Real-Time Data Acquisition, Monitoring, and Control · Real-Time Data Acquisition, Monitoring, and Control Developing better business outcomes through condition monitoring and edge

Real-Time Data Acquisition, Monitoring, and Control Developing better business outcomes through condition monitoring and edge computing

Technical white paper

Technical white paper

Contents Solution Overview ............................................................................................................................................................. 2

Edge Computing using HPE Converged Edge Systems..................................................................................................... 4

HPE Edgeline EL1000 Converged Edge System ............................................................................................................ 4

HPE Edgeline EL4000 Converged Edge System ............................................................................................................ 5

Remote Management using HPE Integrated Lights-Out (iLO) Server Management™ ..................................................... 6

Data Acquisition ................................................................................................................................................................ 8

CompactDAQ™ ............................................................................................................................................................. 8

CompactRIO™ .............................................................................................................................................................. 9

Direct PXI/PXIe ............................................................................................................................................................ 10

NI LabVIEW™ ............................................................................................................................................................. 11

Data Analysis using National Instruments Software ......................................................................................................... 13

National Instruments InsightCM™ Enterprise ............................................................................................................... 13

National Instruments DIAdem™ and DataFinder Server Edition™ ................................................................................ 13

Conclusion ...................................................................................................................................................................... 16

References ..................................................................................................................................................................... 17

Learn more at ................................................................................................................................................................. 17

Technical white paper 2

Solution Overview Terabytes of data from “connected” systems are constantly being generated out on the edge. Gaining knowledge from this data is key in driving better business outcomes through IoT. For example, most of the world’s power is generated through turbines and rotating equipment. The ability to gain insight from rotating equipment can lower operational and maintenance costs associated with equipment failures. This and dozens of other use cases point to the need for processing horsepower and storage at the network edge for real-time insights that were previously only available from data center-based compute platforms.

IDC1 predicts that by 2019 at least 40% of IoT-created data will be stored, processed, analyzed, and acted upon close to, or at the edge of, the network. In so doing, streaming analytics will merge with machine learning and be trained on data lakes, marts, and content stores, accelerated by discrete or integrated processors. Companies are now transforming their businesses by leveraging IoT to accelerate business by transforming data into insight (Moor Insights & Strategy, 2016).

1 IDC FutureScape: Worldwide Internet of Things 2017 Predictions, Carrie MacGillivray, VP, IoT & Mobility, IDC Web Conference, November 8, 2016

As data generated from smart connected “things” such as rotating equipment grows and the frequency at which it is collected increases, so will the need for systems that can acquire and analyze the data in real-time. Real-time analysis is enabled through edge computing. HPE Edgeline Converged Edge Systems converge real-time analog data acquisition with data center level computing and manageability, all within the same ruggedized chassis. This document defines the role of HPE Converged Edge Systems in real-time data acquisition, monitoring, and control for IoT applications.

HPE and National Instruments have collaborated to combine real-time data acquisition with edge analytics to meet the needs of IoT-created data. HPE Converged Edge Systems include multiple features to enable data acquisition, analysis, visualization, and access within the same unified platform, all in real-time in order to design, test, and deploy an IoT solution. National Instruments data acquisition systems, PXI technology, and analysis software will be described in detail at each stage as they are used within the solution. An example of this solution will also be described using industrial rotating equipment with the use case of providing real-time vibration monitoring. There were multiple data acquisition methods tested in this example varying from NI cDAQ and RIO chassis for remote data acquisition as well as PXI/PXIe for direct data ingest on HPE Converged Edge Systems.

The following is a solution example to analyze data and gain insight from rotating equipment. This process starts with the instrumentation of the rotating equipment and connectivity. During instrumentation, sensors collect and forward data from the rotating equipment through various methods to data acquisition devices. Once data is acquired, it is accessed through edge systems for analysis and monitoring of the equipment.

“By 2019 at least 40% of IoT-created data will be stored, processed, analyzed, and acted upon close to, or at the edge of, the network.”

