Elsevier

Fuel

Volume 235, 1 January 2019, Pages 317-325
Fuel

Full Length Article
Artificial intelligence based gene expression programming (GEP) model prediction of Diesel engine performances and exhaust emissions under Diesosenol fuel strategies

https://doi.org/10.1016/j.fuel.2018.07.116Get rights and content

Abstract

The study explores the affectability of oxygenated fuel on the performances and exhaust emissions of adulterated Diesel fueled engine. Diesel adulteration decreases brake thermal efficiency (Bth), brake specific energy consumption (BSEC), unburned hydrocarbon (UHC) and carbon monoxide (CO) emissions with significant reduction in NOX emissions. Ethanol blending improves the engine exhaust emissions without altering performance parameters of adulterated Diesel. In perspective of the experimental data, multi parametric artificial intelligence (AI) based gene expression programming (GEP) models have been developed for mapping the input (engine load, Kerosene share and Ethanol share) and output (Bth, BSEC, NOX, UHC and CO) relationship under Diesosenol platforms. The model predicted output has been validated with experimentally measured data and some statistical measures. The predicted model matched the experimental data with very lower mean square error (0.00002-0.00031). The statistical results such as correlation coefficient (0.99910-0.99995), absolute fraction of variance (0.99821-0.99989), Nash–Sutcliffe coefficient of efficiency (0.992-0.99974) and Kling–Gupta efficiency (0.98091-0.99736) obtained from the GEP model, along with mean absolute percentage error, mean squared relative error and prediction model uncertainty betokened itself as a real time robust machine identical tool under various Diesosenol stages. In addition, Pearson’s chi-square test or goodness of fit measurement elevates the GEP model prediction quality to a higher level.

