COMBUSTION MONITORING THROUGH VIBRATIONAL DATA IN A TURBOCHARGED CITY CAR ENGINE

International Journal of Mechanical Engineering and Technology (IJMET) Volume 8, Issue 3, March 217, pp. 197 28 Article ID: IJMET_8_3_22 Available online at http://www.iaeme.com/ijmet/issues.asp?jtype=ijmet&vtype=8&itype=3 ISSN Print: 976-634 and ISSN Online: 976-6359 IAEME Publication Scopus Indexed COMBUSTION MONITORING THROUGH VIBRATIONAL DATA IN A TURBOCHARGED CITY CAR ENGINE G. Chiatti, O. Chiavola, E. Recco Department of Engineering, ROMA TRE University Rome, Italy ABSTRACT Condition monitoring and optimization of diesel engine has been the focus of a wide research approaches. Techniques have been developed in which in-cylinder pressure measurements are used to calculate peak pressure and burn rates. In the recent past, vibration, acoustic and speed measurements have received considerable attention to this purpose. Methodologies have been developed in which these nonintrusive measurements are employed to estimate the combustion progress. This work is devoted to assess the potential application of a methodology developed by the authors, in which the engine block vibration is used to estimate indicators able to characterize the combustion development. Previous research activity demonstrated that an accelerometer sensor placed in a selected position of the engine block is quite sensitive to the combustion process in a naturally aspirated two-cylinder common rail diesel engine mainly used in micro cars. The objective of this work is to evaluate the applicability of the methodology to a more complex engine architecture (the same engine was downsized by equipping it with a small turbocharger). Measurements were performed in the engine operative field in which the turbocharger is truly effective. The acquired signals were processed in time and frequency domains. Obtained results proved the good accuracy of the estimation of combustion indicators (crank angle corresponding to start of combustion, 5% of mass fraction burnt) via accelerometer signal processing. Key words: Diesel engine, combustion process, engine vibration, combustion indicators. Cite this Article: G. Chiatti, O. Chiavola, E. Recco, Combustion Monitoring Through Vibrational Data in a Turbocharged City Car Engine. International Journal of Mechanical Engineering and Technology, 8(3), 217, pp. 197 28. http://www.iaeme.com/ijmet/issues.asp?jtype=ijmet&vtype=8&itype=3 1. INTRODUCTION Performance, fuel consumption, noise and pollutant emissions in diesel engines can be improved via the closed-loop control of the injection strategy based on the location of combustion progress indicators. http://www.iaeme.com/ijmet/index.asp 197 editor@iaeme.com

Combustion Monitoring Through Vibrational Data in a Turbocharged City Car Engine Start of combustion (SOC) and center of combustion (usually named MFB5, the crank angle position corresponding to the 5% of burnt mass fraction) provide important information on combustion position that has demonstrated to be profitable in combustion control methodologies. In-cylinder pressure sensors have been widely used for combustion investigation. Yang et al. [1] developed a combustion process control based on the start of injection and fuel rail pressure. The controlled combustion parameters are combustion phasing and combined information of ignition delay and combustion width. Lujàn et al. [2] presented a method to detect sudden changes in the cylinder pressure and to evaluate the start of combustion, heat release peak and end of combustion. Lee at al. [3] studied the reduction of engine pollutant emission via a closed-loop control system that receives feedback from 5 % of the mass. Al- Durra et al. [4] proposed an algorithm that extracts from the in-cylinder sensor, the pressure signal allowing for the estimation of 5% burnt rate location. D Ambrosio et al. [5] used incylinder pressure measurements to obtain detailed information on the combustion development. At the moment, problems related to measurement long term reliability and cost proved that the use of in-cylinder pressure transducer is not suitable for on-board installation to estimate in real-time the combustion progress. To overcome this limitation, research activities over the past years have been