By Andrew West
As technology rapidly advances, SCADA and other data management systems are continuing to evolve to meet the needs of a modern utility sector. SCADA communications expert Andrew West provides his take on the current market for control systems, and provides insight into the road ahead for these technologies.
Most people reading this probably already have some idea about the differences between various kinds of control systems and system elements, such as distributed control systems (DCS), supervisory control and data acquisition (SCADA), programmable automatic controllers (PAC) and programmable logic controllers (PLC). But to others, those distinctions are not so clear.
For example: politicians and the journalists of the popular press have been trained to think of SCADA as being a general term for any kind of industrial control system (ICS). And even amongst engineers and technicians, the term SCADA means different things to different people. For some it is the computer that sits on the factory manager’s desk and tells him how many widgets his plant made today. A more sophisticated version might also report other key performance indicators such as the cost to make each widget, and how this is affected by various factors.
Some will see SCADA as part of the factory automation system that lets the operators view and modify process parameters. Others will use the term to describe all the components of a network that monitor and operate a nationwide system such as the power grid, or a gas pipeline or a water or wastewater utility. Even within these groupings, each industry has its own unique requirements that drive subtle (and sometimes not-so-subtle) distinctions in the systems that are designed to serve those requirements.
Each point of view is valid, and different. Whatever we think of as SCADA, that’s what it is. When we discuss our systems, it often happens that we use the same terms to describe different kinds of systems and functionalities, with the result that we often misunderstand each other.
Any conversation about SCADA should consider the distinguishing features that various products and systems have to offer. Don’t assume that any particular product or system supports the same requirements or offers the same functionality that may be familiar from your own work. SCADA systems span an enormous range of capabilities and have evolved to meet the distinct requirements of many different industries. Systems that are suitable for one application might not suit another, or they may provide the required functionality, but be unacceptably expensive. The nuances are sometimes subtle. It is rarely a case of one-size-fits-all. Vendors and end users should continue to remind themselves of this.
The history and evolution of various SCADA products is usually determined by their target application. Many master station devices that began as small factory automation human machine interfaces have grown into systems that are capable of supporting many thousands of variables. The larger systems that managed large distributed networks have also become more powerful.
Nonetheless, many SCADA master stations are still compute-bound: SCADA is one of the few areas of computing technology where state-of-the-art systems are still often limited by the performance of the hardware platforms on which they run.
The remote terminal units (RTU) of yesterday are now typically highly-capable devices that support multiple communication interfaces, perform automation processes and support distributed inputs and outputs. Many vendors now give their products different names in order to differentiate their perception of the device capability from the ‘data multiplexer’ that was the traditional RTU. Even so, as long as there is a need to monitor signals and control processes via a remote communication link, there will continue to be a device that does the job of an RTU, even if it is given a different name.
Standards in SCADA
The specific needs of SCADA systems have spurred the development of specific standards suited to the needs of those systems. Some of these are communication protocols (IEC 60870 and IEEE 1815 (DNP3) for example); some are in middleware and data modelling areas (CIM, IEC 61850 and OPC). Each meets specific needs and usually does not serve well in applications far from its original targets.
SCADA depends on communication systems, and this is an area where standardisation is well established (e.g. interfaces), but evolution is causing rapid change, especially in areas of wireless technologies and broadband ethernet-based systems. Many of these changes are driven by the economics of internet adoption and mobile phone use, not the needs of critical infrastructure operators; so it is little wonder these technologies do not always suit the needs of SCADA.
Even for experienced professionals it takes considerable time and effort to learn the intricacies of different standards and products in order to make good design and implementation decisions. New factors such as cybersecurity continually bring change to the landscape; and this in particular will drive changes to traditional SCADA system maintenance and lifetime planning. There is no single solution or technology that seems to suit every application, although some seem to have wider utility than others.
In the same way that there are a range of requirements and interpretations that fall under the umbrella of SCADA, the differences between SCADA and other kinds of control systems are often blurred. Specialist knowledge is needed to do SCADA well, and typically this is only found in companies that focus solely on SCADA.
For many years, SCADA products have been undervalued relative to equivalent control products from other areas. In this environment of low profit margins, little capital is available for ongoing product development. This has led to a decline in product quality and the demise of a number of vendor companies. Some other vendors have been subsumed into larger conglomerates. There is a tendency for design and manufacturing to be moved to low-cost countries with an attendant loss of expert domain knowledge.
Large companies that have SCADA as a small part of their portfolio typically do not appreciate the distinctive nature of SCADA and merge SCADA into their mainstream product suite, almost invariably leading to a loss of functionality. There are a number of well-known control system companies that have products called RTUs that are just PLCs and have products labelled SCADA that are just DCS systems with some communication driver software. These products may be suitable for some applications, but many of them cannot provide the functionality that was, for example, commonly found in power industry SCADA in the 1980s and 1990s.
Legislation and technology
The introduction of Restriction of Hazardous Substances Directive (RoHS) legislation in Europe has caused a shift in the materials used for electronic manufacturing. Nominally, control system devices are exempt from this legislation; but, practically, it is uneconomical to continue to use traditional materials and techniques. RoHS-compliant products tend to have lower reliability and shorter service lives than products assembled from traditional materials. Systems installed today typically have orders-of-magnitude more complexity than similar systems they are replacing. Sometimes this extra complexity arises from provision of additional functionality; sometimes it arises from the migration of hardware-based architectures to designs that are largely based on software. The 20 to 30-year ‘install and forget’ service lives typical of products produced in the last millennium are unlikely to be achieved today. Many end users have noted the decline in product quality in recent years that can be attributed to these kinds of changes.
Making SCADA sexy
For the foreseeable future, systems will still need careful design and engineering in order to achieve reliable economic operation. Products and vendors still need to be chosen carefully. A knowledgeable user knows their system’s requirements and should specify the functionality they need – preferably in terms of outcomes rather than how they think it should be designed!
The future for the industry holds promise. The adoption of standards will assist the integration of a wider range of products and features. The new development centres are relearning the lost tribal wisdom of SCADA. Multi-national conglomerates are learning that they need to make allowance for the special needs of SCADA systems. End users have realised that they cannot continue to de-skill and rely on outside suppliers to ‘own’ the design of their systems. Users and vendors have realised that many of their system experts are approaching retirement, and so are now introducing programs to induct a new cohort of engineers and technicians. The challenge will be to make control systems sexy enough to hold Gen-Y’s interest.