A network under new demands
Electrical networks were designed around a simple pattern: large generators at one end, consumers at the other and power flowing in one direction between them. That pattern is changing. Generation is becoming more distributed and more variable. Storage is being added at many points. Demand is growing from electrified transport, heating and computing. Power now flows in directions and at times the original designs did not anticipate.
The equipment carrying those flows was, in many cases, installed long ago. Operators are being asked to deliver more through assets that are aging, and to do it with very high reliability. Knowing the true condition of each asset has become central to doing that safely.
Assets that work alone
The defining feature of utility equipment is distance. Substations, transformers and generation sites are spread across wide areas and are mostly unattended. Inspection means sending a crew, and crews can visit each site only occasionally. Between visits an asset is largely on its own.
Major equipment at large substations is often well monitored. The same cannot be said of the far more numerous smaller assets: distribution transformers, equipment at remote sites, and the balance-of-plant machinery at generation and storage facilities. For much of that population, the first clear indication of a problem can still be an outage.
What equipment reveals
Power equipment degrades in ways that produce physical evidence. A transformer under stress runs warmer and may change the character of its hum as its core and windings respond. Partial discharge within insulation produces acoustic and electromagnetic emissions. Cooling fans and pumps wear like any other rotating machine. Loose connections generate heat. Inverters and other power electronics show thermal and electrical changes as components age.
Wind, solar and storage installations add rotating machinery, cooling systems and power conversion equipment at sites that are frequently remote and exposed to weather.
How AFFERENT is designed to help
NODE is designed to be mounted on or beside an asset and to observe it continuously through several kinds of sensing. For energy applications, three aspects of the design are especially relevant.
Edge intelligence. Remote sites often have limited or intermittent connectivity. A device that analyzes signals locally and reports findings needs far less from the network than one that streams raw data, and it keeps working when the link is down. Read about edge AI.
Observation without control. NODE is a sensing device. It is designed to observe equipment and not to operate it, which keeps it outside the control and protection systems on which grid safety depends.
Security. Energy infrastructure is a recognized target. Device identity, encrypted communication and a design that does not require inbound network access are treated as basic requirements. Read our security approach.
Transformers in more detail
Transformers illustrate why continuous observation matters. They are costly, slow to replace and expected to serve for decades. Their life is governed largely by the condition of their insulation, and insulation ages faster the hotter it runs. A transformer that is loaded heavily, cooled poorly or exposed to high ambient temperature is consuming its life more quickly than one that is not, without any visible sign.
Several parts of a transformer give physical evidence of their state.
- Core and windings. Mechanical loosening changes the vibration and sound the unit produces under load.
- Cooling. Fans and oil pumps are rotating machines with bearings that wear. Their failure raises operating temperature and accelerates aging.
- Tap changers. These are among the few moving parts in a transformer and a frequent source of trouble. Their operation has an acoustic and vibration signature that changes as contacts and mechanisms wear.
- Bushings and connections. Deterioration can produce heating and partial discharge, both of which emit detectable energy.
Utilities already use established methods for major units, including periodic oil sampling and dedicated online monitors. Those methods are valuable and are generally reserved for the largest transformers. The much larger population of smaller units is where an economical, externally mounted sensing device could add awareness that does not exist today.
Renewables and storage
Wind, solar and battery installations bring their own equipment and their own difficulties. Sites are often remote, spread over large areas and visited rarely. Wind turbines contain gearboxes, bearings and generators working under variable load. Solar plants depend on inverters and transformers with cooling systems that run hard in hot conditions. Battery installations rely on thermal management plant, including pumps, chillers and fans, to keep cells within safe limits.
In each case the balance-of-plant equipment is conventional machinery in an unconventional location. Its failure can take generating or storage capacity offline, and the cost of a visit is high. Edge-based monitoring is well matched to those conditions.
Working within utility practice
Utilities operate under strict engineering, safety and regulatory disciplines, and any new technology has to fit within them. AFFERENT's design choices reflect that. The device observes and does not control. It is intended to mount externally, without intrusive work on energized equipment. It is designed to operate with limited connectivity and to remain separate from protection and control networks.
We also recognize what we have not yet done. Equipment used on utility systems is commonly subject to specific standards, testing and approval. Meeting those requirements is a precondition for deployment, and we will describe our standing against them only when it can be documented.
Toward a grid that knows its own state
If the condition of individual assets is known continuously, a network can be managed differently. Maintenance crews can be sent where evidence points. Equipment can be loaded according to its actual health. Replacement programs can be ordered by need instead of age alone. Faults can be anticipated in assets that currently give no warning.
That is the sense in which a grid might come to feel itself: thousands of receptors reporting on the state of the equipment they are attached to, so that the whole can be understood.
