• Nuclear Engineering and Technology
• /
• v.56 no.2
• /
• pp.558-567
• /
• 2024

Diagnostic support systems are being researched to assist operators in identifying and responding to abnormal events in a nuclear power plant. Most studies to date have considered single abnormal events only, for which it is relatively straightforward to obtain data to train the deep learning model of the diagnostic support system. However, cases in which multiple abnormal events occur must also be considered, for which obtaining training data becomes difficult due to the large number of combinations of possible abnormal events. This study proposes an approach to maintain diagnostic performance for multiple abnormal events by training a deep learning model with data on single abnormal events only. The proposed approach is applied to an existing algorithm that can perform feature selection and multi-label classification. We choose an extremely randomized trees classifier to select dedicated monitoring parameters for target abnormal events. In diagnosing each event occurrence independently, two-channel convolutional neural networks are employed as sub-models. The algorithm was tested in a case study with various scenarios, including single and multiple abnormal events. Results demonstrated that the proposed approach maintained diagnostic performance for 15 single abnormal events and significantly improved performance for 105 multiple abnormal events compared to the base model.

• 제어로봇시스템학회:학술대회논문집
• /
• 1992.10a
• /
• pp.453-456
• /
• 1992

Diagnostic prototype expert system was developed by analyzing the measured acoustical data of automobile. For the utilities of this system, 1/3 octave filter(band-pass filter) and A/D converter were used for data acquisition and then information was analyzed using signal processing technique and pattern recognition by Hamming network algorithm. In order to raise the reliability of the diagnostic results, fuzzy inference technique was applied and, the results were displayed as graphical method to help the novice in diagnostic field. The validation of this diagnostic system was checked through experiments and it showed and acceptable performance for diagnostic process.

• Journal of Haehwa Medicine
• /
• v.23 no.1
• /
• pp.13-20
• /
• 2014

Objectives : Clinical data in traditional medicine, such as Korean medicine, traditional Chinese medicine have a long history of accumulating evidence and these rich data are recorded in classic literature. We have conducted a study of developing an algorithm that support clinical diagnosis with composing both users knowledge and data obtained from literature. In order to define necessary information and required steps in diagnosis procedure, we have established a clinical diagnostic procedure including a step of collecting patients symptoms, a step of determining candidates, a step of diagnostic decisions, a step of deciding of treatment and a step of adjusting medicinal treatment. Methods : Our study have been based on the following premises. 1. Using data obtained from literature contributes to accurate diagnosis 2. Displaying the data before users request contributes to accurate conclusion. Displaying before users request enable users to recognize their overlooking a fact on purpose or not. 3. Checking symptoms that are commonly accompanied with a group of diseases that accompany symptoms appealed by a patient contributes to accurate conclusion. These symptoms are worthy of checking. 4. Comparing more than two candidates contributes to accurate conclusion. Users can compare their accompanied symptoms with patients symptoms and this helps users to make a decision. Results : Based on the above premises, we have developed an literature based algorithm to provide various functions, such as recommending symptoms to check, comparing groups of symptoms, differential diagnosis, recommending medicinal materials to prescribe, and more. Conclusions : By the results of simulation with virtual diagnostic scenario, we concluded this algorithm is useful helping clinician in diagnosis procedure.

• Journal of the Korean Institute of Intelligent Systems
• /
• v.22 no.6
• /
• pp.675-680
• /
• 2012

Wind energy is currently the fastest growing source of renewable energy used for electrical generation around the world. Wind farms are adding a significant amount of electrical generation capacity. The increase in the number of wind farms has led to the need for more effective operation and maintenance. CMS(Condition Monitoring System) can be used to aid plant operator in achieving these goals. Its aim is to provide operators with information regarding th e health of their machine, which in turn, can help them improve operation efficiency. In this work, wind turbine fault diagnostic algorithm which can diagnose the mass unbalance and aerodynamic asymmetry of the blades is proposed. Proposed diagnostic algorithm utilizes both FFT(Fast Feurier Transform) of the signal from accelerometers installed inside of nacelle and simple diagnostic logic. Furthermore, to verify the applicability of the proposed system, 3W small sized wind turbine system is tested and physical experiments are carried out.

