Transparent formulas. Explicit uncertainty.

The calculation model

Energy is measured in kilowatt-hours. For stable loads, we convert watts to kilowatts and multiply by active runtime. Cost then equals energy multiplied by the user’s marginal electricity rate.

Weekly schedules use 52 weeks divided across 12 months, avoiding the common assumption that every month has exactly four weeks.

Provenance classes

• Estimated: entered or inferred without direct observation.
• Label-derived: taken from a rating plate or technical document.
• Measured: observed through a compatible meter, charger, or utility dataset.
• Outcome-verified: measured before and after a documented change.

Uncertainty

PowerCost Lab presents planning ranges where duty cycle, charging loss, or usage pattern can materially change results. The range is not a statistical confidence interval in the first version; it is a transparent sensitivity band around the user’s scenario.

A result may display two decimal places because currency requires them, not because every input is known to that precision. Readers should pay more attention to assumptions that materially change the decision than to the final decimal.

Monthly and annual normalization

A weekly pattern is normalized using 52 weeks divided by 12 months. A daily projection uses 365 days divided by 12 months. A 30-day table is labelled explicitly as a 30-day example rather than silently treated as an average calendar month. This prevents small calendar conventions from being hidden inside otherwise transparent results.

Tariff selection

Appliance calculations normally use the marginal energy rate: the price of the next kWh consumed. Daily service charges and meter fees usually remain outside the appliance result because the household pays them even when that appliance is not used. Tiered and time-of-use tariffs are modelled as separate scenarios or structured bands rather than collapsed into a false universal average.

Worked-example verification

Every worked example is derived from the visible figures on the page. Energy is calculated before currency. Unit conversions are shown in the formula, and the tariff used for a monetary example is stated beside the result. Examples are starting scenarios, never claims about the consumption of every device in a category.

Publication and correction controls

Market facts require an identifiable source, scope, and review date. Tariff defaults additionally require a validity window and confidence threshold. A source can establish a rate, but PowerCost Lab remains responsible for the calculation, explanation, and any conversion error. Corrections should identify the page, input, expected result, and non-sensitive evidence.

The future Energy Cost Graph

Explicitly contributed measurements may improve cohort benchmarks and estimate calibration. A cohort holds device class, grid convention, climate context, and rated-power band constant. Only approved observations with privately verified meter evidence can enter a published cohort. Contributions are stored without names, email addresses, or exact household addresses.