Predicting the fuel flow rate of commercial aircraft via multilayer perceptron, radial basis function and ANFIS artificial neural networks

2020 ◽  
pp. 1-19
Author(s):  
T. Baklacioglu
Keyword(s):  
Radial Basis Function ◽  
Flow Rate ◽  
Basis Function ◽  
Multilayer Perceptron ◽  
Ann Model ◽  
Commercial Aircraft ◽  
Fuel Flow Rate ◽  
Fuel Flow ◽  
Ann Models ◽  
Hidden Layer

ABSTRACT A first attempt is made to use recently developed, non-conventional Artificial Neural Network (ANN) models with Multilayer Perceptron (MLP), Radial Basis Function (RBF) and Adaptive Neuro-Fuzzy Interference System (ANFIS) architectures to predict the fuel flow rate of a commercial aircraft using real data obtained from Flight Data Records (FDRs) of the cruise, climb and descent phases. The training of the architectures with a single hidden layer is performed by utilising the Delta-Bar-Delta (DBD), Conjugate Gradient (CG) and Quickprop (QP) algorithms. The optimum network topologies are sought by varying the number of processing elements in the hidden layer of the networks using a trial-and-error method. An evaluation of the approximate fuel intake values against the ideal fuel intake data from the FDRs indicates a good fit for all three ANN models. Thus, more accurate fuel intake estimations can be obtained by applying the RBF-ANN model during the climb and descent flight stages, whereas the MLP-ANN model is more effective for the cruise phase. The best accuracy obtained in terms of the linear correlation coefficient is 0.99988, 0.91946 and 0.95252 for the climb, cruise and descent phase, respectively.

scholarly journals Specialios struktūros daugiasluoksnis perceptronas daugiamačiams duomenims vizualizuoti

2009 ◽  
Vol 50 ◽  
pp. 358-364
Author(s):  
Laura Ringienė ◽  
Gintautas Dzemyda
Keyword(s):  
Radial Basis Function ◽  
Basis Function ◽  
Multilayer Perceptron ◽  
Data Clustering ◽  
Multidimensional Data ◽  
Feed Forward Neural Network ◽  
Data Set ◽  
Multidimensional Data Visualization ◽  
Radial Basis ◽  
Hidden Layer

Pasiūlytas ir ištirtas radialinių bazinių funkcijų ir daugiasluoksnio perceptrono junginys daugiamačiams duomenis vizualizuoti. Siūlomas vizualizavimo būdas apima daugiamačių duomenų matmenų mažinimą naudojant radialines bazines funkcijas, daugiamačių duomenų suskirstymą į klasterius, klasterį charakterizuojančių skaitinių reikšmių nustatymą ir daugiamačių duomenų vizualizavimą dirbtinio neuroninio tinklo paskutiniame paslėptajame sluoksnyje.Special Multilayer Perceptron for Multidimensional Data VisualizationLaura Ringienė, Gintautas Dzemyda SummaryIn this paper a special feed forward neural network, consisting of the radial basis function layer and a multilayer perceptron is presented. The multilayer perceptron has been proposed and investigated for multidimensional data visualization. The roposedvisualization approach includes data clustering, determining the parameters of the radial basis function and forming the data set to train the multilayer perceptron. The outputs of the last hidden layer are assigned as coordinates of the visualized points.

Dynamic Simulation of a HRSG System for a Given Start-Up/Shut Down Curve

2011 ◽  
Author(s):  
Hun Cha ◽  
Yoo Seok Song ◽  
Kyu Jong Kim ◽  
Jung Rae Kim ◽  
Sung Min KIM
Keyword(s):  
Dynamic Analysis ◽  
Gas Turbine ◽  
Flow Rate ◽  
Exhaust Gas ◽  
Fuel Flow Rate ◽  
Fuel Flow ◽  
Start Up ◽  
Gas Turbine Exhaust ◽  
Main Steam ◽  
Duct Burner

An inappropriate design of HRSG (Heat Recovery Steam Generator) may lead to mechanical problems including the fatigue failure caused by rapid load change such as operating trip, start-up or shut down. The performance of HRSG with dynamic analysis should be investigated in case of start-up or shutdown. In this study, dynamic analysis for the HRSG system was carried out by commercial software. The HRSG system was modeled with HP, IP, LP evaporator, duct burner, superheater, reheater and economizer. The main variables for the analysis were the temperature and mass flow rate from gas turbine and fuel flow rate of duct burner for given start-up (cold/warm/hot) and shutdown curve. The results showed that the exhaust gas condition of gas turbine and fuel flow rate of duct burner were main factors controlling the performance of HRSG such as flow rate and temperature of main steam from final superheater and pressure of HP drum. The time delay at the change of steam temperature between gas turbine exhaust gas and HP steam was within 2 minutes at any analysis cases.

