Nairobi · Release certification
Power — Release certification
Power's own structural and realism results on Nairobi, pulled straight from the domain's own certification write-up.
Documented release scope · Internal checks of synthetic records, with source dates and known limits. These are not live reruns or independent certification.
12 of 12 structural checks pass, live.
| Structural check | Result |
|---|---|
| Substation count is above the domain's minimum population floor | PASS |
| Transmission-line count is above the domain's minimum population floor | PASS |
| Meter count is above the domain's minimum population floor | PASS |
| Billing amounts fall within a plausible range | PASS |
| Outage start/end chronology is valid (end after start) | PASS |
| Reported grid-frequency readings fall within a valid engineering range | PASS |
| Substation and transmission-line install dates are not in the future | PASS |
| Billing-month values are valid calendar months | PASS |
| Paid-date is consistent with billing status | PASS |
| Foreign-key integrity holds across Substation, TransmissionLine, Meter, Billing, Outage and DemandReading | PASS |
| Ward-level service coverage is present | PASS |
| Outage-affected household counts are non-negative | PASS |
6 of 7 realism checks pass; 1 disclosed WARN.
No implausible residential bills
Household-linked meter coverage
Outage duration ceiling
Substation peak load vs. design capacity
Billing occurs only after meter install
Frequency readings are physically plausible
Formal housing connects at a higher rate than informal housing
This is a Power-scoped extract of what's already published on the domain page. The six-category certification framework (Structural, Statistical, Historical, Privacy, Reproducibility, Explainability) organizes reported checks. Available results depend on the release and domain. See Release certification explorer →.
No tapped example yet on Nairobi.
These published synthetic record examples show the dated history available for this domain. Longer histories are sampled across their full span; each record states how many events are displayed.