– Carrie MacGillivray, VP, IoT & Mobility


Instrumented Rotating Equipment

HPE Edgeline Converged Edge Systems

Edge Compute for Real-Time Analysis

Sensor Data Path

CompactRIO CompactDAQ PXI/PXIe Data

Data Acquisition Methods

Analysis, Monitoring, and Control

LabVIEW DIAdem InsightCM

Figure 1. Data path from sensors to edge analytics systems


Edge Computing using HPE Converged Edge Systems Driving business decisions based on real-time data can be accelerated through edge computing and data analysis. HPE has created unique systems, purpose built for the edge which converge real-time data acquisition, enterprise-class computing, and remote manageability needed for data analysis at the IoT edge. HPE Edgeline systems are low-power consuming, ruggedized platforms with multiple network connectivity and data acquisition options to aggregate and analyze data for industrial systems. HPE Edgeline systems can be sized to meet the demands of IoT-created data by customizing the type of chassis, number of compute-servers, CPU cores, memory, and local storage for each modular compute-server within the chassis.

HPE Edgeline EL1000 Converged Edge System A single HPE Converged Edge System EL1000 chassis with an HPE ProLiant m710x server cartridge was tested in this solution to provide the data acquisition, monitoring, visualization, and control capabilities. The HPE ProLiant m710x server cartridge is a modular server that can be sized accordingly with memory and local storage space required for maximum performance of the monitoring solution. The HPE ProLiant m710x server cartridge also has a built-in GPU with the Intel Iris Pro Graphics P580 for HD and 3D quality data visualization. Up to 64GB memory may be required for advanced data analysis and visualization. Up to 4TB of local NVMe SSD storage is available for historical data storage. The HPE ProLiant m510 server cartridge is also available if additional memory, of up to 128GB, and CPU cores, up to 16 Intel Broadwell-D cores, are required for data analysis applications.

The HPE Edgeline chassis is versatile in its ability to operate in a wide range of environments, and be installed in multiple locations through rack and wall-mount options. HPE Edgeline EL1000 and EL4000 enables enterprise-class remote manageability through HPE Integrated Lights-Out technology which is embedded directly on the HPE ProLiant server cartridges.

For more information about HPE Edgeline EL1000 see: https://www.hpe.com/us/en/product-catalog/servers/edgeline-systems/pip.hpe-edgeline-el1000-converged-iot-system.1008670396.html

Figure 2. HPE Edgeline EL1000 Converged Edge System

https://www.hpe.com/us/en/product-catalog/servers/edgeline-systems/pip.hpe-edgeline-el1000-converged-iot-system.1008670396.html

https://www.hpe.com/us/en/product-catalog/servers/edgeline-systems/pip.hpe-edgeline-el1000-converged-iot-system.1008670396.html


The HPE EL1000 with an HPE ProLiant m710x server cartridge was configured with the following specifications for use with National Instruments CompactRIO and CompactDAQ systems:

Qty Description HPE Part Number

1 HPE EL1000 1Gb System 853995-B21

1 HPE 500W Flex Slot Platinum Hot Plug Power Supply Kit 720478-B21

1 HPE Edgeline EL1000 Rack Rail Kit 866690-B21

1 HPE ProLiant m710x 1P E3-1585Lv5 CPU Configure-to-order Server

Cartridge

833105-B21

2-4 HPE Moonshot 16GB DDR4 DIMM 863953-B21

1 HPE Moonshot 240GB M.2 SATA 2242 Solid State Device 866844-B21

0-4 HPE Moonshot 1024GB M.2 NVMe 2280 Solid State Device 862163-B21

1 HPE m710x Front I/O Cable Kit 867893-B21

The HPE EL1000 with an HPE ProLiant m710x server cartridge was configured with the following specifications when using the PXI/PXIe enabled data acquisition method:


1 HPE EL1000 PXI System Contact HPE1


1 HPE Edgeline EL1000 Rack Rail Kit 866690-B21


Cartridge

833105-B21

2-4 HPE Moonshot 16GB DDR4 DIMM 863953-B21


0-4 HPE Moonshot 1024GB M.2 NVMe 2280 Solid State Device 862163-B21


1 PXI Available 2017

HPE Edgeline EL4000 Converged Edge System An HPE Converged Edge System EL4000 chassis with up four HPE ProLiant m710x server cartridges or up to four HPE ProLiant m510 server cartridges should be considered if scaling out the monitoring solution across a large number of data acquisition systems or for configuring redundancy. The HPE EL4000 operates in the same manner as the HPE EL1000 with the exception that it hosts up to four HPE ProLiant server cartridges per chassis as well as having redundant power options.


