Vibration analysis and combustion parameter evaluation of CI engine based on Fourier Decomposition Method

Abstract

As a non-intrusive method for engine working condition detection, the engine surface vibration contains rich information about the combustion process and has great potential for the closed-loop control of engines. However, the measured engine surface vibration signals are usually induced by combustion as well as non-combustion excitations and are difficult to be utilized directly. To evaluate some combustion parameters from engine surface vibration, the tests were carried out on a single-cylinder diesel engine and a new method called Fourier Decomposition Method (FDM) was used to extract combustion induced vibration. Simulated and test results verified the ability of the FDM for engine vibration analysis. Based on the extracted vibration signals, the methods for identifying start of combustion, location of maximum pressure rise rate, and location of peak pressure were proposed. The cycle-by-cycle analysis of the results show that the parameters identified based on vibration and in-cylinder pressure have the similar trends, and it suggests that the proposed FDM-based methods can be used for extracting combustion induced vibrations and identifying the combustion parameters.

An Analytical Approach to Converting Vibration Signal to Combustion Characteristics of Homogeneous Charge Compression Ignition Engines

“…In this respect, contemporary black-box methods are capable to reproduce relevant combustion indicators (CA50 or CA of peak PRR) with accuracy sufficient for closedloop combustion control applications, i.e. with a mean error below 1 CAD [21,22] For multi-cylinder CI and SI engines the accuracy deteriorates slightly with corresponding errors not exceeding 1.5 CA [24,40].…”

“…The post-processed in-cylinder pressure is used directly to calculate the PRR and IMEP. PRR is an important parameter determining mechanical stress and is considered a primary excitation for combustioninduced vibration [22]. The CA-dependent PRR function and its peak value (PRR max ) have been calculated directly as a discrete derivative of in-cylinder pressure, smoothed over a single CA measurement interval.…”

“…The authors introduced a model to filter inertia force components that overlapped combustion signals, striving to increase the accuracy of combustion parameters estimation. Yang and Yong [22] used the Fourier decomposition method to extract combustion components in the engine-head acceleration signal. Isolated Fourier intrinsic band functions were reassembled to yield solely combustion-induced vibration signals.…”

“…The experiments, performed at different operating conditions, consistently showed good agreement between pressure-derived and vibration-derived combustion timing data. Furthermore, Yang and Yong [22], using the previously mentioned decomposition approach, were able to couple characteristic points on the vibration wave with locations of combustion onset, maximum PRR and peak pressure. All values were predicted with a maximum ± 0.9 CAD error.…”