• Proceedings of the Korea Society for Simulation Conference
• /
• 2003.11a
• /
• pp.51-57
• /
• 2003

In operating distributed systems, proactive management is one of the major concerns for better quality of service and future capacity planning. In order to handle this management problem effectively, it is necessary to analyze performances of the distributed system and events generated by components in the system. This paper provides a rule-based event parsing engine for proactive management. Our event parsing engine uses object hooking-based and event-token approaches. The object hooking-based approach prepares new conditions and actions in Java classes and allows dynamically exchange them as hook objects in run time. The event-token approach allows the event parsing engine consider a proper sequence and relationship among events as an event token to trigger an action. We analyze the performance of our event parsing engine with two different implementations of rule structure; one is table-based and the other is tree-based.

• Journal of the Korea Institute of Information and Communication Engineering
• /
• v.14 no.6
• /
• pp.1429-1434
• /
• 2010

Generally, the simulator for mobile device tests automatically an application software after instilling again the event, which is saved in log file according to an event generation sequence, into the application software of device. But, the simulator performance for mobile device can be different according to a extraction layer even if the events are same. And, the sequence of events extracted from an application is changeable in the environment that multiple applications are operated concurrently. Therefore, even though the same applications is executed to the same sequence, the generation sequence of events is revised in accordance with the state of mobile device system, and whether the errors occur according to circumstances or not. This kind of application software error is very difficult to perform a debugging operation. In this paper, the execution state of various applications is verifiable through the re-editing of events after analyzing the events which is generated in application, kernel, middleware layer, and the event arrange/editor is designed and implemented to understand efficiently the influence on application, kernel, and middleware layer for events.

• Korean Journal of Computational Design and Engineering
• /
• v.17 no.2
• /
• pp.97-110
• /
• 2012

High performance sensors and modern data logging technology with real-time telemetry facilitate system fault diagnosis in a very precise manner. Fault detection, isolation and identification in fault diagnosis systems are typical steps to analyze the root cause of failures. This systematic failure analysis provides not only useful clues to rectify the abnormal behaviors of a system, but also key information to redesign the current system for retrofit. The main barriers to effective failure analysis are: (i) the gathered data (event) logs are too large in general, and further (ii) they usually contain noise and redundant data that make precise analysis difficult. This paper therefore applies suitable pre-processing techniques to data reduction and feature extraction, and then converts the reduced data log into a new format of event sequence information. Finally the event sequence information is decoded to investigate the correlation between specific event patterns and various system faults. The efficiency of the developed pre-filtering procedure is examined with a terminal box data log of a marine diesel engine.

• Nuclear Engineering and Technology
• /
• v.49 no.2
• /
• pp.306-312
• /
• 2017

The purpose of this research is to introduce the technical standard of accident sequence precursor (ASP) analysis, and to propose a case study using the dynamic-probabilistic safety assessment (D-PSA) approach. The D-PSA approach can aid in the determination of high-risk/low-frequency accident scenarios from all potential scenarios. It can also be used to investigate the dynamic interaction between the physical state and the actions of the operator in an accident situation for risk quantification. This approach lends significant potential for safety analysis. Furthermore, the D-PSA approach provides a more realistic risk assessment by minimizing assumptions used in the conventional PSA model so-called the static-PSA model, which are relatively static in comparison. We performed risk quantification of a steam generator tube rupture (SGTR) accident using the dynamic event tree (DET) methodology, which is the most widely used methodology in D-PSA. The risk quantification results of D-PSA and S-PSA are compared and evaluated. Suggestions and recommendations for using D-PSA are described in order to provide a technical perspective.

• Journal of the Korean Society of Safety
• /
• v.31 no.3
• /
• pp.34-41
• /
• 2016

The purpose of this paper is to conduct ETA on six items of ESS: the whole system, battery, BMS, PCS, ESS and cable. To achieve that, ESS work flow and its components are categorized. Based on performance, human, environmental, management, and safety, this paper drew initiation events (IE) and end states (ES). ETA is applied to the main functions of each item, and the end states that may occur in one initiation event are suggested. In addition, detailed classification was performed to induce various end states on the basis of the suggested initiation events ; loss of grid electricity of ESS, loss of battery electricity(DC) of battery, impairment of electric function of BMS, loss of grid electricity(AC) of PCS, loss of data of EMS, Mechanical damage of cable, event sequence analysis conducted on the basis of event trees. If the suggested IEs and ESs are applied on the basis of ESS event cases, it is expected to prevent the same kinds of accident and operate ESS safely.

