top of page

Cooling-Load Data Timestamps: Align Interval Boundaries Before Combining Loads

Oct 27
4 min read
AI-generated illustrative scene: engineering desk with blank graph paper, a clock and a weather sensor. Not product or project evidence.

AI-generated illustrative scene, not a product screenshot or a completed-project photograph. The worked calculation below uses explicit example inputs.


A cooling-load review combines an envelope-load export with an internal-load schedule and produces a sharp midday peak. Both source files appear complete and their daily totals reconcile. Yet one file labels each hour by its ending time and the other by its starting time. Joining the visible timestamps has shifted one load series, creating a coincident peak that the original data do not contain.


This is a data-reconciliation example, not a method for calculating building cooling loads from weather alone. The invented component loads below are already calculated interval averages. Real cooling-load work must retain the appropriate thermal-response model, design conditions and project assumptions. Correct timestamps do not validate an otherwise incomplete load calculation.


Replace a single label with an explicit interval


For each source, record the time zone, calendar, reporting interval, timestamp convention and field meaning. Distinguish an instantaneous reading at 12:00 from an average over 11:00–12:00 or 12:00–13:00. Store interval start and interval end explicitly. A display label is convenient for a chart, but it is not enough information for a reliable join.


EnergyPlus documentation illustrates why field-specific interpretation matters. The official Auxiliary Programs documentation, reviewed on 27 October 2026, uses an hour field from 1 through 24; its temperature field describes a value at the indicated time. The output documentation separately identifies start and end minutes for reporting intervals. These examples are not a claim that every export or software version uses the same convention.


Original example: a false 80 kW peak


Take three consecutive one-hour intervals: 11:00–12:00, 12:00–13:00 and 13:00–14:00. Suppose the envelope component averages 20, 50 and 20 kW respectively. A separate internal-load contribution averages 30, 0 and 0 kW. The envelope file labels these rows 11:00, 12:00 and 13:00 because they are hour-starting. The internal file labels them 12:00, 13:00 and 14:00 because they are hour-ending.


Correct combined loads by interval = [20 + 30, 50 + 0, 20 + 0]


Correct combined loads = [50, 50, 20] kW


Correct three-hour energy = (50 + 50 + 20) × 1 h = 120 kWh


If the internal series is shifted one interval later: [20 + 0, 50 + 30, 20 + 0]


Misaligned combined loads = [20, 80, 20] kW


Misaligned three-hour energy = 120 kWh; apparent peak = 80 kW instead of 50 kW


Joining the raw labels puts the internal file’s 30 kW row labeled 12:00 beside the envelope file’s 50 kW row labeled 12:00. These rows actually describe different intervals. Treating the internal ending label as a starting label moves that contribution one interval later. The correct repair is to map both files to their documented intervals, not to prescribe a universal offset for imported files.


The energy totals remain equal because the example shifts the 30 kW interval within the same three-hour window and assumes zero outside its original active interval. Real joins near day boundaries may lose or duplicate energy as well. Therefore a daily-total check is useful but insufficient: inspect the interval alignment and the timing of the component peaks.


Correct three-hour totals are 50, 50 and 20 kW; misaligned internal-load entries produce 20, 80 and 20 kW. Both sum to 120 kWh.

Synthetic interval-average loads only; not a building simulation or equipment selection.


Do not shift every weather field blindly


A weather row can contain fields with different temporal meanings. The EPW dictionary describes radiation quantities accumulated during the minutes preceding the indicated time, while dry-bulb temperature is a state-point value at that time. Relabeling all fields as hourly averages can create a new error. Read the relevant format documentation and the importer’s handling of those fields.


Similarly, distinguish a rate from an accumulated quantity. An interval-average load in kW can be multiplied by the interval duration in hours to obtain kWh. A radiation entry already expressed in Wh/m² is an interval energy per area, not automatically a W/m² instantaneous reading. Unit matching and time matching are separate checks and both must pass.


Use a small boundary test before the full-year merge


Build a test containing midnight, a date change and at least one known schedule transition. Include daylight-saving transitions only where the source data use them, and preserve the stated time-zone basis rather than guessing from the file name. Check for duplicate intervals, gaps, unexpected durations and rows that move outside the review window. Never silently fill an unmatched interval with zero.


The review-ready output is a short mapping table showing each source label, original convention and normalized interval, plus a before-and-after peak comparison. The cooling-load design handbook is a related reference for the broader workflow; this article does not claim that it imports EPW files or repairs timestamp joins automatically.



Continue the calculation review



Sources and calculation basis


Primary references checked on 27 October 2026. The worked examples and diagrams are original illustrative calculations, not manufacturer selections or project approvals. Apply the stated limitations and the requirements governing the actual project.




Comments

Rated 0 out of 5 stars.
No ratings yet

Add a rating
bottom of page