MATLAB For Data Analytics And Machine Learning

2y ago
72 Views
9 Downloads
1.92 MB
34 Pages
Last View : 1m ago
Last Download : 2m ago
Upload by : Rafael Ruffin
Transcription

MATLAB for Data Analytics and Machine LearningAd-hoc Desktopdata analysisSuite of SharedMATLABAnalyticsEnterpriseIntegration forReal-timeanalyticsSundar UmamaheshwaranAmit DoshiApplication Engineer-Technical Computing 2016 The MathWorks, Inc.1

Data Analytics WorkflowBusinessSystemsSmart ConnectedSystemsData Acquisition Engineering, Scientific, and Field Business and TransactionalData Analytics Data Pre-processing Feature Extraction Building algorithms, math models Making business decisionsAnalytics Integration Integrate algorithms with IT Analytics run on Embedded targets2

Integrated Analytics– Success Stories1. Taking Business Decisions Using Historical Data2. Condition Monitoring On Live Data3. Taking Analytics To Embedded Device3

Success Story 1: Daimler - Data Driven Fuel Cell Vehicle DesignChallenge Understand vehicle usage patterns Plan hydrogen refueling infrastructure Understand how driving patterns affect vehicleperformanceSolution Connect to data using Database ToolboxVehicle health & troubleshootingPlanning hydrogen fuel-station locations Use MATLAB to explore data and identify insights Visualize data on charts and maps and share viaautomated reports and web applicationsResultsOptimized engine control systemsbased on how people drive Millions of miles of drive files translated intomeaningful insights4

Success Story 2: Safran Online Engine Health Monitoring Solution Monitor Systems– Detect failure indicatorsPerformance– Predict time to maintenance- Modular analysis- Thermodynamic cycleGeneral– Identify components - Anomaly detection- Decision support- Fleet monitoringControl System- Sensors- Actuators- Troubleshooting assistance- Errors and WarningsImprove Aircraft Availability– On time departures and arrivalsOil System- Debris- Smart filter- Consumption– Plan and optimize maintenance– Reduce engine out-of-service time Fuel SystemReduce Maintenance Costs- Smart filter- Fuel pumpMechanical Health- Imbalance- Vibration- Transient events– Troubleshooting assistance (isolate faulty element)Liftoff– Limit secondary damage- Tracking- Monitoring ignitionCompiledSharedDesktop Ad-hoc data analysisAnalytics to predict failure Suite of MATLAB AnalyticsShared with other teamsProof of readinessEnterpriseIntegration Real-time analyticsIntegrated with maintenanceand service ferences/matlab-virtual-conference/5

Success Story 3: iSonea Cloud and Embedded AnalyticsChallenge Develop an acoustic respiratory monitoring system forwheeze detection and asthma managementSolution - Analytics in cloud and embedded Captures 30 seconds of windpipe sound and processesthe data locally to clean up and reduce ambient noise Invokes spectral processing and pattern-detectionanalytics for wheeze detection on iSonea server in thecloud Provides feedback to the patient on their smartphoneResults Eliminates error-prone self-reporting and visits to thedoctor6

AeronauticsRailway rognosticsPredictiveMaintenanceRetail AnalyticsFleet AnalyticsIndustrial AutomationOperationalAnalyticsProcess AnalyticsInternetRisk AnalysisSupply ChainHealth MonitoringLogisticsMfg Process AnalyticsAsset AnalyticsOil & GasHealthcare AnalyticsClean EnergyMedical DevicesHealthcareManagement7

Example 1: Predictive Maintenance of Turbofan EngineBackground: Sensor data from 100 engines of the same model The manufacturer recommends that we perform maintenance afterevery 125 flightsQuestions: Are we wasting money by doing maintenance more often than needed? Is there a better way to identify when servicing is needed so we can besmarter about scheduling our maintenance.Alarm?WarningNormal Data provided by NASA PCoE ata-repository/8

Example 2: Sensor Data Analysis and et courtesy of:Davide Anguita, Alessandro Ghio, Luca Oneto, Xavier Parra and Jorge L. Reyes-Ortiz.Human Activity Recognition on Smartphones using a Multiclass Hardware-Friendly Support Vector Machine.International Workshop of Ambient Assisted Living (IWAAL 2012). Vitoria-Gasteiz, Spain. Dec 2012http://archive.ics.uci.edu/ml/datasets/Human Activity Recognition Using Smartphones9