• Journal of the Society of Naval Architects of Korea
• /
• v.59 no.3
• /
• pp.164-172
• /
• 2022

A HAZID is a brainstorming workshop to identify hazards in an early phase of a project. It should be flexible to capture all probable accidents allowing experienced participants to exploit their expertise and experiences. A bowtie analysis is a graphical representation of major accident hazards elaborating safety measures i.e. barriers. The result of these workshops should be documented in an organized manner to share as good as possible details of the discussion through the lifetime of the project. Currently results are documented using a three-step representation of an accident; causes, top event and consequences, which cannot capture correctly sequence of events leading to various accidents and roles of barrier between two events. Another problem is that barriers would be shown repeatedly leading to a misunderstanding that there are an enough number of safety measures. A new bowtie analysis method is proposed to describe an accident in multiple steps showing relations among causes or consequences. With causes and consequences shown in a format of a tree, the frequencies of having the top event (Fault tree analysis) and various consequences (Event tree analysis) are evaluated automatically based on the frequency of initiating causes and the probabilities of failure of barriers. It will provide a good description of the accident scenario and help the risk to be assessed transparently.

• Journal of the Korean Institute of Intelligent Systems
• /
• v.8 no.6
• /
• pp.91-98
• /
• 1998

A DEDS is a system whose stated change in response to the occurrence of events from a predefined event set. A major difficulty in developing analytical results for the system is the lack of appropriate modeling techniques. In this paper, we consider the modeling and control problem for Discrete Event Dynamic Systems(DEDS) in the Temporal Logic framework(TLF) which have been recently defined. The traditional TLF is enhanced with time functions for real time control of Discrete Event Dynamic Systems. A sequence of event which drive the system from a given initial state to a given final state is generated by pertinently operating the given plants. This paper proposes the use of Real-time Temporal Logic as a modeling tool for the analysis and control of DEDS. An given example of fixed-time traffic control problem is shown to illustrate our results with Real-time Temporal Logic Framework.

• Korean Journal of Computational Design and Engineering
• /
• v.18 no.4
• /
• pp.250-257
• /
• 2013

This paper introduces a failure analysis procedure that underpins real-time fault prognosis. In the previous study, we developed a systematic eventization procedure which makes it possible to reduce the original data size into a manageable one in the form of event logs and eventually to extract failure patterns efficiently from the reduced data. Failure patterns are then extracted in the form of event sequences by sequence-mining algorithms, (e.g. FP-Tree algorithm). Extracted patterns are stored in a failure pattern library, and eventually, we use the stored failure pattern information to predict potential failures. The two practical case studies (marine diesel engine and SIRIUS-II car engine) provide empirical support for the performance of the proposed failure analysis procedure. This procedure can be easily extended for wide application fields of failure analysis such as vehicle and machine diagnostics. Furthermore, it can be applied to human health monitoring & prognosis, so that human body signals could be efficiently analyzed.

• Proceedings of the Korean Institute of Industrial Safety Conference
• /
• 1998.11a
• /
• pp.219-224
• /
• 1998

확률론적 안전성 평가(Probabilistic Safety Assessment PSA)나 정량적인 위험도 평가(Quantitative Risk Assessment: QRA)에서 인간신뢰도분석(human reliability analysis)은 인간행위를 기기처럼 생각하여 전체 시스템의 안전성에 중요한 초기사건(initiating event) 이전이나 초기사건 이후 또는 초기사건을 유발하는 인간행위를 파악하고 정량화하여, 확률론적 평가의 논리구조인 사건 및 고장수목(event tree 및 fault tree)이나 사고경위 단절집합 (accident sequence outsets)에 포함시키는 것이다. (중략)

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.11 no.3
• /
• pp.1670-1683
• /
• 2017

Mitotic event recognition is a crucial and challenging task in biomedical applications. In this paper, we introduce the slow feature analysis and propose a fully-automated mitotic event recognition method for cell populations imaged with time-lapse phase contrast microscopy. The method includes three steps. First, a candidate sequence extraction method is utilized to exclude most of the sequences not containing mitosis. Next, slow feature is learned from the candidate sequences using slow feature analysis. Finally, a hidden conditional random field (HCRF) model is applied for the classification of the sequences. We use a supervised SFA learning strategy to learn the slow feature function because the strategy brings image content and discriminative information together to get a better encoding. Besides, the HCRF model is more suitable to describe the temporal structure of image sequences than nonsequential SVM approaches. In our experiment, the proposed recognition method achieved 0.93 area under curve (AUC) and 91% accuracy on a very challenging phase contrast microscopy dataset named C2C12.