Modelling and Simulation in Engineering

Modelling and Simulation in Engineering / 2012 / Article

Research Article | Open Access

Volume 2012 |Article ID 830365 |

J. Mohammadhassani, Sh. Khalilarya, M. Solimanpur, A. Dadvand, "Prediction of NOx Emissions from a Direct Injection Diesel Engine Using Artificial Neural Network", Modelling and Simulation in Engineering, vol. 2012, Article ID 830365, 8 pages, 2012.

Prediction of NOx Emissions from a Direct Injection Diesel Engine Using Artificial Neural Network

Academic Editor: Azah Mohamed
Received03 Apr 2012
Accepted25 Apr 2012
Published28 Jun 2012


In the present study, artificial neural network is used to model the relationship between NOx emissions and operating parameters of a direct injection diesel engine. To provide data for training and testing the network, a 6-inline-cylinder, four-stroke, diesel test engine is used and tested for various engine speeds, mass fuel injection rates, and intake air temperatures. 80% of a total of 144 obtained experimental data is employed for training process. In addition, 10% of the data (randomly selected) is used for network validation and the remaining data is employed for testing the accuracy of the network. The mean square error function is used for evaluating the performance of the network. The results show that the artificial neural network can efficiently be used to predict NOx emissions from the tested engine with about 10% error.

1. Introduction

Direct injection diesel engines are used as propulsion systems with low fuel consumption and very high efficiency for automotive applications. Any attempts to use their privileges require considering emissions stringent disciplines which enforce engine manufacturers to tender their productions with lower emissions [1]. Recently, Lenz and Cozzarini [2] have presented statistics showing that the worldwide passenger car and commercial vehicle traffic contribute 20% of the total anthropogenic emissions of nitrogen oxides (NOx). These emissions have damaging effects upon the environment and people. Therefore, how to control the exhaust emissions especially NOx from diesel engines has become an essential subject for researchers of the automotive field in the world.

The diesel engine industry has undergone a great technical development in the last few years, creating a number of new strategies such as electronic control units (ECUs) and/or engineering management systems (EMSs) as well as new injection systems [37]. They all use some kinds of artificial intelligence (AI) techniques such as artificial neural network (ANN) to process the engine operating conditions and prognosticate the fairly best values of the controlling parameters with the aim of optimizing the engine characteristics.

Digital computers have provided a rapid means of performing many calculations involving the ANN methods. Along with the development of high-speed digital computers, the application of the ANN approach could be outspread in a very impressive rate in several fields. One of the major applications of ANN is industrial pollutants control. Kalogirou [8] presented an elaborated review on the recent applications of AI in environmental pollutants control.

Neural networks are powerful modeling techniques with the ability of identifying cryptic nonlinear highly complex relationships between their input and output data [9]. ANN describes such relations by updating network weights using a trial-and-error-based arithmetic method and a training algorithm such as Levenberg-Marquardt (LM).

A number of studies have been conducted to predict the characteristics of internal combustion engines (ICE) by using ANN approach. This approach has been used by Xu et al. [10] to predict engine systems reliability. The injection characteristics of direct injection (DI) diesel engines have been investigated by Yang et al. [11]. In [12], the effects of NOx and soot level in the case of high-pressure fuel injection have been investigated in a single-cylinder DI diesel engine. ANN has been used to predict the exhaust emissions and performance of a diesel engine taking into account several operating conditions such as the percentage of throttle opening, injection time, engine speed, and fuel compositions as the network inputs [1315]. However, there is no literature that reports the application of neural network to predict and model NOx emissions in terms of engine speed, intake air temperature, and mass fuel injection (MFI) rate. In this study, NOx emissions from a diesel engine are investigated using ANN. For this purpose, experimental tests have been conducted for 144 engine speeds ranging from 591 to 2308 rpm.