Data Analytics Workflow: Data Acquisition10

Data Analytics Workflow: Data AcquisitionServers and DatabasesHardwareCJavaFortranPythonSoftware11

Data Analytics Workflow: Data AcquisitionServers and DatabasesStructuredUn-structured.12

Data Analytics Workflow: Data Analytics13

Example 1: Predictive Maintenance of Turbofan EngineBackground: Sensor data from 100 engines of the same model The manufacturer recommends that we perform maintenance afterevery 125 flightsQuestions: Are we wasting money by doing maintenance more often than needed? Is there a better way to identify when servicing is needed so we can besmarter about scheduling our maintenance.Alarm?WarningNormal Data provided by NASA PCoE ata-repository/14

Why perform predictive maintenance? Example: faulty braking system leads towindmill disaster– https://youtu.be/-YJuFvjtM0s?t 39s What could have caused this?– No scheduled maintenance OR– Edge case scenarios might not taken intoaccount OR– Anything else Things under control:– Carry on maintenance15

Types of Maintenance Reactive – Do maintenance once there’s a problem– Example: replace car battery when it has a problem– Problem: unexpected failures can be expensive and potentially dangerous Scheduled – Do maintenance at a regular rate– Example: change car’s oil every 5,000 miles– Problem: unnecessary maintenance can be wasteful; may not eliminate all failures Predictive – Forecast when problems will arise– Example: certain GM car models forecast problems with the battery, fuel pump, andstarter motor– Problem: difficult to make accurate forecasts for complex equipment16

Data Analytics Workflow: Data Analytics17

Monitoring Equipment Health We have clean data. How can we usethese signals to determine if theequipment is in normal conditions?– Control Charts Challenge:– Number of signals -14– Difficult to say when do we have a problem Is 1 sensor going outside the bounds for 1 point aproblem?5 sensors for 3 points?10 sensors for 20 points?– Control charts become difficult to use in thesecases, so we will bring in dimensionreduction techniques to help us.18

Data Analytics Workflow: Data Analytics19

Principal Components Analysis – what is it doing?20

Summary: Data Analytics for Predictive Maintenance of TurbofanEngine“Yes, theseengines indeedneededmaintenance”Maintenance engineer21

Example 2: Sensor Data Analysis and sificationDataset courtesy of:Davide Anguita, Alessandro Ghio, Luca Oneto, Xavier Parra and Jorge L. Reyes-Ortiz.Human Activity Recognition on Smartphones using a Multiclass Hardware-Friendly Support Vector Machine.International Workshop of Ambient Assisted Living (IWAAL 2012). Vitoria-Gasteiz, Spain. Dec 2012http://archive.ics.uci.edu/ml/datasets/Human Activity Recognition Using Smartphones22

Why Machine Learning?23

Machine Learning WorkflowTrain: Iterate till you find the best model using historical SSIFICATIONREGRESSIONPredict: Integrate trained models into applicationsNEWDATAPREDICTION24

Different Types of Machine LearningType of LearningCategories of AlgorithmsClassification Output is a choice between classes (True, False) (Red, Blue, Green)SupervisedLearningMachineLearningDevelop predictivemodel based on bothinput and output dataUnsupervisedLearningRegressionClustering Output is a real number (temperature,stock prices). Discover a good internal representation Learn a low dimensional representationGroup and interpretdata based onlyon input data25

Different Types of Machine Mixed effects)Neural NetworksDecision TreesEnsemble MethodsSupport VectorMachines (SVM)DiscriminantAnalysisNaive BayesNearest Neighbor26

Summary: Machine learning for Sensor Data ClassificationAccessPreprocess DataPredictive/LearningModelShare/Deploy K-Mean clustering Naïve Bayes SVM Classification Trees KNN Neural Networks Evaluation metrics 𝐴𝑐𝑐𝑢𝑟𝑎𝑐𝑦 𝑅𝑂𝐶𝑇𝑃 𝑇𝑁𝑇𝑃 𝑇𝑁 𝐹𝑃 𝐹𝑁27

Learn Further: MATLAB for Machine LearningGo to MATLAB Help Functions Classes Examples and How-To Concepts28

Data Analytics Workflow: Analytics Integration29

Integrate analytics with your enterprise systemsMATLAB Compiler and MATLAB CoderMATLAB Compiler & SDKMATLAB Coder.exe.lib.dll30

