• Title/Summary/Keyword: Load Detection

Search Result 588, Processing Time 0.031 seconds

Improving the Performance of Machine Learning Models for Anomaly Detection based on Vibration Analog Signals (진동 아날로그 신호 기반의 이상상황 탐지를 위한 기계학습 모형의 성능지표 향상)

  • Jaehun Kim;Sangcheon Eom;Chulsoon Park
    • Journal of Korean Society of Industrial and Systems Engineering
    • /
    • v.47 no.2
    • /
    • pp.1-9
    • /
    • 2024
  • New motor development requires high-speed load testing using dynamo equipment to calculate the efficiency of the motor. Abnormal noise and vibration may occur in the test equipment rotating at high speed due to misalignment of the connecting shaft or looseness of the fixation, which may lead to safety accidents. In this study, three single-axis vibration sensors for X, Y, and Z axes were attached on the surface of the test motor to measure the vibration value of vibration. Analog data collected from these sensors was used in classification models for anomaly detection. Since the classification accuracy was around only 93%, commonly used hyperparameter optimization techniques such as Grid search, Random search, and Bayesian Optimization were applied to increase accuracy. In addition, Response Surface Method based on Design of Experiment was also used for hyperparameter optimization. However, it was found that there were limits to improving accuracy with these methods. The reason is that the sampling data from an analog signal does not reflect the patterns hidden in the signal. Therefore, in order to find pattern information of the sampling data, we obtained descriptive statistics such as mean, variance, skewness, kurtosis, and percentiles of the analog data, and applied them to the classification models. Classification models using descriptive statistics showed excellent performance improvement. The developed model can be used as a monitoring system that detects abnormal conditions of the motor test.

Understanding the Experience of Visual Change Detection Based on the Experience of a Sensory Conflict Evoked by a Binocular Rivalry (양안경합의 감각적 상충 경험에 기초한 시각적 변화탐지 경험에 대한 이해)

  • Shin, Youngseon;Hyun, Joo-Seok
    • Science of Emotion and Sensibility
    • /
    • v.16 no.3
    • /
    • pp.341-350
    • /
    • 2013
  • The present study aimed to understand the sensory characteristic of change detection by comparing the experience of detecting a salient visual change against the experience of detecting a sensory conflict evoked by a binocular mismatch. In Experiment 1, we used the change detection task where 2, 4, or 6 items were short-term remembered in visual working memory and were compared with following test items. The half of change-present trials were manipulated to elicit a binocular rivalry on the test item with the change by way of monocular inputs across the eyes. The results showed that change detection accuracy without the rivalry manipulation declined evidently as the display setsize increased whereas no such setsize effect was observed with the rivalry manipulation. Experiment 2 tested search efficiency for the search array where the target was designated as an item with the rivalry manipulation, and found the search was very efficient regardless of the rivalry manipulation. The results of Experiment 1 and 2 showed that when the given memory load varies, the experience of detecting a salient visual change become similar to the experience of detecting a sensory conflict by a binocular rivalry.

  • PDF

Sag & Swell Detection by Phase Voltage Compensation in 3-Phase Unbalanced Grid (3상 불평형 계통에서의 상 전압 보상을 통한 Sag 및 Swell 검출)

  • Kim, Min-Gi;Kim, Jun-Gu;Jung, Yong-Chae;Won, Chun-Yuen
    • Proceedings of the KIPE Conference
    • /
    • 2013.11a
    • /
    • pp.258-259
    • /
    • 2013
  • Load connection or disconnection makes instantaneous sag & swell in 3-phase grid. When unbalance state occurs, between sensed phase voltage and actual phase voltage may have discrepancy. It makes difficult to detect accurate sag & swell, so it is hard to satisfy the standard for switching ESS system to UPS mode. In this paper, we analyzed unbalanced 3-phase voltage, and compensated the actual sag & swell magnitude.

  • PDF

Small Fatigue Crack Measurement and Crack Growth Characteristics for Smooth and Notch Specimens (평활 및 노치재의 미소피로균열측정과 성장특성)

  • 이종형
    • Transactions of the Korean Society of Mechanical Engineers
    • /
    • v.17 no.9
    • /
    • pp.2145-2152
    • /
    • 1993
  • The objective of this paper is to examine the detection limit, growth characteristics and notch curvature radius in short crack problem. Measurement techniques such as ultrasonic method and back-face strain compliance method were adopted. The fatigue crack growth rate of the short crack is slower than that of a long crack for a notched specimen. The characteristic of crack growth and crack closure is same as the case of a delay of crack growth caused by constant amplitude load for an ideal crack or single peak overload for a fatigue crack. The short crack is detected effectively by ultrasonic method. A short surface crack occurs in the middle of specimen thickness and is transient to a through crack depth is larger than the notch curvature radius.

Antenna array for estimation of direction of arrival utilizing modified minimum eigenvalue searching (개선된 MES 방법을 이용한 신호의 도래각(DOA) 추정을 위한 배열안테나)

  • 이현배;최승원
    • Journal of the Korean Institute of Telematics and Electronics B
    • /
    • v.33B no.4
    • /
    • pp.164-173
    • /
    • 1996
  • This paper presents an alternative technique for DOA (direction-of-arrival) estimation. For generating a weight vector orthogonal to the signal subspace, a modified version of MES (minimum eigenvalue searching ) method is introduced. The performance of the proposed technique is compared to that of the conventional ED (eigen decomposition) method in terms of angle resolution for a number of snapshots during agiven observation period as well as various SNR's. In addition, the superiority of the suggested technique is shown, by analyzing the required computational load of the proposed MES and conventional ED method. A novel procedure of simplifying the MES proposed in [1] is presented on that purpose. Another advnatage of the proposed technique is that it is performed independently of the detection of the number of signal components, which makes it possible to estimate the DOA's of clusters consisting of infinite number of inseparable signal components.

