Intelligent automation series part 2: Automation Pyramid’s first two levels

Part 2: This series explains the automation pyramid: A framework for process intelligence and improvement. The first two levels: process equipment/instrumentation and control systems are explained.

By Alex Marcy, Corso Systems January 20, 2016

The foundation of any process is the equipment. In refineries, the equipment starts with the pumps, valves, and piping. At its most basic level, a refinery receives its raw materials, which are stored in large tanks. These materials are pumped into distillation units, heated, and fractionated. Each fraction is then processed further to remove impurities, or further distilled into other products. Finished products are blended with additives, or shipped to other facilities for more processing.

Conditions must be monitored while processes run and operators must control valve positions and pump speeds to ensure flow rates throughout the system and levels in tanks are maintained. Temperatures and pressures in the process must be monitored. This is where instrumentation comes into play.

Instruments consist of various sensors and transmitters that provide data to the control system. Pump status comes from variable frequency drives or motor starters. Pressure, temperature, and level data come from sensors in the field. Instruments are usually hardwired to a programmable logic controller (PLC) or a distributed control system (DCS). After the data are in the control system, operators can control the process based on current conditions and requirements. While DCS systems are common in large refineries, advances in PLC hardware and software, such as redundancy, object oriented programming languages, and database integration, make PLCs a viable technology for any size system.

Layer two: Control systems

The second level of the pyramid is the hardware that allows operators to control the process. In most large-scale refineries, this is a DCS. In other cases-including smaller facilities-this is a PLC combined with supervisory control and data acquisition (SCADA).

Currently, DCS and PLC/SCADA are functionally similar. They automate process equipment and give operators a software interface to monitor and control the process. Process data can be stored in historical databases for further analysis and reporting including trends, downtime monitoring, and integration with quality control test data. Advances in PLC/SCADA are bridging the gap with DCS with improvements in database integration, standardized control logic, process graphics, and hardware redundancy.

The major difference between DCS and PLC/SCADA systems is the architecture. A DCS is an integrated system with controllers, I/O, a process historian, and an operator interface included in one package. A DCS has many controllers operating specific equipment or areas of the process. Many vendors offer standard controller and I/O hardware options, which simplifies system specification. This approach makes it easy to implement redundancy, ensuring consistent process operation. The operator interface is built directly into the DCS by design. The top-to-bottom integration of a DCS can simplify system development compared to a PLC/SCADA.

PLC and SCADA are individual components that must be integrated as a system. This provides many options in hardware and software choices compared to a DCS. The PLC and associated I/O can be specified to meet specific process requirements, with many levels of cost and functionality available. A process controlled by a PLC typically uses one or two controllers to control the entire process.

The SCADA provides the operator interface and can be easily integrated with the PLC if they are from the same vendor. That said, the SCADA system and PLC are two separate entities requiring their own implementation and maintenance.

Selecting a system

There are several considerations when choosing between a DCS or PLC/SCADA system. First up is the overall cost of the system, including implementation and long-term maintenance obligations, such as training operators and maintenance staff, and ongoing support costs.

Related to cost, another major factor is what systems are in use at other facilities. Using similar systems at multiple facilities can reduce training and support costs for a company.

On a technical level, there can be components in a DCS that are not available in a PLC/SCADA, or vice versa. This can be something simple, such as a predefined template for a specific piece of equipment. It can be something complicated, such as adaptive process optimization algorithms in the controller itself. As technology improves, this is becoming less of an issue between the two systems or even specific vendors.

From a nontechnical perspective, system design choices are also dictated by past experience. If someone worked with a particular system in the past, it will influence his or her decision-making process. This can be a preference to use a particular system based on a good experience, or one to avoid based on a bad experience. All else being equal, this is an important part of selecting a system. With control systems staying in operation for many years, it is important to start on a positive note. 

Wrapping up

In most existing plants, the decision between a DCS and PLC/SCADA has already been made. Expansions made to an existing facility will usually integrate into the existing infrastructure. New facilities will require some research into the right type of system for the job. Regardless of the type of system in place, the control system forms the foundation of automation.

Next part: Process intelligence tools which build on the control system to turn data into useful information.

– Alex Marcy, P.E. is the co-owner and President of Corso Systems, a system integration firm based in Chicago, Ill. Edited by Jack Smith, content manager, Control Engineering,

ONLINE extra

– See part 1 of this series linked below.

Original content can be found at Oil and Gas Engineering.