基于Taguchi和ANN的优化方法预测柴油、生物柴油和天然气驱动的VCR柴油机的最大性能和最小排放

IF 1.4 Q2 ENGINEERING, MULTIDISCIPLINARY

World Journal of Engineering Pub Date : 2023-08-16 DOI:10.1108/wje-04-2023-0116

Taraprasad Mohapatra, S. Mishra, Mukesh Bathre, S. S. Sahoo

{"title":"基于Taguchi和ANN的优化方法预测柴油、生物柴油和天然气驱动的VCR柴油机的最大性能和最小排放","authors":"Taraprasad Mohapatra, S. Mishra, Mukesh Bathre, S. S. Sahoo","doi":"10.1108/wje-04-2023-0116","DOIUrl":null,"url":null,"abstract":"\nPurpose\nThe study aims to determine the the optimal value of output parameters of a variable compression ratio (CR) diesel engine are investigated at different loads, CR and fuel modes of operation experimentally. The output parameters of a variable compression ratio (CR) diesel engine are investigated at different loads, CR and fuel modes of operation experimentally. The performance parameters like brake thermal efficiency (BTE) and brake specific energy consumption (BSEC), whereas CO emission, HC emission, CO2 emission, NOx emission, exhaust gas temperature (EGT) and opacity are the emission parameters measured during the test. Tests are conducted for 2, 6 and 10 kg of load, 16.5 and 17.5 of CR.\n\n\nDesign/methodology/approach\nIn this investigation, the first engine was fueled with 100% diesel and 100% Calophyllum inophyllum oil in single-fuel mode. Then Calophyllum inophyllum oil with producer gas was fed to the engine. Calophyllum inophyllum oil offers lower BTE, CO and HC emissions, opacity and higher EGT, BSEC, CO2 emission and NOx emissions compared to diesel fuel in both fuel modes of operation observed. The performance optimization using the Taguchi approach is carried out to determine the optimal input parameters for maximum performance and minimum emissions for the test engine. The optimized value of the input parameters is then fed into the prediction techniques, such as the artificial neural network (ANN).\n\n\nFindings\nFrom multiple response optimization, the minimum emissions of 0.58% of CO, 42% of HC, 191 ppm NOx and maximum BTE of 21.56% for 16.5 CR, 10 kg load and dual fuel mode of operation are determined. Based on generated errors, the ANN is also ranked for precision. The proposed ANN model provides better prediction with minimum experimental data sets. The values of the R2 correlation coefficient are 1, 0.95552, 0.94367 and 0.97789 for training, validation, testing and all, respectively. The said biodiesel may be used as a substitute for conventional diesel fuel.\n\n\nOriginality/value\nThe blend of Calophyllum inophyllum oil-producer gas is used to run the diesel engine. Performance and emission analysis has been carried out, compared, optimized and validated.\n","PeriodicalId":23852,"journal":{"name":"World Journal of Engineering","volume":" ","pages":""},"PeriodicalIF":1.4000,"publicationDate":"2023-08-16","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"0","resultStr":"{\"title\":\"Taguchi and ANN-based optimization method for predicting maximum performance and minimum emission of a VCR diesel engine powered by diesel, biodiesel, and producer gas\",\"authors\":\"Taraprasad Mohapatra, S. Mishra, Mukesh Bathre, S. S. Sahoo\",\"doi\":\"10.1108/wje-04-2023-0116\",\"DOIUrl\":null,\"url\":null,\"abstract\":\"\\nPurpose\\nThe study aims to determine the the optimal