• The Transactions of The Korean Institute of Electrical Engineers
• /
• v.58 no.7
• /
• pp.1276-1280
• /
• 2009

This paper presents a fault diagnosis expert system for 765kV substation. The proposed system includes the topology processor and intelligent alarm processing subsystems. This expert system estimates the fault section through the inference process using heuristic knowledge and the output of topology processor and intelligent alarm processing system. The rule-base of this expert system is composed of basic rules suggested by Korea Electric Power Corporation and heuristic rules. This expert system is developed using PROLOG language. Also, user friendly Graphic User Interface is developed using visual basic programming in the windows XP environment. The proposed expert system showed a promising performance through the several case studies.

• Proceedings of the KIEE Conference
• /
• 2001.07b
• /
• pp.699-701
• /
• 2001

The expert system monitoring and diagnosing 765kV power apparatus was described in this paper. To develop this expert system, we studied the knowledge bases and data bases for 765kV transformer and GIS. In order to make the reliable inference of knowledge base and the good MMI(Man Machine Interface), the data bases were consisted of the tables of power apparatus information, limit level value, measured input data, inference result and diagnosis result. The knowledge base had various rules to infer the conditions of transformer and GIS. We applied both the forward chaining and backward chaining methods to these rules of system for good inferences. This paper describes the applied methods for expert system. Also, this developed system was tested with dissolved gas analyzing result and the result was shown.

• Proceedings of the KIEE Conference
• /
• 2001.05a
• /
• pp.319-321
• /
• 2001

When the fault occurred at distribution system, the restoration was late. There are 2 reasons The one is the error of fault location the other is multiple possible candidates of fault location. This paper presents two of new techniques for diagnosing fault regions. The proposed diagnosis scheme is capable of accurately identifying the location of fault upon its occurrence. based on the integration of information available from protective devices and measured load current change at the substation. In this paper expert system for real fault region is presented using these facts. Testing of the developed system using EMTP Simulation Model has demonstrated.

• Proceedings of the KIEE Conference
• /
• 1999.07c
• /
• pp.1394-1396
• /
• 1999

It is important the protective relay should have high selectivity and sensitivity performance and it is achieved by correct setting of the relay parametes. The verification of accurate setting for the protective relay is very difficult before a actual fault occurs. This paper reports a diagnosis expert system to verify the correctness of the relay setting by using the operation results in the power system fault simulation.

• Proceedings of the IEEK Conference
• /
• 2002.06e
• /
• pp.269-272
• /
• 2002

In this study, we designed the expert system for the diagnosis of stroke. The causes of stroke in central nervous systems are very diverse, so a doctor who treats the patients with stroke must have the expert knowledge for the quick and correct diagnosis and for the adequate medical management. But the primary physician who engaged in the primary care of the patient with stroke does not have the export knowledge for the stroke. So, we need to develop the expert system for assisting the diagnosis of stroke. Also the diagnosis system can be used as simulator for the medical students who study the neurology. In this study, we developed the diagnosis expert system that offer a pathological name provided by artificial neural networks. And we designed the inference engine and interfaces. The artificial neural network is a system that provide a possible diagnosis of stroke. We implemented the system using Windows2000 Server, IIS5.0 and ASP.

• Proceedings of the KIEE Conference
• /
• 1997.11a
• /
• pp.190-192
• /
• 1997

In this paper we developed an object-based expert system for electric power system fault diagnosis. The object corresponds to a hardware or event in real world and they are independent each other. The expert system is designed to estimate the fault sections and identify the false operation, nonoperation of protective devices. The expert system was developed using C++ language in pentium PC. We applied the expert system in various sample systems and showed that making up a knowledge base is easier and diagnosis was done in a approximately real time.

• Journal of Korean Institute of Industrial Engineers
• /
• v.23 no.1
• /
• pp.95-113
• /
• 1997

This paper outlines a framework for a diagnosis of a complex system with uncertain information. Sensor validation ploys a vital role in the ability of the overall system to correctly determine the state of a system monitored by imperfect sensors. Here, emphases are put on the heuristic technology and post-processor for reasoning. Heuristic Sensor Validation (HSV) exploits deeper knowledge about parameter interaction within the plant to cull sensor faults from the data stream. Finally the modified probability distributions and validated data are used as input to the reasoning scheme which is the runtime version of the influence diagram. The output of the influence diagram is a diagnostic mapping from the symptoms or sensor readings to a determination of likely failure modes. Once likely failure modes are identified, a detailed diagnostic knowledge base suggests corrective actions to improve performance. This framework for a diagnostic expert system with sensor validation and reasoning under uncertainty applies in $HEATXPRT^{TM}$ a data-driven on-line expert system for diagnosing heat rate degradation problems in fossil power plants [1].

• Proceedings of the KIEE Conference
• /
• 1999.11d
• /
• pp.1019-1021
• /
• 1999

In this paper, we describe the design and appliance of the preventive diagnostic expert system of gas in oil for power transformer. This expert system is developed to use expert system development tool; Element Expert (NEURON DATA Inc.) Analysis is developed by four diagnose methods. In first, the standard of KEPCO is applied. It classifies the state of transformer by four level: Normal, warning, abnormal and danger. And the others are gas pattern methods, IEC code method, and Dornen & Roger Ratio method and applied later. These latter methods analyse the cause of result. Inference engine was designed with Element Expert. In last, we operate the system with sample data and we can obtain the correctly inferred result for the state of transformer.

• Journal of Physiology & Pathology in Korean Medicine
• /
• v.16 no.3
• /
• pp.528-531
• /
• 2002

In Traditional Oriental Medicine, there has been a growing needs for computerized diagnosis expert system, which can implement pre-diagnosis and correct the errors of practitioners. Therefore, we developed the expert system (KHU-PIPE : Kyung Hee University - Pattern Identification and Prescription Expert) for diagnosis and treatment. It has three characteristics as following. First. this system has the knowledge base which modified the standardized data designed by Chinese government during 1980s. Second, it provides the objective and standardized diagnosis as the results of pattern identification and their appropriate prescriptions for treatment. Third, it is applied to both LAN system and internet. Furthermore, it can be used as an educational methods for the practices of pattern identification and prescription, and provide the objective criteria for clinical studies and promote the Traditional Oriental Medicine as an evidence-based medicine.

• Tribology and Lubricants
• /
• v.18 no.2
• /
• pp.117-126
• /
• 2002

Abstract - An integrated condition monitoring algorithm based on an expert system was implemented in this work in order to monitor effectively the machine conditions. The knowledge base was consisted of numeric data which meant the posterior probability of each measurement parameter for the representative machine failures. Also the inference engine was constructed as a series of statistical process, where the probable machine fault was inferred by a mapping technology of pattern recognition. The proposed algorithm was, through the user interface, applied for an air compressor system where the temperature, vibration and wear properties were measured simultaneously. The result of the case study was found fairly satisfactory in the diagnosis of the machine condition since the predicted result was well correlated to the machine fault occurred.