Original practice scenario: A power-network simulator streams measurements from many devices. Some links disconnect and later replay buffered readings. Design the monitoring ingestion path.
Read model answer +
Distinguish transport availability from the physical condition being measured. A missing measurement is an unknown state, not a safe zero or evidence that equipment stopped.
Work through it
Identify device sessions and sequence numbers, deduplicate replay and preserve event and ingestion timestamps with quality indicators.
Compute provisional windows with freshness thresholds and late-event correction rules; separate connectivity alerts from measured-condition alerts.
Test burst replay, clock errors, long gaps and invalid units while bounding queues and prioritizing critical synthetic signals under overload.
Tradeoffs to discuss
Longer lateness windows improve historical completeness but delay stable aggregates; real-time views need visible provisional status.
Common pitfalls
Do not interpolate through every gap without a validated assumption.
Keep this monitoring design separate from real grid control or protection logic.
Original practice scenario: A hypothetical wind operator wants a model using weather predictions and turbine history. Explain a fair evaluation for newly added sites.
Read model answer +
Use the weather forecast version available at prediction time and evaluate both future periods and unseen sites. Actual future weather or revised forecasts would make an unrealistically easy test.
Work through it
Build time-indexed features with issue times, forecast horizons, turbine availability and site metadata, explicitly separating curtailment from weather-driven variation.
Use rolling temporal backtests and held-out-site evaluation, comparing to simple persistence and seasonal baselines at each horizon.
Report error and interval coverage by site, weather regime and availability, including missing forecast inputs and limited-history sites.
Tradeoffs to discuss
Shared models borrow strength across locations but can underfit local terrain effects; site-specific models need sufficient representative history.
Common pitfalls
Do not score only on periods when turbines operated normally.
An aggregate average can conceal poor performance during high-impact forecast errors.
