Deploying adaptive AI in distributed water plants - Barbara Accoina

Three key considerations for selecting edge computing hardware for the Oil and Gas Industry

Categories Edge Computing News  |  Guest Posts
Three key considerations for selecting edge computing hardware for the Oil and Gas Industry

This is a guest post by Cole Wangsness, a Solutions Specialist at OnLogic.

In today’s Industry 4.0 world, operational and business decisions need to be made quickly and based upon the best available data to enable the right decision at the right time. This means that IT equipment is being integrated at the edge of networks in almost all industries, including the entire oil and gas industry. When selecting the best PC for use on the edge in the oil and gas industry, it’s important to understand the unique requirements and what solutions can address them.

Top Functional Requirement? Availability and Reliability 

The deployment environments for edge computing in the oil and gas industry are not conducive to your typical desktop PC. These applications require maximized availability of equipment, meaning the reliability of IT technologies is just as important as operational technology (OT) reliability. So what are the key considerations to keep in mind when deploying PCs at the edge, and what PCs are a good fit?

Environmental Considerations for the Oil and Gas Industry

To maximize equipment uptime, we need to consider the environmental factors that need to be addressed. Some common challenges are:

– Wide-operating temperature: Deploying at the edge means dealing with the temperature swings at production sites. Commercial PCs may not start at temperatures below 32°F and overheat at temperatures starting around 80°F (27°C). They may also have component failure over time due to hot or cold operating temperatures or if they experience wide temperature fluctuations. Check for rugged edge computers that can be rated down to -40°F (-40°C) and up to 158°F (70°C) operating temperatures, so you are covered wherever your deployments take you.

– Vibration/Shock: Mounting a PC directly on machinery? A typical off-the-shelf PC with moving parts such as fans and HDDs can fail over time. Look for a fanless computer with solid-state hard drives that contain no moving parts for increased reliability. Systems tested using IEC and MIL-STD procedures will also increase reliability.

– Input Voltage: A home computer needs AC power from a wall outlet. When deploying at the edge, there isn’t always have access to 120VAC power. Industrial PCs can be equipped to take direct DC input from the industrial power supply.

– Particulate/Dust: A fanned PC isn’t going to last when you have dust and debris getting into the system through the vents and electrically shorting out the components or getting in the way of moving parts. Look for fanless PCs that are ventless to protect against dust and particulate ingress. Need a GPU for machine vision or other image-intensive tasks? Fanless hybrid solutions allow a fan to actively cool the graphics card or other expansion card in a separate compartment while keeping the motherboard and other sensitive internal components protected in the ventless and fanless portion of the chassis.

– Enclosure Design: Depending on the deployment environment, a customer may want to deploy your equipment in an enclosure. Most IT equipment does not carry a water rating, so you will want to protect them from rain and snow. Look for equipment that can easily be housed in an enclosure. If needs dictate a system that is IP water rated, expect to pay several thousand dollars more and use non-typical (i.e. expensive) connectors.

– Regulatory/Compliance: Depending on the application, you may require Class 1, Division 2 rating, a hazardous environment rating for flammable gases and liquids. Just like IP rated computers, these cost thousands more than a normal rugged computer, so consider whether the equipment can be located outside of those environments or use a C1D2 compliant enclosure for local equipment. (Applicability will depend on the environment, so always consult a regulatory engineer).

Options to Configure the Best PC to Meet your Oil and Gas Application Requirements

There is no single correct approach to deploying edge computers in oil or gas applications. Look for a configurable solution based on your system requirements:

– System Performance: It is important to consider how much performance that is needed. A data-logging appliance may only require minimal processing power and storage, whereas a surveillance NVR doing machine learning inferencing to monitor remote sites for intrusion detection may require much more processing power, AI accelerators, and larger storage drives.

– I/O at the Edge: For edge applications, data can come from multiple sources. This means lots of I/O options to connect to local devices such as PLCs, sensors, and cameras are needed. For data collection, serial connectivity to take in data from Modbus devices, LAN ports to connect to your OPC-UA enabled PLCs, Power over Ethernet (PoE) for your cameras, or Digital input-output (DIO) for sensors might be required. It’s also important to select local devices that will standard up to the site environment.