Universal Approximation Using Radial-Basis-Function Networks

1991 ◽  
Vol 3 (2) ◽  
pp. 246-257 ◽  
Author(s):  
J. Park ◽  
I. W. Sandberg
Keyword(s):  
Finite Number ◽  
Radial Basis Function ◽  
Basis Function ◽  
Universal Approximation ◽  
Radial Basis Function Networks ◽  
Feedforward Networks ◽  
Rbf Networks ◽  
Smoothing Factor ◽  
Radial Basis ◽  
Hidden Layer

There have been several recent studies concerning feedforward networks and the problem of approximating arbitrary functionals of a finite number of real variables. Some of these studies deal with cases in which the hidden-layer nonlinearity is not a sigmoid. This was motivated by successful applications of feedforward networks with nonsigmoidal hidden-layer units. This paper reports on a related study of radial-basis-function (RBF) networks, and it is proved that RBF networks having one hidden layer are capable of universal approximation. Here the emphasis is on the case of typical RBF networks, and the results show that a certain class of RBF networks with the same smoothing factor in each kernel node is broad enough for universal approximation.

Model Analysis of Syngas Combustion and Emissions for a Micro Gas Turbine

2015 ◽  
Vol 137 (6) ◽  
Author(s):  
Chi-Rong Liu ◽  
Hsin-Yi Shih
Keyword(s):  
Gas Turbine ◽  
Flow Rate ◽  
Heat Input ◽  
Emission Characteristics ◽  
Nox Emissions ◽  
Fuel Flow Rate ◽  
Micro Gas Turbine ◽  
Fuel Flow ◽  
Temperature Distributions ◽  
Flame Structures

The purpose of this study is to investigate the combustion and emission characteristics of syngas fuels applied in a micro gas turbine, which is originally designed for a natural gas fired engine. The computation results were conducted by a numerical model, which consists of the three-dimension compressible k–ε model for turbulent flow and PPDF (presumed probability density function) model for combustion process. As the syngas is substituted for methane, the fuel flow rate and the total heat input to the combustor from the methane/syngas blended fuels are varied with syngas compositions and syngas substitution percentages. The computed results presented the syngas substitution effects on the combustion and emission characteristics at different syngas percentages (up to 90%) for three typical syngas compositions and the conditions where syngas applied at fixed fuel flow rate and at fixed heat input were examined. Results showed the flame structures varied with different syngas substitution percentages. The high temperature regions were dense and concentrated on the core of the primary zone for H2-rich syngas, and then shifted to the sides of the combustor when syngas percentages were high. The NOx emissions decreased with increasing syngas percentages, but NOx emissions are higher at higher hydrogen content at the same syngas percentage. The CO2 emissions decreased for 10% syngas substitution, but then increased as syngas percentage increased. Only using H2-rich syngas could produce less carbon dioxide. The detailed flame structures, temperature distributions, and gas emissions of the combustor were presented and compared. The exit temperature distributions and pattern factor (PF) were also discussed. Before syngas fuels are utilized as an alternative fuel for the micro gas turbine, further experimental testing is needed as the modeling results provide a guidance for the improved designs of the combustor.

scholarly journals A Method to Measure the Fuel Distribution and the Fuel Captured by V–Gutter Flameholder in High Speed Airstream

1985 ◽  
Author(s):  
Gu Shan-Jian ◽  
Yang Mao-Lin ◽  
Li Xiang-Yi
Keyword(s):  
Flow Rate ◽  
Experimental Error ◽  
High Speed ◽  
Fuel Flow Rate ◽  
Fuel Distribution ◽  
Fuel Flow ◽  
Sampling Condition ◽  
Detailed Determination

A method to measure the fuel distribution and the percentage of fuel flow rate captured by a V-gutter flameholder in a high speed airstream has been developed. The effects of configuration and size of the probe and temprature of the sample mixture in the probe on measurement have been investigated. The detailed determination of isokinetic sampling condition is described. The effects of V-gutter geometry on flowfield have been considered. The total experimental error is of the order ±5%.