devoted to the development of innovative methodologies. Low-cost non-intrusive sensors such as accelerometers, microphones and crank angle speed sensors have been proposed to characterize the combustion process bypassing the employment of in-cylinder pressure transducers. Engine vibration, acoustic emission and speed fluctuation sensors offer the advantages of guaranteeing the absence of any type of interaction with the engine operation. The sensor can be installed in any type of engine without the need of a modification. Instantaneous engine speed signal has been used for in-cylinder pressure analysis. Charchalis et al. [6] implemented a method for angular speed signal processing to detect engine failure. Moro et al. [7] investigated the use of the engine speed signal to evaluate the in-cylinder pressure by means of the frequency response function of the engine. Tagliatela et al. [8] used the crankshaft speed to reconstruct the in-cylinder pressure development. Liu et al. [9] presented an experimental study on a multi-cylinder diesel engine for the estimation of the in-cylinder pressure from the crankshaft speed fluctuation. Noise radiation contains information about several processes taking place within the engine. Kual et al. [1] used acoustic emission sensors to identify combustion and various engine events. Chiatti et al. [11, 12] demonstrated the applicability of a technique in which the engine sound emission was related back to the combustion process development. Torii [13] presented a methodology to separate engine noise radiation into mechanical and combustion contributions. Gu et al. [14] and Ball et al. [15] modelled the sound generation of a diesel engine based on the in-cylinder pressure trend and used it for engine condition monitoring. Vibration based methodologies have been proposed and have demonstrated that the engine vibration can be strongly related to the combustion progress if a proper choice of the transducer position is accomplished. Jia et al. [16] presented a method to reconstruct the incylinder pressure from the accelerometer signal via the frequency response function of the engine block. Massey et al. [17] explored the use of accelerometers for combustion phasing sensing in advanced combustion. Morello et al. [18] analyzed the relationship between heat release rate and accelerometer measurements in a large displacement inline 6-cylinder diesel engine. Merkisz et al. [19] assessed the combustion process correctness in a direct injection http://www.iaeme.com/ijmet/index.asp 198 editor@iaeme.com

G. Chiatti, O. Chiavola, E. Recco compression ignition engine on the basis of vibrational signals. Sharma et al. [2] used accelerometer signal for internal combustion engine misfire detection. Previous research by the authors has been devoted to investigate if and how the combustion process affects the block vibration of a naturally aspirated two cylinder common rail diesel engine. Different positions and orientations of a piezoelectric mono-axial accelerometer were tested in order to find the optimal location for the transducer which makes it sensitive to the combustion event [21]. The analysis of the acquired accelerometer and incylinder pressure signals in time and frequency domain demonstrated that a correlation exists between combustion development and engine block vibration [22]. A methodology was implemented in which the accelerometer trace is processed; the vibration components related to the combustion event are insulated and used to evaluate some indicators of the combustion development [23]. This work is devoted to assess the potential application of the developed methodology to a small turbocharged diesel engine. The engine, in naturally aspirated configuration, is mainly used for urban vehicles, microcars, small commercial and leisure vehicles applications. Accelerometer and in-cylinder pressure signals were acquired and analyzed with the objective of evaluating the accuracy of indirect combustion sensing in a complex engine configuration (the trapped mass varies with engine speed and load according to the turbocharged/engine matching; the turbocharged contributes with additional components to the overall vibration signal). Start and center of combustion evaluated via processed accelerometer signals were compared with those obtained by heat release law. The accuracy of the estimations demonstrated that indirect combustion sensing via structure vibration measurements has the potential of being used to provide a feedback signal for control purposes. By comparing the combustion indicators estimated via the accelerometer trace to set-points previously defined, it is possible to generate a correction of the injection settings thus to guarantee the optimal engine performance in terms of fuel consumption, pollutants emission, noise radiation. 