The basic pillars on which the automation system rests are safety (interlock, protection function), efficiency (the closed loop controls), reliability (always doing it’s intended functions the way its supposed to perform) and availability (doing these functions round the clock). Around these core functions of automation system, there is a layer of innovative concepts, namely enhanced connectivity, process optimisation, data analytics, IT/OT convergence and inherent cyber security. It can be termed as Automation System 2.0, if the earlier developments are put as 1.1, 1.2, 1.3, etc.
If one has to describe in a single word the information driven enterprise of today, probably there cannot be any better word than connectivity. Advances in communication infrastructure have made all systems to be connected seamlessly. This has made a difference in automation systems too, and made it possible to make data available anywhere, irrespective of its source, not just for monitoring, but also for operation and analysis. One of the off-shoots of this enhanced connectivity is interoperability of control system and HMI.
Usually, in a DCS, the control system and the HMI are tightly coupled, which means the control system is made operable only from this HMI and not from the HMI of any other DCS. As a result, a multiple package situation, which is the usual case in a power plant, leads to multiple DCS in the main plant and multiple HMI in the main plant control room for the same equipment.
For example, the startup/shutdown of the mill & operation of its binary drives is done from one HMI and the analog controls of the mill from another HMI. To add to the woes, even a situation of multiple DCS and multiple HMI in the same package is also encountered. Here, the integral controls of the main equipment are operated from one HMI and the balance equipment from another HMI (Fig 1.1).
Operation personnel have always demanded a single HMI / a single GUI for all the equipment in the control room. In very naive parlance, a car should have only one steering! This concept of “one car one steering” can be implemented, either by having a single DCS, which is possible in case of a single main plant EPC or using advanced connectivity between DCS to have a unified HMI, which is the technological solution.
The Unified HMI concept illustrated in Fig 1.2 involves the connectivity of the controllers of one DCS (DCS Y) to the control system of DCS X for operation from the HMI of DCS X. All the operation/monitoring possible with HMI of DCS Y is done from the HMI of DCS X, which is designated the master. The HMI of DCS Y (i.e. its native HMI) becomes redundant.
The most critical aspect of this concept is the response time, which is achieved by a detailed design & testing of the communication links between the two control systems. Obviously, this calls for real hard work on the part of the designers of the two DCS and also on the part of the engineering team of the end customer. But such hard work is required for breaking the barriers of conventional thinking. Eventually, the operator, who is oblivious of the underlying dynamics of the unified HMI design, is provided a single HMI.
In today’s regime, operating cost has a direct bearing on the profitability of the company. Further, we have to reduce carbon footprint as much as possible. Hence, process optimisation assumes great significance in today’s business operations and can be a game changer. Advisory systems have a limitation as it is not possible to follow these continuously at all times. Advanced Process Control (APC) is basically optimisation intervening directly on the control system by automatically changing bias and set points. Thanks to the power of today’s controllers, many of the multi-variable predictive adaptive controls can be executed in the control system itself.
This can not only enable tighter controls and better setting times, but also improve the controllability of the modulating controls at lower loads and at faster ramp rates.
The Internet of Things (IoT) has made data analytics a very prominent tool in the information-driven enterprise, giving useful insights for improving bottomlines. Automation systems are a storehouse of enormous real-time data, and falls in the definition of Big Data (i.e. data so huge that humans cannot comprehend). Analysis tools can help in bringing out actionable information out of this.
One such tool is Advanced pattern Recognition (APR) that can help in deriving early warnings of equipment problems much before the routine alarm in DCS, which is based on fixed alarm limits given by the equipment OEM. On a fleet monitoring level, these early warnings are used as a precursor of impending equipment failures and help in taking actions for predictive maintenance. But on a DCS HMI level, this can also be used to reduce the surge of alarms by providing increased action window for rectification to plant personnel, thereby, improving the operator efficiency. The maximum manageable target as per ISA standard is 12 alarms per hour and real time analytics can play a major role in achieving this target.
Besides alarm management, analytics could also improve trip analysis for necessary corrections in plant operation/maintenance strategy. Going by this concept, software for replaying the operation scenario before a trip gives useful insights into what should have the operator done to avoid the trip. Such software, coined ‘Power Replay’, replays the plant mimics/operations with options like rewind or backtrack, slider, etc.
Further, the real-time unit performance data coupled with advanced algorithms can provide guidance on the best combination of load allocation among units for a given station demand, i.e. merit order rating among the generating units. This software provided as part of DCS HMI has become particularly significant in today’s scenario, where thermal power stations are running continuously at part load.
The increasing convergence between IT (Information Technology) and OT (Operational Technologies) has made the advances in IT technology enter the automation world. One such technology is virtualisation. In traditional hardware, we have an operating system and various applications, which are designed keeping in view the OS and the hardware. In virtualisation, we have a hypervisor, which runs simultaneously many OS and applications. The benefits include platform independence, lesser footprint in terms of power, space and air conditioning, easier backup & restoration.
The main advantage of virtualisation is that it makes migration to a new hardware with new OS possible without upgrading the existing application. In DCS HMI, this is very significant to combat obsolescence as it makes possible to avoid a costly HMI software upgrade.
The flip side of open architecture, open protocols and IT/OT convergence is the looming cyber security threats. Targeted attacks like the Stuxnet and Black Energy not only result in production downtimes, but can compromise equipment & human safety. These can also jeopardise the functioning of critical infrastructure, a sort of cyber war.
There are three important aspects to cyber security for DCS/SCADA; first, technology which involves mainly the system architecture, then the people aspect, which are the security procedures to be followed and third, the enforcement i.e. audits. The importance of cyber security as a design aspect cannot be undermined as it enforces security irrespective of the level of awareness of the end user. An international standard like IEC 62443-2-4 and 62443-3-3 aims to address this aspect & institutionalises the collective knowledge of end users, product suppliers and system integrators. For instance, features like multi-factor authentication, white listing, patch management, secure remote access are in-built in the design of today’s automation systems. In this manner, security also becomes inherent to the DCS like quality & reliability. Similar to safety integrated level (SIL) for safety systems, a concept called SAL (security assurance level) enforces graded security, right at the product development stage.
Growing demands will continue to enhance the power and capability of automation systems. Technologies like IIoT, cloud computing and mobile visualisation are already disrupting or has the potential to disrupt the automation field. Yes, the list is endless and so is the nature of the automation discipline.