scispace - formally typeset
Search or ask a question
Journal ArticleDOI

Predicting the methane number of gaseous fuels using an artificial neural network

26 Nov 2021-Vol. 12, Iss: 10, pp 1191-1198
TL;DR: The Methane Number (MN) is a critical gas quality parameter for gaseous-fueled engines as discussed by the authors, and it is a measure of knock resistance for Gaseous fuels.
Abstract: Methane number (MN) is a critical gas quality parameter for gaseous-fueled engines. It is a measure of knock resistance for gaseous fuels, as is the octane number for gasoline. Therefore, a priori ...
Citations
More filters
Journal ArticleDOI
26 Sep 2020-Energies
TL;DR: In this paper, an analysis of deviation of liquefied natural gas (LNG) composition and its impact on LNG quality as an engine fuel was presented, where the authors considered the impact of higher hydrocarbons and nitrogen content on the LNG parameters.
Abstract: The one of main quality requirements of natural gas as an engine fuel is the methane number (MN). This parameter indicates the fuel’s capability to avoid knocking in the engine. A higher MN value indicates a better natural gas quality for gas engines. Natural gas with higher methane content tends to have higher MN value. This study presents analysis of deviation of liquefied natural gas (LNG) composition and its impact on LNG quality as an engine fuel. The analysis of higher hydrocarbons and nitrogen content impact on LNG parameters was considered for several samples of LNG compositions. Most engine manufacturers want to set a new, lower limit value for methane number at 80. This fact causes significant restrictions on the range of variability in the composition of liquefied natural gas. The goal of this study was to determine the combination of the limit content of individual components in liquefied natural gas to achieve the strict methane number criterion (MN > 80). To fulfill this criterion, the methane content in LNG would have to exceed 93.7%mol, and a significant part of the LNG available on the market does not meet these requirements. The analysis also indicated that the methane number cannot be the only qualitative criterion, as its variability depends strongly on the LNG composition. To determine the applicability of LNG as an engine fuel, the simultaneous application of the methane number and Wobbe index criteria was proposed.

17 citations

Journal ArticleDOI
TL;DR: In this paper, performance characteristics of a spark-ignition (SI) engine operated under methane (baseline case) and biogas are compared at the compression ratio (CR) of 8.5:1.
Abstract: Biogas, which is a renewable alternative fuel, has high antiknocking properties with the potential to substitute fossil fuels in internal combustion engines. In this study, performance characteristics of a spark-ignition (SI) engine operated under methane (baseline case) and biogas are compared at the compression ratio (CR) of 8.5:1. Subsequently, the effect of CR on operational limits, performance, combustion, and emission characteristics of the engine fueled with biogas is evaluated. A variable compression ratio, spark-ignition engine was operated at various CRs of 8.5:1, 10:1, 11:1, 13:1, and 15:1 over a wide range of operating loads at 1500 rpm. Results showed that the operating range of the engine at 8.5:1 CR reduced when biogas was utilized in the engine instead of methane. However, the operating range of the engine for biogas extended with an increase in CR—an increase from 9.6 N-m-16.5 N-m to 2.8 N-m-15.1 N-m was observed when CR was increased from 8.5:1 to 15:1. The brake thermal efficiency improved from 13.7% to 16.3%, and the coefficient of variation (COV) of indicated mean effective pressure (IMEP) reduced from 12.7% to 1.52% when CR was increased from 8.5:1 to 15:1 at 8 N-m load. The emission level of carbon dioxide was decreased with an increase in CR due to an improvement in the thermal efficiency and the combustion process.

8 citations

Journal ArticleDOI
TL;DR: In this paper , the effect of hydrogen enrichment on performance, combustion and emission characteristics of a single-cylinder, four-stroke, water-cooled, biogas fuelled spark-ignition engine operated at the compression ratio of 10:1 and 1500 rpm has been evaluated using experimental and computational (CFD) studies.

7 citations

Journal ArticleDOI
TL;DR: In this article , the role of biogas in achieving sustainable development goals with an emphasis on its utilization in gaseous fuelled spark-ignited engines is discussed in detail with a note on engine operating parameters.

