Pendekatan Markov-Switching Autoregressive Model Untuk Monitoring Tekanan Pada Sistem Hidrolik

Sufianoor, Muhammad Febrio Putra (2022) Pendekatan Markov-Switching Autoregressive Model Untuk Monitoring Tekanan Pada Sistem Hidrolik. Other thesis, Institut Teknologi Sepuluh Nopember.

Text 06211840000034-Undergraduate_Thesis.pdf Restricted to Repository staff only Download (2MB)

Abstract

Hidrolik adalah sebuah sistem yang menghasilkan energi mekanis dengan cara kerja berupa pemindahan daya menggunakan zat pengantar berupa zat cair. Sistem kerja hidrolik memiliki beberapa fitur yang dipantau untuk mengetahui kinerja sistem tersebut. Fitur tersebut diantaranya adalah tekanan, tenaga motor, aliran volume, temperatur, getaran, efisiensi pendingin, tenaga pendingin, dan faktor efisiensi. Sistem secara berulang melakukan siklus beban konstan selama 60 detik dan mengukur nilai proses seperti tekanan, volume aliran, dan temperatur sementara kondisi 4 komponen hidrolik (cooler, valve, pump, accumulator) bervariasi secara kuantitatif. Pada pengamatannya, didapatkan bahwa fitur tekanan sistem hidrolik terdeteksi oleh sensor memiliki pergerakan nilai yang cukup besar sehingga menarik untuk diamati model siklus nya dari waktu ke waktu. Berdasarkan deskripsi tersebut, diketahui bahwa data pengamatan berbasis time-series. Maka perlu digunakan metode yang mendukung pemodelan monitoring berbasis time-series. Oleh karena itu, penelitian ini bertujuan untuk memodelkan hasil monitoring fitur tekanan pada sistem hidrolik melalui pendekatan Markov-Switching Autoregressive Model. Pemodelan dengan Markov-Switching Autoregressive menghasilkan 2 regime AR(2) sebagai model yang terbaik mengacu pada perbandingan nilai AIC. Dalam pemodelannya, 2 regime ini mengindikasikan antara pergerakan nilai tekanan sistem hidrolik yang kecil dan pergerakan tekanan sistem hidrolik yang besar dari suatu waktu proses. Pada model tersebut, didapatkan nilai AIC sebesar 18433,04 sebagai nilai AIC minimum dibandingkan model lain dengan nilai RMSE sebesar 0,081 atau sebesar 8,1 persen.
==============================================================================================================================
Hydraulics is a system that produces mechanical energy by means of work in the form of transferring power using a conveying substance in the form of a liquid. The hydraulic working system has several features that are monitored to determine the performance of the system. These features include pressure, motor power, volume flow, temperature, vibration, cooling efficiency, cooling power, and efficiency factor. The system repeatedly cycles a constant load for 60 seconds and measures process values such as pressure, flow volume, and temperature while the condition of the 4 hydraulic components (cooler, valve, pump, accumulator) varies quantitatively. In his observations, it was found that the hydraulic system features such as pressure detected by the sensor have a large enough movement value so it is interesting to observe the cycle model from time to time. Based on the description, it is known that the observation data is based on time series. So it is necessary to use a method that supports time-series-based monitoring modeling. Therefore, this study aims to model the results of monitoring the pressure features in the hydraulic system through the Markov-Switching Autoregressive Model approach. The MSAR model produces 2 AR(2) regimes as the best model referring to the comparison of AIC values, where these 2 regimes indicate the small value movement and large value movement of hydraulic system pressure over a period of time. In this model, the AIC value of-18433,04 is obtained as the minimum AIC value compared to other models with an RMSE value of 0,081 or 8,1 percent.

Item Type:	Thesis (Other)
Additional Information:	RSSt 519.536 Suf p-1 2022
Uncontrolled Keywords:	Autoregressive. Markov Switching. Tekanan Hidrolik. Autoregressive. Hydraulic Pressure. Markov Switching.
Subjects:	H Social Sciences > HA Statistics
Divisions:	Faculty of Science and Data Analytics (SCIENTICS) > Statistics > 49201-(S1) Undergraduate Thesis
Depositing User:	Mr. Marsudiyana -
Date Deposited:	10 Jun 2026 04:45
Last Modified:	10 Jun 2026 04:45
URI:	http://repository.its.ac.id/id/eprint/133692

Actions (login required)

View Item