Figure 3. HPE Edgeline EL4000 Converged Edge System

The HPE EL4000 with up to four HPE ProLiant m710x server cartridges for maximum performance and availability:


1 HPE Edgeline EL4000 10GbE Switch System 847535-B21


1 HPE Edgeline EL4000 Short Rack Rail Kit 868575-B21


Cartridge

833105-B21

16 HPE Moonshot 16GB DDR4 DIMM 863953-B21


16 HPE Moonshot 1024GB M.2 NVMe 2280 Solid State Device 862163-B21

4 HPEm710x Front I/O Cable Kit 867893-B21

The HPE EL4000 with PXI/PXIe and up to four HPE ProLiant m710x server cartridges for maximum performance and availability:


1 HPE EL4000 PXI System Contact HPE1


1 HPE Edgeline EL4000 Short Rack Rail Kit 868575-B21


Cartridge

833105-B21

16 HPE Moonshot 16GB DDR4 DIMM 863953-B21


16 HPE Moonshot 1024GB M.2 NVMe 2280 Solid State Device 862163-B21



Remote Management using HPE Integrated Lights-Out (iLO) Server Management™ As data acquisition and analysis systems are deployed to the edge, the need for remote access and management becomes critical for maintaining and operating of the solutions. HPE Edgeline Converged Edge Systems provide best in-class remote management to accompany their enterprise-class edge compute, and real-time data acquisition capabilities.

HPE has included enterprise-class remote management capabilities through HPE Integrated Lights-Out (iLO) Server Management technology which is embedded into HPE ProLiant m710x and m510 server cartridges. HPE iLO provides server health, reliability, management and performance tracking anytime, anywhere.


HPE iLO automatically monitors, controls, and can even issue alerts based on system health of the server. HPE Integrated Lights-Out is accessible from any web browser, SSH client, or from the iLO Mobile App. HPE iLO technology enables remote management not only through health monitoring, but also through provisioning, firmware updates, virtual media connectivity, operating system installation, integrated remote console, and access to BIOS settings. HPE iLO is also Redfish compliant through the Redfish API and RESTful communication protocol. Refer to the following for more information on HPE iLO technology: https://www.hpe.com/us/en/servers/integrated-lights-out-ilo.html

Figure 4. HPE Integrated Lights-Out (iLO) Server Management

https://www.hpe.com/us/en/servers/integrated-lights-out-ilo.html

https://www.hpe.com/us/en/servers/integrated-lights-out-ilo.html


Data Acquisition Connected sensors require a data acquisition device for acquisition and conversion of analog to digital data. The analog to digital conversion occurs on controller modules connected to each sensor input and then presented upstream for interfacing with data acquisition software. Data acquisition software, for example National Instruments LabVIEW, is used in the solution as the foundation to acquire data and interface with the FPGA within the data acquisition device.

Data acquisition systems from National Instruments come in many different forms based on how data will be acquired and converted for upstream analysis systems. Data acquisition methods used with Edgeline are covered in this document and include CompactDAQ, CompactRIO, and direct data ingest using PXI/PXIe. Each method will be described in detail in this section as well as how they are connected to and accessed by HPE Edgeline Converged Edge Systems.

Many different sensors can be used to measure analog data depending on the type of data to be measured, for example temperature, voltage, pressure, vibration, etc. In the case of vibration logging, two industrial 2-pin accelerometers were used and connected using BNC terminating cables to the data acquisition system. One accelerometer was mounted vertically and the other mounted horizontally on the industrial rotating equipment in order to measure vertical and horizontal movement during equipment operation.

Figure 5. IMI Industrial, 2-Pin Accelerometer

CompactDAQ™ National Instruments CompactDAQ systems may be used in this system as a remote data acquisition device. The CompactDAQ chassis is a ruggedized chassis that can support anywhere from a single data input module to up to eight data input modules. cDAQ systems can interface with HPE Edgeline Converged Edge Systems either through wired Ethernet or through wireless protocols depending on the environmental requirements for data transmission. (National Instruments, n.d.)

