통계적 분석기법을 이용한 선박디젤기관의 고장진단 및 예측시스템에 관한 연구

Abstract

Recently acceleration of ship automation makes ship's crew decreased and ship's schedule fast. That results in making harbour time short and lack of ship's maintenance time. Therefore prediction maintenance using fault diagnosis system is more important to prevent accidents by shortage of maintenance. While monitoring points for ship machineries were about 600 in 1980's ship, but now over 10,000 points. So almost all systems of ship can be monitored by central control and monitoring system. With this background and circumstance, various kinds of study for fault diagnosis of machineries are carried out.

Almost ship monitoring systems are event driven alarm system which warn only when the signal is over or under set point. These kinds of system cannot warn while signal is growing to abnormal state until the signal is over or under the set point and cannot play a role for preventive maintenance system. And fault diagnosis is started by expert engineer after warning from the monitoring system. There is few study which automatically diagnose the fault from ship's monitored signal. The bigger control and monitoring system is, the more important fault diagnosis and predictive maintenance is to reduce damage brought forth by system fault.

This paper proposes fault diagnosis and prediction system which is able to diagnose and forecast the fault from present operating condition by analyzing monitored signals with present ship monitoring system without additional sensors. For this all kinds of ship's engine room monitored data are classified with combustion subsystem, heat exchange subsystem and electric motor and pump subsystem by analyzing ship's operation data. To extract dynamic characteristics of these subsystems, log book data of container ship of H shipping company are used. Even though almost all machineries installed on the ship including main propulsion diesel engine and various auxiliary machineries have non linear characteristics and produce different output data dependent on the operating environment, if those machineries are operating under normal condition state, correlation coefficient(CC) between monitored data of related machine each other will be high. From analyzing this data having high CC, correlation level of interactive data can be understood. Knowledge base of abnormal detection can be built by referring level of CC(Fault Detection CC, FDCC) to detect abnormal data among monitored data from monitoring system and knowledge base of diagnosis built by referring CC among interactive data for related machine each other to diagnose fault part. Fault place can be ascertained by investigating specific data of fault part with decision making tree like answer tree. And fault prediction can be made by regression analysis of monitored data.

To verifying capability of fault detection, diagnosis and prediction, Fault Management System(FMS) is developed by C++. Simulation experiment by FMS is carried out with population data set made by log book data of 2 months duration from a large full container ship of H shipping company. For fault detecting and diagnosing experiment from population data set, three kinds of random number are generated by computer. One is generated on the base of average and covariance of population data set, other on the base of parts of population data and the other on the base of fault detection range(FDR). In the simulation experiment FMS is ascertained to detect abnormal data from monitored data set including generated random number by abnormal detection module with abnormal detection knowledge base and diagnose the fault by abnormal diagnosis module with abnormal diagnosis knowledge base and also forecast predictive fault by fault prediction module.

If the FMS is developed to include maintenance manual and ship's inventory database inside near future, the system will be able to recommend how to maintain the diagnosed fault and necessary spare parts.