2. Experimental Setup

The test engine used to conduct the experiments is a heavy duty (HD) six-cylinder, direct-injection, four-stroke diesel engine. The technical specifications of the engine are given in Table 1. Standard laboratory procedures are used to measure the engine operating parameters and its tailpipe emissions (see Figure 1). The engine is connected to the data acquisition systems, so that several operating parameters could be simultaneously measured and precisely controlled. The ST10 fuel controller sensor is used to measure the mass fuel injection rate in the range of 0.39–10.31 g/sec. An electrical dynamometer is assembled on the engine and used to measure the speed, brake power, and torque of the engine. The engine speeds are recorded between 591 and 2308 rpm. Simultaneously, other engine properties such as exhaust emissions (NOx, soot, HC, CO, CO2), air-fuel ratio (AFR), and intake air temperature are measured by various connected instruments. The range of variations of the operating parameters and the corresponding values of NOx emissions have been listed in Tables 2 and 3, respectively [16, 17].

Engine typeOM355 Mercedes-Benz S. 355093

Number of injector holes4
Injection pressure195 bar
Max output power240 hp
Intermediate power speed1400 rpm
Number of cylinders and layout6-inline
Compression ratio16.1 : 1
Injection timing (Deg BTDC)18
Displacement11580 cc
Stroke150 mm
Bore128 mm

Operating parameterUnitRange of variations

𝑁 rpm [ 5 9 1 , 2 3 0 8 ]
𝑀 𝑓 g/sec [ 0 . 3 9 , 1 0 . 3 1 ]
T °C [ 1 5 . 4 , 2 8 . 3 ]

OP* and NOxEN**

Engine speed (RPM)591592601601602602603604604605606606608611612612615616616617618621623626627628628631631632
Intake air temperature (°C)25.525.123.428.327.617.31620.517.320.415.928.123.821.52224.725.320.820.922.520.827.12322.815.915.926.126.626.523
Mass fuel (g/sec)0.430.430.420.420.410.440.420.410.430.430.420.440.430.390.390.390.390.440.40.430.40.420.470.470.440.430.450.440.460.47
NOx (g/kw·hr)18.9619.9721.5629.0147.7118.6919.4224.4921.5124.0918.542922.7521.4218.2620.7320.1819.1324.0525.2320.4531.5622.5321.6427.9524.8828.7437.9128.2315.14

OP* and NOxEN**

Engine speed (RPM)632633636638640645138913911396139613971398139813991399139913991400140114011401140114021402140214021402140314031403
Intake air temperature (°C)24.922.926.223.718.618.61615.815.515.623.324.623.720.424.821.723.82019.321.719.915.821.219.717.216.925.523.815.619.9
Mass fuel (g/sec)0.40.460.440.410.480.491.531.541.481.482.242.343.656.72.332.192.246.722.
NOx (g/kw·hr)26.714.0939.3227.0424.2123.3110.2110.6711.0211.2513.113.0111.988.90512.8912.713.138.75311.612.2811.0812.69.39811.212.6312.7910.554.439.2429.963

OP* and NOxEN**

Engine speed (RPM)140314031403140414041404140414041404140414041404140514051405140614061406140614061406140714071407140714071407140814081411
Intake air temperature (°C)18.920.125.519.621.115.61815.415.424.617.716.724.315.818.324.215.717.515.716.521.624.615.426.921.717.718.621.218.121.4
Mass fuel (g/sec)3.593.546.652.26.522.196.713.553.553.86.675.010.842.183.60.832.176.693.515.022.180.853.
NOx (g/kw·hr)10.239.97210.549.4669.3469.4699.56510.3710.6112.5110.6412.394.0559.40110.283.9359.86310.511.0512.1112.394.10510.7512.4712.5315.8816.688.1189.60511.73
OP* and NOxEN**

Engine speed (RPM)141114121414200120022005205521042106214821542194219421952196219621982200220022002200220022012201220122012201220122012203
Intake air temperature (°C)
Mass fuel (g/sec)1.533.491.549.
NOx (g/kw·hr)1011.8610.0310.412.2612.1311.8511.4711.4911.1311.310.969.07310.5611.069.1179.3111.068.40711.968.999.00410.7710.7312.2310.899.1699.3589.3718.712

OP* and NOxEN**

Engine speed (RPM)220322042204220522052205220522062207220722072207220722082210221022112215225022542303230423052308
Intake air temperature (°C)23.720.718.519.920.817.321.520.221.822.121.72424.420.917.924.120.821.219.920.518.618.816.417.2
Mass fuel (g/sec)6.310.1110.135.985.9910.173.773.825.725.724.224.374.424.1510.1613.7510.110.3110.39.018.889.939.92
NOx (g/kw·hr)8.9111.6411.129.219.35410.3212.9410.6711.5411.1511.559.579.53811.1110.85.88511.8311.911.0611.418.0957.3059.63110.19

*Operating parameter, **experiment number.