The following cDAQ-9191 system was tested with the LabVIEW Vibration Logging solution:


Figure 6. cDAQ-9191 CompactDAQ Chassis

Referenced cDAQ-9191 Configuration:

Qty Description NI Part Number

1 cDAQ-9191 CompactDAQ Chassis (1 Slot, 802.11G, US Only) 781497-01

1 Power Cord, AC, U.S., 120 VAC, 2.3 meters 763000-01

1 NI 9234, 4 Input, 24-Bit, 51.2 kS/s, SW Selectable IEPE & AC/DC 779680-01

2 IMI Industrial, 2-Pin Accelerometer, 100mV/g, ICP® (IEPE) 780985-01

2 IMI Industrial Accelerometer Cable, 10 ft w/BNC termination 780984-01

CompactRIO™ Another data acquisition example is to use CompactRIO from National Instruments. The CompactRIO includes an integrated controller that can run LabVIEW either in an embedded fashion or interface with an upstream HPE Converged Edge System. There are two processing targets on the CompactRIO, one is for real-time communication and signal processing, and the other is a user-programmable FPGA for high-speed control, timing, and triggering that can be interfaced with using a remote instance of LabVIEW. (National Instruments, n.d.)

The following cRIO-9068 system was tested with the LabVIEW Vibration Logging solution:

Figure 7: cRIO-9068 Integrated Controller and Chassis System


Referenced cRIO-9068 Configuration: http://ohm.ni.com/advisors/crio/pages/common/intro.xhtml?configid=CR5175111


1 cRIO-9068 Integrated Controller and Chassis System, Artix-7 FPGA 782663-01

1 NI PS-15 Power Supply, 24 VDC, 5 A, 100-120/200-240 VAC Input 781093-01

1 NI 9915 DIN Rail Mount Kit for 8-slot cRIO/cDAQ Chassis 779018-01

1 NI 9234, 4 Input, 24-Bit, 51.2 kS/s, SW Selectable IEPE & AC/DC 779680-01



Direct PXI/PXIe An alternate approach to remote data acquisition systems is found using the HPE Edgeline Converged Edge Systems with integrated PXI/PXIe1 slots in order to directly connect sensors such as industrial 2-pin accelerometers. In this case, data acquisition is performed directly on the HPE Edgeline Converged Edge System using a variety of PXI/PXIe modules depending on the data being measured. In the case of the vibration monitoring for industrial rotating equipment, the PXIe-4464 module from National Instruments would be installed in the Edgeline chassis. This option provides real-time data acquisition, also referred to as direct data ingest, condition monitoring, and system control all in real-time and converged into the same edge-compute system. National Instruments PXIe-4464 4-Input Dynamic Signal Analyzer:

Figure 8. PXIe-4464, 24-Bit Dynamic Signal Acquisition

Referenced PXIe-4464 Configuration:


1 PXIe-4464, 24-Bit Dynamic Signal Acquisition - BNC Connector 783087-01




http://ohm.ni.com/advisors/crio/pages/common/intro.xhtml?configid=CR5175111


NI LabVIEW™ NI LabVIEW system design software is a graphical programming platform which simplifies the process of visualizing, creating, and coding of data acquisition, monitoring, and visualization solutions. LabVIEW includes all the drivers and interfaces required to quickly and easily access data collected from National Instruments data acquisition systems. For this solution, NI LabVIEW and required components were installed on HPE Edgeline Converged Edge Systems in order to acquire data from either remote data acquisition systems or directly from sensors. (National Instruments, n.d.)

The following LabVIEW software and drivers were installed on the HPE ProLiant m710x server cartridge running in the HPE Edgeline EL1000 Converged Edge System:

Software Usage

LabVIEW 2016 (64-bit) - English Platform for data acquisition, measurement, and control

NI-DAQmx 16.1.0 Drivers for NI-cDAQ systems

LabVIEW Sound and Vibration Measurement Suite Used to acquire, generate, analyze, compare,

import, log, and play back sound and vibration signals

In the following example, LabVIEW was used to acquire and log vibration data from rotating equipment. The following figure depicts the continuous vibration monitoring under normal operating conditions of two rotating turbines where four channels of vibration data are logged and displayed in real-time.

