2. Need for Vibration Analysis
Vibration measurement is
operated for a sample of mass-
produced or all completed
products to eliminate inferior
pieces and produce satisfactory
ones
The production system is kept in
operation by detecting the
deterioration or possible failure of
a part through adverse vibration
patterns
Predictive Maintenance Improve Quality of
Products
Research and
Development
Develop and design more reliable
products. Vibration measurement
is operated in testing samples or
finished products
3. Need for Vibration Analysis
To prevent the breakdown and
trouble from machinery,
vibration is measured
periodically and compared to
the vibration at initial
condition or at normal
condition after the
maintenance for judging the
deteriorate of equipment
Safety Management in
Factories
Development
& Design
Production
Check &
Quality
Shipping
4. Benefits of Vibration Analysis
Reduce Equipment Costs
Instead of replacement of the entire piece of equipment due to critical
failure, a repair is made prior to failure, and cost is minimized to the
price of the component and the labor needed for the repair.
Reduce Labor Costs
When repairs are scheduled, the amount of time needed for repair is
reduced because of a smaller number of component replacements
instead of entire equipment replacement. Also, the frequency of repair
for critical failure of equipment will be reduced and the amount of
"critical callouts" will be greatly reduced.
Reduce Lost Production Time
Component-only replacement is scheduled with production to take
place during scheduled downtime. Unscheduled downtime may cost
thousands of dollars per hour. A proactive maintenance department
can head off critical failure downtime by scheduling repair during
nonproductive times.
5. Benefits of Vibration Analysis
Increase Safety
Predictive maintenance allows potential problems to be fixed before
failure occurs, which would create safer driving conditions for
employees and customers.
Increase Revenue
With less maintenance on good components and quicker repair of
faulty components, repairs can be more effectively handled, thereby
reducing repair time.
Increase Efficiency of Employee
Time
Identifying the precise repair task needed to correct deficiencies, as
well as the parts, tools, and support needed to correct the problem,
can dramatically increase effective "wrench time."
6. High Level Solution Overview Components
BellaDati IoT Platform
IoT Collector
Data Analytics & Visualisations
Machine Learning Module
REST API & SDK – Integration of solution in
existing system, customizations
Client API – Custom console for your industry
Value Proposition
Device agnostic platform for collecting and analyzing
the machine data
Manage million of device inputs and multiple data
sources
Complete overview of machines
Efficient production line
Higher quality of products
Sensor Technology IoT Gateway IoT Platform Data Analytics & ML
Predictive
Maintenance
Increased Uptime
8. Sensors for Vibration Analysis
Goal: Measuring the vibration of rotating machinery
Sensor: Accelerometer
Key Parameters:
Bandwidth – how many records can sensor measure per second (in Hz)
Sampling Frequency – how many records can contain the output from the sensor
Number of Axis – 1-axis or 3-axis
Mounting Options – stud mount, adhesive, magnetic, handheld or probe tips
Durability (e.g. IP67, IP65, IP61, etc.)
Charging (cable, battery)
Connection (UDP, BLE, etc.)
9. Accelerometer Insights
Transform raw sensor data into insights
Algorithm can run on:
Sensor
Gateway
Backend
Comparison of actual values with the normal status
Maintenance planning
Machine status monitoring
11. IVAS | Intelligent Vibration Analysis Sensor
3-axis vibration sensor up to 3 kHz
identify main axes of vibration.
mounting of IVAS along main vibration axes
vibration sensor up to 20 kHz
pick up vibro-acoustic signals
structure born sound up to threshold of audibility
onboard micro controller
Timely handling of vibro-acoustic data
Fast-Fourrier Tranformation
Real-time handling of large data amounts
Ethernet connection to gateway
12. Use Cases for Intelligent Vibration Analysis Sensor
Tele-Maintenance
Wear Detection
Predictive
Maintenance
Process Monitoring
Monitoring of remote
production equipment
and tools
Reduction of downtimes
by spare part planning
Monitor health status of
equipment and tools like
motors, pumps, drives,
linear sledges …
Monitoring of process
parameters and
equipment conditions for
quality control
Use Case
Unknown health status of
remote equipment,
maintenance staff must
be on-site
High costs due to
unforeseen downtimes
and long spare part
delivery times
Cost intensive time based
preventive maintenance
Improve quality of goods,
reduce process parameter
variation, avoid low yield,
high scrap rate
Problem to Solve
Sensor based Cyber-
physical system enabling
tele-maintenance
Increase machine uptime
and OEE with IVAS
Condition based
Predictive maintenance
by means of structure
born sound analysis
Early detection of process
deviations
Possible Benefits of
IVAS
13. Typical Bath Tub Curve of Failure Rate of Equipment Lifetime
early signal:
pattern change in vibro-acoustics / structure born sound
early signal:
increased vibrations
increase in current consumption
increase in temperature
equipment
failure / downtime
cost / risk
asset condition
Proactive, preventive predictive reactive
14. Typical Use Cases for Different Bandwidth Levels
1E+0Hz 1E+1Hz 1E+2Hz 1E+3Hz 1E+4Hz 1E+5Hz 1E+6Hz
Equipment Vibration
1 – 10 Hz
Typical period for timely
unfluctuating data like
temperature, pressure,
…
300 Hz – 2 kHz
Condition / process
monitoring in
rotating
equipment.
