Grid asset monitoring · Ghana

The regulator should not learn about a fault from the customer.

GridWatch instruments distribution transformers and tells you what has failed, where, and what equipment the crew needs — before the phone starts ringing. Built in Accra for the Energy Commission, ECG, NEDCo and GRIDCo.

Built by POA Nexgen Technologies, Accra.
Registered PPA vendor · aligned to Act 541, Act 538, Act 843, L.I. 1816.
5 faults
Overload, imbalance, LV fuse, HV cutout, tamper
Zero outages
Clip-on sensors install on live lines
±2.3%
Measurement uncertainty after calibration
3 intervals
A fault must hold before anything alarms
One platform

See the fleet. Name the fault. Predict the failure.

National oversight for the regulator, real-time detection at the distribution transformer, condition diagnostics for transmission assets, and the field application that turns an alarm into a repaired transformer.

🗺️

National oversight

Load, capacity headroom and fleet health across all sixteen regions and every licensed operator — an independent view the Commission controls, not one the utility submits.

Fault discrimination

An LV fuseway fault, an HV cutout failure and a device sensor fault look identical from the ground. GridWatch separates them and issues a different dispatch for each.

🔔

Alerts that earn trust

Nothing alarms on a single reading. A fault must hold across three consecutive intervals before a ticket is raised — the restraint that stops crews ignoring the system by month two.

🧪

Condition diagnostics

Dissolved gas, hot-spot temperature, ageing acceleration and insulation life consumed on oil-monitored assets — replacement planning becomes a schedule, not a reaction.

📱

Field engineer app

Dispatch with the equipment list, live readings at the pole, install calibration and provisioning. Encrypted offline buffer for dead zones, SMS fallback where data fails.

📊

Reliability & reporting

SAIDI, SAIFI and CAIDI computed from measured outage events by region and operator, with CSV export for the Commission's own analysis.

🔐

Roles & scope

The Commission sees every operator. A utility sees only its own fleet, its own alerts, its own districts. Access scoped by organisation, as control-room software should be.

🧾

Transparent value case

Every assumption behind the benefit model is exposed and editable, so the Commission can test the case against its own figures rather than accepting ours.

🇬🇭

Ghana integrations

MTN and Telecel private APN, NCA type approval, GSA device approval, NITA or on-premise hosting, and alignment to L.I. 1816 technical rules.

How it works

Clamp it on. No customer loses power.

Split-core sensors read current through the cable's magnetic field. No conductor is cut, no bare metal is exposed, and no neighbourhood is disconnected to install a monitor.

1

Install without an outage

Clip-on current transformers hinge around the existing insulated cable; insulation-piercing connectors take voltage and unit power through the sheath. Certified crews, live-line procedure.

2

Calibrate on site

The engineer captures a per-phase correction against a reference instrument. This is what takes combined uncertainty from about ±5.8% to ±2.3% — and what makes every threshold defensible.

3

Let the platform hold its nerve

Readings stream over a private cellular tunnel. A fault must persist across three consecutive intervals before it escalates the asset or raises a ticket.

4

Dispatch with the right kit

The crew is told which fault it is, at what height, and what to carry — LV fuse link, HV cutout work, or a device service call that needs no line crew at all.

Stated up front

What GridWatch will not claim

A monitoring platform is only as trustworthy as its willingness to say what it cannot do. These three appear in the proposal, not buried in a footnote.

Theft detection needs metering that does not yet exist

Energy-balance reconciliation requires interval reads from every customer meter plus an accurate transformer-to-customer mapping. Until those exist, the capability cannot be built — and we will not sell it.

Asset protection alone does not pay for itself

On our own figures the annual benefit is smaller than the annual running cost. The case for funding is regulatory capability and public safety, not financial return.

The data directs an inspection, it is not evidence

These are Class 1–3 instruments, not revenue meters. GridWatch targets an inspection; the inspection produces the evidence. It must not be the sole basis for a penalty.

Start with 150 units and a control group.

Not a national rollout. An eighteen-month controlled pilot that measures the failure rates nobody currently knows, so the scale decision is taken on evidence.

And one decision that costs nothing: every transformer already being replaced can be fitted on the way in, at a third of the retrofit cost and with no live-line work at all.

Open the console → Read the proposal
Note on the demonstration. The console runs against simulated telemetry generated in your browser. No physical device exists at this stage and no live network data is involved. The decision logic is built and tested; the device is not.