Skip to content
MathWorks - Mobile View
  • MathWorks 계정에 로그인합니다.MathWorks 계정에 로그인합니다.
  • Access your MathWorks Account
    • 내 계정
    • 나의 커뮤니티 프로필
    • 라이선스를 계정에 연결
    • 로그아웃
  • 제품
  • 솔루션
  • 아카데미아
  • 지원
  • 커뮤니티
  • 이벤트
  • MATLAB 받기
MathWorks
  • 제품
  • 솔루션
  • 아카데미아
  • 지원
  • 커뮤니티
  • 이벤트
  • MATLAB 받기
  • MathWorks 계정에 로그인합니다.MathWorks 계정에 로그인합니다.
  • Access your MathWorks Account
    • 내 계정
    • 나의 커뮤니티 프로필
    • 라이선스를 계정에 연결
    • 로그아웃

비디오 및 웨비나

  • MathWorks
  • 비디오
  • 비디오 홈
  • 검색
  • 비디오 홈
  • 검색
  • 영업 담당 문의
  • 평가판 신청
4:44 Video length is 4:44.
  • Description
  • Full Transcript
  • Related Resources

Feature Extraction Using Diagnostic Feature Designer App

Melda Ulusoy, MathWorks

Use Diagnostic Feature Designer app to extract time-domain and spectral features from your data to design predictive maintenance algorithms.

In this example, measurements have been collected from a triplex pump under different fault conditions. The app lets you import this data and interactively visualize it. You can group the measurements by different fault conditions. After the time-domain and spectral features are extracted from the data, you can evaluate the effectiveness of the extracted features using histograms. You can also rank them to determine numerically which features are likely to best discriminate healthy and faulty behavior. Finally, the most effective features are exported to Classification Learner app for further evaluation of feature effectiveness and for training machine learning models.

Feature Extraction Using the Diagnostic Feature Designer App

In this video, we’re going to demonstrate how you can use Diagnostic Feature Designer app to extract features for developing a predictive maintenance algorithm.

We start by importing our data set into the app. The data has been collected from a triplex pump under different fault conditions. It is stored in an ensemble which is a specialized datastore for developing predictive maintenance algorithms. The ensemble datastore contains 1.2 seconds long measurements of flow and pressure and also the fault codes for each of these measurements. After we import the data set, it shows up in the data browser. To visualize the flow signal, we select it and click Signal Trace. This plots all the measurements with different fault conditions. We can now group the measurements by the fault codes by selecting this option. If we zoom in using the panner strip below, we can better see how measurements are highlighted with different colors based on different fault types. Next, we’ll extract time-domain and spectral features from this data. We go back to the Feature Designer tab, and under this menu, we select signal features to generate statistics features. We’ll first use the flow data and later extract features from the pressure signal. Here, we have commonly used time-domain features such as the mean, standard deviation, kurtosis, and skewness. Now that we computed the time-domain features, we’ll continue with extracting spectral features. The app can use the time-domain data to estimate the signal spectra of these signals which can be then used to extract spectral features. We select spectral estimation and click power spectrum. Here, you can try out nonparametric or parametric methods to compute the spectrum and compare their results. We choose the auto-regressive model with a model order of 20. Next, to compute spectral features, we click here. We select the frequency band such that it includes the first four peaks. The reason is that due to noisy data at higher frequencies, it’s harder to distinguish the spectral peaks. Therefore, any features extracted from higher frequencies won’t contribute to the performance of machine learning models.

 So far, we identified time and spectral features from the flow data. You can repeat the same process with the pressure data and extract some additional features. Now, all the extracted features from flow and pressure data are stored in the FeatureTable1. After selecting this table, we can click on the Feature Table View that shows all the computed feature values in a tabular form. Different features are displayed on different columns. We can also use histograms which show distributions of the computed features. On these plots, different colors indicate different faults. Due to overlapping distributions of different fault types and high number of features, it’s hard to decide on which features are more separable and distinctive. The app lets us rank all the features to identify the ones that effectively separate different types of faults. On the Feature Designer tab, when we click Rank Features, the app uses one-way ANOVA to calculate ranking scores for all the features. The results of the ANOVA test are displayed on the right-hand side, whereas the bars on the left shows the normalized scores for different features. We can view the feature names by hovering over the bars. The features with a higher score are good candidates for training a machine learning model. For further evaluation of the extracted features, we can now export them to the Classification Learner, where we can train machine learning models for fault classification.

