1. Start with the handover question, not the device list

At a substation BESS, teams can have plenty of data and still be unable to answer a basic operating question: what did the site exchange with the grid, what passed through the PCS, and what did the site itself consume?

That gap usually appears because different teams inherited different views of the same asset. Electrical teams may focus on a substation feeder. The PCS supplier may report conversion values. The BMS may report state of charge (SoC) and battery status. SCADA may present a combined dashboard. None of those sources automatically becomes the authoritative record for every purpose.

Write a one-line purpose next to each measurement point before equipment is selected or configuration is frozen. Typical purposes include:

• recording net import and export at a defined grid or project boundary;
• observing power entering or leaving the PCS AC side;
• identifying separately supplied auxiliary consumption;
• reconciling a defined energy balance during commissioning; and
• providing operational data to a gateway, EMS or SCADA system.

If a value will be used for a formal commercial, utility or compliance purpose, identify the accepting party and its specification separately. An internal monitoring point should not be described as an accepted revenue, settlement or compliance meter unless the applicable requirements say so.

2. Freeze a boundary point list against the final single-line diagram

A useful point list is not a generic product schedule. It is a controlled map between the final single-line diagram, a physical location, a data tag and a business or engineering question.

The point list should make direction explicit. “Charge” and “discharge” may be intuitive to an operator, but each data source still needs a documented sign convention, register mapping and unit. Record whether data is stored as separate import/export registers or as signed values; do not assume two systems use the same convention.

3. Commission the complete measurement chain, not only the meter

Every electrical value travels through a chain. Depending on the selected architecture, that chain can include voltage wiring, current transformers (CTs), potential transformers (PTs), sensors or shunts, a meter, a gateway and a consuming system. A correct device configuration cannot compensate for a reversed CT, an incorrect ratio, a wrong phase mapping or an incorrectly scaled register upstream or downstream.

For each point, the commissioning record should identify:

1. the physical boundary and associated drawing revision;
2. the sensing method and configured ratio or scaling, where applicable;
3. phase mapping, polarity and expected import/export direction;
4. the meter or data-source tag, firmware/configuration reference and measured units;
5. the source system that consumes the data;
6. timestamp basis, time zone and synchronization method; and
7. the person or party responsible for approving the record for its intended use.

The U.S. Department of Energy's FEMP evaluation method is a useful reminder that metered charge/discharge data and controller data play different roles. Its method uses actual time-series charge and discharge meter records to estimate selected BESS KPIs, but it also states that meter data alone cannot determine availability. [2] A sound handover preserves that distinction instead of forcing one dashboard value to prove everything.

4. Test direction, time and cross-system behaviour under safe operating conditions

Commissioning tests should be agreed with the responsible electrical and operations teams, and carried out only under the project's approved safety and switching procedures. The objective is not to demonstrate a universal acceptance test. It is to show that the named measurement chain behaves as expected for the purpose assigned to it.

A practical test log can include the following checks:

8. Drawing-to-field check. Confirm that the installed point, phases and sensing arrangement match the approved single-line diagram and latest point list.
9. Scaling and polarity check. Confirm CT/PT, sensor or shunt information, configured multipliers and expected direction using an approved test method.
10. Zero and status check. Distinguish a genuine zero reading from an offline, stale, estimated or substituted value.
11. Import/export check. Where safe and relevant, observe a controlled condition that confirms the documented directional convention.
12. Timestamp check. Compare meter, gateway and receiving-system timestamps. Keep measurement refresh, communication polling, stored interval, upload delay and dashboard refresh as separate fields.
13. Cross-system comparison. Compare values at the same boundary and aligned time. If they differ, record the boundary, units, averaging interval, source hierarchy and expected uncertainty before calling it a fault.
14. Communications recovery check. Test the agreed response to a link interruption, including buffer behaviour, duplicate records, recovery order and data-quality flags.
15. Evidence retention check. Save test steps, as-left configuration, register maps, screenshots or exports where allowed, and discrepancy resolutions in a retrievable handover package.

It is tempting to reconcile every number to a single perfect balance during a short test. That can be misleading. Differences can arise from non-identical boundaries, transformer or cable losses, conversion losses, auxiliary loads, time misalignment, different averaging intervals, sensor accuracy and calculation methods. The useful outcome is a documented explanation of the differences within the project's agreed method, not an unsupported promise of zero variance.

5. Separate field measurement, controller records and calculated values

This separation matters most when an issue occurs after handover.

For example, a PCS may show a power command and a power reading, while a grid-side measurement shows the net exchange after the defined transformer, feeder and auxiliary boundaries. Both can be useful. Neither should be relabelled as the other. Likewise, an SoC trend can explain operating context but does not replace an electrical energy record.

