Open AccessPredicting the Emissive Characteristics of an IC Engine Using DNN
Author(s) -
M. C. Pravin,
M. Mukilan,
G. Vishnu Prakash,
P. Nithish,
B. Monish Kanna,
E. Logesh
Publication year - 2020
Publication title -
iop conference series. materials science and engineering
Language(s) - English
Resource type - Journals
eISSN - 1757-899X
pISSN - 1757-8981
DOI - 10.1088/1757-899x/995/1/012010
Subject(s) - python (programming language) , diesel fuel , automotive engineering , internal combustion engine , diesel engine , automotive industry , computer science , combustion , fuzzy logic , artificial intelligence , engineering , chemistry , programming language , organic chemistry , aerospace engineering
Biodiesel is the new form of automotive fuel that the world is now concerning and several researches are going on for the production of an efficient form of Bio-Diesel because of the fact that Diesel and Petrol are going to be exhaustible in nearly 60 years. In order to produce an efficient fuel, it is inevitable to calculate the emission characteristics concerning the fuel. This project deals with the efficient and intelligent way of analyzing and calculating the engine emission characteristics of Bio-diesel operated IC engines. A Machine Learning based model using TensorFlow library has been developed using python programming for the calculation of emission characteristics such as Carbon-monoxide (CO) and Carbon-dioxide (CO 2 ) of an IC engine upon injection of Bio-diesel as fuel in different proportions. These investigations and data-sets are considered for a four stroke internal combustion engine. In this Machine Learning model TensorFlow library has been used for the better visualization of the results and error rectification. The results of the developed TensorFlow model are then compared with an existing Fuzzy model for the same application. The results predicted by this model clearly are in good correlation with the actual values which depicts that this method is effective and the total error of the developed model was found to be ±0.02 which is comparatively lower than that of the existing Fuzzy model. Concludingly, the Machine learning model using TensorFlow was found to be the best model for the calculation of engine emission characteristics of Bio-diesel operated IC engines as it offers more visualization tools and better predictive analysis.