Every Meter, Tag, and Outage Log Tells Part of the Story
The historian holds SCADA rollups, the billing system holds metering and AMI reads, the outage log lives somewhere else, and weather and maintenance records sit apart again. Each is accurate on its own and blind to the rest.
So when losses creep up on a feeder or demand spikes off pattern, the explanation is split across systems that never meet. QuantumLayers merges them, runs real correlation and group-difference tests across the combined data in minutes, and keeps checking for drift, built for hourly and daily cadences rather than sub-second control.
Connect the Network Without an Integration Project
Point QuantumLayers at what you already have. Connections stay read-only, and nothing is copied into a separate warehouse.
SQL
Metering, billing, and asset databases (MySQL, PostgreSQL, SQL Server), queried live.
REST API
SCADA or historian rollups from any endpoint that returns JSON, structure auto-detected.
SFTP
Interval and AMI meter exports as CSV, matched on wildcard filenames each period.
Google Sheets
Field logs, outage notes, and manual readings, connected directly.
CSV
One-off uploads for ad-hoc studies and back-data.
Manual Reconciliation vs. One Network Dataset
The difference between reconciling meters after the fact and seeing the network as it runs.
| Before | After |
|---|---|
| Metering, telemetry, and outages live in separate systems | One merged dataset links load, generation, loss, and cost |
| Loss creep spotted only when a bill looks wrong | Outlier tests and feeder comparisons surface it as data lands |
| “Demand was high because it was hot,” unquantified | Weather explains a measured share of the swing, tested |
| Outage causes buried in a log nobody reads | Causes ranked by lost supply and tested across feeders |
| Multi-site comparison means a manual pivot table | Sites and feeders compared statistically, updated hourly |
| Drift goes unnoticed until performance drops | Monitors email you when loss or load factor breaks pattern |
What the Statistics Reveal Across the Network
Same connectors, same engine, pointed at generation, consumption, and outages together.
Load & Demand Trends
Consumption and peak demand trended over time and compared across sites and feeders, with periods that break pattern flagged automatically.
Weather-to-Load Correlations
Correlation and regression connect temperature and irradiance to consumption and generation, quantifying how much of a swing the weather actually explains.
Loss & Anomaly Flags
Outlier detection and feeder-level group tests highlight where losses run high, separating expected variation from consumption that does not add up, useful leads for non-technical loss.
Outage & Downtime Drivers
Outage causes ranked by their share of lost supply and tested across feeders and assets, so the few that dominate stand out from the log.
Generation Performance & Drift
Regression compares solar, genset, or plant output against expected drivers, and monitors email you when loss percentage or load factor shifts before it reaches a bill.
Frequently Asked Questions
Yes, through their databases or a REST API. Sources backed by MySQL, PostgreSQL, or SQL Server connect directly, and interval or AMI data exported to CSV on SFTP is pulled in automatically.
No. Every connection is read-only. It reads your data to analyze and explain it, and never writes back to operational systems.
Connected sources sync every hour, and SQL and API connections are queried live. It is built for hourly and daily analysis rather than sub-second SCADA control loops.
It flags feeders and accounts whose consumption breaks statistical pattern and ranks differences across the network, tested leads to investigate rather than automatic verdicts.
Yes. Group-difference tests tell you whether the gaps between sites, feeders, or assets are statistically real, and rank them by size.
Explore Further
How the rest of the platform works, and where the analysis comes from.