Perancangan Sistem Kendali Cerdas Electric Power Steering (EPS)Dengan Software Mathlab Pada Kendaraan Kelas Menengah (Medium Vehicle)

Rizky, Muhammad Syahrial (2025) Perancangan Sistem Kendali Cerdas Electric Power Steering (EPS)Dengan Software Mathlab Pada Kendaraan Kelas Menengah (Medium Vehicle). Other thesis, Institut Teknologi Sepuluh Nopember.

Text 2039211023-Undergraduate_Thesis.pdf - Accepted Version Restricted to Repository staff only Download (4MB) | Request a copy

Abstract

Perkembangan teknologi otomotif pada kendaraan terus mengalami peningkatan dengan tujuan untuk meningkatkan performa dan keselamatan berkendara. Salah satu perkembangannya pada inovasi dibidang power steering dengan adanya teknologi motor listrik atau electric power steering (EPS). Penelitian mengenai EPS sebelumnya telah dilakukan oleh(Rahman,2023),yang mengembangkan sistem EPS pada kendaraan PHEV ITS,melalui pemodelan sistem kontrol berbasis PID-BPNN dan Fuzzy, namun penelitian mengenai EPS pada kendaraan kelas menengah masih terbatas. Untuk itu, dalam proyek akhir ini, dilakukan perancangan sistem kendali cerdas EPS pada kendaraan kelas menengah dengan aplikasi mathlab dengan sistem pengendali menggunakan PID dan pengendali berbasis back propagation neural network.Optimasi
parameter pengendali PID dan PID BPNN dilakukan melalui simulasi simulink mathlab.Simulasi dilakukan dengan menggunakan input sinyal step dan sine input dengan nilai error histogram 0,004385 yang sangat kecil dan terdistribusi secara merata yang menandakan performa pelatihan yang baik, Pada hasil performansi algoritma BPNN berupa grafik respon
torsi mencapai torsi target 5 Nm dengan nilai rise time 0,02 s dan nilai overshoot 11,2% bahwa performa PID-BPNN lebih responsif dan mencapai kestabilan, grafik respon tegangan dengan puncak sekitar stabil 25V dalam dengan nilai rise time 0,01 s dengan nilai overshoot 13,8%,dan grafik respon arus maksimum lebih cepat memberikan bantuan puncak bantuan arus maksimum sekitar 56A dengan nilai overshoot 8,75% yang lebih kecil daripada sistem kendali EPS berbasis PID.
Oleh karena itu, Sistem kendali EPS dengan PID BPNN terbukti lebih responsif dan stabil dalam mengoptimalkan parameter PID secara real time pada kendaraan kelas menengah(medium vehicle).
=====================================================================================================================================
The development of automotive technology in vehicles continues to improve with the aim of enhancing performance and driving safety. One such advancement is in the field of power steering innovation, specifically with the introduction of electric motor technology or electric power steering (EPS). Previous research on EPS has been conducted by (Rahman, 2023), who developed an EPS system for PHEV ITS vehicles using a control system modeling approach based on PID-BPNN and Fuzzy logic. However, research on EPS for mid-range vehicles remains limited. Therefore, in this final project, a smart EPS control system was designed for mid-range vehicles using MATLAB, with a control system employing PID and a backpropagation neural network (BPNN)-based controller. Optimization of PID and PID-BPNN controller parameters was performed through MATLAB Simulink simulations. The simulation was conducted using step and sine signal inputs with a very small error histogram value of 0.004385, which was evenly distributed, indicating good training performance. The performance results of the BPNN algorithm in the form of a torque response graph reached a target torque of 5 Nm with a rise time value of 0.02 s and an overshoot value of 11.2%, indicating that the PID performance -BPNN is more responsive and achieves stability. The voltage response graph shows a stable peak of approximately 25V with a rise time of 0.01 s and an overshoot value of 13.8%. The maximum current response graph provides faster assistance, with a maximum current peak of approximately 56A and an overshoot value of 8.75%, which is smaller than the PID-based EPS control system. Therefore, the EPS control system with PID BPNN has proven to be more responsive and stable in optimizing PID parameters in real time for medium-sized vehicles.

Item Type:	Thesis (Other)
Uncontrolled Keywords:	Pengendali PID, Mathlab, Electric Power Steering, Back Propagation Neural Network, Medium Vehicle, PID Controller, Mathlab, Electric Power Steering, Back Propagation Neural Network, Medium Vehicle
Subjects:	T Technology > TJ Mechanical engineering and machinery > TJ217 Adaptive control systems T Technology > TJ Mechanical engineering and machinery > TJ223 PID controllers
Divisions:	Faculty of Vocational > Mechanical Industrial Engineering (D4)
Depositing User:	Muhammad Syahrial Rizky
Date Deposited:	11 Aug 2025 05:54
Last Modified:	11 Aug 2025 05:54
URI:	http://repository.its.ac.id/id/eprint/128057

Actions (login required)

View Item