Susanto, Adi (2021) Optimasi Kinerja dan Kestabilan Unmanned Aerial Vehicle Melalui Pengaturan Parameter Dihedral dan Tip-twist Sayap dengan Menerapkan Metode Artificial Neural Network-Genetic Algorithm. Masters thesis, Institut Teknologi Sepuluh Nopember.
![[thumbnail of thesis-AdiSusanto-v4_edit.pdf]](http://repository.its.ac.id/style/images/fileicons/text.png) |
Text
thesis-AdiSusanto-v4_edit.pdf - Accepted Version Restricted to Repository staff only until 1 October 2023. Download (9MB) | Request a copy |
Abstract
Teknologi unmanned aerial vehicle atau AUV telah menjadi kebutuhan dalam
berbagai elemen masyarakat, mulai dari hobi hingga memenuhi kebutuhan
pengintaian militer. UAV memiliki berbagai jenis, dimana salah satunya berupa
small UAV yang terbang pada tinggi jelajah hingga 300 m dan beroperasi pada
kecepatan 100 knot. Sayap merupakan bagian utama yang memungkinkan sebuah
UAV terbang. Sayap yang baik dapat menentukan kinerja aerodinamis dan
keseimbangan UAV. Penulisan tesis ini bertujuan untuk mencari pengaruh sudut
dihedral (Γ) dan tip-twist (�) terhadap kinerja sayap dan kestabilan UAV,
menentukan besar sudut dihedral dan tip-twist yang menghasilkan kinerja sayap
dan kestabilan UAV optimal, serta membandingkan karakteristik aliran fluida
antara UAV dengan sayap baseline dan optimal.
Data awal yang diambil berupa informasi berat, letak CoG tiap komponen,
dilengkapi dengan pemilihan airfoil berperforma tinggi dijadikan dasar simulasi
XFLR5 sebagai metode yang dipilih sebagai CFD solver, dengan memvariasikan
dihedral dan tip-twist sehingga didapatkan sekelompok data yang nantinya
dijadikan sebagai populasi awal dalam proses optimasi menggunakan Genetic
algorithms. Metode back-propagation network dipilih untuk mencari hubungan
antara variabel yang diatur dan respon sehingga didapatkan fitness function untuk
membantu GA menghemat waktu komputasi. Sayap dengan geometri optimal
dibandingkan kinerja aerodinamisnya terhadap UAV baseline.
Hasil ANOVA menunjukkan bahwa kedua faktor berpengaruh signifikan terhadap
respon kinerja UAV antara lain koefisien lift dan drag maksimum (max���/� ),
koefisien drag pada sudut serang 0 derajat (� | ��), juga terhadap kestabilan UAV
yang meliputi kestabilan statis (�|����), kestabilan dinamis Short period (��),
kestabilan Roll damping (��), dan kestabilan Dutch roll mode (��). pada tingkat
kepercayaan 95%. Hubungan faktor terhadap respon dicari menggunakan ANN
dengan arsitektur feed-forward dan algoritma backpropagation. Hubungan antara
faktor dan respon koefisien lift dan drag maksimum diwakili oleh ANN dengan
struktur 2-5-7-1. Hubungan antara faktor dan respon koefisien drag pada sudut
serang nol derajat diwakili oleh ANN dengan struktur 2-4-9-1. Hubungan antara
faktor dan respon nilai sudut serang saat pitching moment bernilai nol diwakili oleh
ANN dengan struktur 2-8-3-1. Hubungan antara faktor dan respon Short Period
diwakili oleh ANN dengan struktur 2-9-3-1. Hubungan antara faktor dan respon
Roll Damping diwakili oleh ANN dengan struktur 2-4-5-1. Hubungan antara faktor
dan respon Dutch Mode diwakili oleh ANN dengan struktur 2-6-5-1. Proses
optimasi menggunakan GA menghasilkan individu optimal dengan nilai integer
dihedral 5° dan tip-twist −2°. Simulasi konfirmasi dilakukan untuk memastikan
hasil optimasi tidak berbeda dari data simulasi, bila nilai error tertinggi terjadi pada
respon � | �� sebesar 0.309677%, nilai error yang lebih kecil dari 5%
menunjukkan proses optimasi berjalan dengan baik. UAV baseline memiliki
perbandingan koefisien lift terhadap drag yang lebih tinggi dibandingkan dengan
UAV dengan sayap optimal. Kontur tekanan dan wall shear stress pada wing-tip
menyebabkan terjadinya tip-vortex pada sayap.