Efficiency point,
balance and
alignment issues.
50 kHz – 1 MHz
Predicting bearing failure and
component wear in specialized
high speed rotating equipment
25 – 100 Hz
Human motions,
gestures, fine
motor skills
2kHz – 10kHz
Predicting
bearing failure
and component
wear in tools
and equipment
vibro-acoustical signals
15. Typical Bath tub Curve of Failure Rate over Equipment Lifetime
Equipment Lifetime (Typically many years)
intervals of
time-based
maintenance
Failure
Rate
16. Examples of Damages to Electro Motors
Vibrations indicate different damages in the operation of electro motors (Source ISO 17359:2018-05):
Rotor winding
Stator winding
Eccentric rotor
Stock harm
Phase input power loss
Unbalance
Alignment
17. Vibration Velocity Limits (DIN ISO 10816-3)
relevant vibration frequencies for damage to
motors
10 – 10 kHz for r > 600 rpm
2 – 1 kHz for r > 1200 rpm
using RMS value to monitor damages
unbalance – 1x frequency of spindle
misalignment – 2x frequency of spindle
damaged bearing / gear box
Vibro-acoustic noise level
foreign particles in bearing
e.g. dust or, corrosion particles
18. Examples of Damages to Pumps
Vibrations indicate different damages in the operation of pumps:
Vane harm
Damaged seals
Eccentric impeller
Stock harm
Bearing wear
Error in storage
Unbalance
Alignment
19. High Bandwidth Vibration Measurement – Use Case
High frequency vibration measurement is needed for bearing wear out
According to ISO 13373-1:2002-07: The monitoring requires sensors for
high-frequency vibration components. The conditions are usually
limited to the surrounding of the damaged stock. The time line of
the vibrations is usually unsteady and increases with time. To
identify the type of damage, additional techniques are required.
And other kinds of machine failure
According to ISO 13373-1:2002-07: high frequencies correspond with the
harmonic of meshing frequencies and side bands with the frequencies of
rotation. Detection requires sensors for high-frequency vibration components.
Defective tooth: one or multiple. Meshing frequency with sideband and higher
harmony
IVAS is mounted on the harvester motor and pump
20. Vibro-Acoustical Data Path
Raw Signal
• Acquisition
• Pre-filtering
Pre-processing
• Detect relevant phases in
time-varying processes
• Find suitable frequency
bands
Feature
extraction
• Calculate features
(time/frequency domain)
Features
selection /
classification /
evaluation
• Which features describe process best?
• Which classifier gives best prediction
accuracy?
