NOx Prediction by Cylinder Pressure Based on RBF Neural Network in Diesel Engine

Wang, Jun; Zhang, Youtong; Xiong, Qingyu; Ding, Xuhui

doi:10.1109/icmtma.2010.621

Cited by 12 publications

(7 citation statements)

References 2 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Wang et al analyzed a marine 2-stroke diesel engine's emissions based on the modeling of a RBF NN [19]. Wang et al predicted the NO x emissions using cylinder pressure based on the RBF and BP NN in the diesel engine [20]. Manjunatha et al studied a diesel engine, fueled with a biodiesel blend, and predicted the NO x , CO 2 , CO, HC, and smoke emissions by means of the RBF and BP NN [21].…”

Section: Introductionmentioning

confidence: 99%

Prediction of emissions and exhaust temperature for direct injection diesel engine with emulsified fuel using ANN

Kökkülünk¹,

Akdoğan²,

Ayhan³

2013

Turk J Elec Eng & Comp Sci

View full text Add to dashboard Cite

Exhaust gases have many effects on human beings and the environment. Therefore, they must be kept under control. The International Convention for the Prevention of Pollution from Ships (MARPOL), which is concerned with the prevention of marine pollution, limits the emissions according to the regulations. In Emission Control Area (ECA) regions, which are determined by MARPOL as ECAs, the emission rates should be controlled. Direct injection (DI) diesel engines are commonly used as a propulsion system on ships. The prediction and control of diesel engine emission rates is not an easy task in real time. Therefore, in this study, an artificial neural network (ANN) structure using the back propagation (BP) learning algorithm and radial basis function (RBF) has been developed to predict the emissions and exhaust temperature for DI diesel engines with emulsified fuel. In order to show the ANN performance, the network outputs and experimental results of the BP and RBF have been compared in this paper. The experimental results were obtained from a real diesel engine. The results showed that the emissions and exhaust temperature were estimated with a very high accuracy by means of the designed neural network structures and the RBF is more reliable than the BP.

show abstract

Section: Introductionmentioning

confidence: 99%

Prediction of emissions and exhaust temperature for direct injection diesel engine with emulsified fuel using ANN

Kökkülünk¹,

Akdoğan²,

Ayhan³

2013

Turk J Elec Eng & Comp Sci

View full text Add to dashboard Cite

show abstract

“…As interest in N 2 O emissions from agricultural fields has grown, researchers have turned to data-driven models to enhance predictions. Early efforts have involved a variety of machine learning models [14], including the Multilayer Perceptron (MLP) [15], Support Vector Machine (SVM) [16], Random Forest (RF) [17], Radial Basis Function Neural Network (RBFNN) [18], Deep Belief Network (DBN) [19], and Long Short-Term Memory (LSTM) [4], to establish connections between input variables and N 2 O emissions.…”

Section: A N 2 O Emission From Farmingmentioning

confidence: 99%

An Explicit Improvement on Generative Adversarial Network-Based Time Series Generation: Applying Synthetic Data to N2O Emission Prediction in Farming

Lin,

Killeen,

et al. 2023

Preprint

View full text Add to dashboard Cite

<p>Traditionally, time series data augmentation has primarily focused on improving the architecture of Generative Adversarial Network (GAN), with the aim of closely matching the original data distribution while also preserving the dynamic behavior of the original data. However, even state-of-the-art GAN models like TimeGAN fall short in preserving the temporal dynamics present in the original time series due to the absence of first-order difference information. To address this limitation, this study proposes a novel process for generating multivariate time series data. The proposed process comprises four essential modules: a) the GAN module for generating multivariate time series data, b) the sampling module for preserving the first-order difference distribution, c) the smoothing module for refining the generated data, and d) an evaluation module using the Kolmogorov-Smirnov Test (KS-test) and Hilbert-Schmidt Independence Criterion (HSIC), along with other metrics to test the synthetic time series data. This comprehensive approach ensures that the synthetic time series data maintains both the distribution and the dynamic behavior of the original data.</p> <p>We extensively discuss the role of the β factor in the modified Metropolis-Hastings algorithm (in the sampling module), which controls the level of information preservation from the original time series. Our experiments reveal that with small β values, periodic information can be retained effectively. The joint distribution of the first-order difference of the synthetic time series data remains consistent when the same β value is applied in the modified Metropolis-Hastings algorithm. However, we observe that β has no impact on the partial autocorrelation functions. Nevertheless, the generated data from the sampling module maintains the memoryless property of the Markov Chain. Therefore, in the smoothing module, we apply the exponential moving average (EMA) method to simulate the long-term relationships within the original time series, and find that an optimal α value is approximately 0.4 or 0.5. Lastly, we employ the synthetic time series data to train a neural network model developed in another work. Our findings indicate that the neural network model trained on synthetic time series data exhibits performance comparable to that of a model trained on the original data.</p>

show abstract

“…Traditionally, researchers utilized different approaches from the machine learning field, including Multiple Layer Perceptron (MLP) [14], Support Vector Machine (SVM) [15], Random Forest (RF) [16], Radial Basis function neural network (RBFNN) [17], or Deep belief network (DBN) [18], to predict the emission of greenhouse gas (including Carbon Dioxide, Nitrous Oxide), viewing the entire process as a static mapping from different variables to the amount of N 2 O emitted. However, the static type approach missed the dynamic information of the system; therefore, the simulation accuracy is poor.…”

Section: Introductionmentioning

confidence: 99%

Stacked Bidirectional LSTM for Predicting Emission of Nitrous Oxide

Lin¹,

Yeap²,

Kiringa³

2022

Proceedings of the Canadian Conference on Artificial Intelligence

View full text Add to dashboard Cite

In this paper, a stacked Bidirectional Long Short-Term Memory (BDLSTM) is proposed to predict the emission of Nitrous Oxide (N 2 O) from agricultural soils. The model is trained using the data collected by LI-COR soil-gas measurement equipment in Area X.O. Ottawa, Canada. With MSE as the loss function and Adam as the optimizer, the model is evaluated against mean absolute error(MAE), mean absolute percentage error (MAPE), and root mean square error(RMSE). In comparison with the MLP model, it is observed that the stacked BDLSTM model has superior performance.The model is trained on two thousand data points with an early stopping technique, which, in general, is used to avoid overfitting in a highly complex model. Although the overfitting is depressed with the early stopping technique, it is necessary to consider the trade-off between the computation efficiency and the prediction accuracy. In this study, two BDLSTM layers (each with 100 hidden nodes) would have achieved the optimal balance. Also, the experiment shows that if an unrelated feature is added to the input features, it will degrade the performance of the model, and the simulation also indicated that it is necessary to choose an appropriate number of time steps (12 time steps) to obtain the best prediction accuracy.

show abstract

NOx Prediction by Cylinder Pressure Based on RBF Neural Network in Diesel Engine

Cited by 12 publications

References 2 publications

Prediction of emissions and exhaust temperature for direct injection diesel engine with emulsified fuel using ANN

Prediction of emissions and exhaust temperature for direct injection diesel engine with emulsified fuel using ANN

An Explicit Improvement on Generative Adversarial Network-Based Time Series Generation: Applying Synthetic Data to N2O Emission Prediction in Farming

Stacked Bidirectional LSTM for Predicting Emission of Nitrous Oxide

Contact Info

Product

Resources

About