2. EXPERIMENTAL SETUP The experimentation was performed in the test cell of ROMA TRE University. Some preliminary tests were devoted to investigate the turbine behavior embedded in the engine system; the compressor was not connected to the intake system of the engine, but was connected to an instrumented pipe in which mass flow, pressure and temperature were measured. A spherical valve was placed at the end of the pipe thus to simulate the pressure losses caused by the engine intake system. 2.1. The Engine The tested engine is a common rail two cylinder diesel engine. Its technical specifications are summarized in Table 1. It is mainly used for urban vehicles, microcars, small commercial and leisure vehicles applications. The engine is manufactured in naturally aspirated configuration; its intake and exhaust systems were modified in order to equip the engine with a turbocharger (IHI RMB31). The engine geometrical compression ratio was not modified. The engine was connected to a SIEMENS 1PH7 asynchronous motor (nominal torque 36 Nm, power 7 kw) and was fully instrumented for pressure and temperature measurements along intake and exhaust systems. The engine under investigation was equipped with incylinder pressure transducers (AVL GU13P). An accelerometer (Endevco 724C) was used for vibration measurements: it is a mono-axial piezoelectric transducer whose sensitivity is 3pC/g and resonance frequency 9 khz. Its signal was conditioned via B&K Nexus device. The accelerometer was mounted with a threaded pin on one stud of the engine block, on the http://www.iaeme.com/ijmet/index.asp 199 editor@iaeme.com

Combustion Monitoring Through Vibrational Data in a Turbocharged City Car Engine side of one cylinder. It was oriented in line with the cylinders axis(figure 1). This position was selected based on a previous experimentation that highlighted that the vibration signal was able to ensure the combustion sensing in both cylinders [21]. Table 1 Engine main characteristics Two cylinder common rail diesel engine displacement 44 cm 3 stroke bore 6.6 mm 68 mm compressionratio 2:1 (a) (b) Figure 1 (a) Test bed; (b) Accelerometer installation on the engine block. The engine speed was measured by using an angular sensor (AVL 364C) with 288 pulses/revolution. All signals were simultaneously digitized by National Instruments data acquisition devices (boards type 611 for analogical signals and type 6533 for digital signals). A custom program was developed by the authors to manage the data monitoring and acquisition. The sampling frequency was varied based on the engine speed value, thus to ensure a fixed crank angle resolution of all signals. 2.2. Tests In order to demonstrate the correlation between the accelerometer trace and combustion progress when the engine is in turbocharged configuration, several tests have been carried out by imposing to the engine different values of load and speed. Steady state tests were performed from 28 to 44 rpm, with a step of 4 rpm (this field corresponds to the expected engine operational range). Three loads were imposed, corresponding to 6, 8 and 1% of the maximum torque output. During tests, ECU managed the injection strategy, thus to set a two-pulse injection mode for all engine operation range. During pre-injection, a fixed amount of 1 mm 3 /str of fuel was delivered. Main injection was set in order to guarantee the imposed value of load. ECU strategy was maintained unchanged both for naturally aspirated and turbocharged configuration in order to test how the increase of air mass affects the combustion process and the engine vibration. Data were collected only after the engine had completed its warm up and had reached nominally stationary conditions (coolant temperature reached 8 C). During all tests, the inlet air temperature and humidity were about 23 C and 45%, respectively. http://www.iaeme.com/ijmet/index.asp 2 editor@iaeme.com

