IoT Collaboration Bridging Edge to Cloud: Inside Petasense & OSIsoft’s Effort to Improve Manufacturing

October 102017
IoT Collaboration Bridging Edge to Cloud: Inside Petasense & OSIsoft’s Effort to Improve Manufacturing

By Bill Lydon, Editor, Automation.com

Reducing downtime has been one of the most relevant, shining examples of how Internet of Things (IoT) technology can have a real impact on today’s factory. One method to accomplish this task is to bridge sensors, from the network edge to the cloud, in order to improve industrial efficiency. This specific effort has beem made real, through a recent partnership between OSIsoft and Petasense, and I had the opportunity to  talk with Arun Santhebennur, the Petasense co-founder, and OSIsoft, Enrique Herrera, Market Principal at OSIsoft, to discuss the application and the architecture.

The aim of this collaboration is to provide customers a complete solution to monitor assets, and predict the health of their machines, in real-time, with Petasense edge devices. The Petasense system includes a patent-pending wireless vibration sensor, cloud software and machine learning (ML) based analysis of vibration patterns.  Ideally, this system will reduce unplanned downtime and lower repair costs. Petasense’s clients already include JLL, C&W Services, Silicon Valley Power and Stanford University.

 

Edge Devices Boost Communication

As part of this effort, Petasense has developed network edge vibration sensors - named MOTES - that are 2 inches in height and 1.25 inch in diameter.  The MOTEs are wireless field devices that have internal antennas to communicate full vibration waveform data to the Petasense Cloud. 

The company currently offers two models:

  • Vibration Mote Model 1 Vibration Sensor for predictive maintenance features a tri-axial MEMS vibration sensor with a 1.6 kHz bandwidth targeted for detecting machine imbalance, shaft misalignment, structural or rotating looseness, pump cavitation and late stage bearing wear.
  •  
  • Vibration Mote 2 features a triaxial piezoelectric vibration sensor with a 5.5 kHz bandwidth. This device is ideal for detecting high frequency defects like early stage bearing defects and gearbox problems in addition to low frequency defects like machine imbalance, shaft misalignment, structural or rotating looseness, pump cavitation and late stage bearing wear. 

 

Inside the Protective MOTE

The Mote has an IP69K ingress protection rating. Its enclosure consists of a polycarbonate cover and a corrosion-resistant, stainless steel base (SS316), making it suitable for use in a wide variety of industrial environments. The units are powered by a standard 3Volt CR123a lithium battery that will last 2 years under normal operating conditions.  MOTEs are mounted on machines by applying industrial steel epoxy or alternatively stud mounted with a ¼” 28 UNF tapped socket on the base of the device.  Further, the units have Bluetooth support for iPhone and iPad for users to activate and configure the MOTE and read on-demand vibration measurements.  The Mote is equipped with TLS 1.0 security. AES encryption keys are stored on microcontroller flash to ensure authorized communication. The device features readback protection (RDP) level 2, protecting encryption keys.  MOTEs can be installed indoor or outdoor and are certified for use in Class 1 Div 2 environments. The devices are certified FCC, CE, UL746C, RoHs and NEMA 250.   The cover material is F1 rated for UV resistance.

Vibration Mote mounted on motor sends vibration data to cloud for analysis.

 

Enhanced Maintenance Through Cloud Analytics

By analyzing vibration characteristics, Petasense software is used to predict common defects like pump cavitation, bearing wear and misalignment. These Machine Learning (ML) algorithms calculate a numerical health score of the machine in real-time, enabling plant managers to make informed maintenance decisions. Even a small plant can generate over 10 million vibration readings daily and Petasense has already logged 25 billion wireless vibration readings across its client base.

Petasense Cloud, operating on the Microsoft Azure public cloud architecture, incorporates the machine learning algorithms to help automate the vibration analysis process. These algorithms set specific baselines and alarm limits for each machine and derive real time machine health scores based on deviations from the baseline. For organizations with trained analysts and established programs, this machine learning software is designed to act as a first line of defense in monitoring machines at the fleet level.

Petasense Cloud integrates with Data Historians, like OSIsoft PI or eDNA, and Enterprise Asset Management (EAM) software from leading vendors like Maximo, SAP or Oracle. This trend vibration data alongside other process data, like pressure, temperature and current, helps to get insights into overall machine and process health.

 

OSIsoft Connected Services partner (CS Partners)

Petasense is an OSIsoft CS Partner. This program is a way for hardware OEMs, software developers and service providers (collectively CS Partners) to add digital services to their products with PI System technology. CS Partners integrate PI System agents into their products to collect operational data, analyze it, and then provide recommendations to their customers so they can improve their operations or reduce costs.

The architecture provides a mechanism for users to have an integrated view of their operations.

Petasense uses the PI System to make its data more visible. Petasense’s sensor collects vibration data and then sends it to its cloud application for signal processing and event analysis. If Petasense detects a problem it sends a signal to a PI System, hosted by OSIsoft. OSIsoft can then send it to the end-user and integrate it within the end-user’s console. Engineers at the end user’s site can see the problem within the context of their operations, rather than as a separate application, and make decisions whether to repair or continue working.

OSIsoft currently has over 70 CS Partners taking advantage of this architecture.

 

Bill’s Thoughts & Observations

This is an excellent example of collaboration leveraging IoT devices, analytics, and cloud services to help users achieve more efficiency and productivity.

 

Related Articles

MORE ARTICLES

VIEW ALL

RELATED