Artificial Neural Network for Airborne Noise Prediction of a Diesel Engine

The engine acoustic character has always represented the product DNA, owing to its strong correlation with in-cylinder pressure gradient, components design and perceived quality. Best practice for engine acoustic characterization requires the employment of a hemi-anechoic chamber, a significant number of sensors and special acoustic insulation for engine ancillaries and transmission. This process is highly demanding in terms of cost and time due to multiple engine working points to be tested and consequent data post-processing. Since Neural Networks potentially predicting capabilities are apparently un-exploited in this research field, the following paper provides a tool able to acoustically estimate engine performance, processing system inputs (e.g. Injected Fuel, Rail Pressure) thanks to the employment of Multi Layer Perceptrons (MLP, a feed forward Network working in stationary points). In particular, the investigation addressed the estimation of direct Combustion Noise (CN), Sound Power Averaged over the main radiating surfaces, Loudness and Modulation. The NN was trained and tested in low to medium load/speed operating conditions of a 4 cylinders inline turbocharged Diesel Engine. The outcome is more than encouraging, achieving less than 0.5 % of Test Root Mean Square Error (RMSE) for the estimation of CN and Sound Power, less than 2 % Test RMSE for Loudness, less than 4 % Modulation. In addition, the same networks were calibrated on single microphones and third of octaves parameters (CN, Sound Power and Loudness) with slightly lower accuracy of the results.