Figure 9. LabVIEW for Vibration Logging during normal operation


In this example an anomalous event can cause the real-time vibration values to increase in amplitude and frequency as shown in the following figure. Since the vibration logging is continuous, the data associated with the anomalous event will also be displayed and logged in real-time. This logged data can then be used in monitoring and analysis applications.

Figure 10. LabVIEW for Vibration Logging during anomalous event

A key benefit to using LabVIEW for data acquisition and control include the ability to easily drag and drop objects into a block diagram in order to make connections from data acquisition devices to logical constructs. With no coding required the data scientist or engineer can build data acquisition and control solutions without having deep programming knowledge. An example block diagram of the NI LabVIEW project for Vibration Logging is shown in the following figure.


Figure 11. Example LabVIEW block diagram for vibration logging

Data Analysis using National Instruments Software Data analysis is a term referring to a broad range of methods used to extrapolate knowledge from acquired data which includes condition monitoring, visualization, and control. Condition monitoring is a process which streamed data is constantly monitored in order to determine the current operational status of the system. Data visualization can be employed in order to easily and quickly access and visualize either real-time or historical data. System control is an advanced method in order to remotely manage monitored systems based on decisions made from the data analysis process. This section describes the data analysis processes that may be used during data acquisition and monitoring, and examples of applications used during the process.

National Instruments InsightCM™ Enterprise NI InsightCM Enterprise is a deployment-ready and scalable software solution for condition monitoring, cost-effective asset health monitoring, and for driving predictive maintenance programs. This software combines data acquisition, analysis, and visualization of sensor data for a complete and in-depth view of the hardware assets being monitored. NI InsightCM Enterprise consists of three components…data acquisition and monitoring systems, NI InsightCM Server, and NI InsightCM Data Explorer. NI InsightCM Server and Data Explorer can be installed on HPE Edgeline Converged Edge Systems, for example the HPE EL1000, in order to enable real-time condition and asset monitoring. The EL1000 would then be deployed at the edge near the NI InsightCM monitoring systems in order to streamline the data acquisition and monitoring solution. NI InsightCM Server running on the EL1000 enables systems management, security, data management, and powerful data analytics using up to 16 Intel Xeon CPU cores on a single EL1000. NI InsightCM Data Explorer running on the EL1000 provides data visualization and examination of live or historical data through customizable user interfaces. This monitoring capability at the edge enables systems operators to consistently and efficiently manage equipment systems and asset health in real-time.

An example of the customer value and scalability obtained through integration of NI InsightCM running on HPE Edgeline can be seen in the following industrial solution: https://www.youtube.com/watch?v=Ha1o5ginodk

More information about InsightCM can be found here: http://www.ni.com/insightcm/

National Instruments DIAdem™ and DataFinder Server Edition™ NI DIAdem is analysis software which enables users to quickly and easily locate, access, visualize, analyze, and report on historical data. One of the main features of DIAdem is the capability to quickly and efficiently analyze or process large volumes of scattered data that may be in many different formats in order to make decisions and drive business outcomes. The NI DataFinder Server Edition is the key component in data management by National Instruments. DataFinder Server Edition allows large numbers of data scientists and engineers to collaborate and access large volumes of data that may be widely dispersed in different locations. DIAdem and DataFinder Server Edition, when combined on HPE Edgeline

https://www.youtube.com/watch?v=Ha1o5ginodk

http://www.ni.com/insightcm/


Converged Edge Systems with real-time data acquisition can enable reliable data management, analysis, visualization, and reporting at the edge.

The following figure shows a sample data analysis report when using DIAdem:

Figure 12. Sample data analysis report when using DIAdem

The following figures show vibration data loaded within DIAdem for further review and analysis:


Figure 13. Vibration data loaded within DIAdem for further review and analysis

Figure 14. Vibration data loaded within DIAdem for further review and analysis

DIAdem securely enables edge analytics and data access for remote users when combined with HPE’s industry leading remote access solution using Citrix XenDesktop for a single user or XenApp for multiple users. For instance, the ability for multiple users to remotely access and analyze against the same data source creates a seamless and robust solution for edge analytics by remote data scientists and engineers. This solution removes the latency, storage, duplication, and security costs associated with sending data off premise to remote data scientists and engineers.


