Application of polynomial vector (pv) processing to improve the estimation performance of bio diesel in variable compression ratio diesel engine

TL;DR

This study applies a polynomial vector back propagation algorithm to estimate engine performance and emissions when using biodiesel blends from waste cooking oil in a variable compression ratio diesel engine.

Contribution

It introduces the use of polynomial vector back propagation for accurate prediction of engine parameters with biodiesel blends, enhancing estimation performance.

Findings

PVBPA effectively predicts engine power, torque, and SFC.

Biodiesel blends from waste cooking oil can be used in VCR diesel engines.

The method improves estimation accuracy over traditional approaches.

Abstract

This paper presents the implementation of polynomial vector back propagation algorithm (PVBPA) for estimating the power, torque, specific fuel consumption and presence of carbon monoxide, hydrocarbons in the emission of a direct injection diesel engine. Experimental readings were obtained using the biodiesel prepared form the waste low quality cooking oil collected from the canteen of Sri Sairam Engineering College, India.. This waste cooking oil was due to the preparation of varieties of food (vegetables fried and non vegetarian). Over more than a week, trans esterification was done in chemical lab and the biodiesel was obtained. The biodiesel was mixed in proportions of 10%, 20%, 30%, 40%, 50% with remaining combinations of the diesel supplied by the Indian government. Variable compression ratio (VCR) diesel engine with single cylinder, four stroke diesel type was used. The outputs of the engine as power, torque and specific fuel consumption were obtained from the computational facility attached to the engine. The data collected for different input conditions of the engine was further used to train (PVBPA). The trained PVBPA network was further used to predict the power, torque and brake specific fuel consumption (SFC) for different speed, biodiesel and diesel combinations and full load condition. The estimation performance of the PVBPA network is discussed.