Related Products

  • Predictive Maintenance Toolbox

MATLAB and Simulink for Predictive Maintenance
MATLAB Tech Talks on Predictive Maintenance (4 videos)
Analyze and Select Features for Pump Diagnostics

Bridging Wireless Communications Design and Testing with MATLAB

Read white paper

Overcoming Four Common Obstacles to Predictive Maintenance with MATLAB and Simulink

Read white paper

Feedback

Featured Product

Predictive Maintenance Toolbox

  • Request Trial
  • Get Pricing

Up Next:

4:50
Feature Detection, Extraction, and Matching with RANSAC

Related Videos:

7:59
New MATLAB Feature in Release 2016a: Pause Button
3:26
New Desktop Plotting Feature in R2009b
36:51
Computational Statistics: Feature Selection,...
0:38
New Publishing to PDF Feature in R2009b

View more related videos

MathWorks - Domain Selector

Select a Web Site

Choose a web site to get translated content where available and see local events and offers. Based on your location, we recommend that you select: .

  • Switzerland (English)
  • Switzerland (Deutsch)
  • Switzerland (Français)
  • 中国 (简体中文)
  • 中国 (English)

You can also select a web site from the following list:

How to Get Best Site Performance

Select the China site (in Chinese or English) for best site performance. Other MathWorks country sites are not optimized for visits from your location.

Americas

  • América Latina (Español)
  • Canada (English)
  • United States (English)

Europe

  • Belgium (English)
  • Denmark (English)
  • Deutschland (Deutsch)
  • España (Español)
  • Finland (English)
  • France (Français)
  • Ireland (English)
  • Italia (Italiano)
  • Luxembourg (English)
  • Netherlands (English)
  • Norway (English)
  • Österreich (Deutsch)
  • Portugal (English)
  • Sweden (English)
  • Switzerland
    • Deutsch
    • English
    • Français
  • United Kingdom (English)

Asia Pacific

  • Australia (English)
  • India (English)
  • New Zealand (English)
  • 中国
    • 简体中文Chinese
    • English
  • 日本Japanese (日本語)
  • 한국Korean (한국어)

Contact your local office

  • 영업 담당 문의
  • 평가판 신청

MathWorks

Accelerating the pace of engineering and science

MathWorks는 엔지니어와 과학자들을 위한 테크니컬 컴퓨팅 소프트웨어 분야의 선도적인 개발업체입니다.

활용 분야 …

제품 소개

  • MATLAB
  • Simulink
  • 학생용 소프트웨어
  • 하드웨어 지원
  • File Exchange

다운로드 및 구매

  • 다운로드
  • 평가판 신청
  • 영업 상담
  • 가격 및 라이선스
  • MathWorks 스토어

사용 방법

  • 문서
  • 튜토리얼
  • 예제
  • 비디오 및 웨비나
  • 교육

지원

  • 설치 도움말
  • MATLAB Answers
  • 컨설팅
  • 라이선스 센터
  • 지원 문의

회사 정보

  • 채용
  • 뉴스 룸
  • 사회적 미션
  • 고객 사례
  • 회사 정보
  • Select a Web Site United States
  • 신뢰 센터
  • 등록 상표
  • 정보 취급 방침
  • 불법 복제 방지
  • 애플리케이션 상태
  • 매스웍스코리아 유한회사
  • 주소: 서울시 강남구 삼성동 테헤란로 521 파르나스타워 14층
  • 전화번호: 02-6006-5100
  • 대표자 : 이종민
  • 사업자 등록번호 : 120-86-60062

© 1994-2022 The MathWorks, Inc.

  • Naver
  • Facebook
  • Twitter
  • YouTube
  • LinkedIn
  • RSS

대화에 참여하기