Artificial Intelligence for the Prediction of Exhaust Back Pressure Effect on the Performance of Diesel Engines

Abstract

The actual trade-off among engine emissions and performance requires detailed investigations into exhaust system configurations. Correlations among engine data acquired by sensors are susceptible to artificial intelligence (AI)-driven performance assessment. The influence of exhaust back pressure (EBP) on engine performance, mainly on effective power, was investigated on a turbocharged diesel engine tested on an instrumented dynamometric test-bench. The EBP was externally applied at steady state operation modes defined by speed and load. A complete dataset was collected to supply the statistical analysis and machine learning phases—the training and testing of all the AI solutions developed in order to predict the effective power. By extending the cloud-/edge-computing model with the cloud AI/edge AI paradigm, comprehensive research was conducted on the algorithms and software frameworks most suited to vehicular smart devices. A selection of artificial neural networks (ANNs) and regressors was implemented and evaluated. Two proof-of concept smart devices were built using state-of-the-art technology—one with hardware acceleration for “complete cycle” AI and the other with a compact code and size (“AI in a nut-shell”) with ANN coefficients embedded in the code and occasionally offline “statistical re-calibration”.

Automated high-throughput genome editing platform with an AI learning in situ prediction model

“…1d) , an AI model was developed by the analysis of the editing results, to predict the outcomes of base editing. The XGB Regressor module of the XGBoost (eXtrened Gradient Boosting) classifier 22 was employed to predict the BE4max editing efficiency, based on information of the 20 bp protospacer sequence and the DNA accessibility value.…”

“…XGBoost Regressor 22 , which is a popular machine learning model due to its advantages in many aspects including flexibility and regularization, was used to construct machine learning models for predicting the in situ base editing results of human cell lines. We previously utilized convolutional neural networks (CNN) 25 to predict GBE base editing efficiency, but we selected XGB Regressor in this study, because our dataset only consisted of editing results for 1134 target sequences, and the XGB Regressor commonly exhibits better performance than the deep neural networks when dealing with small datasets 22 . Based on the above 1134 valid editing results, with the context-sequence information and chromatin accessibility value of each endogenous target sites, we constructed machine learning models to predict the in situ base editing results of human cell lines.…”

Advancing genome editing with artificial intelligence: opportunities, challenges, and future directions

“…CAELM was shown to accurately forecast the outcome of in-situ BED. They previously used CNN ( Feng et al, 2019 ) to predict the efficiency of GBE base editing, but in this study, they chose the XGB Regressor because their dataset only included the editing results for 1134 target patterns, and the XGB Regressor frequently performed better than DNNs when working with small datasets ( Fernoaga et al, 2020 ). The model’s accuracy was evaluated using Pearson’s correlation value, which yielded a r value of 0.64 within the predicted and measured values.…”

A study on the significance of exhaust manifold’s bending angle to the brake torque of 115cc SI engine

“…In previous studies, higher exhaust backpressure reduced engine power and torque while increasing fuel consumption. Several studies have been conducted to discover a strong correlation between exhaust backpressure [4]and IC engine performance (power, torque, and fuel consumption). Bhure et al (2018) documented that there is a 2 % increase in fuel consumption at 87.5% valve opening, brake-specific fuel consumption (BSFC) increases by 10% at 75% valve opening, 42% of Hydro Carbon (HC) Emission, HC decrease at 75% Back Pressure Control Valve, BPCV opening [2].…”