Physical-oriented and machine learning-based emission modeling in a diesel compression ignition engine: Dimensionality reduction and regression

Abstract

The development of internal combustion engines is affected by the exhaust gas emissions legislation and the striving to increase performance. This demands for engine-out emission models that can be used for engine optimization for real driving emission controls. The prediction capability of physically and data-driven engine-out emission models is influenced by the system inputs, which are specified by the user and can lead to an improved accuracy with increasing number of inputs. Thereby the occurrence of irrelevant inputs becomes more probable, which have a low functional relation to the emissions and can lead to overfitting. Alternatively, data-driven methods can be used to detect irrelevant and redundant inputs. In this work, thermodynamic states are modeled based on 772 stationary measured test bench data from a commercial vehicle diesel engine. Afterward, 37 measured and modeled variables are led into a data-driven dimensionality reduction. For this purpose, approaches of supervised learning, such as lasso regression and linear support vector machine, and unsupervised learning methods like principal component analysis and factor analysis are applied to select and extract the relevant features. The selected and extracted features are used for regression by the support vector machine and the feedforward neural network to model the NOx, CO, HC, and soot emissions. This enables an evaluation of the modeling accuracy as a result of the dimensionality reduction. Using the methods in this work, the 37 variables are reduced to 25, 22, 11, and 16 inputs for NOx, CO, HC, and soot emission modeling while maintaining the accuracy. The features selected using the lasso algorithm provide more accurate learning of the regression models than the extracted features through principal component analysis and factor analysis. This results in test errors RMSETe for modeling NOx, CO, HC, and soot emissions 19.22 ppm, 6.46 ppm, 1.29 ppm, and 0.06 FSN, respectively.

A misfire-integrated Gaussian process (MInt-GP) emulator for energy-assisted compression ignition (EACI) engines with varying cetane number jet fuels

“…Machine Learning (ML)-based surrogate modeling of internal combustion engines (ICE) has been widely used for a broad range of applications. 25–29 Data-driven ML approaches, in particular, are popular for building ICE surrogate models; such approaches include neural networks (NN), 30–45 Support Vector Machines (SVM), 46–49 Gaussian Processes (GPs, 50–60 also known as kriging 61 ), and other learning models. 62–68 In surrogate modeling applications with limited training runs from expensive simulators, GPs (and its recent non-stationary extensions) have several key advantages over alternate deep learning models.…”

“…A similar study was conducted by Mishra and Subbarao, 44 where they employed physics-based models with ML models (specifically, NNs trained with experimental data) to improve predictions of the engine CA50, IMEP, and Start of Combustion (SOC) for a Reactivity Controlled Compression Ignition (RCCI) engine. Mohammad et al 49 explored the use of supervised learning (LASSO-based regression and SVM) and unsupervised learning (principal component analysis and factor analysis) for feature extraction and selection, which were then combined with feed-forward NNs and SVMs for predicting NOx, HC, CO, and soot emissions; they found that supervised learning performed better feature selection and extraction than unsupervised learning. These methods all required the generation and availability of experimental data under a wide range of input parameter changes.…”

A component sizing prediction study for a series hybrid electric vehicle based on artificial neural network

“…With its powerful parallelism ability and inherent contextual information processing toward the problem, this tool first experienced a peak during the 1940s to solve many complex problems. 30 Mohammad et al 31 utilized the support vector machine (SVM) and the feedforward Neural Network to model the NOx and other emissions of a diesel Compression-Ignition (CI) engine. Khamesipour et al 32 predicted the components sizes of a series HEV based on 3000 experimental data and utilizing the ANN; their results revealed that 1.44 kW/h for battery size and 80 kW for electric motor power were optimal for the investigated cycles.…”

Statistics-Guided Accelerated Swarm Feature Selection in Data-Driven Soft Sensors for Hybrid Engine Performance Prediction

“…Filter approaches to feature selection problems rely on an external measure calculated from the data that must be defined to select a subset of features [13]. Mohammad et al use the least absolute shrinkage and selection operator (Lasso) algorithm to select features used for training emission models of a diesel engine, in which the 37 variables are reduced to 25, 22, 11, and 16 inputs for NOx, CO, HC, and soot emission modeling while maintaining the accuracy [16]. Besides, the superiority of the Lasso algorithm was confirmed on predicting the fuel consumption of ship engines [17], [18].…”