Investigation and back-propagation modeling of base pressure at sonic and supersonic Mach numbers

Afzal, Asif; Khan, Sher Afghan; Islam, Md. Tariqul; Jilte, Ravindra; Khan, Aijaz Ahmed; Soudagar, Manzoore Elahi M.

doi:10.1063/5.0022015

Cited by 43 publications

(21 citation statements)

References 30 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Normalized root mean square was used to check and evaluate the performance of the ANN [ 33 ]. Three inputs (L: load, S: speed, D: distance) were used to construct the ANN architecture for all the samples in which weight loss (w) was the output, and the hidden layer indicated that an interaction between the neurons was not visible [ 54 , 55 , 56 , 57 ].…”

Section: Resultsmentioning

confidence: 99%

Multi Ceramic Particles Inclusion in the Aluminium Matrix and Wear Characterization through Experimental and Response Surface-Artificial Neural Networks

et al. 2021

Self Cite

View full text Add to dashboard Cite

Lightweight composite materials have recently been recognized as appropriate materials have been adopted in many industrial applications because of their versatility. The present research recognizes the inclusion of ceramics such as Gr and B4C in manufacturing AMMCs through stir casting. Prepared composites were tested for hardness and wear behaviour. The tests’ findings revealed that the reinforced matrix was harder (60%) than the un-reinforced alloy because of the increased ceramic phase. The rising content of B4C and Gr particles led to continuous improvements in wear resistance. The microstructure and worn surface were observed through SEM (Scanning electron microscope) and revealed the formation of mechanically mixed layers of both B4C and Gr, which served as the effective insulation surface and protected the test sample surface from the steel disc. With the rise in the content of B4C and Gr, the weight loss declined, and significant wear resistance was achieved at 15 wt.% B4C and 10 wt.% Gr. A response surface analysis for the weight loss was carried out to obtain the optimal objective function. Artificial neural network methodology was adopted to identify the significance of the experimental results and the importance of the wear parameters. The error between the experimental and ANN results was found to be within 1%.

show abstract

Section: Resultsmentioning

confidence: 99%

Multi Ceramic Particles Inclusion in the Aluminium Matrix and Wear Characterization through Experimental and Response Surface-Artificial Neural Networks

et al. 2021

Self Cite

View full text Add to dashboard Cite

show abstract

“…The ANN used 520 experimental results, split into 3 segments: the training set (65% data), the test set (25% data), and the validation set (10% data). The importance of determining the best ANN architecture is critical because it has a significant impact on the results [50][51][52]. The optimisation of ANN variables is achieved by minimising the mean square error (MSE) after examining a large number of distinctly configured neural networks [53].…”

Section: Results Of Ann Modellingmentioning

confidence: 99%

Influence of Heat Treatment and Reinforcements on Tensile Characteristics of Aluminium AA 5083/Silicon Carbide/Fly Ash Composites

et al. 2021

Self Cite

View full text Add to dashboard Cite

The effect of reinforcements and thermal exposure on the tensile properties of aluminium AA 5083–silicon carbide (SiC)–fly ash composites were studied in the present work. The specimens were fabricated with varying wt.% of fly ash and silicon carbide and subjected to T6 thermal cycle conditions to enhance the properties through “precipitation hardening”. The analyses of the microstructure and the elemental distribution were carried out using scanning electron microscopic (SEM) images and energy dispersive spectroscopy (EDS). The composite specimens thus subjected to thermal treatment exhibit uniform distribution of the reinforcements, and the energy dispersive spectrum exhibit the presence of Al, Si, Mg, O elements, along with the traces of few other elements. The effects of reinforcements and heat treatment on the tensile properties were investigated through a set of scientifically designed experimental trials. From the investigations, it is observed that the tensile and yield strength increases up to 160 °C, beyond which there is a slight reduction in the tensile and yield strength with an increase in temperature (i.e., 200 °C). Additionally, the % elongation of the composites decreases substantially with the inclusion of the reinforcements and thermal exposure, leading to an increase in stiffness and elastic modulus of the specimens. The improvement in the strength and elastic modulus of the composites is attributed to a number of factors, i.e., the diffusion mechanism, composition of the reinforcements, heat treatment temperatures, and grain refinement. Further, the optimisation studies and ANN modelling validated the experimental outcomes and provided the training models for the test data with the correlation coefficients for interpolating the results for different sets of parameters, thereby facilitating the fabrication of hybrid composite components for various automotive and aerospace applications.

show abstract

“…The six BPMs with two hidden layers containing four neurons were determined to be the best suited for regression analysis. The very non-linear values of the base and wall pressure are correctly predicted by BPM 5 and BPM 6 [167]. Figure 12 depicts a broad schematic depiction of the back-propagation model (BPM).…”

Section: Machine Learningmentioning

confidence: 90%

A Critical Review of Supersonic Flow Control for High-Speed Applications

2021

Self Cite

View full text Add to dashboard Cite

In high-speed fluid dynamics, base pressure controls find many engineering applications, such as in the automobile and defense industries. Several studies have been reported on flow control with sudden expansion duct. Passive control was found to be more beneficial in the last four decades and is used in devices such as cavities, ribs, aerospikes, etc., but these need additional control mechanics and objects to control the flow. Therefore, in the last two decades, the active control method has been used via a microjet controller at the base region of the suddenly expanded duct of the convergent–divergent (CD) nozzle to control the flow, which was found to be a cost-efficient and energy-saving method. Hence, in this paper, a systemic literature review is conducted to investigate the research gap by reviewing the exhaustive work on the active control of high-speed aerodynamic flows from the nozzle as the major focus. Additionally, a basic idea about the nozzle and its configuration is discussed, and the passive control method for the control of flow, jet and noise are represented in order to investigate the existing contributions in supersonic speed applications. A critical review of the last two decades considering the challenges and limitations in this field is expressed. As a contribution, some major and minor gaps are introduced, and we plot the research trends in this field. As a result, this review can serve as guidance and an opportunity for scholars who want to use an active control approach via microjets for supersonic flow problems.

show abstract

Investigation and back-propagation modeling of base pressure at sonic and supersonic Mach numbers

Cited by 43 publications

References 30 publications

Multi Ceramic Particles Inclusion in the Aluminium Matrix and Wear Characterization through Experimental and Response Surface-Artificial Neural Networks

Multi Ceramic Particles Inclusion in the Aluminium Matrix and Wear Characterization through Experimental and Response Surface-Artificial Neural Networks

Influence of Heat Treatment and Reinforcements on Tensile Characteristics of Aluminium AA 5083/Silicon Carbide/Fly Ash Composites

A Critical Review of Supersonic Flow Control for High-Speed Applications

Contact Info

Product

Resources

About