Introduction

Conventional Diesel fueled engines have already been acknowledged as prime suspect for producing higher harmful air pollutants. Because of the higher poisonous exhaust emissions, conventional Diesel fueled vehicles have been authoritatively restricted in few places particularly in Indian capital Delhi [1], [2]. Nevertheless, higher thermal efficiency could be gained from Diesel energized vehicle with the penalties of higher NOX and UHC footprints [2], [3]. Global industrialization and rising fossil fuel energy demand ceaselessly dwindles the present fuel reserves. The global fossil fuel demand-supply balance is astonishingly fluctuating in the present millennium. The whole world is completely relying upon only couples of oil supply places that dictate the global fuel price. Inadequate fuel supply, abruptly rising global fossil fuel price and crisis have enhances the path of fuel adulteration. Fuel adulteration is a significant matter which specifically influences the economy of any country side by side debilitates the environment. Instead of the fact that, fuel adulteration, mostly done by Kerosene, in gasoline as well as Diesel has some features like improves chemical properties for the blended fuel and profitable business for the suppliers, but produces higher air pollutants [1]. Moreover, few countries are providing subsidy in household Kerosene which also enhances the path of fuel adulteration. However, a blue dye is included in subsidized Kerosene for identification concern, but it is fully miscible in conventional Diesel [2]. Kerosene blended conventional Diesel fuel has some focal points, for example, lower fuel density, increase fuel lubricity, atomization of charge and lower kinematic viscosity [4], [5] that increases its possibilities in a CI engine. Kerosene improves the cold flow properties by diluting the paraffin waxes of conventional Diesel fuel [6]. A Study of Yadav et al. [4] resolutely reported that Kerosene incorporation to conventional Diesel fuel decreases fuel density and kinematic viscosity. A recent study by Singh et al. [5] concluded that the volatility characteristic of Kerosene supplies homogeneous blend which helps in the diminution of soot emission better than conventional Diesel fuel operation. Reduced soot emission of Kerosene blended Diesel has also been reported by Bergstrand [7]. Patil et al. [8] in their studies deduced that Kerosene inclusion to conventional Diesel ameliorates NOX emissions of compression ignition (CI) engine at the cost of lower Bth as contrasted with pure Diesel operation. Adulterated Diesel especially utilized in off-road application based construction and agriculture sectors, wherein indirect injection (IDI) combustion type engine mainly engaged [2], [9], [10]. IDI engine contains a special type of chamber called pre-combustion chamber where homogeneous mixture is produced for the combustion. Kerosene mixed Diesel replaces some elements of required oxygen because of higher volatility characteristics and results in essential reduction of NOX emissions of the present day along with the penalties of higher UHC and CO footprints [1]. Juxtaposing to the above mentioned paradoxical issues and present day tough vehicle emission standards, automotive analysts need to adopt bio-based, renewable, oxygen containing fuel in order to fulfill the present day fossil fuel demand, help in meeting emission standard and support Diesel power trains. Ethanol, an oxygenated fuel, effectively substantiated itself as a effective additive for fossil fuel in IC engine domain, firmly utilized in the research works [11], [12], [13] devoted by the similar researchers. Owing to higher oxygen molecule in its chemical structure and other inherent favorable fuel properties, Ethanol blending elevates the combustibility of conventional Diesel fuel by providing superfluous amount of oxygen. Although, only a limited amount of Ethanol can be supported in CI engine fuel system, but this limited addition can demonstrate tremendous results towards performances as well as exhaust emissions improvements [14], [15], [16]. Paul et al. [17] in their research work unquestionably clarified a fact that, maximum 10% Ethanol can be added to conventional Diesel fuel at preferred experimental conditions, more than that amount of percentage displays post phase separation as a result of natural surfactant of Ethanol. Xu et al. [18] concluded that Ethanol addition to conventional Diesel noticeably improved Bth together with soot emission of Diesel engine. Works of Yan et al. [19] irradiated that, oxygenated Ethanol significantly reduces NOX emissions of Diesel engine compared to baseline Diesel fuel operation. Kim and Choi [20] in their investigation utilized Diesel-Ethanol blends in CI engine and firmly manifested that Ethanol inclusion to conventional Diesel fuel drastically improves CO and UHC emissions of the engine. The ameliorations with Ethanol blended Diesel in parallel exhaust fumes viz. UHC and NOX have been demonstrated by Hardenberg and Schaefer [21] without any substantial alteration in performance parameters. Accordingly, the present investigation aims from the above cited contemporary research works in IC engine domain to harness the positive aspects of bio-based oxygen containing Ethanol as a stronger inexhaustible platform in supplying a supportable arrangement of fuel additive to compensate for adulterated Diesel under the simultaneous end goal of similar or better performance-exhaust emission benchmark.

Concerned with unique selling point (USP) of consumers, the traditional based Diesel energized engine is designed with a predefined objective that is least number of operational complexities under standardized vehicle emission legislation. However, experimentations on traditional with concerning present day converging emission standards and multi-fuel blend are quite strenuous, time consuming and costlier. One-factor-at-a-time (OFAT) based traditional engine experiments need to be uprooted by designed experiments that can vary multiple factor at a time as it is turned out to be less feasible in the face of modern generation multi-objective requirements under multi-fuel blend. Such headways in Diesel power train under multi-fuel blend requires to control several independent parameters with the goal that the essential USP ought to be maintained. Machine identical tool (MIT) based computational models are the only possible route to solve the above discussed paradoxical issues. Among the different kinds of MIT based computational models, GEP based meta-model has already been revealed its strength in IC engine domain to harness the intrinsic prospects of multi-fuel blend [22], [23]. GEP based meta-model requires less number of resources, time and cost compared to other MIT models such as computational fluid dynamics and chemical kinetic. Additionally, GEP models are not influenced by model iteration, topology variation, transfer functions and learning algorithms as it is formed by fully evaluation principle. A GEP model has been formulated by Roy et al. [22] in order to foresee Diesel engine performance and exhaust emission parameters under CNG dual-fuel operation with R value ranges of 0.99936-0.99999. Research studies dedicated by the numerous analysts [23], [24], [25] have also been exhibited the feasibility of GEP model for mapping the respective output parameters under consideration and accordingly developed its intrinsic quality as wide-ranging MIT.

