Your Loss Number Is a Measurement Claim: Quantum Sensing and the DISCOM Accounting Layer

Executive Summary

Most quantum conversations in Indian power circles start with optimisation and end with a slide about qubits. I want to start somewhere less glamorous and far closer to our P&L: measurement. Every loss number a DISCOM reports — feeder-level, DT-level, board-level — is not a fact. It is a claim, assembled from thousands of current transformers, potential transformers and meter classes, each carrying its own error band. We spend enormous effort arguing about what the number means and almost none on how well it was measured in the first place.

Quantum technology is going to reach the distribution business through that door. Not through a quantum computer solving unit commitment, but through instruments — magnetometers, atomic clocks, diamond-based sensors — that measure current, field and time with an accuracy our present hardware cannot approach. The demonstration hardware exists. India has a national hub building it. The question for a DISCOM leader in 2026 is not whether to buy quantum sensors. It is whether the measurement layer underneath your energy accounting is disciplined enough to benefit when they arrive.

The Number Nobody Audits

India's distribution reform has produced real movement. AT&C losses improved from 21.91% in FY2021 to 15.04% in FY2025, and 3.90 crore smart meters had been installed under RDSS against 20.34 crore sanctioned as of 31 December 2025. Those are the headline figures, and they are earned.

But look at what sits below them. A feeder meter reads what its CT tells it. A Class 0.5S meter on a 200/5 CT with a burden mismatch, a decade of thermal cycling and an unverified ratio is a precise instrument reporting an imprecise quantity. At low load — which is most of the night on most rural feeders in North Bihar — CT accuracy degrades exactly where our theft and reconciliation questions live. When I look at a DT with a 9% gap between input and billed units, I have learned to ask three questions before I send anyone to site: is the meter right, is the CT right, and are the two clocks agreeing?

That third question matters more than people expect. Energy accounting is an integral over time. If the feeder meter and the consumer meters disagree by ninety seconds, block-level reconciliation smears. We currently solve this with NTP and periodic drift correction, and mostly it works. "Mostly" is the operative word when you are attributing crores of rupees.

What Quantum Instruments Actually Change

Quantum sensing exploits the fact that atoms and defects in crystals respond to magnetic fields and time in ways governed by physical constants, not by manufacturing tolerance. A nitrogen-vacancy centre in diamond has the same properties in Mumbai as in Muzaffarpur, this year and in fifteen years. That is the real proposition: drift-free, calibration-light measurement.

Note what is not in that table. Nothing here is a quantum computer. These are sensors that happen to use quantum states, and the honest framing from EPRI's work on utility quantum sensing is that they are not expected to replace classical sensors wholesale. They will sit where classical instruments are weakest, and where the cost of being wrong is highest.

The Houdian Signal — Read It Soberly

In November 2024, State Grid Anhui Electric Power commissioned a 220 kV quantum application demonstration substation at Houdian in Hefei. By August 2026, reporting described 85 domestically developed quantum devices across 18 categories running there — diamond-based sensors, quantum-dot sensors, quantum lidar, fibre-based quantum-secure communications, and links to a superconducting quantum computer. The facility estimates the technology could reduce annual electricity measurement errors at that site by more than 500,000 kWh.

I would not put that number in a board paper. It is a site estimate without independent verification, and no performance comparison against conventional instrumentation was published. But the signal is not the number. The signal is that a state utility built a full substation as an instrumented testbed and ran eighteen categories of device in it for nearly two years. That is an industrial learning curve, and learning curves compound.

The lesson I take is procedural, not technical. They created a place where the technology could fail safely against real 220 kV conditions. Most Indian DISCOMs have no equivalent — no designated substation where new measurement technology can be run alongside the incumbent for a year and judged on evidence.

India Already Has the Supply Side

This is not a technology we would import wholesale. Under the National Quantum Mission, approved with an outlay of ₹6,003.65 crore for 2023-24 to 2030-31, quantum sensing and metrology is one of four thematic hubs. It is led by the Qmet Tech Foundation at IIT Bombay, which spans 16 institutes and around 40 researchers, and has built a quantum diamond microscope using NV centres in lab-grown CVD diamond along with a portable magnetometer intended for chip-scale deployment.

Read that as an industrial policy fact, not a physics fact. The instruments India is developing under NQM are magnetometers and atomic clocks — precisely the two categories that map onto current measurement and time synchronisation in a distribution network. The supply side is being built with public money. The demand side, which is us, has said almost nothing about what it needs.

That is the gap worth closing, and it costs a DISCOM almost nothing to close. A written statement of requirement — accuracy class at 5% of rated current, ambient range of 0–55°C, unattended operation for five years, cost ceiling per DT — is more useful to a national hub than another memorandum of understanding.

A four-rung measurement ladder

I use this with my teams to decide where quantum is even a question.

1. Rung 1 — Inventory truth. Do you know the CT ratio, class and install date of every feeder and DT meter? Most utilities cannot answer this for 100% of assets.
2. Rung 2 — Time truth. Are all interval meters synchronised within a defined tolerance, with drift monitored as an operational KPI?
3. Rung 3 — Error budget. Can you state, per feeder, the measurement uncertainty on your loss figure? If your loss is 14% ± 4%, you are managing noise.
4. Rung 4 — Instrument substitution. Only here does quantum sensing become a live procurement question, because only here can you prove it improved anything.

Almost every Indian DISCOM, mine included, is working somewhere between rungs one and three. That is not a criticism. It is the actual work, and it is the work that makes rung four worth reaching.

What This Means for Leaders

1. Publish an error budget before you publish a loss target. For your ten worst feeders, state the measurement uncertainty alongside the loss percentage. The conversation changes immediately.
2. Treat clock drift as an operational KPI. Add meter time-sync deviation to the daily MIS the way you track communication success rate. It is cheap and it protects every downstream reconciliation.
3. Designate one instrumented substation. Pick a 33/11 kV station with good communications and make it the place where new measurement technology runs in parallel with the incumbent for twelve months. Budget it as a capability, not a project.
4. Write the requirement, then send it. Give NQM's sensing hub and domestic vendors a one-page utility specification. The supply side cannot design for a demand signal it has never received.
5. Keep quantum in the metrology budget, not the IT budget. This is instrumentation. Funding it from a digital transformation line invites the wrong evaluation criteria and the wrong owners.

Key Takeaways

• Every AT&C loss figure is a measurement claim, and the instrument layer beneath it is rarely audited.
• Quantum reaches distribution utilities through sensors and clocks first, long before it reaches them through computation.
• China's Houdian substation is a credible signal of industrial intent, though its 500,000 kWh claim remains unverified.
• India's supply side is funded and active under NQM; the utility demand signal is missing.
• Rungs one to three — inventory, time, error budget — deliver value now and are prerequisites for anything quantum.

Written from the field in North Bihar, where the argument about a loss number is usually an argument about a current transformer.

Sources: PIB — Progress on Smart Meter Installation under RDSS · DST — National Quantum Mission · DST — Journey of NQM's Quantum Sensing and Metrology Hub, IIT Bombay · The Quantum Insider — China Tests Quantum Technology in Power Grid · POWER Magazine — Quantum Sensor Ambitions: A New Horizon for Utility Innovation