Load profiles and diversity
The key idea
The plant supplies the loads that operate at the same time. Adding every appliance rating gives connected load; it does not automatically give the coincident peak.
Start with what you know
A measured demand profile records use over time. A synthesized profile estimates use from customer counts, appliance ratings and operating schedules. Both can support design, but they carry different evidence. A smooth curve is not proof that demand was measured.
For a surveyed site, record the customer classes, counts and expected energy use. A school, workshop and household can have different daily patterns. If a site map supplies only building footprints, confirm how buildings correspond to connections and customers before you turn the count into demand.
Timing changes the peak
In this example, 20 customers each operate a 1 kW load for four hours. Their combined energy is 80 kWh. If every load starts together, the peak is 20 kW. When start times differ, fewer loads may overlap and the peak can fall.
Increase the start-time spread. Watch the total energy as well as the peak. Use the seed controls to compare repeatable sets of start times.
- Daily energy
- 80 kWh
- Coincident peak
- 17 kW
- Connected load
- 20 kW
- Coincident demand
- All customers aligned
Read the exact values
| Time / step | Coincident demand | All customers aligned |
|---|---|---|
| 00:00 | 0.00 | 0.00 |
| 01:00 | 0.00 | 0.00 |
| 02:00 | 0.00 | 0.00 |
| 03:00 | 0.00 | 0.00 |
| 04:00 | 0.00 | 0.00 |
| 05:00 | 0.00 | 0.00 |
| 06:00 | 0.00 | 0.00 |
| 07:00 | 0.00 | 0.00 |
| 08:00 | 0.00 | 0.00 |
| 09:00 | 0.00 | 0.00 |
| 10:00 | 0.00 | 0.00 |
| 11:00 | 0.00 | 0.00 |
| 12:00 | 0.00 | 0.00 |
| 13:00 | 0.00 | 0.00 |
| 14:00 | 0.00 | 0.00 |
| 15:00 | 0.00 | 0.00 |
| 16:00 | 3.00 | 20.00 |
| 17:00 | 5.00 | 20.00 |
| 18:00 | 10.00 | 20.00 |
| 19:00 | 14.00 | 20.00 |
| 20:00 | 17.00 | 0.00 |
| 21:00 | 15.00 | 0.00 |
| 22:00 | 10.00 | 0.00 |
| 23:00 | 6.00 | 0.00 |
Model note · Constructed example: every customer uses 1 kW for four hours. Start times use a seeded uniform draw from 16:00 to the selected latest start. This is not the full Phasor load-synthesis model.
A seed identifies one repeatable random draw. Reusing the same seed helps compare design changes without also changing the demand sample. It does not make the sample representative of a real community; that requires evidence about actual behaviour.
Preserve the shape and the basis
A daily energy estimate fixes an area under the curve. It does not fix the evening peak, seasonal change or day-to-day variation. Check whether the input describes an average day, a peak day or a whole year. Preserve the time zone and sample interval when you import measurements.
The coincidence factor is the simultaneous group peak divided by the sum of individual peaks. It is at most one for the same set of non-negative loads. The related diversity factor is the reciprocal when the group peak is nonzero. Name the quantity you use; the two terms are easy to reverse.
The math, if you want itOptional — the page reads completely without it
At each hour the group demand is the sum of whatever each customer is drawing at that hour:
coincident demand at one hour
Pgroup = Σ Pcustomer
daily energy
E = Σ Pgroup · Δt
Every customer here contributes 1 kW × 4 h = 4 kWh, so 20 customers always contribute 80 kWh regardless of start time. Only the peak of Pgroup moves.
coincidence factor
CF = max PgroupΣ max Pcustomer ≤ 1
The example selects integer start times uniformly from the chosen window. It omits appliance combinations, seasonal behaviour, correlations, growth and measured variability. A wider window usually reduces coincidence, but one random sample does not guarantee a strict reduction at every slider step.
See it in Phasor
Choose the load method that matches your evidence: an existing profile, customer mix or detailed appliance use. Inspect the preview, preserve its seed and save the profile before sizing the plant.
Continue the design path
Review power and energy if the energy total and peak still seem interchangeable. Then add the other time-varying input: solar resource and weather.