1© 2015 The MathWorks, Inc.

Big Data and Machine Learning

for Predictive Maintenance

Paul Peeling

2

Agenda

▪ The Predictive Maintenance Opportunity

▪ Exploring Big Data

▪ Machine Learning Approaches

▪ Deep Learning

▪ Fault Modelling

▪ Deploying to the Edge and the Cloud

3

Aaron “tango” Tang on Flicker

React or Prevent?

4

Predictive Maintenance software

Sense

Perceive

Decide & Plan

Act

Temperature

sensors

Pressure

sensors

Vibration

sensors

Total of 25 sensors - but which ones were the best predictors?

5

Predictive Maintenance software

Sense

Perceive

Decide & Plan

Act

6

What do we mean by Predictive Maintenance?

▪ Monitor equipment to avoid future failure.

▪ Schedule maintenance when it’s needed.

▪ Identify the root cause of issues.

▪ How?

– Predictive models and sensor data.

– Deploying to the equipment and cloud.

7

Why is Predictive Maintenance Important?

8

Why is Predictive Maintenance Important?

Source: GE Oil & Gas

9

Deploying Predictive Maintenance Algorithms

10

Aside: What if … ?

▪ I’m not in the business of Predictive Maintenance

▪ I don’t have big data

▪ I don’t have any data

▪ I don’t have a computing cluster

▪ I need a simpler solution

11

Integrate Analytics

with Systems

Enterprise Scale

Systems

Embedded

Devices/Hardware

Files

Sensors

Access and

Explore Data

Develop

Predictive Models

Machine learning

Model

Validation

Preprocess Data

Visualizing

Data

Data Reduction/

Transformation

Workflow

12

Files

Sensors

Access and

Explore Data

Workflow

13

Sensor data from 100 engines of the same model

– Maintenance scheduled every 125 cycles

– Only 4 engines needed maintenance after 1st round

Predict and fix failures before they arise

– Import and analyze historical sensor data

– Train model to predict when failures will occur

– Deploy model to run on live sensor data

– Predict failures in real time

Predictive Maintenance of Turbofan Engine

Data provided by NASA PCoEhttp://ti.arc.nasa.gov/tech/dash/pcoe/prognostic-data-repository/

14

Working with Big Data

Where is the data?

What code can I write?

How big is the data?

15

Machine

Memory

Tall ArraysScaling your code to big data

▪ Automatically optimize data access bottlenecks

– Write code the same way you've always written it

– MATLAB automatically reorders operations to

minimize disk access

▪ Applicable when:

– Data is columnar – with many rows

– Overall data size is too big to fit into memory

– Operations are mathematical/statistical in nature

▪ Statistical and machine learning applications

– Hundreds of functions supported in MATLAB and

Statistics and Machine Learning Toolbox

Tall Data

16

Filtering Data

17

Integrate Analytics

with Systems

Enterprise Scale

Systems

Embedded

Devices/Hardware

Files

Sensors

Access and

Explore Data

Develop

Predictive Models

Machine learning

Model

Validation

Preprocess Data

Visualizing

Data

Data Reduction/

Transformation

Workflow

18

Visualizing Big Data Using tall

▪ Support for:

– histogram

– histogram2

– ksdensity

– plot

– scatter

– binscatter

19

Visualizing Big Data Using tall

scatter binscatter

20

Standardizing Data

21

Deferred evaluation and gatheringWhat does “gather” do?

1. Evaluate any pending

operations

2. Collect the partitioned data

into MATLAB main memory

3. Unwrap the data into an array

or table

22

Integrate Analytics

with Systems

Enterprise Scale

Systems

Embedded

Devices/Hardware

Files

Sensors

Access and

Explore Data

Develop

Predictive Models

Machine learning

Model

Validation

Preprocess Data

Visualizing

Data

Data Reduction/

Transformation

Workflow

23

Traditional Approaches

24

Use historical data to predict when failures will occur

?

His

torical

Liv

e

Engine 1

Engine 2

Engine 100

Initial Use/

Prior Maintenance

Cycles

(Time)

Engine X

Recording Starts Failure Maintenance

Schedule Maintenance

⁞

?

25

Principal Components Analysis

26

Dimensionality Reduction with PCA

27

Early Warning System

28

Start of

Engine Life

Preprocessing and Classifying our Input Data

Engine 1

Engine 2

Engine 3

Start of

Recorded Data

Cycles

(Time)

Engine 100

Recording Starts

Failure

⁞

Cycle 0

29

Classification Learner App

30

Convolutional Neural Network

31

Pretrained Networks

32

LSTM Network

33

Useful Life Estimation Simulink Model

34

Integrate Analytics

with Systems

Enterprise Scale

Systems

Embedded

Devices/Hardware

Files

Sensors

Access and

Explore Data

Develop

Predictive Models

Machine learning

Model

Validation

Preprocess Data

Visualizing

Data

Data Reduction/

Transformation

Workflow

35

Internet of Things

36

▪ Run in parallel on Spark clustersMATLAB Distributed Computing Server

▪ Deploy MATLAB applications as standalone applications on Spark clustersMATLAB Compiler

▪ Run in parallel on compute clustersMATLAB Distributed Computing Server

▪ Tall arraysMATLAB

▪ 100’s of functions supportedMATLABStatistics and Machine Learning Toolbox

▪ Run in parallelParallel Computing Toolbox

Using Tall Arrays

Spark + Hadoop

Compute ClustersLocal diskShared folders

Databases

37

Working with GPU Coder: Deep Learning Workflow

Access Data Preprocess Select Network Train

Image

Acquisition

Tbx

Image

Processing Tbx

Computer

Vision System

Tbx

Neural

NetworkParallel

Computing Tbx

GPU

Coder

Deploy

38

Machine Learning on MATLAB Production Server

Shell analyses big data sets to

detect events and abnormalities at

downstream chemical plants using

predictive analytics with MATLAB®.

Multivariate statistical models

running on MATLAB Production

Server™ are used to do real-time

batch and process monitoring,

enabling real-time interventions

when abnormalities are detected.

39

Where Next?

Talks

▪ MatConvNet: Deep Learning

Research in MATLAB

▪ Introduction to Machine & Deep

Learning

▪ Scaling MATLAB for your

Organisation and Beyond

Demo Stations

▪ Big Data with MATLAB

▪ Deep Learning with MATLAB

▪ Predictive Maintenance with

MATLAB and Simulink

▪ Deploying Video Processing

Algorithms to Hardware

▪ Using MATLAB and ThingSpeak

to Explore the Internet of Things

40

Thank You!