Pemodelan dan Pengendalian Sistem Inertia Wheel Pendulum

Hidayati, Alfiana Nur (2021) Pemodelan dan Pengendalian Sistem Inertia Wheel Pendulum. Masters thesis, Institut Teknologi Sepuluh Nopember.

Text 02111950080008-Master_Thesis.pdf - Accepted Version Restricted to Repository staff only until 1 October 2023. Download (4MB) | Request a copy

Abstract

Perkembangan teknologi self-balancing dalam dunia otomotif mulai terlihat dimana mulai dikembangkan kendaraan roda dua menggunakan metode two-wheeled balancing. Selain digunakan pada motor listrik, self-balancing juga dikembangkan sebagai penstabil pada transportasi portable seperti Segway. Segway memiliki persamaan dinamika dengan permasalahan kontrol klasik, yaitu inverted pendulum (IP). Inverted pendulum merupakan sistem pendulum yang memiliki kestabilan pada posisi terbalik dimana pendulum akan berada pada posisi tegak. Pada prinsipnya, sistem IP merupakan sistem multivariabel yang nonlinear dan tidak setimbang. Hal tersebut menjadikan sistem ini banyak dipakai sebagai platform suatu penelitian maupun sebagai verifikator berbagai teori pengendalian. Pendulum dapat dipertahankan tetap pada posisi tegak dalam keadaan stabil dengan menerapkan sistem kendali. Sistem kendali tersebut akan memantau besarnya sudut pendulum dan membantu membawanya kembali ke posisi tegak.

Ada beberapa sistem pengendalian IP, diantaranya yaitu dengan menggunakan kereta (cart), stabilisasi gyroscopic, dan inertia wheel. Pada penelitian ini digunakan sistem inertia wheel karena tidak membutuhkan ruang besar seperti penggunaan sistem IP dengan cart. Selain itu IW memiliki model yang lebih sederhana jika dibandingkan dengan gyroscope dan hanya membutuhkan satu inputan, yaitu input dari putaran flywheel. Tujuan dari penelitian ini yaitu mendesain sistem kendali untuk menyeimbangkan IP agar stabil dalam posisi tegak dengan menggunakan roda reaksi. Pada penelitian ini telah dirancang sistem inertia wheel pendulum, dimana langkah awal yang dilakukan yaitu penurunan persamaan gerak sistem dengan metode Lagrange. Hasil penurunan persamaan gerak digunakan untuk mendesain sistem kendali. Sistem kendali yang digunakan yaitu LQR, PID dan kombinasi antara PID dan LQR. Sistem kendali dites pada model nonlinear yang dibuat dengan menggunakan Sismscape Multibody.

Hasil analisis menunjukkan bahwa sistem kendali LQR, PID, dan LQR-PID mampu mempertahankan IP dalam posisi tegak dengan asumsi tidak ada saturasi. Sistem kendali LQR-PID menghasilkan respons paling bagus dimana overshoot yang dihasilkan lebih kecil dibandingkan dengan PID dan lebih cepat kembali ke posisi ekuilibrium dibandingkan dengan LQR. Selain itu, analisis juga dilakukan pada kasus input dengan saturasi dimana terdapat batasan voltase dari aktuator maksimal 24V. Hasil respons menunjukkan bahwa ketiga sistem kendali mampu membawa sistem IWP kembali ke posisi ekuilibrium untuk sudut awal pendulum maksimal 14,5 derajat.
=====================================================================================================
Self-balancing technology in the automotive world is growing where two-wheeled vehicles began to be developed. Self-balancing has been used in electric motors and also developed as a stabilizer in portable transportation such as the Segway. The Segway has a dynamic equation with the classical control problem, namely the inverted pendulum (IP). An inverted pendulum is a pendulum system that has stability in an inverted position where the pendulum will be in an upright position. In principle, the IP system is a nonlinear and unbalanced multivariable system. This system is widely used as a research platform as well as a verifier of various control theories. The pendulum can be maintained in an upright position in a stable state by applying a control system. The control system will monitor the magnitude of the pendulum angle and bring it back to an upright position.

There are several IP systems used, including using a cart (cart), gyroscopic stabilization, and inertia wheel (IW). In this study, the IW system is used because it does not require large space like an IP system with a cart. In addition, IW has a simpler model when compared to a gyroscope and only requires one input, namely the input of the flywheel rotation. The purpose of this study is to design a control system to balance the IP. So that, it is always stable in an upright position using a reaction wheel. In this research, an inertia wheel pendulum system has been designed where the first step is to derive the equation of motion of the system by using the Lagrange method. The result of the derivation of the equation of motion is used to design the control system. The control system used is LQR, PID, and a combination of PID and LQR. The control system is tested on a nonlinear model that has been made using Sismscape Multibody.

The results of the analysis show that the LQR, PID, and LQR-PID control systems are able to maintain IP in an upright position with the assumption there is no saturation. The LQR-PID control system produces the best response where the resulting overshoot is smaller than the PID and returns to the upright position faster than the LQR. In addition, the analysis is also carried out in the case of input with saturation where the maximum voltage f the actuator is 24V. The response results show that the three control systems can bring the IWP system back to the upright position for a maximum pendulum initial angle of 14.5 degrees. The best response is produced by the LQR-PID control system, where the resulting overshoot is smaller than the PID and can return to an upright position faster than the LQR.

Item Type:	Thesis (Masters)
Uncontrolled Keywords:	Inertia wheel pendulum, Self-balancing, Saturasi, LQR, PID, Inertia wheel pendulum, Self-balancing, Saturation, LQR, PID.
Subjects:	T Technology > TJ Mechanical engineering and machinery T Technology > TJ Mechanical engineering and machinery > TJ223 PID controllers
Divisions:	Faculty of Industrial Technology and Systems Engineering (INDSYS) > Mechanical Engineering > 21101-(S2) Master Thesis
Depositing User:	Alfiana Nur Hidayati
Date Deposited:	24 Aug 2021 08:11
Last Modified:	24 Aug 2021 08:11
URI:	http://repository.its.ac.id/id/eprint/89235

Actions (login required)

View Item