Dr Yair Poleg, PhD, Chief Technology Officer (CTO)
For more than 50 years, supervisory control and data acquisition (SCADA) systems have been at the forefront of monitoring industrial infrastructure through computerized means. Comprising a collection of cyber-secure, industrial-grade hardware and software to collect, analyze, and control information from sensors, SCADA systems are deployed around the world to monitor and control complex, fast-moving industrial processes that exceed humans’ abilities for manual oversight. SCADA applications include providing monitoring and control capabilities in power generation plants, HVAC systems, and large-scale manufacturing processes such as automated production lines.
The Challenges of Reaching Beyond the Perimeter
Traditionally, remote monitoring (also called telemetry) has been predominantly employed for ‘inside the fence’ scenarios such as the above in which both the sensors and the control systems are co-located within the same facility or situated nearby. ‘True’ telemetry, involving input from dispersed assets, and integrating its output into SCADA platforms, has been deployed on a far lesser scale.
Several factors help explain why telemetry has been slow to take the leap beyond the confines of the network perimeter.
First, remote monitoring has historically necessitated the building of large, unwieldy, ‘monitoring stations’ to contain the equally cumbersome communications and power equipment needed to power it. Installing such sites involves logistical and bureaucratic considerations, such as coordinating the construction effort of these stations and obtaining the requisite planning permissions from local authorities.
In communications, the higher the packet size and more frequent the transmission cycle, the greater the power and data overheads the devices incur. Assuring a reliable power supply for the installations was seldom easy and, until recently, mobile data networks were both expensive and not universally available. No utility wants to economize on either solely to extend battery life. Therefore, if remote monitoring were employed at all, transmitting by short message service (SMS) – a medium which cannot be encrypted – was the preferred medium.
The second factor limiting the integration of beyond-the-fence assets with SCADA systems has been the difficulty in providing the reliable and secure communications networks the systems need to relay their readings back to the control room. Unlike networks that run solely within a site perimeter, deploying a completely proprietary network to physically isolate channels that reach remote assets to prevent cyber-attacks (a strategy called air-gapping) is unfeasible for all but the largest of utilities. In addition, networks optimized (and suitable for) the Internet of Things’ (IoT) ultra-low-power requirements have recently come into play. The traditional collection of GSM-based networks, such as 2G and 3G, were designed with consumer data applications (such as video streaming) in mind, and therefore incurred power overheads that were wholly unacceptable for IoT’s humble data requirements.
For these reasons, among others, industries that require the managing of widely dispersed assets have largely chosen to leave their remote assets out of their computerized monitoring strategy altogether. They instead rely on intermittent ‘field readings’ taken by technicians sent to physically inspect the status of the remote reaches of their infrastructure.
For such operators, employing a partially complete monitoring strategy necessarily results in a serious knowledge gap about their networks’ health. Water supply networks, for instance, typically involve clusters of centralized infrastructure, such as water treatment plants, in addition to a widespread network of midpoints, such as reservoirs and pumping stations, and a vast amount of system endpoints at homes or industrial sites.
Without real-time pressurization information from the network edge, water network operators cannot employ Automated Demand Response (ADR) systems to optimize water pressure for current demand and must instead over-pressurize by default. Such over-pressurization is the major driver of water’s enormous energy footprint (a connection termed the ‘water energy nexus’). In this case, simply not integrating the network edge with central data processing programs on SCADA platforms has far-reaching ramifications for the systems’ power consumption, their running cost, and the environment.
No matter what the application, however, utilities bereft of information about the state of edge infrastructure can only be partially aware of the state of their networks under management at best. Proactive and predictive monitoring strategies – vital for efficient system maintenance – cannot be implemented. Inefficiencies increase with the size of the network.
Connecting to SCADA Through the IIoT Can Bridge the Gap
Finding reliable power and communications sources remain a challenge for integrating SCADA systems with remote assets such as in-field sensors. However, both are rapidly diminishing in magnitude as increasingly sophisticated solutions come to market.
The rise of the IoT and its industrial variant, the Industrial Internet of Things (IIoT), means that for the first time in the history of infrastructure monitoring, deploying widespread telemetric networks is an entirely feasible endeavor from both financial and technological perspectives.
The monitoring cabinet of a few decades ago is quickly being displaced by a new breed of IoT gateways. These allow operators to communicate data at a fraction of cost such transmission would have incurred formerly. Such gateways, often no larger than a desktop telephone, can be installed virtually anywhere, including within the confines of underground infrastructure or, in some cases, such as digital oilfields, even housed offshore. These gateways harness the flexibility of modern, LP-WAN and 3GPP-based IoT-specific networks (such as Sigfox, LoRa, and the NB-IoT protocol currently being rolled out in Europe), to transmit data from the edge with minimal power overheads. Demands for interconnectedness among control systems in the modern IT environment have rendered the air gap an almost obsolete security practice. Holistic cyber security strategies, particularly involving advanced cryptography, have taken their place.
The rapid advance of edge computing means that the process of analyzing gathered information is also increasingly being carried out in situ aboard the devices on the network edge themselves. Not sending unnecessary information to central servers obviates the need for analyzing unnecessary information within SCADA systems. This further reduces power overheads, can lengthen battery life by extending transmission cycles, and improves both responsiveness and system throughput rates down the stack.
The Final Hurdles to Cross for Seamless Integration
With so much progress already made, it’s fair to ask, “what’s left to do to achieve the dream of integrating SCADA systems with the full extent of operators’ monitoring networks?”
Although this will be the year in which the first commercial NB-IoT networks go live, the protocol was only formally standardized last summer. Significant milestones remain to be passed before it emerges as the preeminent communications network to drive IIoT connectivity. The current, fragmented state of the network market for IIoT connectivity has also created significant uncertainty among manufacturers about which of the competing protocols, Sigfox, LoRa, and NB-IoT, should be supported and prioritized. For those responsible for integrating the sensors’ outputs into SCADA systems, such a multitude of networking options does not make this process any easier. Compared to the state of affairs just a few years ago, however, the problem of excessive choice is an enviable situation.
As network roll-outs continue at breakneck pace, the IIoT will continue to grow and meet the ambitious forecasts for its expansion. Coupled with the rapid rise in computing power at the edge, utilities will come to regard full remote integration with their SCADA systems as a norm rather than an expansion. How operators will utilize this to drive efficiencies will undoubtedly also be impressive.
Dr Yair Poleg, PhD, is the Chief Technology Officer (CTO) of Ayyeka. With a PhD in Computer Science from the Hebrew University of Jerusalem, and as a former team leader and project manager with the Israeli Defense Force’s elite 8200 signals intelligence unit, Yair specializes in signals processing and spearheads Ayyeka’s technological and intelligence R&D innovation.
For more information visit www.ayyeka.com.