HARMAN Quick Predict was developed in collaboration with Intel® intellectual property and design expertise

 

On November 1, 2016, HARMAN Connected Services announced an Industrial Internet of Things (IoT) solution at the China International Industry Fair (CIIF) in Shanghai. The solution, HARMAN Quick Predict, was engineered by the Mobile and Communications Services team in collaboration with Intel and relies on Intel’s intellectual property and design expertise. HARMAN Quick Predict is an end-to-end Industrial IoT solution that provides early detection of problems with rotating equipment in industrial settings.

 

Engineering for Industrial Automation

The solution was originally designed by Intel for its fabrication plants to improve maintenance and uptime. Intel engaged with HARMAN Connected Services to convert its vision and intellectual property into an operational product. Once the solution was developed, deployed, and had demonstrated results, Intel and HARMAN entered into an agreement to jointly take the solution to market. The announcement and demonstrations in Shanghai at the Intel booth this week were a direct result of the joint go-to-market efforts.

 

Maintenance of rotating equipment that powers pumps, blowers, and fans is expensive and resource-intensive for industrial users. In continuous operations, organizations need a spare stocked due to poor reliability so they may quickly repair those components. Even with a full stock of spares, a pump failure on a line can cause costly production delays leading to emergency work orders and hurried scheduling of maintenance crews. Manual vibration readings collected, under preventative maintenance programs, on a weekly or monthly basis by technicians simply do not provide the data needed to identify all problems early enough to allow for planned repair.

 

Unlike other systems, HARMAN Quick Predict generates predictions based on real-time signature analysis instead of historical data collected over time. The early detection of potential breakdowns is based on analysis of abnormal vibration patterns. The problems are then flagged over email, text, or control system alarms and displayed in an easy-to-use, web-based interface to the responsible team leads.

 

Benefits to Industrial Users

Predictive maintenance is desired in industrial settings. HARMAN Quick Predict is a major step forward in Industrial IoT and can provide substantial savings for end users by reducing process downtime and repair costs. The system is completely scalable and customizable for industrial users.

 

Averting a single catastrophic failure would pay for the cost of the solution many times over. If maintenance departments could communicate in advance to operations about when a rotating equipment needed to be taken offline for repair – while also being able to plan the duration and type of desired equipment repair – there would be substantial financial benefits resulting from:

  • Reduced unplanned downtime
  • Reduced preventative maintenance
  • More efficient use of maintenance crews
  • Improved capital equipment utilization
  • Fewer costly repairs
  • Smaller spare parts inventory

 

The Value of Real-time Data and Edge Analytics

HARMAN Quick Predict collects the high-resolution vibration data needed to detect problems early and provides learning analytics that help map abnormal vibration and rotation speed patterns associated with failures. Vibration sensors are closely coupled with a HARMAN data acquisition board and intelligent HARMAN gateway with edge analytics. The data processed in the gateway is communicated to the cloud, where the learning algorithms reside. This architecture allows HARMAN Quick Predict to deliver the benefits of 24/7 continuous monitoring without requiring the full cost of networking, storage, and data center infrastructure.

 

By implementing statistical machine learning algorithms on gateways at the edge, the solution greatly reduces the amount of data that must be transmitted to the cloud. Continuous monitoring of all rotating equipment in a manufacturing facility would result in terabytes of data per day. By leveraging edge computing, the amount of data transmitted, stored, and analyzed is reduced by more than 99.9 percent to only megabytes per day. This is achieved by real-time analysis of time-series data on the gateway. Data can be sent from the gateway using hardwired connections into existing control systems, or to the cloud through Ethernet, Wi-Fi, or cellular networks, depending on end user preference.

 

iot-vibration-analytics.jpg

 

The system can also be configured to send alerts and show only the data that indicates an issue, saving the organizations significant time and effort. The solution may be integrated with specified sensors, data acquisition boards, gateways, and server hardware to provide customer solutions.

 

The Business Case in Industrial IoT

Research and deployments have shown that high-resolution vibration data collected continuously provides ample information to predict many types of failures early enough to schedule repair. HARMAN Quick Predict was designed to serve as a robust solution for high-resolution, 24/7 data collection and analysis with a simple user interface to allow industrial users improved predictability in their maintenance regimen. 

 

We invite you to learn more on how HARMAN Quick Predict delivers that level of predictability in industrial settings and discuss your maintenance concerns with us.