• Journal of the Korean Institute of Illuminating and Electrical Installation Engineers
• /
• v.15 no.5
• /
• pp.68-74
• /
• 2001

This paper describes the new algorithm method for detecting abnormal causes within power transformers. Generally, the gas analysis has been proved the most confident method of many transformer diagnostics. The proposed algorithm is adapted to the international codes of IEC, Dornenburg, Gas Pattern including the DEPCO´s gas analysis method for the improvement of diagnostic efficiency. Specially, this algorithm is programmed by the tool of Element Expert developed Neuron DATA Inc. in USA. Also, it was confirmed that the developed algorithm is proved the confidence by the use of real data in fault power transformers.

• Transactions of the Korean Society of Machine Tool Engineers
• /
• v.12 no.4
• /
• pp.29-35
• /
• 2003

The rotating machineries always generate harmonic frequencies of their own rotating speed, and increment of vibration amplitude affects to the equipments which connected to the vibrational source and causes industrial calamities. The life cycle of equipments can be extended and damages to the human beings could be prevented by identifying the cause of malfunctions through prediction of the increment of vibration and records of vibrational history. In this study, therefore, diagnostic expert algorithm for the centrifugal pump is developed by integrating fuzzy inference method and signal processing techniques. And the validity of the developed diagnostic system is examined via various computer simulations.

• Journal of the Korean Institute of Intelligent Systems
• /
• v.13 no.5
• /
• pp.600-605
• /
• 2003

PCA(Principle Component Analysis) has emerged as a useful tool for process monitoring and fault diagnosis. The general approach requires the user to identify the root cause by interpreting the residual or principle components. This could be tedious and often impossible for a large process. In this paper, PCA scheme is combined with the FCM-based fault diagnostic algorithm to enhance the diagnostic results. The implementation of the FCM-based fault diagnostic system by using PCA is done and its application is illustrated on the two-tank system.

• 대한상한금궤의학회지
• /
• v.14 no.1
• /
• pp.41-50
• /
• 2022

Objective : The purpose of this study is to explore the most suitable machine learning model algorithm for Shanghanlun diagnostic system classification using natural language processing (NLP). Methods : A total of 201 data items were collected from 『Shanghanlun』 and 『Clinical Shanghanlun』, 'Taeyangbyeong-gyeolhyung' and 'Eumyangyeokchahunobokbyeong' were excluded to prevent oversampling or undersampling. Data were pretreated using a twitter Korean tokenizer and trained by logistic regression, ridge regression, lasso regression, naive bayes classifier, decision tree, and random forest algorithms. The accuracy of the models were compared. Results : As a result of machine learning, ridge regression and naive Bayes classifier showed an accuracy of 0.843, logistic regression and random forest showed an accuracy of 0.804, and decision tree showed an accuracy of 0.745, while lasso regression showed an accuracy of 0.608. Conclusions : Ridge regression and naive Bayes classifier are suitable NLP machine learning models for the Shanghanlun diagnostic system classification.

• Journal of the Korean Institute of Intelligent Systems
• /
• v.24 no.1
• /
• pp.78-83
• /
• 2014

Recently, a rapid growth of wind power system as a leading renewable energy source has compelled a number of companies to develop intelligent monitoring and diagnostic system. Such systems can detect early mechanical faults, which prevents from costly repairs. Generally, fault diagnostic system for wind turbines is based on vibration and process signal analysis. In this work, different type of mechanical faults such as mass unbalance and shaft misalignment which can always happen in wind turbine system is considered. The proposed intelligent fault diagnostic algorithm utilizes artificial neural network and Wavelet transform. In order to verify the feasibility of the proposed algorithm, mechanical fault generation experimental system manufactured by Gaon corporation is utilized.

• Nuclear Engineering and Technology
• /
• v.47 no.5
• /
• pp.624-632
• /
• 2015

A self-diagnostic monitoring system is a system that has the ability to measure various physical quantities such as temperature, pressure, or acceleration from sensors scattered over a mechanical system such as a power plant, in order to monitor its various states, and to make a decision about its health status. We have developed a self-diagnostic monitoring system for an air-operated valve system to be used in a nuclear power plant. In this study, we have tried to improve the self-diagnostic monitoring system to increase its reliability. We have implemented three different machine learning algorithms, i.e., logistic regression, an artificial neural network, and a support vector machine. After each algorithm performs the decision process independently, the decision-making module collects these individual decisions and makes a final decision using a majority vote scheme. With this, we performed some simulations and presented some of its results. The contribution of this study is that, by employing more robust and stable algorithms, each of the algorithms performs the recognition task more accurately. Moreover, by integrating these results and employing the majority vote scheme, we can make a definite decision, which makes the self-diagnostic monitoring system more reliable.