G. Chiatti, O. Chiavola, E. Recco For each tested engine operating condition, 25 engine cycles were acquired; an algorithm was developed to elaborate the data in order to compute the average signals thus to attenuate the engine cyclic irregularities. 3. RESULTS AND DISCUSSION The first part of this section is devoted to present some crank angle evolutions of in-cylinder pressure and accelerometer signal. Data related to both naturally aspirated and turbocharged configurations are presented. The second part of this section focuses on the engine turbocharged configuration; results of data processing in time and frequency domain are presented. Figure 3 shows the comparison between in-cylinder pressure obtained at 4 rpm, full load condition for naturally aspirated and turbocharged configuration. Motored trace is also reported. pressure [bar] 1.E+2 9.E+1 8.E+1 7.E+1 6.E+1 5.E+1 4.E+1 3.E+1 2.E+1 1.E+1.E+ -36-27 -18-9 9 18 27 36 Figure 2 In-cylinder pressure at 4 rpm: 1% load in turbocharged configuration aspirated configuration, motored, 1% in It is possible to observe the effect of turbocharged on the pressure trace; since the engine geometrical pressure ratio was not modified during tests, the increase of trapped mass is responsible of the increase of the maximum pressure value. In Figures 3 and 4, these in-cylinder pressure traces are shown with the corresponding acquired accelerometer signals. The plots highlight that the engine block vibration is characterized by a contribution at low frequency that corresponds to the double of the engine frequency speed. High frequency components are also exhibited. They are caused by many mechanical sources, such as intake and exhaust valve opening and closing, fuel injection, piston slap; combustion process also contributes to the overall signal. The combustion effect on the vibration trace is highlighted by high frequency and high amplitude oscillations. The combustion related vibration components can be observed not only in the crank angle interval in which the combustion takes place in the cylinder where the pressure transducer was installed. Combustion signature is pointed out also in the crank angle range in which combustion process takes place in the other cylinder (combustion events are spaced 36 cad apart). http://www.iaeme.com/ijmet/index.asp 21 editor@iaeme.com

Combustion Monitoring Through Vibrational Data in a Turbocharged City Car Engine pressure [bar] 1,E+2 8,E+1 9,E+1 6,E+1 8,E+1 4,E+1 7,E+1 6,E+1 2,E+1 5,E+1,E+ 4,E+1-2,E+1 3,E+1-4,E+1 2,E+1 1,E+1-6,E+1,E+ -8,E+1-36 -27-18 -9 9 18 27 36 accelerometer signal [m/s 2 ] Figure 3 In-cylinder pressure, accelerometer signal at 4 rpm, 1% load (aspirated configuration). pressure [bar] 1,E+2 9,E+1 8,E+1 7,E+1 6,E+1 5,E+1 4,E+1 3,E+1 2,E+1 1,E+1 8,E+1 6,E+1 4,E+1 2,E+1,E+ -2,E+1-4,E+1-6,E+1,E+ -8,E+1-36 -27-18 -9 9 18 27 36 accelerometer signal [m/s 2 ] Figure 4 In-cylinder pressure, accelerometer signal at 4 rpm, 1% load (turbocharged configuration). Figure 5 presents the accelerometer signals acquired during motored and fired tests. Aimed at highlighting the vibration contribution related to the combustion event, the trace obtained by subtracting to the in-cylinder pressure trend acquired in fired condition that one acquired during motored test is also shown. pressure [bar] 1,E+2 9,E+1 8,E+1 7,E+1 6,E+1 5,E+1 4,E+1 3,E+1 2,E+1 1,E+1 8,E+1 6,E+1 4,E+1 2,E+1,E+ -2,E+1-4,E+1-6,E+1,E+ -8,E+1-36 -27-18 -9 9 18 27 36 accelerometer signal [m/s 2 ] Figure 5 Difference between in-cylinder pressure at 4 rpm during fired (1% load) and motored condition, accelerometer signal at 1% load, accelerometer signal in motored test. http://www.iaeme.com/ijmet/index.asp 22 editor@iaeme.com

