Electric utilities and substations
Transmission assets are numerous, unmanned and geographically dispersed, and they sit under a physical security standard that already requires assessment. That combination shapes the whole problem.
Key facts
- NERC Reliability Standard CIP-014 requires utilities to identify transmission stations, substations and control centres whose loss could result in instability, uncontrolled separation or cascading outages, and to develop physical security plans for them.
- Aerial reconnaissance of a substation supports an attack conducted from the ground, so the security question is what imagery of the site enables rather than what the aircraft itself can do.
- Transmission substations are generally unstaffed, so detection alerts must route to a control centre or dispatch function rather than to anyone on site.
- Detection is lawful for a utility to deploy; mitigation is not available to a private operator, so a utility cannot act on the aircraft itself.
- Because assets are numerous and dispersed, the practical question is which sites justify sensors at all, which is a prioritisation exercise rather than a coverage design.
What makes this facility type different
The threat model
The concern is reconnaissance ahead of a physical attack rather than the aircraft itself. A drone photographs transformer banks, control houses, relay panels and access routes, and the resulting imagery supports an attack conducted from the ground. Cascading-outage risk means the consequence of a successful attack on a small number of assets is disproportionate to their size.
The regulatory position
NERC Reliability Standard CIP-014 requires identification and protection of transmission stations, substations and control centres whose loss could cause instability, uncontrolled separation or cascading outages. It already mandates risk assessment and physical security planning, so aerial exposure is an input to an obligation the utility carries regardless of whether it ever buys a sensor.
What constrains the response
Transmission assets are numerous, unstaffed and spread across wide service territories, so per-site staffed response is not available and per-site sensor deployment does not scale economically. Coverage decisions become a prioritisation exercise across a fleet of sites rather than a design for one, which is the opposite of the single-site problem every other facility type here presents.
The aircraft is not the weapon
The instinctive concern about drones over a substation is that the aircraft will do something to the equipment. That is rarely the realistic case.
What a drone does well at a substation is photograph it: transformer banks, control house, relay and protection equipment, the fence line, the access road, and where the cameras are and are not pointed. That imagery supports an attack conducted from outside the fence with entirely conventional means.
So the security question is not what the aircraft can carry. It is what knowing the site in that detail enables.
The standard is already there
Utilities differ from almost every other facility type in this respect: an obligation to assess physical security risk already exists.
NERC CIP-014 requires identification of transmission stations, substations and control centres whose loss could result in instability, uncontrolled separation or cascading outages, and the development of physical security plans for those assets. Aerial exposure is an input to that existing process rather than a new compliance category.
This matters commercially as well as legally, because it means the analysis has somewhere to land. A finding about aerial observation of a critical substation attaches to an assessment the utility is already producing.
The standard does not mandate drone detection, and a reading of it that arrives at a required technology goes further than the text supports. The standard itself governs rather than any summary of it.
Dispersal is the real constraint
Every other facility type on this site presents a single perimeter to be designed around. A transmission utility presents hundreds of sites, most of them unstaffed, spread across a service territory measured in counties or states.
Two consequences follow, and they dominate everything else.
The first is economic. Per-site sensor deployment does not scale across a fleet, so the honest question is which sites justify it. The cascading-consequence subset the standard already identifies is the sensible starting point, not an even spread.
The second is operational. An alert at an unstaffed substation reaches nobody on site. It has to route somewhere with a response capability attached, and if that path does not exist the detection produces a log rather than a defence. Deciding where alerts go, and what happens when they arrive, is worth settling before any sensor is specified.
Related work
- Site assessment & design Vulnerability assessment, RF survey, sensor siting and coverage modelling.
- Procurement advisory Specification and evaluation against the Authorized Systems List.
- Detection system integration Bringing sensors, command-and-control and existing security systems together.
FAQ
▸Does CIP-014 require drone detection?
No. It requires identification of critical transmission assets and physical security risk assessment and planning for them. Aerial exposure is one input to that assessment. Reading the standard as mandating a particular technology would be wrong, and a vendor presenting it that way is overreaching. The obligation is to assess and to plan, and the standard itself governs, not this summary.
▸We have hundreds of substations. Where do we start?
With the prioritisation the standard already drives, rather than with sensors. The assets whose loss carries cascading consequence are a small subset, and those are where aerial exposure is worth assessing first. Deploying detection evenly across a fleet is the most reliable way to spend a lot and protect the wrong things.
▸What use is an alert at an unstaffed site?
That is the right question to settle before buying anything. An alert at an unstaffed substation is only worth generating if it reaches a control centre or dispatch function that can do something with it, whether that is a camera slew, a patrol dispatch or a law enforcement call. Where no such path exists, the detection spend is producing a log rather than a defence.
▸Can we stop a drone over a substation?
Not as a private utility. Mitigation is reserved to authorised federal agencies and, since 1 July 2026, to certified state, local, tribal and territorial law enforcement and correctional agencies. A utility may detect, identify, track and report. Where a persistent pattern exists, the productive route is usually the relationship with the law enforcement agency that does hold the authority.