Data cleaning
AuraScore 79/100

Smart Meter Telemetry Ingestion and Imputation Checklist

Systematically validate, scrub, and impute smart meter interval reads before billing calculation cycles.

Use this template when preparing raw AMI interval consumption data for downstream billing engines or customer portals. It ensures nulls, spikes, and daylight saving shifts are resolved without distorting baseline consumption.

Template

Role: Lead Meter Data Operations Engineer with 12 years of experience in utility AMI pipelines and MDM systems.

Context

  • Operating Utility: {{utility_name}}
  • Head-End / AMI System Vendor: {{ami_vendor}}
  • Telemetry Recording Interval: {{telemetry_interval_minutes}}
  • Maximum Continuous Missing Threshold: {{imputation_threshold_hours}}
  • Primary Outage Data Source: {{outage_log_source}}
  • Target Billing Cycle Cutoff: {{billing_cycle_cutoff}}

Task

Generate an operational data cleaning checklist to audit, sanitize, and impute raw interval energy consumption records prior to the downstream billing cycle, preventing billing disputes and tariff calculation failures.

Method

  1. Check timestamp consistency across {{telemetry_interval_minutes}} intervals, identifying clock drift, duplicate intervals, and UTC/local daylight saving offsets.
  2. Cross-reference zero-consumption windows against {{outage_log_source}} to distinguish valid grid outages from disconnected meter hardware.
  3. Identify negative volume anomalies, reverse-flow flags, and register rollovers, flagging potential meter tamper or bidirectional solar misconfigurations.
  4. Detect extreme consumption spikes using historical standard deviation thresholds for the specific customer rate class.
  5. Apply linear or historical proxy interpolation for gaps under {{imputation_threshold_hours}} while isolating longer outages for manual field verification.
  6. Validate data completeness percentages against the minimum threshold needed before {{billing_cycle_cutoff}}.
  7. Generate audit trail logs capturing all replaced, imputed, or discarded interval values with source attribution.

Constraints

  • Checklists MUST categorize each action item by operational phase (Pre-Validation, Scrubbing, Imputation, Sign-off).
  • Imputation rules MUST NOT alter known zero-consumption records verified by {{outage_log_source}}.
  • Every verification step MUST include an explicit pass/fail acceptance criterion.
  • Do not include general IT infrastructure maintenance tasks outside the meter telemetry scope.

Output format

1. Pre-Cleaning Validation Checks (4-5 markdown checklist items)

2. Anomaly Scrubbing & Filtering (4-5 markdown checklist items)

3. Imputation and Reconciliation Protocol (3-4 markdown checklist items)

4. Final Data Gate Sign-off Criteria (3 markdown checklist items with pass/fail metrics)

Self-review

  • Did I incorporate all six contextual variables directly into the checklist logic?
  • Are outage-driven zeroes clearly protected from improper estimation?
  • Does every checklist item contain a concrete verification mechanism?
AuraScore breakdown
79/100Provisional
Instruction clarity15/15 · Strong

Explicit role, a named task, and discrete steps the model can follow.

Context architecture12/12 · Strong

Background, inputs and variables the model needs before it starts.

Constraint engineering10/12 · Adequate

Hard boundaries — what the model must and must not do.

Output specification6/14 · Thin

A named, field-level shape for the response.

Reasoning structure10/10 · Strong

Ordered work items that force analysis before an answer.

Model compatibility10/10 · Strong

Length and structure that travel across frontier models.

Token efficiency5/10 · Thin

Signal density — instruction weight without padding.

Reusability7/7 · Strong

Documented variables so the scaffold adapts to new inputs.

Robustness3/5 · Adequate

Quality bar, assumptions and behaviour when inputs are thin.

Observed performance1/5 · Thin

How much real usage the template has behind it.

data-analytics
data-cleaning
energy-utilities
smart-meter
ami
meter-data