Optimisasi Sistem Kontrol Kaskade Logika Fuzzy dan PID pada Kemudi Kendaraan Sistem Steer By Wire menggunakan Quantum Behaved Particle Swarm Optimization

-, Fachrudin (2017) Optimisasi Sistem Kontrol Kaskade Logika Fuzzy dan PID pada Kemudi Kendaraan Sistem Steer By Wire menggunakan Quantum Behaved Particle Swarm Optimization. Doctoral thesis, Institut Teknologi Sepuluh Nopember.

[img]
Preview
Text
2209301010-Disertation.pdf - Published Version

Download (6MB) | Preview

Abstract

Penelitian ini bertujuan untuk optimisasi parameter sistem kontrol pada sistem kemudi kendaraan dengan fully automatic of steer by wire system. Sistem kontrol kemudi dibangun untuk mengendalikan dinamika kendaraan yang disusun secara kaskade yaitu Fuzzy Logic Control (FLC) sebagai kontrol utama untuk menghilangkan kesalahan lateral motion terhadap desired trajectory, serta untuk menentukan setting point pada kontrol berikutnya yaitu Proportional–Integral– Derivative controller (PID) sebagai kontrol penyempurna dari kontrol utama yang berfungsi untuk menekan kesalahan yaw motion. Parameter sistem kontrol baik FLC maupun PID dioptimisasi menggunakan Quantum Behaved Partikel Swarm Optimization (QPSO) sebagai metode optimisasi yang lebih menjamin tercapainya global convergence dari pada menggunakan Partikel Swarm Optimization (PSO) sehingga diperoleh parameter yang optimal untuk menentukan lebar dan posisi dari membership function pada FLC dan konstanta gain dari kontroler PID. Dinamika kendaraan direpresentasikan dalam model kendaraan yang dibangun berdasarkan 10 Degree of Freedom (DOF) dari dinamika kendaraan, terdiri dari 7-DOF vehicle ride model (pergerakan kendaraan; rolling, pitching, bounching dan vertical displacement each wheel) dan 3-DOF vehicle handling model (longitudinal motion, lateral motion dan yaw motion). Pengujian dilakukan menggunakan Software In the Loop Simulations (SILS) dengan input desired trajectory berupa look up table (x – y) yang merepresentasikan manuver kendaraan berliku (double lane change). Hasil pengujian Software In the Loop Simulation (SILS) menggunakan sistem FLC dan PID yang di-tuning dengan QPSO pada kecepatan kendaraan antara 10 – 70 Km/h dapat dipertahankan menurut desired trajectory dengan rata-rata C-RMS error yang lebih kecil dibanding dengan menggunakan sistem FLC dan PID yang di-tuning dengan PSO (FL-PID tuned PSO), yaitu rata-rata lateral motion error = 0,010562 dan rata-rata yaw motion error = 0,065053. Hasil pengujian SILS selanjutnya diverifikasi menggunakan pengujian Hardware In the Loop Simulations (HILS) yang merepresentasikan pergerakan steer pada kendaraan yang sebenarnya. Hasil pengujian HILS menunjukkan bahwa pergerakan kendaraan dapat dipertahankan menurut desired trajectory (double lane change) dengan rata-rata C-RMS error 0,015965 untuk lateral motion dan 0,095685 untuk yaw motion, nilai rata-rata C-RMS error tersebut lebih besar 51.16% untuk lateral motion dan 47.08% untuk yaw motion terhadap hasil pengujian SILS. ====================================================================================== This study to optimize the parameters of the control system on the steering system of a vehicle with a fully automatic steer-by-wire system. Steering control system constructed to control the dynamics of the vehicle are arranged in a cascade consists of Fuzzy Logic Control (FLC) as the main control to eliminate errors lateral motion of the desired trajectory, as well as to determine the setting point in the the next control is Proportional-Integral-Derivative controller (PID) as falsifies control of the main control that serves to suppress the yaw motion errors. Both FLC and PID optimized using Quantum behaved Particle Swarm Optimization (QPSO) as a method of optimization which better ensure the achievement of global convergence than using Particle Swarm Optimization (PSO), thus gained control system parameters that optimal, namely parameter to determine the width and position of the membership function in the FLC and the constant gain of the PID controller. Vehicle dynamics represented in the model vehicle built by 10 Degree of Freedom (DOF) from the dynamics of a vehicle, consisting of a 7-DOF vehicle ride model (the movement of vehicles; rolling, pitching, bounching and vertical displacement of each wheel) and 3-DOF vehicle handling models (longitudinal motion, lateral motion and yaw motion). The test were performed using Software In the Loop Simulations (SILS) with input desred trajectory in the form of look-up tables (x - y), which represents the winding vehicle maneuver (double lane change). The results of Software In the Loop Simulation (SILS) using a system of FLC and PID tuned by QPSO in the vehicle speed between 10-70 km/h can be maintained according to the desired trajectory with an average C-RMS error is smaller compared with the system of FLC and PID tuned by PSO (FL- PID tuned by PSO), ie an average of the lateral motion error = 0.010562 and the average of yaw motion error = 0.065053. Then performed the Hardware In the Loop Simulations (HILS) representing the movement of a steer in the the actual vehicle. The results of HILS shows that the movement of the vehicle can be maintained according to the desired trajectory (double lane change) with an average CRMS error to lateral motion 0.015965 and 0.095685 for the yaw motion, the average value of the C-RMS error 51.16 % greater for lateral motion and 47.08% for the yaw motion of the test results of SILS.

Item Type: Thesis (Doctoral)
Uncontrolled Keywords: Fuzzy Logic; PID; QPSO; Lateral motion; Yaw motion
Subjects: Q Science > QA Mathematics > QA248_Fuzzy Sets
T Technology > TK Electrical engineering. Electronics Nuclear engineering
Divisions: Faculty of Industrial Technology > Electrical Engineering > (S3) PhD Theses
Depositing User: Mrs Anis Wulandari
Date Deposited: 15 May 2017 07:47
Last Modified: 15 May 2017 07:47
URI: http://repository.its.ac.id/id/eprint/41266

Actions (login required)

View Item View Item