– Wireless: What happens after data is collected? Remote sites may not have a hardwired copper or fiber connection. Using wireless technology such as CAT M1 or 4G LTE allows data to be sent over the network using nationwide carriers. This can remove the need for expensive infrastructure development or the need to send out technicians to collect data on-site. Wi-fi and Bluetooth can also be used to connect to local networks and sensors.

– Displays: Does the data need to be displayed to operators locally? For a plant supervisor, this might mean connecting an OnLogic PC to multiple screens so they can view plant processes on their control system. This will usually require multiple display outputs such as DisplayPort or HDMI. For local human-machine interfaces (HMIs), another option is to use an industrial Panel PC to provide an industrial touchscreen computer for operators.

– Security: With more devices on industrial networks than ever before, ensuring network and device security is imperative. A 2017 security study on IoT for Oil and Gas by Deloitte showed that the risk for the most vulnerable and severe cyberattacks occurred in production environments, which are the areas with the most IoT device integration. Securing your network and edge devices with technologies such as Intel’s TPM (Trusted Platform Module) and firmware PTT (Platform Trust Technology) can help. These modules provide intrusion resistance at network endpoints where cyber-attacks are most common either due to lack of monitoring, failure of physical security, or vulnerability versus virtualized resources.

Software for Oil and Gas

In addition to the hardware, IT buyers need to consider what software to load onto a system to meet project needs. Many customers will select the software they want before considering hardware, but it is much better to consider them at the same time since one cannot work without the other. Here are a few key vendors to consider:

Inductive Automation’s Ignition SCADA, (Supervisory Control And Data Acquisition) platform allows the operators to gather and act on data in realtime. Ignition SCADA empowers plant supervisors, line operators, and business decision-makers to make smarter decisions. For example, an oil processing facility can take data in and display it to machine operators at HMIs. This provides a view of the entire plant to supervisors, and generate analytical reports for decision-makers all along the production process—from the intake pipelines to filling your delivery trucks. Ignition allows you to connect to different PLCs so you can usually use what has already been deployed onsite.

AWS Greengrass is an Amazon cloud service designed to facilitate edge deployments. If AWS is already integrated into the IT infrastructure, this can be a great option. For example, deploying Greengrass Core Gateways to monitor a fleet of natural gas delivery trucks by sending back data such as GPS and tank information, facilitated by a 4G enabled edge PC, enables the accurate monitoring of the fleet and enables smarter operational decisions. Once this data is in the cloud, existing AWS services such as AWS Quicksight can be used to visualize data in dashboards or Sagemaker if CAN bus-based data back is drawn back from an edge PC to do health monitoring and predictive monitoring on the fleet.

EdgeIQ is designed for IoT Orchestration, managing a network of devices from a single pane of glass; A major benefit when you have thousands of devices on the network. For example, one can easily differentiate between edge computers monitoring well sites, pipelines, vehicles, and storage tanks by creating device profiles to easily provision devices with specific system images. This allows users to monitor device health and make policy-based decisions using a centralized command software suite. EdgeIQ is designed to integrate with a private cloud, or with existing cloud providers such as AWS, Azure, and Google.

What PCs are best for the Oil and Gas Industry?

In summary, IT buyers need to look beyond the typical off-the-shelf computer when looking for the best PC for the oil and gas industry. Industrial computers that offer hardshell construction and solid-state cooling that is both fanless and ventless for ingress protection will offer the best solution. For more heavy-duty use cases, look for a rugged computer that offers extended operating temperature, shock and vibration protection and the ability to work with wide power input.

About the Author

Cole Wangsness is a Sales Engineer at OnLogic. OnLogic is a global industrial computer manufacturer that designs highly-configurable, solution-focused computers engineered for reliability for the IoT edge. The company has offices in the U.S., Netherlands, Taiwan, and Malaysia, and has helped more than 70,000 customers worldwide.

DISCLAIMER: Guest posts are submitted content. The views expressed in this blog are that of the author, and don’t necessarily reflect the views of Edge Industry Review (EdgeIR.com).

Article Topics

 |   |   |   |   |   |   | 

Comments

Leave a Reply

This site uses Akismet to reduce spam. Learn how your comment data is processed.

Featured Edge Computing Company

Edge Ecosystem Videos

Machine learning at the Edge

“Barbara

Latest News