On collocating the ever increasing Diesel adulteration, fuel crisis and conscious present day vehicle emission legislations, the enforceable path for supporting Diesel power train along with future energy reserve and meeting arduous emission standard is by the deployment of oxygen containing renewable Ethanol. Studies devoted in the above discussed literatures [11], [12], [13], [14], [15], [16], [17], [18], [19], [20], [21] have confidently been exploited the potentiality of bio-based Ethanol in reducing Diesel engine exhaust emissions without substantial decrement in performances and supporting the Diesel power train. Moreover, conventional Diesel engine needs to be performed in effective way so that the various independent parameters can be varied at one time side by side the present day fossil fuel crisis and converging vehicle norms can be fulfilled along with the required USP. AI based GEP model has been effectively used by the various researchers to map the performance and exhaust emission characteristics in IC engine domain in real time. Adulterated Diesel mostly used in rural areas wherein IDI engines are mainly employed. Research investigation with adulterated Diesel on IDI engine along with the applicability of GEP model under Ethanol enrichment has not been revealed. Hence, an endeavor is prepared to fill this gap.

In perceptive of the above motivation, the main objective of present study thus focuses on the followings:

  • To study the impacts of oxygenated Ethanol on the performance-emission footprints of adulterated Diesel fueled IDI engine.

  • To evaluate AI based GEP model to accurately map the performance-emission profile of the existing engine under Diesosenol platforms.

  • To develop correlation between input–output paradigms and quantify its cause and effect, and validate against experimentally measured output under Diesosenol.

Section snippets

Experimental setup and procedure

The experimental study has been conducted on a single cylinder four-stroke horizontal IDI engine mounted on sturdy base frame that fabricated with mild steel “C” channel. An eddy current dynamometer has been coupled to the output shaft of IDI engine to deliver engine load. A crank angle encoder has been used to determine the engine rpm, which has 10 of accuracy. An AVL Digas 444 exhaust gas analyzer, conforming to ARAI certification-ARAI/TA (4G-RV)/AVL/DiGas 444/0910-12, has been attached to

Gene expression programming model

A GEP model, in its quintessence, is exactly similar to the genetic material recombination in living organisms based on Darwinian principle [25]. AI based GEP models are the evolutionary algorithms that structure complex tree called expression trees (ET). ET learns and adjusts by changing their sizes, shapes and creation. Prior to the formation of a complex ET, it is encoded as linear chromosomes of preset size [24]. Basically, AI based GEP model contains one chromosome. This single chromosome

Result, discussion and validation

Fig. 2, Fig. 3, Fig. 4, Fig. 5, Fig. 6(a–c) depicts the ETs, comparison between model predicted and experimental data alongside experimental uncertainty, and R value between model predicted and experimental data for the output parameters of IDI engine under Diesosenol platforms, respectively. Fig. 7, Fig. 8 shows the comparison of correlation coefficients and error matrices for the model output parameters, respectively.

Conclusion

In this present study, the effects of oxygenated fuel inclusion on the performances and exhaust emissions of an existing IDI engine under adulterated Diesel stages have been surveyed. In view of the IDI engine experimental performances and exhaust emissions data, AI affiliated GEP models have been developed to replicate input–output relationship under Diesosenol strategies. The key upshots of the study are abridged below:

  • Adulterated Diesel deteriorates Bth, BSEC, UHC and CO emission with the

References (25)

  • A. Nazari et al.

    Modeling the compressive strength of geopolymeric binders by gene expression programming-GEP

    Expert Syst Appl

    (2013)
  • S. Bhowmik et al.

    Artificial Neural Network Prediction of Diesel engine performance and emission fueled with Diesel-kerosene-ethanol blends: a fuzzy based optimization

    J Energy Res Technol

    (2017)
  • Cited by (0)

    View full text