PROCESSING RELIABILITY with automatic diagnosis algorithm

Intelligent condition monitoring to everyone

Artificial intelligence
to enhance analyst intelligence

More value to your business:

Asensiot Algorithm

Screening power to your condition monitoring data

The algorithm processes vibration-based condition monitoring data as “invisible buddy” and screens out the sophisticated alarms.

Asensiot FRT

Practical first response tool to pinpoint where the analysts should focus their effort

The FRT™ First Response Tool is a responsive web-based application with features to support vibration analysts.

Asensiot ForeC

Fully scalable IIoT asset health monitoring concepts

ForeC Electric Motors
ForeC Pumps
ForeC Fans
ForeC Gears (ready during 2021)

Asensiot IIoT Testing

Independent partner to evaluate the performance of IIoT technology

Special field test program to ensure the tested IIoT tecnology meets the requirements for the realiable and scalable automatic results.

Independent

Asensiot Algorithm

Ready-made models provide immediate results and fast scalability

Ready-made models provides an unique basis and easier scalability for any AI endeavours.

Data science cannot address the actual problem to be solved without application expertise.

Asensiot AI is the best combination of Asensiot’s theoretical and real world expertise with data science in the unique algorithm.

Machine learning is a good tool to train AI forward in terms of coverage and performance.

Practical

Asensiot FRT

Practical tools to support analysts to focus on what really matters

The vibration-based condition monitoring data from rotating machines as such has no value. Only correctly measured data interpreted as comprehensible results brings real added value. FRT™ provides added value through efficient data screening i.e. first response. This allows the end user to react on failures earlier than ever before.

Verified data collectors for FRT:
Prüftechnik Vibscanner 2
Prüftechnik Vibxpert II
SKF Microlog*
Emerson CSI*

*With recommended accelerometer and magnet combination

Scalable

Deep understanding of data

The time waveform is the truth of machine´s vibration behaviour.

SPECIFICATION OF REQUESTED MINIMUM DATA QUALITY:
Raw data from all verified condition monitoring systems:
Acceleration time waveform min. 1.28 s (5.12 s) / 12.8 kHz
Bearing or rpm information not needed

Raw data

Acceptable | defect not started

Warning | 2-9 months before failure

Danger | 1-2 months before failure

Realiable data is a source for the realiable automatic results in condition monitoring

Vibration data is digitalized version of machine vibration, therefore it is essential to collect data using verified technology.

It is a well known fact that most of the automatic solutions are failing due to:

  • Lack of data
  • Fault condition classification is incomplete
  • Noise created by measurement systems
  • Poor data quality
  • Noise created by human
  • Lack of knowledge
  • Technology is chosen to solve secondary problem

Our competency is to help and define the optimal setup for the data collection to ensure ability for early stage fault detection – automatically.

WE KNOW HOW IT WORKS!

01 I DATA COLLECTION
Asensiot algorithm processes the acceleration raw data from all systems with sufficient performance and enables more independent condition monitoring program. The algorithm is system independent.

02 I AUTOMATIC FIRST RESPONSE
The algorithm screens with more than 99% accuracy the machines that don’t need any attention. This is what we call first response. By using the first response, concrete work can be focused on real problems.

03 | SUMMARY OF MACHINE FAILURES
In addition to machines in good condition, Asensiot solution provides a first response summary of fault conditions without rotational speed and bearing information with more than 85% accuracy. The results can be integrated into external systems.

Fast

NEW AI IIoT CONCEPT!

Asensiot ForeC

Electric Motors, Pumps and Fans

Potential: Electric motors represents approx. 60-80 % of all fault conditions of all rotating machines. 

Approx. 60-80 % of detected fault conditions in electric motors are in early stage, 6-12 months before potential failure. 

Typical early stage fault conditions: Inadequate lubrication, bearing fault early stage 

Data collection: only with verified* wireless IIoT 5-10 kHz sensors or route measurements. 

Horizontal DE (> 100 kW also NDE), 3-axis 

Results: Machine health status and fault condition. Intelligent algorithm-based change detection and temperature.  

Solution value: Comprehensive automatic asset health monitoring detects faults in the early stage, enabling better predictability. This helps optimize lifecycle services and improves cost efficiency as well as less risk to unexpected production downtime.

Scalability: Also to pumps and fans

*Verified by Asensiot IIoT field test prochedure

Reliable

Example cases with the algorithm

Electrical motors 202 pcs
First response acceptable 99 %
First response fault types 95 %

Pumps, fans and electrical motors 255 pcs ​
First response acceptable 99 %​
First response fault types 94 %

Pumps, fans and electrical motors 59 pcs
First response acceptable 99,9 %
First response fault types 98 %​​

Pumps, fans and electrical motors 144 pcs ​
First response acceptable 99 %​
First response fault types 87 %

Asensiot Value

Numerous options to integrate with existings systems using REST API

 A system free approach that provides fast,  reliable and scalable automatic results to all users willing to take a leap towards the automatic asset health monitoring in the future.

Verified

Asensiot IIoT Testing

Asensiot Live

Asensiot technology is chosen to the Nordic service concept NiVia from MLT Machine & Laser Technology

MLT NIVIA IN A NUTSHELL:
  • Over 4000 monitored machines
  • You measure we analyze
  • Immediate first response summary 
  • Specialist supported recommended actions
  • No hidden costs – you pay only for the results
  • MLT delivers route analyzers and data collectors, online systems and startups/trainings

Mitattavan raakadatan määrä tuplaantuu digitalisaation myötä joka toinen vuosi, mutta pelkkä data ei ole itseisarvo. Vasta oikein mitattu data ymmärrettäviksi tuloksiksi tulkittuna tuottaa todellista lisäarvoa. Tekoälyn hyödyntäminen kunnonvalvonnassa on uusi tehokas tapa säästää rahaa ja resursseja. Nivia Powered by FRT on pohjoismainen palvelukonsepti, joka …(lue lisää tästä)

Käynnissäpitoon liittyy usein oleellisesti kunnonvalvonta ja diagnostiikka. Eräs yleisimmistä tässä yhteydessä käytetyistä menetelmistä ovat värähtelymittaukset, jotka toteutetaan nykyään yleensä kiihtyvyysantureilla. Lähes jokainen meistä kantaa mukanaan yhtä tai useampaa kiihtyvyysanturia – esimerkiksi mobiililaitteissa – mutta tuskin koskaan tulee pohtineeksi, sopisiko tämä hyvin arkipäiväsessä käytössä oleva tekniikka myös käynnissäpitosovelluksiin.(lue lisää tästä)

Asensiot Proof of Concept

1. Verified data quality
2. Verified data collection
3. Automatic data processing

RESULTS ON THE SAME DAY!