Proportional impact prediction model of animal waste fat-derived biodiesel by ANN and RSM technique for diesel engine. (15th January 2022)

• Record Type: Journal Article
• Authors: Simsek, Suleyman; Uslu, Samet; Simsek, Hatice
• Is Part Of: Energy. Volume 239:Part D(2022)
• Publication Date: 2022-01-15
• Subjects: Artificial neural network -- Response surface methodology -- Animal fat biodiesel -- Prediction -- Diesel engine; Power resources -- Periodicals; Power (Mechanics) -- Periodicals; Energy consumption -- Periodicals; 333.7905
• DOI: 10.1016/j.energy.2021.122389
• Languages: English
• ISSNs: 0360-5442

Abstract

Abstract: Instead of many experimental studies made for the suitability of biodiesel for use in diesel engine, it has become easier to determine by fewer experiments with the development of computer applications. In this research, it was aimed to determine the optimum ratio of animal waste fat biodiesel (AWFBD) and the corresponding engine responses by using artificial neural network (ANN) and response surface methodology (RSM). In addition, a comparison was made with test results to evaluate the performance of ANN and RSM. According to the regression results obtained from RSM, absolute fraction of variance (R 2 ) values greater than 0.95 emerged for all answers. Correlation coefficient (R) values obtained from ANN were found to be higher than 0.97. The developed ANN model was able to predict engine responses with mean absolute percentage error (MAPE) in the range of 3.787–10.730%. MAPE values for RSM were obtained between 2.004 and 11.461%. Combined desirability factor obtained from RSM was found as 0.72288% and optimum engine parameters were found as 22% AWFBD ratio and 1350-Watt engine load. In addition, according to the verification test between the optimum results and the prediction results, it was concluded that there is a good agreement with a maximum error rate of 3.863%. Highlights: CI Engine characteristics study on animal waste fat biodiesel. Development of RSM regression and optimization model. ANN based prediction model development. Optimized results are 22% biodiesel ratio and 1350-Watt load.