G. Chiatti, O. Chiavola, E. Recco Differences in block vibration signals are exhibited mainly in the crank angle intervals in which the combustion takes place (around and 36 cad), thus pointing out the accelerometer sensor ability to sense the combustion process in both the cylinders. Aimed at examining the relation between the engine block vibration and the combustion event, an analysis in the frequency domain of the signals was performed. According to previous works [22, 23], coherence function between in-cylinder pressure trace and engine vibration was computed in terms of power spectral densities and cross-power spectral density of the signals. In the computation of the coherence function, windowed signals have been used, in order to isolate the contribution of the combustion process in the cylinder under investigation. Figure 6 shows the coherence function trends obtained for naturally aspired and turbocharged configurations at the engine condition 4rpm, full load. The curves highlight the existence of a frequency range around 2 Hz, in which coherence is characterized by the highest values. 1.8 coherence [-].6.4.2 1 2 3 4 5 6 frequency [Hz] Figure 6 Coherence function at 4 rpm, 1% load: naturally aspirated and turbocharged The analysis of coherence function trends and spectrograms in the engine operational range allowed to figure out that the in-cylinder pressure curve exhibits good correlation with accelerometer trend in a specific frequency band, that has demonstrated to be reliant on the engine speed value, load condition, injection setting [24]. In order to remove from the vibration all the components due to sources other than those caused by the combustion process and select only the combustion related contributions, accelerometer signals have been band-pass filtered in the range of frequencies where the traces demonstrated to be highly correlated with in-cylinder pressure development. Figure 7 shows the filtered accelerometer trace obtained at 4 rpm and full load condition. Data were normalized with respect to their maximum value. In-cylinder pressure is also plotted to highlight the crank angle intervals in which combustion process takes place. http://www.iaeme.com/ijmet/index.asp 23 editor@iaeme.com

Combustion Monitoring Through Vibrational Data in a Turbocharged City Car Engine pressure [bar] 1.E+2 9.E+1 8.E+1 7.E+1 6.E+1 5.E+1 4.E+1 3.E+1 2.E+1 1.E+1.E+ -1.5-36 -27-18 -9 9 18 27 36 Figure 7 In-cylinder pressure, normalized filtered accelerometer signal at 4 rpm, 1% load. The plot highlights that the filtering allowed isolating the components of the vibration related to the combustion process in both cylinders. In order to relate the engine vibration to the combustion progress, some indicators were computed by means of the rate of heat release (ROHR). It was evaluated starting from the incylinder pressure measurements through a thermodynamic model in which the rate of heat transfer to the walls was accounted for. Once ROHR was computed, the cumulated sum of rate of heat release was evaluated. It provides important information about the combustion evolution, such as start of combustion, center of combustion and combustion duration. By comparing CHR curve to filtered accelerometer signal, it was highlighted that some points of this curve are always able to locate the crank angle values corresponding to the SOC and MFB5. Figure 8 shows the filtered accelerometer signal overlapped to the rate of heat release. Figure 9 presents the cumulated sum of rate of heat release. Both plots are related to the engine operating condition 4 rpm, 1% of load. SOC corresponds to the crank angle value of a zero-crossing in the filtered accelerometer trace (it is the start of a negative oscillation that comes before the maximum oscillation). MFB5 corresponds to the crank angle value in which the filtered accelerometer trace reaches is minimum value. In the plots, circles have been used to highlight the crank angle corresponding to the SOC and MFB5. 1.5 1.5 -.5-1 normalized filtered accelerometer signal [-] normalized ROHR [-] 1.9.8.7.6.5.4.3.2.1-1 -45 45 9 135 1.8.6.4.2 -.2 -.4 -.6 -.8 normalized filtered accelerometer signal [-] Figure 8 Rate of heat release curve, normalized filtered accelerometer signal at 4 rpm, 1% load. http://www.iaeme.com/ijmet/index.asp 24 editor@iaeme.com

