An advanced meter delivers a series of interval data instead of a monthly reading. To make use of it, you have to validate it, fill in what is missing, store it and analyze it: load curves, balance by transformer, detection of abnormal consumption and voltage tracking. Regulations commonly set the recording interval and protect the customer's personal data.
What data an advanced meter delivers
The most important difference between a traditional meter and an advanced one is the form of the data. The traditional one delivers an accumulated number. The advanced one delivers a time series.
Meter specifications commonly require recording data at 15-minute intervals and storing energy, voltage and current data for a minimum period. Besides energy, the meter reports events and alarms: energy data are the records of voltage, current, consumption or production, plus the meter alarms.
| Data | What it is for |
|---|---|
| Energy per interval | Load curve, billing, balance |
| Voltage per interval | Checking whether the customer receives voltage within the expected range |
| Current per interval | Detecting overload and abnormal consumption |
| Events and alarms | Cover opening, loss of voltage, tampering attempts |
| Disconnect relay state | Confirming disconnects and reconnects |
The path of the data
- Acquisition. The central system receives the meter's records
- Validation. Impossible values, repeated values or wrong timestamps are detected
- Estimation. Missing intervals are filled in and flagged as estimated
- Storage. Interval history, with version control
- Publication. The data manager delivers the information to whoever must receive it
- Analysis. Balance, anomalies, demand, service quality
The structure of the systems that perform each step is in AMI architecture. Where a data manager exists, its task is to collect, administer and publish the energy data, with neutrality and independence. Availability is also often regulated: the usual target is that the data of a day is available early the next morning for a high percentage of the data.
The volume justifies the system
With 15-minute recording there are 96 intervals per day for each quantity. One million meters with a single quantity generate 96 million records a day. If energy, voltage and current are stored, the volume multiplies. That is why data management is a design component of the system and not an add-on.
Data quality
An analysis is only as good as the data that feeds it. The most common quality controls are:
- Completeness: how many of the expected intervals arrived.
- Time consistency: a series with drifted time assigns consumption to the wrong time bands.
- Range and jump: a negative value where there should not be one, or an impossible jump between intervals.
- Counter resets: an accumulated value that goes down indicates a meter change or a failure.
- Flagging: every data item must indicate whether it is measured, estimated or corrected, so as not to mislead the analyst or the customer.
What analysis it enables
Load curves
With the interval series you see when each customer consumes and when the demand of the transformer or feeder is concentrated. That allows equipment to be sized with real data instead of estimates, and it is the basis of any demand response program or time-of-use tariff.
Energy balance
By cutting the customer meters and the head meter at the same time, the energy balance is calculated by day and by time band, not only by month. If the difference appears only at certain hours, the cause is different from one that appears all day. It is a central tool against non-technical losses.
Detection of abnormal consumption
Each customer’s patterns are compared with their history and with similar customers. The usual signals are a sharp and sustained drop in consumption with no change in activity, a flat profile where there used to be variation, or repeated meter alarms. A signal is a lead, not proof: what confirms an anomaly is a field inspection.
Voltage quality
Because the meter records voltage by interval, the utility can see which areas have voltages outside the expected range without waiting for the customer’s complaint. If you need to study this topic at your plant, see power quality in industry.
Locating supply faults
The «last gasp» notice from the meters, explained in remote reading, disconnect and reconnect, shows which customers lost voltage and outlines the affected area before the calls arrive.
Distributed generation
With the bidirectional meter, what the customer imports is separated from what they export. That is necessary for the balance of a circuit with self-generators and for the settlement of surpluses.
Data protection
A meter’s data reveal people’s habits: what time they get up, when they are away. Regulations commonly recognize this:
- The customer is the owner of the personal data obtained from their meter.
- Service cannot be restricted for not authorizing the processing of the data.
- The utility must meet the obligations on personal data.
- The utility and the data manager are often required to hold ISO/IEC 27001 information security certification.
In the design, this translates into role-based access, a record of who queries what, and analysis with aggregated or anonymized data when there is no need to identify the person.
Frequent mistakes
- Analyzing without validating. An analysis on data with gaps or drifted time produces false conclusions.
- Confusing estimated with measured. If it is not flagged, the balance mixes data of different nature.
- Storing everything without a retention policy. You must define how long each type of data is kept and who can see it.
- Treating an anomaly signal as proof. You end up inspecting customers without cause or, worse, accusing them.
- Not integrating with operations. Data that only reaches billing loses much of its value.
To go further, see what smart metering is, how demand response works and the differences between SCADA, ADMS and DERMS. The phases of an AMI project, from diagnosis to operation, are in How to Implement a Smart Metering Project.
How Tiza Group can help
The communications, integration and security side of a metering data system is designed with the same discipline as that of a substation. At Tiza Group we provide communications engineering, SCADA and gateway integration and power quality studies, worldwide and remotely. Contact us and we will quote.
Related guides
- Smart metering (AMI): what it is and how it works
- AMI architecture: meter, concentrator and central system
- Remote reading, disconnect and reconnect with smart meters
- Power quality in industry
- Bidirectional meter for small self-generators
- AMI Project Cost-Benefit Analysis
- How to Implement a Smart Metering Project
References
- Local metering regulation and utility requirements: recording interval, minimum storage, definition of energy data, data manager, personal data protection and information security certification vary by market.
Frequently asked questions
How much data does an advanced meter generate?
With 15-minute recording there are 96 intervals per day for each stored quantity. For one million meters and a single quantity, that is 96 million records per day. That volume is why a data management system is required and a spreadsheet is not enough.
What is done with the intervals that did not arrive?
The data management system flags them as missing and fills them with a defined estimation method, recording that they are estimated. If the meter stored the data, it is recovered when communication is restored. Meter specifications commonly require the meter to keep at least a minimum period of energy, voltage and current data.
Who owns the meter data?
In many jurisdictions the customer is the owner of the personal data obtained from their meter, and service cannot be restricted because the customer does not authorize the processing of their data. Where a data manager exists, it collects and publishes the energy data with neutrality and independence.