7 citations

Journal ArticleDOI
TL;DR: In this paper , the authors investigated the impact of gaseous fuel energy share on combustion noise in dual fuel diesel engines and found that an increment in methane share is effective in controlling the sound pressure levels to 95 dB (at 1.5 Newtonmeter [Nm] and 13.5 Nm loading conditions) and it is lower than the diesel only operational levels at all loading conditions.
Abstract: Rising pollution levels and stringent emission regulations have raised the requirement for the utilization of alternative green fuels as a sustainable energy source for engine applications. The economic and environmental benefits of using biomethane or renewable natural gas (>95% methane) in diesel engines have made it a promising solution. Sound levels caused by combustion noise, however, is an area of concern in the early stages of engine development process, and to the best of our knowledge studies related to the impact of gaseous fuel energy share on combustion noise in dual fuel diesel engines have not been reported in the literature. In the present work, experimental research on the combustion noise of a biomethane augmented dual fuel common‐rail direct injection (CRDI) diesel engine has been carried out with a correlational design of its performance and combustion characteristics at various engine operating conditions. Comparative results on combustion and performance characteristics of different fuel compositions at various engine loads revealed that there is a better utilization of gaseous fuel at full load conditions resulting in improved power output and reduced heat transfer losses. Cylinder pressure spectra derived from recorded in‐cylinder pressure were used for carrying out the quantitative analysis of sound levels. The results suggested that an increment in methane share is effective in controlling the sound pressure levels to 95 dB (at 1.5 Newton‐meter [Nm] and 13.5 Nm loading conditions) and it is lower than the diesel‐only operational levels at all loading conditions.

5 citations

References
More filters
Journal ArticleDOI
01 May 1993
TL;DR: The architecture and learning procedure underlying ANFIS (adaptive-network-based fuzzy inference system) is presented, which is a fuzzy inference System implemented in the framework of adaptive networks.
Abstract: The architecture and learning procedure underlying ANFIS (adaptive-network-based fuzzy inference system) is presented, which is a fuzzy inference system implemented in the framework of adaptive networks. By using a hybrid learning procedure, the proposed ANFIS can construct an input-output mapping based on both human knowledge (in the form of fuzzy if-then rules) and stipulated input-output data pairs. In the simulation, the ANFIS architecture is employed to model nonlinear functions, identify nonlinear components on-line in a control system, and predict a chaotic time series, all yielding remarkable results. Comparisons with artificial neural networks and earlier work on fuzzy modeling are listed and discussed. Other extensions of the proposed ANFIS and promising applications to automatic control and signal processing are also suggested. >

15,085 citations

Journal ArticleDOI
TL;DR: Experiments show that SCG is considerably faster than BP, CGL, and BFGS, and avoids a time consuming line search.

3,882 citations

01 Jan 1977
TL;DR: A robust and efficient implementation of a version of the Levenberg--Marquardt algorithm is discussed and it is shown that it has strong convergence properties.
Abstract: The nonlinear least-squares minimization problem is considered. Algorithms for the numerical solution of this problem have been proposed in the past, notably by Levenberg (Quart. Appl. Math., 2, 164-168 (1944)) and Marquardt (SIAM J. Appl. Math., 11, 431-441 (1963)). The present work discusses a robust and efficient implementation of a version of the Levenberg--Marquardt algorithm and shows that it has strong convergence properties. In addition to robustness, the main features of this implementation are the proper use of implicitly scaled variables and the choice of the Levenberg--Marquardt parameter by means of a scheme due to Hebden (AERE Report TP515). Numerical results illustrating the behavior of this implementation are included. 1 table. (RWR)

1,837 citations


"Predicting the methane number of ga..." refers methods in this paper

  • ...Various training functions have been applied in previous studies, such as Bayesian regularization [31], gradient descent with adaptive learning rule [32], scaled conjugate gradient [33], and Levenberg–Marquardt (LM) [34]....

    [...]

Journal ArticleDOI
TL;DR: Standard techniques for improved generalization from neural networks include weight decay and pruning and a comparison is made with results of MacKay using the evidence framework and a gaussian regularizer.
Abstract: Standard techniques for improved generalization from neural networks include weight decay and pruning. Weight decay has a Bayesian interpretation with the decay function corresponding to a prior over weights. The method of transformation groups and maximum entropy suggests a Laplace rather than a gaussian prior. After training, the weights then arrange themselves into two classes: (1) those with a common sensitivity to the data error and (2) those failing to achieve this sensitivity and that therefore vanish. Since the critical value is determined adaptively during training, pruning---in the sense of setting weights to exact zeros---becomes an automatic consequence of regularization alone. The count of free parameters is also reduced automatically as weights are pruned. A comparison is made with results of MacKay using the evidence framework and a gaussian regularizer.

362 citations

Journal ArticleDOI
TL;DR: In this article, the ability of an artificial neural network model, using a back propagation learning algorithm, to predict specific fuel consumption and exhaust temperature of a Diesel engine for various injection timings is studied.

210 citations