given. The procedure has been validated comparing the numerical results with the data acquired on-board the Yamaha M1 race motorcycle. Correlation shows a good match between the predicted curves and the actual ones, giving confidence in the capability of the method to evaluate the performance of competition motorcycles.