Implementasi Kalman Filter Pada Pengukuran Massa Secara Dinamis Dalam Proses Pengecekan Massa Produk Pengantongan Urea

Lee, Valiant Tornado (2025) Implementasi Kalman Filter Pada Pengukuran Massa Secara Dinamis Dalam Proses Pengecekan Massa Produk Pengantongan Urea. Other thesis, Institut Teknologi Sepuluh Nopember.

Text 2040211022-Undergraduate_Thesis.pdf - Accepted Version Restricted to Repository staff only Download (3MB) | Request a copy

Abstract

Departemen Pergudangan pada sebuah perusahaan petrokimia bertanggung jawab atas proses pengantongan produk pupuk. Proses pengantongan pupuk terdiri dari proses pengisian atau filling pupuk ke dalam kantong, proses penjahitan kantong pupuk dengan mesin jahit, proses bagcoding pada kantong pupuk, proses pengecekan massa produk, dan proses palletizing. Pada proses pengecekan massa produk, perusahaan masih menerapkan metode manual berupa sampling produk tiap satu jam sekali pada 2 produk. Hal ini menyebabkan massa dari tiap produk tidak diketahui, dengan begitu kualitas menjadi tidak terjamin. Sebagai solusi, diusulkan sebuah sistem pengecekan massa otomatis berbasis checkweigher, yang dapat mengukur massa produk saat bergerak di atas konveyor. Namun, karena pengukuran dilakukan pada objek yang sedang bergerak, sistem ini rentan terhadap gangguan noise yang muncul akibat faktor-faktor seperti getaran, variasi kecepatan, dan ketidakstabilan permukaan konveyor. Gangguan-gangguan tersebut dapat menyebabkan fluktuasi yang besar dalam hasil pengukuran, yang dapat menurunkan akurasi dan kestabilan sistem. Oleh karena itu, diperlukan penerapan mekanisme filter untuk mengurangi dampak gangguan dan memastikan akurasi pengukuran massa yang lebih baik.Mekanisme filter yang digunakan adalah Kalman Filter yang berfungsi untuk mengurangi noise pada pengukuran massa produk secara dinamis. Prototipe yang dirancang tersusun atas ESP32, Sensor load cell, sensor infrared, dan motor DC. Selain itu sistem juga diintegrasikan dengan website sebagai media monitoring. Sistem checkweigher yang dikembangkan sanggup menimbang setiap kantong secara dinamis hanya dalam 0,96 detik per produk. Pengukuran massa secara dinamis tanpa filter menunjukkan estimasi massa cenderung sedikit overestimasi dengan error rata-rata 2,39% atau 126 gram dengan puncak 4,19 %. Penerapan Kalman Filter (Q = 1, R = 147) terbukti secara efektif menekan error hingga 0,08% atau 4,08 gram. Dari 26 pengujian, setengah diantaranya menghasilkan error kurang dari 1%. Namun terdapat nilai error tertinggi sebesar 3,16 % atau sebesar 163,63 gram.
=========================================================================================================================================
The Warehouse Department at a petrochemical company is responsible for the product bagging process of fertilizer. This process consists of filling the fertilizer into bags, stitching the bags with a sewing machine, applying bag coding, checking the product's mass, and palletizing. For the mass checking process, the company still implements a manual method of sampling two products once every hour. This causes the mass of each individual product to be unknown, thus the quality is not guaranteed. As a solution, an automated mass checking system based on a checkweigher is proposed, which can measure the mass of products as they move on a conveyor. However, because the measurement is performed on a moving object, this system is susceptible to noise disturbances that arise from factors such as vibrations, speed variations, and conveyor surface instability. These disturbances can cause large fluctuations in the measurement results, which can decrease the system's accuracy and stability. Therefore, the implementation of a filter mechanism is required to reduce the impact of these disturbances and ensure better mass measurement accuracy. The filter mechanism used is the Kalman Filter, which functions to dynamically reduce noise in the product mass measurement. The designed prototype consists of an ESP32, a load cell sensor, an infrared sensor, and a DC motor. Additionally, the system is integrated with a website as a monitoring medium. The developed checkweigher system is capable of dynamically weighing each bag in just 0.96 seconds per product. Dynamic mass measurement without a filter shows that the mass estimation tends to be slightly overestimated, with an average error of 2.39% or 126 grams, and a peak of 4.19%. The application of the Kalman Filter (Q = 1, R = 147) is proven to effectively reduce the error down to 0.08% or 4.08 grams. Out of 26 tests, half resulted in an error of less than 1%. However, there was a highest recorded error value of 3.16%, or 163.63 grams.

Item Type:	Thesis (Other)
Uncontrolled Keywords:	Checkweigher, Kalman Filter, Noise, Pengukuran Massa, Checkweigher, Kalman Filter, Noise, Mass Measurement.
Subjects:	T Technology > TK Electrical engineering. Electronics Nuclear engineering > TK3070 Automatic control T Technology > TK Electrical engineering. Electronics Nuclear engineering > TK351 Electric measurements. T Technology > TK Electrical engineering. Electronics Nuclear engineering > TK5102.9 Signal processing. T Technology > TK Electrical engineering. Electronics Nuclear engineering > TK7867.5 Noise
Divisions:	Faculty of Vocational > 36304-Automation Electronic Engineering
Depositing User:	Valiant Tornado Lee
Date Deposited:	06 Aug 2025 04:02
Last Modified:	06 Aug 2025 04:02
URI:	http://repository.its.ac.id/id/eprint/127714

Actions (login required)

View Item