ML_Diesel_Engine

The project contains data from the Diesel Engine Testing Laboratory. The goal of the project is to collect and process data from laboratories that test diesel engines and build machine learning models.

Description of the current stage of the object:

Three machine learning (ML) algorithms were used to build models for predicting parameters of a diesel engine using biological and fossil fuels. Datasets from bench tests of a diesel engine were used to fit ML models. The Bench tests were carried out on two types of fuel. First fuel was standard diesel fuel. Second fuel was Rape-seed Methyl Ester (RME).

Programming language Python is applied to proceed data and build models. For data cleaning and visualization were used such Python standard libraries as Numpy, Pandas and Matplotlib. For building models were applied two libraries such as Scikit-learn and Tensorflow. In our study are used three types of ML algorithms: Decision Tree, Random Forest and Deep Neural Network.

Tools:

The project was done using Anaconda.Navigator and JupyterLab.

The list of articals that were published under the project:

1. „Comparison of machine learning algorithms for predicting parameters of a diesel engine using biological and fossil fuels”. Viktar Taustyka, Paweł Krzaczek, Adam Koniuszy, Pavel Navitski, Andrej Skadorva. 2022. XXVIII Międzynarodową Konferencję Naukową „Problemy Zrównoważonego Rolnictwa, Ochrona Obszarów Wiejskich, Zasobów Wodnych i Środowiska”. Instytut Technologiczno-Przyrodniczy Państwowy Instytut Badawczy. Polska.