|
ARPN Journal of Engineering and
Applied Sciences November 2022 | Vol. 17 No. 22 |
Title: |
Provision of Carboxymethyl Cellulose
material based on durian seed powder |
Author (s): |
M. H.
S. Ginting, A. Utama, H. Muhammad, Maulida R. Tambun and A. H.
Rajagukguk |
Abstract: |
Carboxymethyl Cellulose (CMC) is a water-soluble derivative compound.
Synthesis of this CMC includes the stages of cellulose alkalinization
and carboxymethylation reactions. Durian seed flour is reacted with
sodium hydroxide in isopropanol and sodium chloroacetate as a solvent.
This study aims to determine the effect of temperature,
carboxymethylation reaction time, volume variation of sodium hydroxide,
and weight of sodium chloroacetate on the degree of substitution of
resulting CMC. The results are shown that the greater the volume of
20%NaOH solution added and the longer carboxymethylation reaction time,
the higher the carboxymethylcellulose Substitution Degree (DS) was
produced. The most significant degree of substitution was the addition
of 10 ml of 20%NaO Hand the reaction time lasted for 2 hours, namely
0.61. |
|
|
Full Text |
|
|
Title: |
Classification of geopolymer concrete
grade with Convolutional Neural Network using LeNet architecture |
Author (s): |
Agustinus Agus Setiawan, Roesdiman Soegiarso, Harianto Hardjasaputra and
Lina |
Abstract: |
Geopolymer concrete is one of the innovations in the field of
construction materials, this kind of material can reduce the impact of
carbon emissions on the environment. Geopolymer concrete is an
environmentally friendly material, which does not use cement as a base
material. Compressive strength is a quality parameter of geopolymer
concrete as well as normal concrete. This study aims to model the
compressive strength classification of geopolymer concrete using an
artificial neural network. The classification process is based on the
composition of the geopolymer concrete mixture by considering the
geopolymer concrete curing process, including the temperature and
duration of geopolymer concrete curing. Eight independent variables and
one dependent variable were used in this modelling process. The
artificial neural network model developed is a Deep Learning model,
using the Convolutional Neural Network algorithm and LeNet network
architecture. Three variations of hyper parameters were compared in this
study, including variations in the number of epochs, learning rate
values, ??and variations in the optimizer function. From the modelling
results that have been made, the LeNet architectural model with 1000
epochs, a learning rate value of 0.001, and using the Adam optimizer
function is able to produce the best model with a training accuracy rate
of 86.15%, and an R-square value of 0.93. This model is able to produce
a testing accuracy value of 79.80%. As an alternative, the RMSprop
optimizer function is also able to produce an adequate model to classify
the compressive strength of geopolymer concrete. |
|
|
Full Text |
|
|
Title: |
Preliminary evaluation of commercial
additives using four-ball tribotester machine |
Author (s): |
Bukhari Manshoor, Mohd. Zaki Bahrom, Norfazillah Talib and Zulkifli
Mohamed |
Abstract: |
This
study evaluates the kinematic viscosity, coefficient of friction, and
wear scar diameter of the commercial additive (CAA) with engine oil. The
base oil, synthetic engine oil (SEO) SAE 10W40, has blended physically
with selected CAA in a volume ratio of 1:0.06. The results show that one
of the blended SEO and CAA increases the kinematic viscosity value at
temperatures of 40°C and 100°C. However, the value of the coefficient of
friction and wear scar diameter blended between SEO and CAA is higher
compared to pure synthetic engine oil. Based on the finding of this
study, the role of additional commercial additives can be applied to
improve several of the lubricant properties, such as viscosity. It has
been demonstrated that synthetic engine oil is superior without
additional commercial additives for automotive lubrication. |
|
|
Full Text |
|
|
Title: |
Crash investigation on frontal vehicle
chassis frame using Finite Element Simulation |
Author (s): |
Nuruddin Ariffin, Kamarul-Azhar Kamarudin, Ahmad Sufian Abdullah and
Mohd. Idrus Abd Samad |
Abstract: |
Car
chassis can be considered the primary protective shield for the safety
of the passenger during rear-end crashes. This study focuses on the
deformation and failure behavior of the frontal car A-pillar chassis
frame when subjected to collision with a heavy vehicle. Two different
angles of the A-pillar chassis frame used are 45-degree and 70-degree.