G. Chiatti, O. Chiavola, E. Recco normalized CHR [-] -1-45 45 9 135 Figure 9 Cumulated sum of rate of heat release, accelerometer signal at 4 rpm, 1% load. The same processing was performed with all acquired data; plots in Figures 1 and 11 show the relation between the combustion indicators evaluated by means of the computer CHR and estimated via the filtered vibration trace. Figure 1presents the data obtained during tests conducted on naturally aspirated engine configuration. Figure 11 shows the values obtained on the turbocharged engine configuration. In each plot, points are related to a three values of engine speed (36, 4 and 44 rpm), three load conditions (6%, 8% and 1%). Data on x axis show the crank angle value computed via CHR. Crank angle values in y axis were computed via filtered accelerometer trace. In both plots, the interpolation lines and the corresponding R-squared values are shown (they are the square of the correlation coefficients). The obtained R values are in all cases very close to the unity, giving a measure of the very high reliability of the relationship between the combustion indicators estimated via accelerometer transducer and computed by direct in-cylinder pressure measurements. 4 1.9.8.7.6.5.4.3.2.1 1.8.6.4.2 -.2 -.4 -.6 -.8 normalized filtered accelerometer signal [-] 3 2 1 R² =.9951-1 -2-2 -1 1 2 3 4 Figure 1 SOC, MFB5 for engine naturally aspired configuration: 36 rpm, 4 rpm, 44 rpm. http://www.iaeme.com/ijmet/index.asp 25 editor@iaeme.com

Combustion Monitoring Through Vibrational Data in a Turbocharged City Car Engine 4 3 2 1 R² =.9853-1 -2-2 -1 1 2 3 4 Figure 11 SOC, MFB5, engine turbo charged configuration: 36 rpm, 4 rpm, 44 rpm Table 2 reports the mean absolute error values (MAE) computed aimed at obtaining a measure of the deviation of predictions from their values computed by in-cylinder pressure measurements. For each rating-computed pair, the absolute error is evaluated. By first summing these absolute errors of the pairs and then computing the average, MAE was estimated. Table 2 Mean absolute error Naturally aspirated Turbocharged SOI MFB5 SOI MFB5,78,88,84 1,49 The Table highlights that SOI mean absolute error is always less than 1 cad; MFB5 mean absolute error was always less than 1.5 cad, thus providing evidence of the accuracy of the methodology. 4. CONCLUSION An experimental investigation has been carried out on a turbocharged two cylinder common rail diesel engine, with the aim to assess the potential application for combustion control purposes of combustion indicators evaluated by non intrusive measurements. An accelerometer has been placed on the engine block and engine structure vibration measurements have been performed in the engine complete operational range. The acquired signals have demonstrated that the engine vibration is strongly sensitive to the combustion events. The proper processing of the accelerometer signal allowed estimating the combustion indicators that are used in combustion control methodologies (SOC, MFB5). The results obtained were compared to the same indicators computed by means of in-cylinder pressure measurements. The accuracy of the estimations demonstrates that engine block vibration is a promising solution for remote combustion sensing in order to guarantee the combustion effectiveness in terms of fuel consumption, pollutant emission, noise radiation. http://www.iaeme.com/ijmet/index.asp 26 editor@iaeme.com

REFERENCES G. Chiatti, O. Chiavola, E. Recco [1] Yang, Z., Stobart, R. and Winward, E. Online adjustment of start of injection and fuel rail pressure based on combustion process parameters of diesel engine. SAE Technical Paper 213-1-315, 213. [2] Lujàn, J. M., Bermudez, V., Guardiola, C. and Abbad, A. A methodology for combustion detection in diesel engines through in-cylinder pressure derivate signal. Mechanical Systems and Signal Processing,24(7), 21, pp. 2261-2275. [3] Lee, S., Lee, J., Lee, S., Kim, D., Lee, Y., Yu, S. and Choi, H. Study on reduction of diesel engine out emission through closed loop control based on the in-cylinder pressure with EGR model. SAE Technical Paper 213-1-322,213. [4] Al-Durra, A., Canova, M. and Yurkovich, S.A real-time pressure estimation algorithm for closed-loop combustion control. Mechanical Systems and Signal Processing, 38(2), 213, pp.411-427. [5] d Ambrosio, S., Ferrari, A. and Galleani, L. In-cylinder pressure-based direct techniques and time frequency analysis for combustion diagnostics in IC engines. Energy Conversion and Management, 99, 215, pp.299-312. [6] Charchalis, A. and Dereszewski, M.Processing of instantaneous angular speed