An example of this solution using DIAdem, DataFinder Server Edition, Citrix XenApp, and edge analytics was jointly tested by HPE and National Instruments. In this example, data was acquired and indexed by DataFinder Server Edition running on an HPE ProLiant m710 server cartridge at the data source. This data was then accessed by DIAdem installed on a second HPE ProLiant m710 server cartridge in the same chassis at the edge in order to allow a minimum of 60 remote DIAdem users to quickly access, visualize, analyze, and report on the data using Citrix XenApp remoting capabilities.

See the following for more information about Edgeline and Citrix: https://www.citrix.com/blogs/2016/06/09/x-marks-the-spot-xenapp-xendesktop-xenserver-with-intel-xeon-hpe-moonshot/

Conclusion HPE Edgeline Converged Edge Systems integrate unprecedented edge compute, data capture and control, datacenter-class security, device and systems management, as well as large storage capacity to provide heavy-duty analytics and insights to the edge. This enables businesses to make real-time decisions and add value to their operational processes to result in better business outcomes. The Edgeline EL1000 also prevents cloud lock-in by not having to send data back to the cloud or data center. It addresses latency, bandwidth, cost, security, duplication, corruption, and compliance issues, enabling three critical components of savings and success—time, money, and time to action.

The insight gained from improved data management and equipment monitoring is a driving force behind IoT. As this emerging world of IoT-created data becomes visible, technology must continue to evolve and keep pace with the needs of IoT. To this end Hewlett Packard Enterprise and National Instruments are working together to redefine how data acquisition, data management, and analysis processes can coexist at the edge for faster insight into critical business operations.

https://www.citrix.com/blogs/2016/06/09/x-marks-the-spot-xenapp-xendesktop-xenserver-with-intel-xeon-hpe-moonshot/


References

Hewlett Packard Enterprise. (2016). HPE Converged IoT Systems. Retrieved from www.hpe.com: https://www.hpe.com/us/en/servers/edgeline-iot-systems.html

Hewlett Packard Enterprise. (2016). HPE Edgeline EL1000 Converged IoT System. Retrieved from www.hpe.com: https://www.hpe.com/us/en/product-catalog/servers/edgeline-systems/pip.hpe-edgeline-el1000-converged-iot-system.1008670396.html

Hewlett Packard Enterprise. (2016). HPE Edgeline EL4000 Converged IoT System. Retrieved from www.hpe.com: https://www.hpe.com/us/en/product-catalog/servers/edgeline-systems/pip.hpe-edgeline-el4000-converged-iot-system.1008670180.html

National Instruments. (n.d.). CompactDAQ. Retrieved from www.ni.com: http://www.ni.com/compactdaq/

National Instruments. (n.d.). LabVIEW System Design Software. Retrieved from www.ni.com: http://www.ni.com/labview/

National Instruments. (n.d.). The CompactRIO Platform. Retrieved from www.ni.com: ttp://www.ni.com/compactrio/

Learn more at hpe.com/info/edgeline

© Copyright 2016 Hewlett Packard Enterprise Development LP. The information contained herein is subject to

change without notice. The only warranties for Hewlett Packard Enterprise products and services are set forth in

the express warranty statements accompanying such products and services. Nothing herein should be construed

as constituting an additional warranty. Hewlett Packard Enterprise shall not be liable for technical or editorial errors

or omissions contained herein.

The National Instruments logo, LabVIEW, and CompactRIO, are trademarks of National Instruments. Neither Hewlett Packard Enterprise, nor any software programs or other goods or services offered by Hewlett Packard Enterprise, are affiliated with, endorsed by, or sponsored by National Instruments. Citrix and XenDesktop are trademarks of Citrix Systems, Inc. and/or one more of its subsidiaries, and may be registered in the United States Patent and Trademark Office and in other countries.

873731-001 November 2016

http://www.hpe.com/info/edgeline

Documents

Real-Time Data Acquisition, Monitoring, and Control · Real-Time Data Acquisition, Monitoring, and Control Developing better business outcomes through condition monitoring and edge