HABIBULLAH, HIFDZUL LUTHFAN (2020) PENDEKATAN BAYESIAN NETWORK UNTUK PERANCANGAN SISTEM CERDAS DIAGNOSIS KINERJA DAN KERUSAKAN ALAT PENUKAR KALOR TIPE PLAT SH041H-1P-55. S1 thesis, Universitas Mercu Buana Jakarta.
|
Text (HAL COVER)
01 Cover.pdf Download (652kB) | Preview |
|
![[img]](https://repository.mercubuana.ac.id/style/images/fileicons/text.png) |
Text (BAB I)
02 Bab 1.pdf Restricted to Registered users only Download (130kB) |
|
|
Text (BAB II)
03 Bab 2.pdf Restricted to Registered users only Download (500kB) |
|
|
Text (BAB III)
04 Bab 3.pdf Restricted to Registered users only Download (318kB) |
|
|
Text (BAB IV)
05 Bab 4.pdf Restricted to Registered users only Download (790kB) |
|
|
Text (BAB V)
06 Bab 5.pdf Restricted to Registered users only Download (53kB) |
|
|
Text (DAFTAR PUSTAKA)
07 Daftar Pustaka.pdf Restricted to Registered users only Download (56kB) |
|
|
Text (LAMPIRAN)
08 Lampiran.pdf Restricted to Registered users only Download (105kB) |
Abstract
The heat exchanger is very influential in the cooling process and demands maximum performance. Some factors that cause the decline in the performance of the heat exchanger tool are rising pressure drop, decrease in output flow, and leakage in the Plate Heat Exchanger (PHE). This research aims to design a system with the Bayesian Network approach to know the symptoms of damage to the plate heat exchanger SH041H-1P-55. The Bayesian Network is a network that represents the graphics of a knowledge with reasoning of uncertainty. The design of the system uses Microsoft Bayesian Network Editor (MSBNx) and MATLAB software. Starting with specifying variables and related categories in the network, creating causality diagrams, specifying the prior probability variable, populating conditionals probability each variable, and inserting evideence to see the predicted results. After that a case test is performed to display the probabilistic infrention which occurs with the pressure drop case on the heat exchanger. Based on the evidence of the symptoms obtained, the results of the diagnosis Intelligent system concluded that the problem occurs due to the factor of the impurities so that the heat exchanger is blocked and recommend treatment is clean the pipe with a percentage value of 62.1% and clean the plate with a percentage value of 60.15%. Diagnosis test results obtained the appropriate results between the tests conducted manually and by the intelligent system. The result of intelligent system design with the Bayesian Network method approach is expected to make it easier to know the diagnosis of performance and damage that occur and recommend the necessary action against the problematic heat exchanger. Keywords; Heat exchanger, Bayesian Network, software, Microsoft Bayesian Network Editor (MSBNx), pressure drop, maintenance recommendations Alat penukar kalor sangat berpengaruh dalam rangkaian proses pendinginan dan dituntut memiliki kinerja yang maksimal. Beberapa parameter faktor penyebab turunnya kinerja alat penukar kalor adalah naiknya pressure drop, turunnya aliran output, dan kebocoran pada Plate Heat Exchanger (PHE). Penelitian ini bertujuan untuk merancang sebuah sistem dengan pendekatan Bayesian Network guna mengetahui gejala kerusakan pada alat penukar kalor tipe plat SH041H-1P-55. Metode Bayesian Network merupakan sebuah jaringan yang merepresentasikan grafis sebuah pengetahuan dengan penalaran dari sebuah ketidakpastian. Perancangan sistem ini menggunakan software Microsoft Bayesian Network Editor (MSBNx) dan MATLAB. Dimulai dengan menentukan variabel dan kategori yang terkait dalam jaringan, membuat diagram kausalitas, menentukan prior probability variabel, mengisi conditional probability setiap variabel, dan memasukkan evideence untuk melihat hasil prediksi. Setelah itu dilakukan pengujian kasus untuk menampilkan infrensi probabilistik yang terjadi dengan kasus pressure drop pada alat penukar kalor. Berdasarkan bukti gejala yang diperoleh, hasil diagnosis sistem cerdas menyimpulkan bahwa permasalahan terjadi akibat adanya faktor pengotor sehingga alat penukar kalor tersumbat dan merekomendasikan perawatan yaitu bersihkan pipa dengan nilai presentase 62,1% dan bersihkan plat dengan nilai presentase 60,15%. Hasil pengujian diagnosis diperoleh hasil yang sesuai antara tes yang dilakukan secara manual dan oleh sistem cerdas. Hasil dari rancangan sistem cerdas dengan pendekatan metode Bayesian Network diharapkan lebih mempermudah mengetahui diagnosis kinerja dan kerusakan yang terjadi serta merekomendasikan tindakan yang perlu dilakukan terhadap alat penukar kalor yang bermasalah. Kata kunci; Alat penukar kalor, Bayesian Network, software, microsoft bayesian network editor (MSBNx), pressure drop, rekomendasi perawatan
Actions (login required)
 |
View Item |