Summary: Data Analytics WorkflowBusinessSystemsSmart ConnectedSystemsData Acquisition Engineering, Scientific, and Field Business and TransactionalData Analytics Data Pre-processing Feature Extraction Building algorithms, math models Making business decisionsAnalytics Integration Integrate algorithms with IT Analytics run on Embedded targetsMATLAB: Single Platform31

Key Takeaways: Data Analytics with MATLABNo need to be an expert in everything and if you can still develop & test faster! Direct access to sensors/HW (& aggregators) Integrated workflow from a single environment– Access Rapid/Iterative Analysis Deploymentellipfilterrmsperiodogramxcovfindpeaks parfor Leverage parallel computing to scale-up your analytics to large datasets Eliminate need to recode by deploying/embedding algorithms into sensors or doopC/C Java.NET.pyWeb/ApplicationServerC/C 32

MathWorks Services Consulting– Integration– Data analysis/visualization– Unify workflows, models, datawww.mathworks.com/services/consulting/ Training– Classroom, online, on-site– Data Processing, Visualization, Deployment, Parallel Computingwww.mathworks.com/services/training/33

Questions? 2016 The MathWorks, Inc.34

2 Data Analytics Workflow Data Analytics Data Pre-processing Feature Extraction Building algorithms, math models Making business decisions Smart Connected Systems Business Systems Analytics Integration Integrate algorithms with IT Analytics run on Embedded targets Data Acquisition Engineering, Scientific, and Field Business and Transactional

Related Documents:

19 MATLAB Excel Add-in Hadoop MATLAB Compiler Standalone Application MATLAB deployment targets MATLAB Compiler enables sharing MATLAB programs without integration programming MATLAB Compiler SDK provides implementation and platform flexibility for software developers MATLAB Production Server provides the most efficient development path for secure and scalable web and enterprise applications

Bruksanvisning för bilstereo . Bruksanvisning for bilstereo . Instrukcja obsługi samochodowego odtwarzacza stereo . Operating Instructions for Car Stereo . 610-104 . SV . Bruksanvisning i original

Compiler MATLAB Production Server Standalone Application MATLAB Compiler SDK Apps Files Custom Toolbox Python With MATLAB Users With People Who Do Not Have MATLAB.lib/.dll .exe . Pricing Risk Analytics Portfolio Optimization MATLAB Production Server MATLAB CompilerSDK Web Application

3. MATLAB script files 4. MATLAB arrays 5. MATLAB two‐dimensional and three‐dimensional plots 6. MATLAB used‐defined functions I 7. MATLAB relational operators, conditional statements, and selection structures I 8. MATLAB relational operators, conditional statements, and selection structures II 9. MATLAB loops 10. Summary

MATLAB tutorial . School of Engineering . Brown University . To prepare for HW1, do sections 1-11.6 – you can do the rest later as needed . 1. What is MATLAB 2. Starting MATLAB 3. Basic MATLAB windows 4. Using the MATLAB command window 5. MATLAB help 6. MATLAB ‘Live Scripts’ (for algebra, plotting, calculus, and solving differential .

MATLAB tutorial . School of Engineering . Brown University . To prepare for HW1, do sections 1-11.6 – you can do the rest later as needed . 1. What is MATLAB 2. Starting MATLAB 3. Basic MATLAB windows 4. Using the MATLAB command window 5. MATLAB help 6. MATLAB ‘Live Scripts’ (for

Lecture 14: MATLAB I “Official” Supported Version in CS4: MATLAB 2018a How to start using MATLAB: CS Dept. Machines - run ‘cs4_matlab’ Total Academic Handout (TAH) Local Install - software.brown.edu MATLAB Online (currently 2019a) - matlab.mathworks.com Navigating the Workspace (command window, variables, etc.) Data types in MATLAB (everything is a 64-bit .

I. Introduction to Programming Using MATLAB Chapter 1: Introduction to MATLAB 1.1 Getting into MATLAB 1.2 The MATLAB Desktop Environment 1.3 Variables and Assignment Statements 1.4 Expressions 1.5 Characters and Encoding 1.6 Vectors and Matrices Chapter 2: Introduction to MATLAB Programming 2.1 Algorithms 2.2 MATLAB Scripts 2.3 Input and Output