System Level Analysis of Acoustically Forced Nonpremixed Swirling Flames

2014 ◽  
Vol 6 (3) ◽  
Author(s):  
Uyi Idahosa ◽  
Saptarshi Basu ◽  
Ankur Miglani
Keyword(s):  
Heat Release ◽  
Flow Rate ◽  
Ccd Camera ◽  
Time Dependent ◽  
Fuel Flow Rate ◽  
Fuel Flow ◽  
Flame Response ◽  
Rate Oscillations ◽  
Swirling Flames ◽  
Flame Sheet

This paper reports an experimental investigation of dynamic response of nonpremixed atmospheric swirling flames subjected to external, longitudinal acoustic excitation. Acoustic perturbations of varying frequencies (fp = 0–315 Hz) and velocity amplitudes (0.03 ≤ u′/Uavg ≤ 0.30) are imposed on the flames with various swirl intensities (S = 0.09 and 0.34). Flame dynamics at these swirl levels are studied for both constant and time-dependent fuel flow rate configurations. Heat release rates are quantified using a photomultiplier (PMT) and simultaneously imaged with a phase-locked CCD camera. The PMT and CCD camera are fitted with 430 nm ±10 nm band pass filters for CH* chemiluminescence intensity measurements. Flame transfer functions and continuous wavelet transforms (CWT) of heat release rate oscillations are used in order to understand the flame response at various burner swirl intensity and fuel flow rate settings. In addition, the natural modes of mixing and reaction processes are examined using the magnitude squared coherence analysis between major flame dynamics parameters. A low-pass filter characteristic is obtained with highly responsive flames below forcing frequencies of 200 Hz while the most significant flame response is observed at 105 Hz forcing mode. High strain rates induced in the flame sheet are observed to cause periodic extinction at localized regions of the flame sheet. Low swirl flames at lean fuel flow rates exhibit significant localized extinction and re-ignition of the flame sheet in the absence of acoustic forcing. However, pulsed flames exhibit increased resistance to straining due to the constrained inner recirculation zones (IRZ) resulting from acoustic perturbations that are transmitted by the co-flowing air. Wavelet spectra also show prominence of low frequency heat release rate oscillations for leaner (C2) flame configurations. For the time-dependent fuel flow rate flames, higher un-mixedness levels at lower swirl intensity is observed to induce periodic re-ignition as the flame approaches extinction. Increased swirl is observed to extend the time-to-extinction for both pulsed and unpulsed flame configurations under time-dependent fuel flow rate conditions.

scholarly journals A new generation of F1 race engines – hybrid power units

2016 ◽  
Vol 167 (4) ◽  
pp. 22-37 ◽  
Author(s):  
Zbigniew STĘPIEŃ
Keyword(s):  
Flow Rate ◽  
Regulatory Framework ◽  
Fuel Flow Rate ◽  
Maximum Performance ◽  
Engine Efficiency ◽  
Hybrid Power ◽  
Fuel Flow ◽  
New Generation ◽  
Technical Regulations

The paper aims at reviewing the evolution of the F1 engine technology and the associated regulatory framework governing the sport over the last 10 years. Technical regulations, in force since 2014, replaced the 2.4-liter V8 naturally aspirated engines with sophisticated hybrid units such as the 1.6-liter V6 turbocharged engines supported with energymanagement and recovery systems. Since 2014 the fundamental trend in the development of powertrains has been the advancement of their efficiency. Due to the fact that the fuel flow rate has been restricted, the maximum performance is now entirely dependent on the engine efficiency.

scholarly journals Estudo sobre Métricas para Definir Reputação do Autor de Comentários em Sites de Vendas de Produtos

2018 ◽  
Author(s):  
Carlos Augusto De Sá ◽  
Raimundo Santos Moura
Keyword(s):  
Radial Basis Function ◽  
Basis Function ◽  
Multilayer Perceptron ◽  
Radial Basis

Conhecer a reputação do autor de textos opinativos é de suma importância para avaliação de comentários na Web. Este artigo apresenta um estudo sobre medidas usadas no processo de avaliação da reputação do autor em sites de vendas de produtos. Realizou-se dois experimentos com as redes neurais Multilayer Perceptron (MLP) e Radial Basis Function (RBF), sendo que a rede MLP obteve melhor desempenho. Comparou-se também a abordagem TOP(X) original, usada para inferir os melhores comentários, com um novo modelo que utiliza rede MLP na dimensão da reputação do autor. Considerando os comentários excelentes e bons, a nova abordagem apresentou resultados significativamente superiores.

Sign in / Sign up

Export Citation Format

Share Document