value of output parameters of a variable compression ratio (CR) diesel engine are investigated at different loads, CR and fuel modes of operation experimentally. The output parameters of a variable compression ratio (CR) diesel engine are investigated at different loads, CR and fuel modes of operation experimentally. The performance parameters like brake thermal efficiency (BTE) and brake specific energy consumption (BSEC), whereas CO emission, HC emission, CO2 emission, NOx emission, exhaust gas temperature (EGT) and opacity are the emission parameters measured during the test. Tests are conducted for 2, 6 and 10 kg of load, 16.5 and 17.5 of CR.\\n\\n\\nDesign/methodology/approach\\nIn this investigation, the first engine was fueled with 100% diesel and 100% Calophyllum inophyllum oil in single-fuel mode. Then Calophyllum inophyllum oil with producer gas was fed to the engine. Calophyllum inophyllum oil offers lower BTE, CO and HC emissions, opacity and higher EGT, BSEC, CO2 emission and NOx emissions compared to diesel fuel in both fuel modes of operation observed. The performance optimization using the Taguchi approach is carried out to determine the optimal input parameters for maximum performance and minimum emissions for the test engine. The optimized value of the input parameters is then fed into the prediction techniques, such as the artificial neural network (ANN).\\n\\n\\nFindings\\nFrom multiple response optimization, the minimum emissions of 0.58% of CO, 42% of HC, 191 ppm NOx and maximum BTE of 21.56% for 16.5 CR, 10 kg load and dual fuel mode of operation are determined. Based on generated errors, the ANN is also ranked for precision. The proposed ANN model provides better prediction with minimum experimental data sets. The values of the R2 correlation coefficient are 1, 0.95552, 0.94367 and 0.97789 for training, validation, testing and all, respectively. The said biodiesel may be used as a substitute for conventional diesel fuel.\\n\\n\\nOriginality/value\\nThe blend of Calophyllum inophyllum oil-producer gas is used to run the diesel engine. Performance and emission analysis has been carried out, compared, optimized and validated.\\n\",\"PeriodicalId\":23852,\"journal\":{\"name\":\"World Journal of Engineering\",\"volume\":\" \",\"pages\":\"\"},\"PeriodicalIF\":1.4000,\"publicationDate\":\"2023-08-16\",\"publicationTypes\":\"Journal Article\",\"fieldsOfStudy\":null,\"isOpenAccess\":false,\"openAccessPdf\":\"\",\"citationCount\":\"0\",\"resultStr\":null,\"platform\":\"Semanticscholar\",\"paperid\":null,\"PeriodicalName\":\"World Journal of Engineering\",\"FirstCategoryId\":\"1085\",\"ListUrlMain\":\"https://doi.org/10.1108/wje-04-2023-0116\",\"RegionNum\":0,\"RegionCategory\":null,\"ArticlePicture\":[],\"TitleCN\":null,\"AbstractTextCN\":null,\"PMCID\":null,\"EPubDate\":\"\",\"PubModel\":\"\",\"JCR\":\"Q2\",\"JCRName\":\"ENGINEERING, MULTIDISCIPLINARY\",\"Score\":null,\"Total\":0}","platform":"Semanticscholar","paperid":null,"PeriodicalName":"World Journal of Engineering","FirstCategoryId":"1085","ListUrlMain":"https://doi.org/10.1108/wje-04-2023-0116","RegionNum":0,"RegionCategory":null,"ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"Q2","JCRName":"ENGINEERING, MULTIDISCIPLINARY","Score":null,"Total":0}

