SCADA, ICS, and PLCs: A Plain-English Guide to How Energy Infrastructure Runs

At the bottom are field devices. A sensor measures a physical property, such as pressure, temperature, or voltage, and sends the value to a controller. An actuator acts on the process. Valves, breakers, switches, and motors are all actuators. The National Institute of Standards and Technology (NIST) describes the loop this way: the controller takes in the sensor values, works out what the output should be, and sends it to the actuators.

A PLC is a purpose-built industrial computer that runs the logic for a piece of equipment. NIST describes it as having user-programmable memory for logic, timing, and control functions. It reads signals from sensors, applies the program its engineers wrote, and sends signals to actuators. A remote terminal unit (RTU) fills a similar role at a remote site. It controls the local process, monitors local sensors, and reports to a central system.

The two overlap. NIST notes that PLCs used inside a SCADA system can play a role similar to an RTU's, and that in a smaller system a PLC can be the main controller, usually with no central server or operator screen. Electric systems add protective relays. NIST lists them among the field devices that can report to a control server directly or through an RTU.

SCADA is built for equipment spread over a wide area, where collecting data in one place matters as much as controlling it. NIST describes a typical system in three parts: a control server at a control center, communications equipment, and remote field sites with RTUs, PLCs, or both. The software combines data acquisition, data transmission, and human-machine interface (HMI) software. The HMI is the screen where operators view status and history and change set points, which are the target values a controller works to hold.

The control server polls field devices at set intervals, stores and processes what comes back, and handles alarms, trend analysis, and reporting. A field alarm can also interrupt the polling cycle as a priority. Field sites report over telephone, cable, fiber, or radio links, using standard or proprietary protocols. Large systems add sub-control servers, and NIST's example of a large system includes a regional control center and a backup center.

Electric utility transmission and distribution is one of the uses NIST lists. The guide also shows how much the grid leans on these links. In one example, the loss of a microwave link for a transmission SCADA system could take a generating unit offline and lead to a loss of power at a substation, and the problem could spread from there.

In grid control rooms, SCADA sits under another layer. The North American Electric Reliability Corporation (NERC) describes an energy management system (EMS) as a set of computer-aided tools that operators use to monitor, control, and optimize generation and transmission. NERC calls SCADA the heart of EMS architecture. Applications built on that data include state estimation, which works out the grid's present electrical state from measurements, and contingency analysis, which simulates outages to predict conditions. Automatic generation control, which adjusts generator output as load and frequency change, is another. Without SCADA, NERC notes, operators lose device status, critical substation data, and the ability to open or close breakers remotely.

A DCS does a related job inside a single site, such as a refinery, a water treatment plant, or a chemical plant. A supervisory controller sends set points to controllers spread around the process, and each of those controllers manages its own part of it. NIST says dividing the process into modules limits the effect of a single fault. Power plants are a DCS case too. NIST says electric generation facilities are controlled by a DCS, which communicates with SCADA so that the plant's output can match demand on the transmission and distribution system.

A simplified example shows the pieces working together. An operator at a control center needs to open a breaker at a substation. The operator selects the breaker on the HMI. The control server sends the command over the telemetry link to the substation. The RTU or PLC there passes a signal to the breaker, which is the actuator.

The answer comes back the same way. The controller reads the breaker's new status from its sensors and reports it on the next poll, or at once if it raises an alarm. The control server records it, and the HMI shows it to the operator. Real systems add more steps and safeguards. The basic path is the one NIST describes: commands go down from the control center, and data comes back up.

Two other components show up in most diagrams. A data historian collects and logs field data over time. An engineering workstation is the computer engineers use to program PLCs and other controllers. NIST shows both on a local network in its PLC example.

ICS is the whole category. SCADA covers equipment across a wide area, and a DCS covers a single site. PLCs and RTUs are the controllers inside both, and the HMI is the operator's window into them. In grid operations, an EMS sits above SCADA. Operational technology (OT) is wider still. NIST counts building automation, transportation, and physical access control systems alongside industrial control.