3. ANN Approach

The building unit of an ANN is a simplified model of the much more complex one known as organic neuron. This model was introduced by the neurophysiologist McCluch and the logician Pitts in 1943 [18], but its learning behavior was first treated extensively in a book by Rosenblatt in 1962 [19].

One of the main advantages of ANN is its ability to model complex nonlinear relationships between multiple input variables and the required outputs. Another important advantage of the ANN approach is its fast response, which allows one to use it in more complex procedures including optimization applications. Therefore, it offers the advantage of being fast, accurate, reliable, and powerful in dealing with multivariate problems as well as in the prediction or approximation affairs, especially when numerical and mathematical methods fail [20, 21].

To get the best prediction by the network, many parameters should be adjusted such as biases, weights, number of hidden layers, number of hidden layer neurons, and type of transfer function. The biases and weights must be modified in every epoch by using training algorithms such as LM algorithm. The performance of the network is evaluated by comparing the error obtained from converged neural network runs and the measured data. The error of the network is calculated at the end of training, validation, and testing processes based on the differences between the targeted and calculated outputs. The back propagation algorithm is used to minimize the error function, which relates the outputs of each neuron in the output layer and the corresponding desired output. The error function used here is the so-called mean square error (MSE) function given by 1𝐸=𝑄𝑝𝑘𝑦𝑝𝑘𝑓𝑝𝑘2,(1) where 𝑄 represents training pairs of vectors, 𝑘 is the index of elements in the output vector, 𝑦𝑝𝑘 is the 𝑘th element of the 𝑝th desired pattern (target value) vector, and 𝑓𝑝𝑘 is the 𝑘th element of the output vector when pattern 𝑝 is introduced as input to the network. Investigations have proved the accuracy and rapid convergence of LM algorithm for training in engineering applications with limited number of experimental data [22, 23]. In the present work, the LM training algorithm is employed, which uses Hessian matrix approximation. In what follows, a detailed description of this algorithm is presented.

4. LM Algorithm

The LM algorithm is a virtual standard in nonlinear optimization which significantly outperforms gradient descent and conjugate gradient methods for medium-sized problems. It is a pseudo-second-order method which means that it works with only function evaluations and gradient information but it estimates the Hessian matrix using the sum of outer products of the gradients (for more details, see [23]). It also fits a curve on a given dataset by finding the optimum parameters 𝛽 based on a user specified model 𝑓 such that the final parameters can characterize the target function by minimizing the errors. This is analogous to solving the least squares problem, where we want to minimize the sum of squares between the target values 𝑦𝑖 and the output of the user model 𝑓 with the input values 𝑥𝑖 and the estimated parameter vector 𝛽 with 𝑚 number of input data points [24]: 𝑆(𝛽)=𝑚𝑖=1𝑦𝑖𝑥𝑓𝑖,𝛽2.(2)

Here, 𝑆(𝛽) is the function to be minimized. To start a minimization, the user has to provide an initial guess for the parameter vector, 𝛽, which may be critical in convergence if the user model 𝑓 is high dimensional or nonlinear with a large number of parameters. In many cases, an uninformed standard guess like 𝛽𝑇=(1,1,,1) will work fine; in other cases, the algorithm converges only if the initial guess is already somewhat close to the final solution. At each step, the initial parameters are updated by a small amount in the optimum direction by adding the update delta values, 𝛿, such that 𝛽𝑡+1=𝛽𝑡+𝛿.(3)

To find the update delta value, 𝛿, we need to solve for the approximation of the sum of squares function by setting the gradient equal to zero as follows: 𝐽𝑇𝑓𝑥𝑦𝑖,𝛽+𝐽𝑖𝛿=0.(4)

