Combustion parameters identification and correction in diesel engine via vibration acceleration signal

Abstract

Using closed loop control of the internal combustion engine is beneficial to reducing emission and fuel consumption. Accurate combustion parameters are the foundation of effective closed loop control. Some combustion parameters, including the start of combustion, the location of maximum pressure rise rate and the location of peak pressure can be identified from the vibration signals. Empirical Mode Decomposition (EMD) method is introduced to reconstruct the vibration acceleration signal, from which the combustion parameters are identified. However, there are angle deviations between the combustion parameters extracted from the reconstructed vibration acceleration signal and those from the cylinder pressure. Algorithms to correct the angle deviations are introduced. A system deviation value is used to correct the extracted start of combustion with an error bound being within 0.6°CA. Two algorithms are proposed to correct the deviation between the predicted location of maximum pressure rise rate and that from the cylinder pressure. Test results show that the two algorithms are able to correct the deviation within 0.3°CA error bound. The location of peak pressure can be predicted with the knee point following the peak value in the reconstructed vibration acceleration signal. The predicted result is then corrected using a linear regression of the location of peak pressure versus the knee point within 0.5°CA error bound. A real-time monitoring framework is utilized for calculating the combustion parameter prediction.