Elsevier

Applied Acoustics

Volume 71, Issue 12, December 2010, Pages 1132-1141
Applied Acoustics

Acoustic monitoring of engine fuel injection based on adaptive filtering techniques

https://doi.org/10.1016/j.apacoust.2010.07.001Get rights and content

Abstract

Diesel engines injection process is essential for optimum operation to maintain the design power and torque requirements and to satisfy stricter emissions legislations. In general this process is highly dependent upon the injection pump and fuel injector health. However, extracting such information about the injector condition using needle movements or vibration measurements without affecting its operation is very difficult. It is also very difficult to extract such information using direct air-borne acoustic measurements.

In this work adaptive filtering techniques are employed to enhance diesel fuel injector needle impact excitations contained within the air-borne acoustic signals. Those signals are remotely measured by a condenser microphone located 25 cm away from the injector head, band pass filtered and processed in a personal computer using MatLab. Different injection pressures examined were 250, 240, 230, 220 and 210 bars and fuel injector needle opening and closing impacts in each case were thus revealed in the time–frequency domain using the Wigner–Ville distribution (WVD) technique. The energy of 7–15 kHz frequency bands was found to vary according to the injection pressure. The developed enhancement scheme parameters are determined and its consistency in extracting and enhancing signal to noise ratio of injector signatures is examined using simulation and real measured signals; this allows much better condition monitoring information extraction.

Introduction

Improvement of diesel engines performance necessities has arisen as a result of the increasing interest in environmental problems, and reduction of noise and pollutant emissions. Fuel injection and the intake management system are one of the key components, which determine engine torque, emissions, noise quality and specific fuel consumption [1], [2], [3], [4]. Injection pressure, fuel quantity, injector opening and closing timings and other parameters are keys to the ideal injection process condition monitoring system. These parameters should be kept at their optimum values to reduce the fuel consumption and pollutant emissions, and increase the output power [2]. Injection induced acoustic signals have been sporadically studied for many years. However, previous work has focused on topics other than the noise radiated from the injector itself. Much of the work in the literature on diesel fuel injection has been devoted to describing the noise and vibration generated by the fuel pumps and fuel lines and radiated by the engine block [3], [5], [6].

In the medium size, high speed diesel engines, the moving mass inside the injector is small in the order of 15 g, and this mass takes a very short time, in the order of 1–3 ms, from the fully open to fully closed position. As a result, the noise from the diesel injector is a very short click with transient broad frequency content. This noise is radiated from the surface of the injector itself or transmitted through the fuel system or the engine block. There is a distinct opening and closing vibration and noise for most injectors. The opening vibration and noise is due to the moving mass hitting the upper stop and the closing one is due to the moving mass hitting the seat. Unfortunately, the diagnostic signal is dominated by the other sources, corrupted by a background and interference noise. Coherent filtering techniques and adaptive noise cancelling (ANC) could be used to improve the signal to noise ratio of a diagnostic signal. The drawback with these two methods is that the first relies on a synchronizing signal and the second needs a reference signal which is not always available [7], [8]. In Ref. [9], Independent Component Analysis (ICA) technique was used to decompose the air-borne acoustic signals into their sources and to identify diesel engine noise sources; considerable computations are needed for ICA methods.

In the work reported here, the air-borne acoustic signals were used to monitor injector conditions in acoustically untreated laboratory. Firstly the data was high pass filtered to remove the dominant harmonics and the low frequency bands excited by the room modes. Secondly the adaptive filtering techniques were introduced and the adaptive self-tuning scheme was applied to enhance the transient components of a simulated engine air-borne acoustic signals. Finally the injector needle impact excitations were enhanced and extracted from real air-borne acoustic signals, and the time–frequency domain using the Wigner–Ville distribution (WVD) was employed to give better localisation for these impacts.

It is evident that the proposed procedure has given clearer results to enhance the injector needle opening and closing impacts excitations. More importantly, SNR improvement allows various statistical methods to be successfully used in diagnosing injector related faults.

Section 2 in this paper gives a review of the engine air-borne acoustic signals. The adaptive filtering based schemes are introduced in Section 3; results of simulated signals are also presented. The test rig facilities, combustion dependent acoustic signals determination and the exploitation of the proposed scheme in real air-borne acoustic data are described in Section 4 signals. Section 5 summarises the conclusions.

Section snippets

Engine air-borne acoustic modelling

The noise from a diesel engine is composed of many components emitted from different sources, see Fig. 1. These sources include combustion noise, mechanical noise, and a combination of both. Understanding the components of the acoustic signal is essential to identify the requirements for acoustic signal analysis. Table 1 briefly summarises diesel engine noise excitation forces, their generation mechanisms and noise transmission paths. The combustion noise is produced by the rapid rate of

The proposed adaptive scheme

In a situation when there is a spectral overlap between the signal and noise, conventional fixed filters fail and it is necessary for the filter characteristics to be variable so that it can learn to track the signal of interest which is buried in wide-band variable noise. Such filters are defined as adaptive filters.

The adaptive filter is a filter automatically adjusts its coefficients according to the error signal. A particularly interesting class – the self-adapting finite impulse response

Test rig and instrumentation

The experiments were performed with a four-stroke, four-cylinder, in-line OHV, direct injection, Ford FSD 425 type diesel engine. The schematic diagram for the test rig and instrumentation are shown in Fig. 8. Before the pressure sensor and accelerometer signals were fed to the Analogue-to-Digital Converter (ADC), they passed through a B&K type 2635 charge amplifier to condition the signal. The charge amplifier compensates for the reduction in transducer sensitivity due to the use of long

Conclusions

The adaptive filtering based scheme presented in this work shows promising results in detecting and diagnosing injection pressure faults. This technique was employed for the enhancement of diesel common hole injector events extraction. The following few conclusions could be highlighted:

  • a.

    Diesel engine acoustic signals time domain extracted features, spectral analysis and statistical parameters give limited information. The frequency domain analysis gives only information about the frequency

References (15)

There are more references available in the full text version of this article.

Cited by (53)

  • Fault detection of injectors in diesel engines using vibration time-frequency analysis

    2019, Applied Acoustics
    Citation Excerpt :

    Other similar studies have been conducted on the engine vibration. For example, in a same study, the energy of 7–15 kHz frequency bands was varied according to the injection pressure in a research by Albarbar et al. [31]. Jianmin et al. studied the fuel-injection system of a diesel engine by analyzing the cylinder head vibrations which obtained almost the same results [33].

  • Vibro-acoustic condition monitoring of Internal Combustion Engines: A critical review of existing techniques

    2018, Mechanical Systems and Signal Processing
    Citation Excerpt :

    The noise emitted by an ICE engine consists of many components emitted from different sources (Fig. 2). These sources include combustion, mechanical, and aerodynamic noise [20,21]. An interesting classification of noise source types is given in [22] in which the generation and transmission of both load-dependent and load-independent sources are clearly exploited.

View all citing articles on Scopus
View full text