• When possible: use low complexity
features and classifiers
Implementation
• Implement selected
frequency bands, features
and classifier
In real-time from the time domain
frequency to the domain with
Fast Fourrier Transformation (FFT)
Sensor embedded feature
processing on built-in
micro controller
21. Vibration Strength Diagram
RMS of acceleration in m/s2
(RMS within 10 Hz to 10 kHz)
Highest
peak
acceleration
in
m/s
2
2 – normal state
1 – unusual low
(repeat measurement)
3 – warning indicator
4 – Alarm
IVAS is capable of measuring
vibration with an amplitude of ±60 g
or ± 588.6 m/s2
Measurements according to the
vibration strength diagram for
spinning machines according to DIN
ISO 13373-3 can be performed with
IVAS
𝑣𝑟𝑚𝑠 =
1
𝑇 0
𝑇
𝑣2 𝑡 𝑑𝑡
22. Intelligent Vibration Sensor - Parameters
Parameter Range Unit
Measurement range ± 60 g
Sensitivity 8 LSB/g
Acceleration data resolution 10 Bit
Bandpass filter selectable 5…15 (20) kHz
Nonlinearity of sensitivity < ±2 %
Tolerance of sensitivity (integr.) < ±13 %
Cross axis sensitivity < +5 %
23. CISS | Connected Industrial Sensor Solution
Connection to CISS: BLE or USB
Sensors:
Accelerometer
Temperature
Humidity
Digital light sensor
Gyroscope
Magnetometer
Pressure sensor
Microphone
Operation modes:
Data streamer
Data logger
Event mode
Time aggregation
Parameters:
Sampling rate:
Streaming accelerometer (other sensor data
are not acquired): 2 kHz
Standard mode: 100 Hz
24. Bosch XDK110
Connection to CISS: BLE or USB
Sensors:
Accelerometer
Acoustic Sensor
Temperature
Humidity
Digital light sensor
Gyroscope
Magnetometer
Pressure sensor
Parameters:
Sampling rate:
Sampling Rate: 2kHz
26. BellaDati IoT Suite
Collecting data from multiple IoT elements in real-time
Device agnostic
Data Security
Data are routed to BellaDati Platform
Custom actions based on the pre-defined thresholds
can be configured
Intelligent notifications (Email/SMS/Phone Call)
Integration with the 3rd party systems (e.g. Call Center
System)
28. BellaDati Platform Overview
Purely web-based technology
All features available in the web browser. No
need to install other applications.
Agile reporting and
dashboarding layer
Content is created and distributed in the real-
time, no publishing servers required.
Ready-made for embedding
Wide range of APIs allow integrating,
embedding or extending your apps.
29. BellaDati Platform Overview
ETL
Machine Learning Studio
IoT Suite
Collect data from unlimited IoT elements in real-
time. Collected data can be routed to 3rd party
systems, you can trigger automatic actions,
thresholds..
Visual framework to interact with data and
analytical results. Mathematical modeling and
predictions. Integration with R and Python.
Combine data from different data sources and
use BellaDati native features for data enriching,
cleansing and transforming.
30. BellaDati Platform Overview
Cloud & On-Premise Versions
BellaDati can be installed as on-premise or
cloud on all standard platforms (Windows or
Linux).
REST API & SDK
REST API and SDK allow you to integrate
seamless analytics to your own application
(web, mobile or standalone).
Client API
Use Client API to extend BellaDati front-end
functions, create various UI component or
incorporate custom charts.
31. BellaDati Platform Overview
Big Data
Use Big Data features for analysis of hundreds
of millions of records or connect to your big
data sources via native connectors for
Hadoop, MongoDB or Cloudera Impala.
Unstructured and Structured
Data Analysis
Turn unstructured data from text or PDF files
into structured form. Use BellaDati features
for sentiment analysis or web crawling.
100+ Data Connectors
More than 100 native connectors for different
data sources are available out of the box.
Connector SDK can be used for the
implementation of your own data connector.
32. BellaDati Platform Overview
Intelligent Notifications
Data-driven notifications automatically send
email, push notification or trigger custom
action in 3rd party system. Data content is
included in the notifications.
Social Network for Business Data
Discovery
Add comments to the values, share the
content with users or publicly, schedule
automatic email exports or download ad-hoc
report and dashboard exports.
BellaDati Multi-Domain
Multiple environments can run on one
hardware. Each environment provides
separate workspace and can be fully
customized.
33. Reports are Created and Changed in Real-Time
Report Creation in real-time
Combine KPIs without coding
Build your own data drill-down
Pivot, Forecast, Math formulas
Easy filters for business users
Create smart notifications
Analyze un-structured data
Share, collaborate with the team
34. Integrations & Customizations
Integration with 3rd party applications
BellaDati IoT Collector to get the data from
other types of sensors
REST API & SDK allow other applications to
directly access data provided by BellaDati.
Through the API, users can retrieve reports,
charts, reports, entire dashboards or even
source data to integrate with their own client
application
Layout customization via CSS styles
Extensions to customize BellaDati Look&Feel
35. BellaDati Forest Harvester Solution
Data from vibration sensors and other data
from the forest harvester are collected.
Data is transferred from the terminal
installed in the forestry machines to
BellaDati IoT Collector
BellaDati iOT collector sent/select data for
sending to BellaDati advanced analytical
platform or (and) to a customer’s existing
platform
Alghoritms to analyze vibrations are run on
IoT collector or on BellaDati platform to
identify exceptions.
Reference: Komatsu
Data from terminals
Belladati IoT Collector
BellaDati anlytics or customer’s existing platform