Optimization of engine performance parameters and exhaust emissions in compression ignition engine fueled with biodiesel-alcohol blends using taguchi method, multiple regression and artificial neural network. (2020)

• Record Type: Journal Article
• Authors: Agrawal, Tanmaya; Gautam, Raghvendra; Agrawal, Sudeekcha; Singh, Vishal; Kumar, Manish; Kumar, Saket
• Is Part Of: Sustainable futures. Volume 2(2020)
• Publication Date: 2020
• Subjects: Kusum oil ethyl ester -- Taguchi -- Artificial neural network -- Multiple regression -- Optimization; Sustainability -- Periodicals; Sustainable development -- Periodicals; Green technology -- Periodicals; 338.92705
• DOI: 10.1016/j.sftr.2020.100039
• Languages: English
• ISSNs: 2666-1888

Abstract

Abstract: Humanity is on the verge of setting new benchmarks in warming up the planet 4-5°C by triggering a series of cascading tripping points. Rising carbon content in the atmosphere is causing an unprecedented spike in global temperatures leading to drastic climatic changes like massive heat waves, droughts, melting of polar ice caps and a rise in the sea level. Vehicles alone are responsible for at least 12% of the total Carbon dioxide (CO2 ) emissions in the world. Carbon Monoxide (CO), a toxic air pollutant, is produced by the incomplete combustion of carbon-containing fuels like gasoline in automobiles and its emissions are responsible for a wide range of health problems like headache, nausea, vomiting, and dizziness. Nitrogen Oxides (NOx) emissions produced by motor vehicles are instrumental in causing smog, acid rain which can lead to adverse health effects like respiratory disorders, heart problems etc. Alternative fuels will act as a "life jacket" to save humanity from drowning in this ocean of its design. Biodiesel is a "carbon neutral" fuel that can be blended with diesel and used in driving compression ignition engines. In the study engine performance in terms of brake thermal efficiency (BTE), brake specific fuel consumption (BPSC) and CO, HC, CO2, and NOx emissions are optimized for Kusum Oil Ethyl Ester (KOEE) with butanol blend using Taguchi's method, to obtain an optimized dataset of the input parameters viz. engine rpm, fuel properties and lower heating value of fuel which are then fed into the two prediction techniques viz. Artificial Neural Network (ANN) and multiple regression. These techniques are also ranked in terms of precision, based on generated errors.