This relation can readily be reduced to 𝐽𝑇𝐽𝑖𝛿=𝐽𝑇𝑥𝑦𝑓𝑖,𝛽,(5) where 𝐽 is the Jacobian matrix or gradient of the function that describes the user model and the superscript 𝑇 represents its transpose form. Equation (5) yields a set of linear equations in the form 𝐴𝑥=𝑏, where 𝐴 is a square matrix, 𝑥 represents the vector of unknown delta values, and 𝑏 is a known vector of the same length as 𝑥. At each step, the delta values can be obtained by solving this set of linear equations. However, the LM algorithm also adds a regularization parameter 𝜆 that helps as the dampening factor to produce the final LM equation: 𝐽𝑇𝐽𝑖𝐽+𝜆diag𝑇𝐽𝑖𝛿=𝐽𝑇𝑥𝑦𝑓𝑖,𝛽.(6)

Here, diag(𝐽𝑇𝐽𝑖) stands for the principal diagonal elements of the matrix 𝐽𝑇𝐽𝑖. This equation is still a system of linear equations of the form 𝐴𝑥=𝑏 and one can use direct solvers to solve for delta values at each step.

5. Implementation of the ANN to Predict NOx Emissions

The neural network toolbox of MATLAB 7.8 is used to form the ANN. Simple and detailed structures of the employed ANN have been shown in Figures 2 and 3, respectively. According to the Kolmogorov theory, multilayer perceptron algorithms can approximate any complex and nonlinear relation between input and output data, among which the three-layer algorithm is the simplest but efficient one. The three layers include the input layer, the hidden layer, and the output layer. Each layer involves some neurons which should be properly determined. The number of input and output parameters of the system determines the number of neurons in the input and output layers of the network, respectively. Thus, the input layer has three neurons while the output layer has only one neuron. It should be noted that in the present work, nominal 19 neurons (determined by trial and error) are used in the hidden layer.

The number of data patterns required for training the network should be chosen in such a way that the network is properly trained and in the meantime adequate data is remained for testing the network. In addition, it is essential to set aside some data patterns for validating the network during the process. About 80% of the total 144 experimental data (i.e., 116 data) is used for training the network and 10% (i.e., 14 data) of the data is used for validation. The remainder data is left for testing the network. Neural network requires that the range of both the input and output values lies between 0 and 1. For this purpose the following formula is used to normalize these values [25]: 𝑉norm=𝑉𝑎𝑉min𝑉max𝑉min×𝑉𝑉𝑙+𝑉𝑙,(7) where 𝑉norm is the normalized value and 𝑉𝑎 is the actual value of the data. 𝑉min and 𝑉max are the minimum and maximum values of the experimental data, respectively. Also, the values of 𝑉 and 𝑉𝑙 are set to be 0 and 1, respectively.

There are various types of transfer functions such as logsig, tansig, purelin, among others. In the present work, the logsig transfer function is used in both the hidden and output layers. This function is defined as 1logsig(𝑛)=1+𝑒𝑛,(8) where 𝑛 is the weighted sum of the input.

6. Results and Discussions

The artificial neural network is used to predict the NOx emissions from a direct injection diesel engine using LM training algorithm. Technical specifications of the test engine are given in Table 1. The input data of the network are the measured operating parameters of the engine such as mass fuel injection rate, intake air temperature, and speed of engine whose range of variations is given in Table 2.

A MATLAB program has been developed to first obtain the desired correlations for training, validation, and testing stages of the network. Then, the accuracy of the network is evaluated through the comparison of the predicted values of NOx emissions with the experimentally measured ones. The total 144 measured engine’s operating parameters and the corresponding values of NOx emissions are listed in Table 3.

Figures 2 and 3 show the simple and detailed structures of the ANN employed, respectively. These figures demonstrate the three layers of the network, namely, input layer, hidden layer, and output layer. The operating parameters of the engine are fed into the network as inputs and NOx emissions leave the network as outputs. Note that 𝑊 in Figure 3 represents the weight of the layer.

Figure 4 shows a regression analysis between the network response (outputs) and the corresponding targets. According to this figure, the training process has been properly performed, where the correlation factor between outputs and targets is 0.91972. Figures 5 and 6 show the validation and testing results of the network, respectively. It is observed from these figures that the ANN represents the best accuracy in modeling the NOx emissions with correlation factors of 0.98222 and 0.89123, respectively, for the network validation and testing.