6. Make the handover pack usable six months later

The last commissioning meeting should not be the last time the measurement architecture is explained. A usable handover pack allows an O&M or integration team to answer “what changed?” without reconstructing the project from emails and screenshots.

At minimum, retain:

• the final single-line diagram and point list with drawing revisions;
• a boundary/source-of-truth matrix stating purpose, owner, units, direction and intended consumer;
• sensing, scaling and configuration records appropriate to the installed architecture;
• register map or field mapping, including signed/unsigned direction treatment;
• time basis, interval definition, retention responsibility and missing-data treatment;
• commissioning test evidence and the disposition of differences;
• interface responsibility between meter, gateway, EMS, SCADA, PCS and BMS; and
• a controlled change process for configuration, firmware, CT/PT scaling, field mapping and dashboard calculations.

This is especially important for a BESS because its operating state changes. DOE FEMP notes that incomplete or irregular charge/discharge data can produce physically implausible results in analysis, and that data quality needs ongoing checking rather than a one-time installation event. [2]

7. What the Hanoi 110 kV BESS pilot changes - and what it does not

EVNHANOI reported that its five 10 MW / 20 MWh BESS were installed at Bac Thang Long, Quang Minh, Sai Dong 2, Phung Xa and Thanh Oai 110 kV substations. According to EVN, the systems are integrated into the distribution grid to support operation, load regulation and flexibility. [1]

The public notice does not disclose the projects' meter placements, data interfaces, acceptance method, control configuration, suppliers or measured results. It would therefore be inaccurate to infer a specific metering design from it.

The safe engineering lesson is narrower: when BESS is commissioned at multiple substations, a repeatable point-list, direction, time-alignment and handover discipline becomes more valuable. That is an engineering inference, not an EVN requirement or a statement about the project's equipment.

8. Where YTL can support - and where responsibility remains elsewhere

Once the project team has defined its boundaries and data purpose, YTL can discuss the field-measurement and data-output layer: the intended AC or DC measurement context, direct or externally sensed architecture where applicable, import/export direction, register mapping and the interface expected by the receiving system. Exact model, sensing arrangement, measurement range, accuracy, interval behaviour, communication protocol, certification suitability and platform compatibility must be confirmed for the selected model, firmware and project requirements.

YTL does not define the final POI/PCC/POM boundary, grid-code requirements, protection settings, tariff or settlement acceptance, BMS/PCS/EMS/PPC control logic, commissioning approval or final regulatory compliance. Those remain with the relevant owner, EPC, integrator, utility, grid operator, consultant and contractual parties.

CTA

If you are preparing a BESS meter-selection or handover discussion, provide the final single-line diagram, intended boundary, voltage/current and sensing arrangement, required fields, timestamp and retention needs, interface, acceptance purpose and destination market. This lets the measurement and data-output layer be reviewed against the project before a model or configuration is assumed.

FAQ

Does one grid-side meter describe the whole BESS?

No. It can describe electrical exchange at its defined boundary. It does not automatically show PCS-side energy, DC-side throughput, auxiliary consumption, SoC, control response or the reason for an operating outcome.

Can BMS or PCS data replace a field meter?

They provide valuable controller and status information, but they should not be automatically relabelled as independent field measurement. The authoritative source depends on the project purpose and accepted method.

What is the most important commissioning check for bidirectional energy data?

There is no universal single test. The project should confirm the physical point, sensing configuration, polarity, scaling, sign convention, units, timestamps and intended import/export interpretation under an approved safe test method.

Does a fast dashboard mean the data is suitable for every analysis?

No. Measurement refresh, polling rate, stored interval, upload delay, display refresh and any formal reporting interval are different attributes. The needed attribute depends on the project use case.

Can a meter alone prove BESS availability or project acceptance?

No. Availability and project acceptance can require additional dispatch, state, test, contractual and approval evidence. DOE FEMP specifically notes a limitation of meter-only availability analysis. [2]

Official sources

16. Vietnam Electricity (EVN): EVNHANOI puts five battery energy storage systems into operation, 6 July 2026 (https://en.evn.com.vn/d/en-US/news/EVNHANOI-Puts-into-operation-5-battery-energy-storage-systems-60-204-501555)
17. S. Department of Energy, Federal Energy Management Program: Battery Energy Storage System Evaluation Method, 30 January 2024 (https://www.energy.gov/cmei/femp/articles/battery-energy-storage-system-evaluation-method) and report PDF (https://www.energy.gov/sites/default/files/2024-01/bess-evaluation-method.pdf)