#TITLE_ALTERNATIVE#

Advancements in sensor and computer technologies are making it feasible to install sensors and small powerful computers on complex equipment to monitor the general <br /> <br /> <br /> condition or state of that equipment. Methods for analyzing system status and health and for pr...

Full description

Saved in:
Bibliographic Details
Main Author: DANIAL (NIM : 23104004); Pembimbing : Dr. Arief Hariyanto; Prof. Wiranto Arismunandar, DENI
Format: Theses
Language:Indonesia
Online Access:https://digilib.itb.ac.id/gdl/view/15326
Tags: Add Tag
No Tags, Be the first to tag this record!
Institution: Institut Teknologi Bandung
Language: Indonesia
Description
Summary:Advancements in sensor and computer technologies are making it feasible to install sensors and small powerful computers on complex equipment to monitor the general <br /> <br /> <br /> condition or state of that equipment. Methods for analyzing system status and health and for predicting system life expectancy need to be more powerful, insightful, reliable, and robust for data collected onboard systems in real time. <br /> <br /> <br /> This thesis reports on analysis of equipment operation data using prognostic and diagnostic methods as a tool of maintenance activity to improve equipment operation <br /> <br /> <br /> efficiency. Even with relatively new equipment, there is tremendous benefit of extending the time between overhauls and maintenance, reducing the probability of failure in the <br /> <br /> <br /> field, and increasing appropriate preventive repair. Other contributing factor to equipment operation efficiency is the reduction of dependency to maintenance expert. <br /> <br /> <br /> There are three methods of equipment operation data analysis covered in this thesis, each with their own characteristics and results. First method is a prognostic method for smooth slope data using LEAP-Frog algorithm to get characteristic curve of operation parameters. Second method is prognostic method to evaluate remaining useful life of equipment through neural network analysis of operation parameters with short period of normal to failure time. Third method employ neural network algorithm to perform root cause analysis of equipment through parameter classification and Boolean operation. <br /> <br /> <br /> Result of equipment operation data analysis using these methods is satisfactory as a tool for “just-in-time” maintenance implementation. More improvement and tuning can be performed to see the performance of the three methods on more comprehensive and complicated data.