The results show that the ANN with LM training algorithm is an appropriate technique, which can accurately predict NOx emissions for different engine operating parameters including engine speed, intake air temperature, and mass fuel rate. A comparison between the predicted and the measured values of NOx emissions are depicted in Figure 7. There is a good agreement between the predicted values using the neural network model and the measured values obtained from experimental tests. It may be noted that, for medium engine speeds, the agreement is more considerable than that for the medium speeds. For medium speeds the MSE is less than 8%.

7. Conclusions

The operating parameters involving speed, intake air temperature, and mass fuel rate of a DI diesel engine have been used to train the ANN to predict NOx emissions from the engine. The results of this research reveal that a three-layer neural network along with LM training algorithm leads to a desirable mapping between the inputs and outputs of the network case. The proposed ANN model for prediction of the NOx emissions gives the correlation factors of 0.92, 0.98, and 0.89 for training, validating, and testing the network, respectively. It is concluded that, ANN model is a potentially feasible tool for prediction of NOx emissions from a diesel engine with respect to the engine operating parameters, especially in medium engine speeds.


𝑓:ANN output value
𝑀𝑓: Mass fuel injection (g/sec)
𝑁: Engine speed (rpm)
𝑄: Number of pairs
𝑅: Correlation factor
𝑇𝑖: Intake air temperature (C)
𝑥: Input value
𝑦: Target value.

Greek  Symbols

𝛽:Estimated parameter
𝛿:Updated delta value.


AFR:Air-fuel ratio
AI:Artificial intelligence
ANN:Artificial neural network
DI:Direct injection
ECU:Electronic control unit
EMS:Engineering management system
GD:Gradient descent
HD: Heavy duty
MF:Mass fuel injection.


This research has been supported by the Iranian Diesel Engine Manufacturing (IDEM) Company.


