Dewani, Niken Dian Rahma (2020) REAL-TIME RELIABILITY PREDICTION PADA MOTOR DC MENGGUNAKAN PARTICLE FILTERING. Undergraduate thesis, Institut Teknologi Sepuluh Nopember.
Preview |
Text
02311640000054-Undergraduate_Thesis.pdf Download (4MB) | Preview |
Abstract
Keamanan dan kehandalan operasi sistem pada suatu plant sangat berpengaruh terhadap kinerja dan keselamatan proses, sehingga diperlukan suatu cara untuk dapat mengetahui kehandalan dari suatu plant. Salah satu instrumen yang paling sering digunakan di industri adalah motor listrik, dan salah satu jenis dari motor listrik ini adalah motor DC. Penurunan performansi pada motor DC dapat menyebabkan penurunan keandalannya, sehingga diperlukan prediksi reliabilitas untuk motor DC. Tujuan dari Tugas Akhir ini adalah untuk dapat merancang sistem monitoring keandalan pada motor DC dengan algoritma particle filtering dan mengetahui tingkat ketelitian dan performansi sistem monitoring keandalan pada motor DC dengan algoritma particle filtering. Tahapan-tahapan yang dilakukan antara lain perancangan pemodelan sistem kontrol motor DC dengan kesalahan sensor, perancangan algoritma particle filtering untuk mengestimasi kesalahan, perancangan algoritma exponential smoothing, serta perancangan sistem prediksi keandalan. Pengujian dilakukan dengan parameter Holt a=0.1 dan b=0.1 dengan batas atas kecepatan motor sebesar 2,3 m/s. Pada uji algoritma prediksi keandalan dengan 3 jenis step, didapatkan hasil prediksi keandalan dengan sepuluh step memiliki ketelitian paling baik dan ketiga step mencapai R=0 pada detik ke 13.606 detik, lebih lambat 6 detik dari keadaan sebenarnya yaitu pada detik ke-13.600,08.
========================================================================================================================
Security and reliability of a system operation at a plant is very influential on the performance and safety of the process, so we need to find a way to be able to determine the reliability of a plant. One of the most commonly used instrument in the industry is electric motor, and one of them is DC motor. A decrease in the performance of the DC motor can cause the decrease in its reliability, so a reliability prediction system is needed for a DC motor. The purpose of this Final Project is to be able to design a reliability monitoring system on a DC motor with a particle filtering algorithm and determine the level of accuracy and performance of a reliability monitoring system on a DC motor with a particle filtering algorithm. The steps taken include designing a DC motor control system with sensor error, designing a particle filtering algorithm to estimate errors, designing an exponential smoothing algorithm, and designing a predictive reliability system. The test was carried out with Holt’s parameters a = 0.1 and b = 0.1 with the upper limit of the motor speed of 2.3 m/s. From the reliability prediction algorithm test with 3 types of steps, the results obtained from the reliability predictions system are that the ten steps prediction has the best performance from all the steps predictions and the 3 types of prediction steps reach R=0 at 13.606 seconds, 6 seconds slower than the the real condition that reach the same point at 13.600,08 seconds.
| Item Type: | Thesis (Undergraduate) |
|---|---|
| Uncontrolled Keywords: | Particle filtering, exponential smoothing, motor DC, prediksi keandalan, reliabilitas |
| Subjects: | T Technology > T Technology (General) T Technology > T Technology (General) > T57.8 Nonlinear programming. Support vector machine. Wavelets. Hidden Markov models. T Technology > TS Manufactures > TS173 Reliability of industrial products T Technology > TS Manufactures > TS174 Maintainability (Engineering) . Reliability (Engineering) |
| Divisions: | Faculty of Industrial Technology and Systems Engineering (INDSYS) > Physics Engineering > 30201-(S1) Undergraduate Thesis |
| Depositing User: | Niken Dian Rahma Dewani |
| Date Deposited: | 04 Aug 2020 01:58 |
| Last Modified: | 21 May 2023 15:17 |
| URI: | http://repository.its.ac.id/id/eprint/76846 |
Actions (login required)
 |
View Item |