signal for detection of a diesel engine failure. Mathematical Problems in Engineering, 213, 213, pp.659243. [7] Moro, D.,Cavina, N. and Ponti, F.In-Cylinder Pressure Reconstruction Based on Instantaneous Engine Speed Signal. J of Eng for Gas Turbines and Power, 124, 22, pp. 22-225. [8] Taglialatela, F.,Lavorgna, M.,Mancaruso, E., and Vaglieco, B.M. Determination of combustion parameters using engine crankshaft speed. Mechanical Systems and Signal Processing, 38(2), 213, pp. 628 633. [9] Liu, F.,Amaratunga, G.A.J., Collings, N. and Soliman, A. An experimental study on engine dynamics model based in-cylinder pressure estimation. SAE Technical Paper 212-1-896,212. [1] Kaul, B., Lawler, B. and Zahdeh, A. Engine Diagnostics Using Acoustic Emissions Sensors. SAE Int J Engines, 9(2), 216, pp. 684-692. [11] Chiatti, G.,Chiavola, O. and Recco, E. Analysis of the relationship between noise emission and in-cylinder pressure in a small displacement diesel engine. SAE Technical Paper 214-1-1364, 214. [12] Chiatti, G.,Chiavola, O., Palmieri, F. and Piolo, A. Diagnostic methodology for internal combustion diesel engines via noise radiation. Energy Conversion and Management,89, 215, pp. 34-42. [13] Torii, K. Method Using Multiple Regression Analysis to Separate Engine Radiation Noise into the Contributions of Combustion Noise and Mechanical Noise in the Time Domain. SAE Int J Engines,7(3),215, pp. 152-1513. [14] Gu, F., Li, W., Ball, A.D. and Leung, A.Y.T. The condition Monitoring of Diesel Engines Using Acoustic Measurements Part 1: Acoustic Characteristics of the Engine and Representation of the Acoustic Signals. SAE Technical Paper 2-1-73, 2. [15] Ball, A.D., Gu, F. and Li, W. The condition Monitoring of Diesel Engines Using Acoustic Measurements Part 2: Fault Detection and Diagnosis. SAE Technical Paper 2-1- 368, 2. http://www.iaeme.com/ijmet/index.asp 27 editor@iaeme.com

Combustion Monitoring Through Vibrational Data in a Turbocharged City Car Engine [16] Jia, L., Naber, J. D. And Blough, J.R. Frequency response function adaptation for reconstruction of combustion signature in a 9-L diesel engine. Proc IMechE Part C: J Mechanical Engineering Science, 229(17), 215, pp. 371 383. [17] Massey, J. A. and Drallmeier, J.A. Identification of the dominant combustion characteristics on homogeneous charge compression ignition engine noise. International J of Engine Research, 16(4), 215, pp. 518-53. [18] Morello, A.,Blough, J.,Naber, J.and Jia, L. Signal Processing Parameters for Estimation of the Diesel Engine Combustion Signature. SAE Int J Passeng Cars - MechSyst, 4(2), 211, pp. 121-1215. [19] Merkisz, J. and Waligórski, M. Strategy of the combustion process diagnosis in direct injection engines. Procedia Engineering: Modelling of Mechanical and Mechatronic Systems, 96, 214, pp. 294-31. [2] Sharma, A., Sugumaran, V. and Babu Devasenapati, S.Misfire detection in an IC engine using vibration signal and decision tree algorithms. Measurement, 5, 214, pp.37-38. [21] Arnone, L.,Boni, M.,Manelli, S.,Chiavola, O.,Conforto, S. andrecco, E. Block Vibration Measurements for Combustion Diagnosis in Multi-Cylinder Common Rail Diesel Engine. SAE Technical Paper 29-1-646, 29. [22] Chiatti, G., Chiavola, O. And Recco, E. Accelerometer measurements to optimize the injection strategy. SAE Technical Paper 212-1-1341, 212. [23] Chiatti, G.,Chiavola, O. and Recco, E. Combustion diagnosis via block vibration signal in common rail diesel engine. International J of Engine Research, 15,214, pp. 654 63. [24] Chiatti, G., Chiavola, O. and Recco, E.Accelerometer Signal to Characterize the Combustion Development in Multiple Injection Diesel Engine. Proceedings ASME Internal Combustion Engine Division Spring Technical Conference, Torino,ICES212-81235, 212, pp. 659-666. [25] Zawad Abedin, Mousumi Rizia and H M Khairul Enam, Comparative Emission Analysis of CNG and Octane in Retrofitted Car Engines. International Journal of Mechanical Engineering and Technology, 8(3), 217, pp. 123 13 [26] Pundlik R. Ghodke, Dr. J. G. Suryawanshi, Advanced Turbocharger Technologies for High Performance Diesel Engine-Passanger Vehicle Application. International Journal of Mechanical Engineering and Technology, 3 (2), 212, pp. 62-632 http://www.iaeme.com/ijmet/index.asp 28 editor@iaeme.com