  1. A. de Lucas, A. Durán, M. Carmona, and M. Lapuerta, “Modeling diesel particulate emissions with neural networks,” Fuel, vol. 80, no. 4, pp. 539–548, 2001. View at: Publisher Site | Google Scholar
  2. H. P. Lenz and C. Cozzarini, Emissions and Air Quality, SAE Bookstore, 1999.
  3. C. J. Morey and J. Mark, “Diesel passenger vehicles—can they meet air quality needs and climate change goals?” SAE Paper 2000-01-1599, 2000. View at: Publisher Site | Google Scholar
  4. J. Benajes, S. Molina, and J. M. García, “Influence of pre- and post-injection on the performance and pollutant emissions in HD diesel engine,” SAE Paper 2001-01-0526, 2001. View at: Google Scholar
  5. P. L. Herzog, L. Buerguer, E. Winklhofer, P. Zelenka, and W. Caterllieri, “NOx reduction strategies for diesel engines,” SAE Paper 920470, 1992. View at: Google Scholar
  6. C. Havenith, J. R. Needham, A. J. Nicol, and C. H. Such, “Low emission heavy duty diesel engine for Europe,” SAE Paper 932959, 1993. View at: Google Scholar
  7. J. M. Alonso, F. Alvarruiz, J. M. Desantes, L. Hernández, V. Hernández, and G. Moltó, “Combining neural networks and genetic algorithms to predict and reduce diesel engine emissions,” IEEE Transactions on Evolutionary Computation, vol. 11, no. 1, pp. 46–55, 2007. View at: Publisher Site | Google Scholar
  8. S. A. Kalogirou, “Applications of artificial neural networks in energy systems. A review,” Energy Conversion and Management, vol. 40, no. 10, pp. 1073–1087, 1999. View at: Publisher Site | Google Scholar
  9. S. Haykin, Neural Network, Prentice Hall, 2nd edition, 1999.
  10. K. Xu, M. Xie, L. C. Tang, and S. L. Ho, “Application of neural networks in forecasting engine systems reliability,” Applied Soft Computing Journal, vol. 2, no. 4, pp. 255–268, 2003. View at: Publisher Site | Google Scholar
  11. H. C. Yang, H. S. Ryou, Y. T. Jeong, and Y. K. Choi, “Spray characteristics in a direct-injection diesel engine,” Automization and Sprays, vol. 6, no. 1, pp. 95–109, 1996. View at: Google Scholar
  12. K. Nakakita, T. Kondoh, K. Ohsawa, T. Takahashi, and S. Watanabe, “Optimization of pilot injection pattern and its effect on diesel combustion with high-pressure injection,” JSME International Journal B, vol. 37, no. 4, pp. 966–973, 1994. View at: Google Scholar
  13. D. Yuanwang, Z. Meilin, X. Dong, and C. Xiaobei, “An analysis for effect of cetane number on exhaust emissions from engine with the neural network,” Fuel, vol. 81, no. 15, pp. 1963–1970, 2002. View at: Publisher Site | Google Scholar
  14. E. Arcaklioǧlu and I. Çelikten, “A diesel engine's performance and exhaust emissions,” Applied Energy, vol. 80, no. 1, pp. 11–22, 2005. View at: Publisher Site | Google Scholar
  15. D. D. Massie, “Neural network fundamentals for scientists and engineers,” in Proceedings of the International Conference on Efficiency, Costs, Optimization, Simulations and Environmental Impact of Energy Systems (ECOS '01), p. 123, Istanbul, Turkey, July 2001. View at: Google Scholar
  16. V. Pirouzpanah, “Prediction of major pollutants emission in direct-injection dual-fuel diesel and natural-gas engines,” SAE Paper 1999-01-0841, 1999. View at: Google Scholar
  17. M. H. Saidi, K. Eisazadeh Far, and V. Pirouzpanah, “Analysis of combustion process in dual fuel diesel engines: knock phenomenon approach,” SAE Paper 2005-01-1132, 2005. View at: Google Scholar
  18. W. S. McCulloch and W. Pitts, “A logical calculus of the ideas immanent in nervous activity,” The Bulletin of Mathematical Biophysics, vol. 5, no. 4, pp. 115–133, 1943. View at: Publisher Site | Google Scholar
  19. J. Keleman, “From artificial neural networks to emotion machines with marvin minsky,” Acta Polytechnica Hungarica, vol. 4, no. 4, pp. 1–12, 2007. View at: Google Scholar
  20. M. K. Deh Kiani, B. Ghobadian, T. Tavakoli, A. M. Nikbakht, and G. Najafi, “Application of artificial neural networks for the prediction of performance and exhaust emissions in SI engine using ethanol-gasoline blends,” Energy, vol. 35, no. 1, pp. 65–69, 2010. View at: Publisher Site | Google Scholar
  21. S. M. R. Loghmanian, H. Jamaluddin, R. Ahmad, R. Yusof, and M. Khalid, “Structure optimization of neural network for dynamic system modeling using multi-objective genetic algorithm,” Neural Computing & Applications. In press. View at: Publisher Site | Google Scholar
  22. O. Obodeh and C. I. Ajuwa, “Evaluation of artificial neural network performance in predicting diesel engine NOx emissions,” European Journal of Scientific Research, vol. 33, no. 4, pp. 642–653, 2009. View at: Google Scholar
  23. S. Roweis, Levenberg Marquardt Optimization, Cambridge University Press, 2nd edition, 2004.
  24. T. F. Coleman and P. E. Plassmann, “A parallel nonlinear least-squares solver: theoretical analysis and numerical results,” SIAM Journal on Scientific Computing, vol. 13, no. 3, pp. 771–793, 1992. View at: Google Scholar
  25. A. Parlak, Y. Islamoglu, H. Yasar, and A. Egrisogut, “Application of artificial neural network to predict specific fuel consumption and exhaust temperature for a diesel engine,” Applied Thermal Engineering, vol. 26, no. 8-9, pp. 824–828, 2006. View at: Publisher Site | Google Scholar

Copyright © 2012 J. Mohammadhassani et al. This is an open access article distributed under the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.

More related articles

 PDF Download Citation Citation
 Download other formatsMore
 Order printed copiesOrder

Related articles

We are committed to sharing findings related to COVID-19 as quickly as possible. We will be providing unlimited waivers of publication charges for accepted research articles as well as case reports and case series related to COVID-19. Review articles are excluded from this waiver policy. Sign up here as a reviewer to help fast-track new submissions.