  • PDF

Study on Damage Repair of Polymer Composites Using Self-Healing Technique (Self-healing Technique을 적용한 폴리머 복합재의 손상 보수 연구)

  • ;;M.R. Kessler;S.R. White
    • Proceedings of the Korean Society For Composite Materials Conference
    • /
    • 2001.05a
    • /
    • pp.93-96
    • /
    • 2001
  • Structural polymer composites are susceptible to damage in the form of cracks, which form deep within the structure where detection is difficult and repair is almost impossible. A recent methodology for the damage repair of polymer composites using the self-healing technique is reported. The polymerization of the healing agent is triggered by contact with an embedded catalyst, being necessary to damage repair of polymer composites. For this purpose, the self-healing concept is introduced and the manufacturing process of microcapsule with the healing agent is briefly described. The polymerization between the healing agent and the catalyst is verified by the use of ESEM and IR spectroscopy. Finally the efficiency of the self-healing technique is investigated by measuring the critical load of TDCB specimen.

  • PDF

A Data Fault Detection System for Diesel Engines Using Neural Networks (신경회로망을 이용한 디젤기관의 데이터 이상감지 시스템에 관한 연구)

  • 천행춘;유영호
    • Journal of Advanced Marine Engineering and Technology
    • /
    • v.26 no.4
    • /
    • pp.493-500
    • /
    • 2002
  • The operational data of diesel generator engine is two kinds of data. One is interactive the other is non interactive. We can find the fault information from interactive data measured for every sampling time when the changing rate, direction and status of data are investigated in comparition with those of normal status to diagnose the fault of combustion system. The various data values of combustion system for diesel engine are not proportional to load condition. The criterion to decide the level of data value is not absolute but relative to relational data. This study proposes to compose malfunction diagnosis engine using neural networks to decide that level of data value is out of normal status with the data collected from generator engine of the ship using the commercial data mining tool. This paper investigates the real ship's operational data of diesel generator engine and confirms usefulness of fault detecting through simulations for fault detecting.

A Study of Hear Flux and Instantaneous Temperature According to the Initial Tamperature of Combustion Chamber in a Constant Volume Combustion Chamber (연소실 초기온도 변화에 따른 순간열유속에 관한 연구)

  • Lee, Chi-Woo
    • Journal of the Korean Society of Industry Convergence
    • /
    • v.6 no.3
    • /
    • pp.193-200
    • /
    • 2003
  • In the production of internal combustion engines, there has been a move towards the development of high performance engines with improved fuel efficiency, lighter weight and smaller sizes. These trends help to answer problems in engines related to thermal load and abnormal combustion. In order to investigate these problems, a thin film-type probe for instantaneously measuring temperatures has been suggested. A method for manufacturing such a probe was established in this study. The instantaneous surface temperature of a constant volume combustion chamber was measured by this probe and the heat flux was obtained through Fourier analysis. In order to thoroughly understand the characteristics of combustion, the authors measured the wall temperature of the combustion chamber and computed heat flux through a cylinder wall while varying the protrusion height of the probe. For achieving the above goals, a instantaneous temperature probe was developed, thereby making possible the analysis of the instantaneous temperature of wall surface and the detection of unsteady heat flux in the constant volume combustion chamber.

  • PDF

Realization of a Parallel Network System for Image Processing Techniques (영상 처리 기법을 위한 병렬화 네트워크 시스템의 구성)

  • 서원찬;조강현;김우열
    • Journal of Institute of Control, Robotics and Systems
    • /
    • v.6 no.6
    • /
    • pp.492-499
    • /
    • 2000
  • In this paper, realization techniques of the parallel processing and the parallel network system for image processing are described. The parallel image processing system is constructed by the characterization of image processing and processor. Several problems are solved to achieve effective parallel processing and processor networking with the particular properties of image processing, which are reduction of communication quantity, equalization of load and delay depreciation on communication. A parallel image input device is developed for the flexible networking of parallel image processing. An abnormal region detection algorithm which is the basic function in machine vision is applied to evaluate the constructed parallel image processing system. The performance and effectiveness of the system are confirmed by experiments.

  • PDF

A Study on an Equipment Performance Measurement System for Effective Bottleneck Management (병목 설비의 개선 활동에 유용한 설비관리 지표체계에 관한 연구)

  • Lee, Min-Ho;Lim, Sung-Mook
    • Journal of Korean Society of Industrial and Systems Engineering
    • /
    • v.33 no.4
    • /
    • pp.100-113
    • /
    • 2010
  • Manufacturing companies' cost competitiveness with respect to equipment management can be achieved by satisfying additional market demands by their own capacity without purchasing additional equipments. In essence, it can be accomplished by making continuous investigation into bottlenecks and improvement on them. Therefore, equipment performance measure systems should be designed so that they can support manufacturing companies' such endeavors. With the purpose of establishing an effective equipment performance measurement system for detecting and improving bottlenecks, this study (1) suggests some desirable features that such a system should have, (2) evaluates conventional equipment performance measurement systems in terms of their usefulness for the detection and improvement of bottlenecks, and (3) proposes an improved system. We also perform a simulation experiment to demonstrate the limitations of the conventional systems and show how the proposed system can resolve the problems.