引用次数: 0

摘要

目的本研究旨在确定可变压缩比（CR）柴油机在不同负荷、CR和燃料运行模式下的输出参数的最佳值。实验研究了可变压缩比（CR）柴油机在不同负荷、CR和燃料运行模式下的输出参数。性能参数如制动热效率（BTE）和制动比能耗（BSEC），而CO排放、HC排放、CO2排放、NOx排放、废气温度（EGT）和不透明度是测试期间测量的排放参数。测试针对2、6和10 kg负载，16.5和17.5的CR。设计/方法/方法在本研究中，第一台发动机以单燃料模式使用100%柴油和100%藻油。然后将含有生产者气体的Calophyllum inophyllum油加入发动机。在观察到的两种燃料运行模式下，与柴油相比，水藻油的BTE、CO和HC排放量较低，不透明度较低，EGT、BSEC、CO2排放量和NOx排放量较高。使用田口方法进行性能优化，以确定测试发动机最大性能和最小排放的最佳输入参数。然后将输入参数的优化值输入到预测技术中，如人工神经网络（ANN）。发现从多重响应优化中，最低排放量为0.58%的CO、42%的HC、191 对于16.5 CR，10的NOx ppm和21.56%的最大BTE kg负载和双燃料运行模式。基于产生的误差，还对人工神经网络的精度进行了排序。所提出的人工神经网络模型以最小的实验数据集提供了更好的预测。训练、验证、测试和所有测试的R2相关系数分别为1、0.95552、0.94367和0.97789。所述生物柴油可以用作传统柴油燃料的替代品。独创性/价值利用海藻油生产商的混合气体来运行柴油发动机。性能和排放分析已经进行、比较、优化和验证。

本文章由计算机程序翻译，如有差异，请以英文原文为准。

查看原文本刊更多论文

Taguchi and ANN-based optimization method for predicting maximum performance and minimum emission of a VCR diesel engine powered by diesel, biodiesel, and producer gas

Purpose The study aims to determine the the optimal value of output parameters of a variable compression ratio (CR) diesel engine are investigated at different loads, CR and fuel modes of operation experimentally. The output parameters of a variable compression ratio (CR) diesel engine are investigated at different loads, CR and fuel modes of operation experimentally. The performance parameters like brake thermal efficiency (BTE) and brake specific energy consumption (BSEC), whereas CO emission, HC emission, CO2 emission, NOx emission, exhaust gas temperature (EGT) and opacity are the emission parameters measured during the test. Tests are conducted for 2, 6 and 10 kg of load, 16.5 and 17.5 of CR. Design/methodology/approach In this investigation, the first engine was fueled with 100% diesel and 100% Calophyllum inophyllum oil in single-fuel mode. Then Calophyllum inophyllum oil with producer gas was fed to the engine. Calophyllum inophyllum oil offers lower BTE, CO and HC emissions, opacity and higher EGT, BSEC, CO2 emission and NOx emissions compared to diesel fuel in both fuel modes of operation observed. The performance optimization using the Taguchi approach is carried out to determine the optimal input parameters for maximum performance and minimum emissions for the test engine. The optimized value of the input parameters is then fed into the prediction techniques, such as the artificial neural network (ANN). Findings From multiple response optimization, the minimum emissions of 0.58% of CO, 42% of HC, 191 ppm NOx and maximum BTE of 21.56% for 16.5 CR, 10 kg load and dual fuel mode of operation are determined. Based on generated errors, the ANN is also ranked for precision. The proposed ANN model provides better prediction with minimum experimental data sets. The values of the R2 correlation coefficient are 1, 0.95552, 0.94367 and 0.97789 for training, validation, testing and all, respectively. The said biodiesel may be used as a substitute for conventional diesel fuel. Originality/value The blend of Calophyllum inophyllum oil-producer gas is used to run the diesel engine. Performance and emission analysis has been carried out, compared, optimized and validated.

求助全文

通过发布文献求助，成功后即可免费获取论文全文。去求助

来源期刊

World Journal of Engineering ENGINEERING, MULTIDISCIPLINARY-

CiteScore

4.20

自引率

10.50%

发文量

期刊介绍： The main focus of the World Journal of Engineering (WJE) is on, but not limited to; Civil Engineering, Material and Mechanical Engineering, Electrical and Electronic Engineering, Geotechnical and Mining Engineering, Nanoengineering and Nanoscience The journal bridges the gap between materials science and materials engineering, and between nano-engineering and nano-science. A distinguished editorial board assists the Editor-in-Chief, Professor Sun. All papers undergo a double-blind peer review process. For a full list of the journal''s esteemed review board, please see below.