Yazar "Godwin, D. Jesu" seçeneğine göre listele

Listeleniyor 1 - 2 / 2
  • Küçük Resim Yok
    Öğe
    A comparative analysis of advanced machine learning models for the prediction of combustion, emission and performance characteristics using endoscopic combustion flame image of a pine oil–gasoline fuelled spark ignition engine
    (Elsevier Ltd., 2024) Godwin, D. Jesu; Varuvel, Edwin Geo; Jesu Martin, M. Leenus; Jasmine R, Anita; Josephin JS, Femilda
    This research focuses on using machine learning to predict the spark ignition engine's combustion, performance, and emission parameters with bio-fuel blends such as pine oil blend, which significantly diminishes the environmental impact of traditional fuels, reduces the limitations of repeated engine experimentation and addresses the nonlinearities in engine test results contributing to sustainable cleaner fuel and energy solutions. The models used were Ensemble Decision Tree Bagging, Ensemble Least Squares Boosting, Gaussian Process Regression and Support Vector Machine Regression, with good generalization ability. Brake Specific Fuel Consumption data from the test engine trials and endoscopic image flame area data after spark timing at different crank angles (320, 400, 480, 560, and 640 after Spark Timing) were fed into the machine-learning models as predictors. The response variables were Brake thermal efficiency, Unburnt Hydrocarbons, Carbon monoxide, Carbon dioxide, Oxides of nitrogen, maximum In-cylinder pressure, and maximum heat release rate. The bootstrap technique was used to generate numerous datasets from the experimental data for data-driven model training and tested using both interpolative and extrapolative data. The experimental and predicted values for all these algorithms were subjected to repeated hyperparameter optimization trials and the best machine learning method was identified using the performance and error metrics. The Ensemble Least Squares Boost model showed the overall best correlation (R2) in the range of 0.97–0.99 for gasoline and pine oil PN20 blend for the predicted versus experimental engine parameters. The root-mean-squared error (RMSE) at maximum load ranged between 0.0086 and 0.3044 for gasoline and 0.0049–0.2046 for the Pine oil PN20 fuel blend respectively. Therefore, employing an Ensemble Least Squares Boosting machine learning framework can effectively predict the characteristics of gasoline engines using pine oil and blends. This approach serves as a virtual engine model, efficiently overcoming the limitations and complexities inherent in conventional engine experiments. © 2024 Elsevier Ltd
  • Küçük Resim Yok
    Öğe
    Prediction of combustion, performance, and emission parameters of ethanol powered spark ignition engine using ensemble Least Squares boosting machine learning algorithms
    (Elsevier Sci Ltd, 2023) Godwin, D. Jesu; Varuvel, Edwin Geo; Martin, M. Leenus Jesu
    This research concentrates on the application of machine learning techniques to predict combustion, performance, and emission parameters in a dual-fuel spark ignition (SI) engine powered by neat gasoline and E20 ethanol dual fuel. The goal is to overcome the limitations posed by repeated engine experiments and nonlinear test results. In order to optimise engine parameters, the research seeks to develop efficient machine learning models with high generalizability and employ an optimisation strategy to determine the optimal engine settings. Input for training and evaluating machine learning algorithms, such as Artificial Neural Networks (ANN), and Ensemble LS Boosting was derived from experimental data from a combustion test engine, which includes Neat gasoline, and Ethanol dual fuel blend E20 at various load conditions. The dataset includes engine combustion, performance, and emission indices such as brake thermal efficiency (BTE), Exhaust Gas Temperature (EGT), Hydrocarbons (HC), Carbon monoxide (CO), Carbon dioxide (CO2), and Nitrogen oxides (NOx), under various operating conditions. Load and brake-specific fuel consumption (BSFC) were training input attributes. Using a comprehensive experimental database of input-output engine parameters, the Artificial Neural Network (ANN) and Ensemble LS Boosting were constructed. The training data points were resampled to generate multiple training datasets for training different models. 50 test samples were used to evaluate the generalisation capability of the machine learning models, while BTE, EGT, CO, CO2, HC and NOx, were the primary parameters subject to prediction. The optimal machine learning method was determined by comparing R-squared (R-2) values, root mean square error (RMSE), mean square error (MSE), and mean absolute error (MAE). Using multiple hyper-parameter tuning iterations, the agreement between actual and predicted values for diverse Ensemble LS Boost algorithms was evaluated. The Ensemble LS Boost model exhibits the maximum level of agreement between predicted and experimental engine parameters across all datasets when compared to the other ANN models. This finding was corroborated by additional research based on test datasets, specifically the test sample interpolation data, which measures generalisation ability. The study also focuses on developing and applying two unique, interactive Simulink models for the Spark Ignition (SI) engine that are tailored for Neat Gasoline and Ethanol E20 test fuels under all loads. The key component of the model-based development technique in MATLAB and Simulink was the incorporation of sophisticated machine learning algorithms, i.e., Ensemble Least-Squares (LS) Boosting, to the model-based development workflow which produced reliable results. Implementing an Ensemble LS Boost machine learning framework is therefore highly recommended as an efficient method for predicting and optimising the combustion, performance, and emission characteristics of dual-fuel gasoline engines utilising Ethanol-based dual-fuel blends.