The crash simulation is conducted by using Finite Element software under
the explicit dynamic. The car chassis frame geometries are designed by
using SolidWorks 2021 and imported to the finite element software while
a rigid block is designed in the finite element software as a rigid body
to replicate the heavy vehicle. The chassis body is simulated for two
types of materials, Aluminum alloy, and steel. The car speed impacted at
60 km/h. Results show that the intrusion of a rear barrier for 45
degrees of aluminum alloy will stop at 0.03 s but for 70 degrees it will
intrude the car frame until the end. For the steel car frame, 45 degrees
design is capable to withstand the intrusion of a rear barrier from a
serious deform but for 70 degrees the intrusion will continue until the
end. Car frame crush behavior, energy dissipation, and vehicle
decelerations from the crash simulation were observed. |
|
|
Full Text |
|
|
Title: |
Development of hybrid nano cutting fluid
from tri-agro waste synthesized nanoparticles |
Author (s): |
Omolayo M. Ikumapayi, Sunday A. Afolalu, Temitayo S. Ogedengbe, Joseph
F. Kayode, Abayomi Adegbenjo and Tien-Chien Jen |
Abstract: |
Despite the vast opportunities that nanotechnology presents due to its
application in various sectors, these opportunities are still yet to be
maximized in many other areas. Agricultural wastes which ordinarily are
a menace to the environment could be synthesized into nanoparticles and
used to develop cutting fluids. This study highlights the possibility of
the development of such nanofluids from tri-agro wastes. Banana peels,
Coconut shells, and Egg shells were synthesized into nanoparticles using
the centrifugal process and were characterized using a scanning electron
microscope. Nanoparticles were sonicated to ensure homogeneity and mixed
with the base fluid to develop the nanofluids. Performance evaluation on
grinding of stainless steel plates showed that the developed nanofluids
produced a better surface finish of 1.00µm than the conventional cutting
fluid which produced a surface finish of 1.73 µm during the grinding of
galvanized steel. Also, results for mild steel showed a better surface
finish (0.617 µm) when nanofluids were used as against when the
conventional cutting fluid was used (1.857 µm). Hence, the use of
nanofluids developed from tri-agro wastes has not only solved the
problem of environmental pollution but has also proved to be a better
metal working fluid providing improved surface finish during metal
working activities. |
|
|
Full Text |
|
|
Title: |
Comparative study between a neural network
controller and a classic pi applied to an experimental hydraulic system |
Author (s): |
Jhon
Jairo Ramirez Mateus, Francisco Ernesto Moreno Garcia and July Andrea
Gomez Camperos |
Abstract: |
A
large part of the industrial processes, when entering competitiveness,
must be subject to flexibility so that related aspects can be adapted
according to demands at the production level as well as current
technological trends. One strategy to appropriate these processes is to
adopt the use of control techniques such as Artificial Neural Networks
(ANN) inspired by the biological neural networks of the human brain; its
advantage is the ability to provide abstract dynamic features from a
series of experimental data. Under this concept, an ANN controller
system applied to a test hydraulic system was developed, which was
compared with a classic PI strategy. Said comparison at the simulation
level presented satisfactory results, demonstrating the quality and
optimization in the processing, emulation, and control of a physical
system with non-linear characteristics. The performance of the networks
is noteworthy, the Tau response times for both controllers when the
level of the tank decreases are similar, however, the settling time of
the neural network was between 20% and 40% faster than the controller
PI. The presence of overshoot above 20% was identified by the PI control
in response to changes in the setpoint for the size of the tank level. |
|
|
Full Text |
|
|
|
|
|