=====================================================================================================
Unmanned aerial vehicle or AUV technology has become a necessity in various
elements of society, ranging from hobbies to meeting the needs of military
reconnaissance. UAV has various types, one of which is a small UAV that flies at
a cruising height of up to 300 m and operates at a speed of 100 knots. Wings are the
main part that allows a UAV to fly. A good wing can determine the aerodynamic
performance and balance of the UAV. The purpose of writing this thesis is to find
out the effect of dihedral angle (Γ) and tip-twist (ϕ) on wing performance and UAV
stability, determine dihedral and tip-twist angles that produce optimal wing
performance and UAV stability, and compare fluid flow characteristics between
UAVs. with baseline and optimal wings.
The initial data taken in the form of weight information, the location of the CoG of
each component, equipped with the selection of a high-performance airfoil as the
basis for the XFLR5 simulation as the method chosen as the CFD solver, by varying
the dihedral and tip-twist so that a group of data is obtained which will be used as
the initial population in the process. optimization using Genetic algorithms. The
back-propagation network method was chosen to find the relationship between the
regulated variable and the response in order to obtain a fitness function to help GA
save computational time. Wing with optimal geometry compared to its aerodynamic
performance against the baseline UAV.
The ANOVA results show that the two factors have a significant effect on the UAV
performance response, including the maximum lift and drag coefficient, drag
coefficient at 0 degree angle of attack, also on the stability of the UAV which
includes static stability, Short period dynamic stability, Roll damping stability , and
Dutch roll mode, at the 95% confidence level. Factor relationship to response is
searched using ANN with feed-forward architecture and backpropagation
algorithm. The relationship between factor and response coefficient of maximum
lift and drag is represented by ANN with a 2-5-7-1 structure. The relationship
between factor and drag coefficient response at zero degree angle of attack is
represented by ANN with a 2-4-9-1 structure. The relationship between the factor
and the response of the angle of attack when the pitching moment is zero is
represented by ANN with a 2-8-3-1 structure. The relationship between factors and
Short Period responses is represented by ANN with a 2-9-3-1 structure. The
relationship between factors and the response of Roll Damping is represented by
ANN with a 2-4-5-1 structure. The relationship between factors and Dutch Mode
responses is represented by ANN with a 2-6-5-1 structure. The optimization process
using GA produces optimal individuals with a dihedral integer value of 5° and a
tip-twist of -2°. Confirmation simulation is carried out to ensure that the
optimization results do not differ from the simulation data, if the highest error value
occurs in the response drag at zero angle of attack of 0.309677%, an error value of
less than 5% indicates the optimization process is running well. The baseline UAV
has a higher lift to drag coefficient ratio than the UAV with the optimal wing. The
pressure contour and wall shear stress on the wing tips cause tip-vortex on the
wings.
| Item Type: | Thesis (Masters) |
|---|---|
| Uncontrolled Keywords: | sUAV; sayap; kinerja; kestabilan; Back-propagation Neural Network; Genetic Algorithm |
| Subjects: | Q Science > QA Mathematics > QA402.5 Genetic algorithms. Q Science > QA Mathematics > QA76.87 Neural networks (Computer Science) T Technology > TJ Mechanical engineering and machinery U Military Science > U Military Science (General) > UG Military Engineering > UG1242.D7 Unmanned aerial vehicles. Drone aircraft |
| Divisions: | Faculty of Industrial Technology and Systems Engineering (INDSYS) > Mechanical Engineering > 21101-(S2) Master Thesis |
| Depositing User: | Adi Susanto |
| Date Deposited: | 04 Sep 2021 07:09 |
| Last Modified: | 04 Sep 2021 07:09 |
| URI: | http://repository.its.ac.id/id/eprint/91576 |
Actions (login required)
 |
View Item |