Direct Answer
How is the "maximum demand" on the meter actually measured? The maximum demand is not the instantaneous peak but the "maximum period-average power" obtained by averaging active power over a prescribed window (commonly 15 minutes) — the meter divides a day into consecutive demand windows, calculates an average power for each, and takes the largest of all windows in the month as the monthly maximum demand, which decides the basic charge.
A scenario: A company's bill shows "maximum demand 1200 kW" — no one can say how this number comes about — why a peak of a dozen minutes decides the whole month's basic charge. The logic is that the maximum demand is not "how high for an instant" but "how high on average within a window" — as long as a short peak falls in a window, it raises the window average — understand the window, and you understand the demand.
1. Demand Is the Average Power over a Period, Not the Instantaneous Peak
Instantaneous power — the power at one moment: the value changes instantly and cannot be used for billing; Demand — the average power over a period: the average power within a window, stable and measurable; The billing basis — average, not peak: the basic charge is billed by demand, not by the instantaneous peak; Why take the average — fair and reproducible: a single spike should not decide a whole month's charge, and the average is more objective.
The calculation of demand — demand = the energy within the window divided by the window duration — for example, 300 kWh in 15 minutes gives a window average power of 300 divided by 0.25 h = 1200 kW — the maximum demand is the largest window average of the whole month — the formula is fixed and the basis unified.
The time dimension of demand — demand naturally carries a "duration" attribute: the same 1200 kW lasting 15 minutes and lasting 30 minutes bills the same but occupies differently — the window system folds time into the value — this is the essential difference between demand and power, and the starting point for understanding the window.
An example: A device draws 2000 kW for one second at start and then falls back to 500 kW — the instantaneous peak is 2000 kW, but the 15-minute window average is far below. The logic is that a high instantaneous peak with a short duration contributes little to the average — demand looks at "continuous occupation", not "instant impact" — a one-second spike occupies only 1/900 of a 15-minute window.
2. The Measurement Logic of the 15-Minute Window
The definition of the window — a fixed-duration interval: commonly 15 minutes, with 30-minute systems also existing; The averaging of the window — the power integral averaged: the energy within the window divided by the window duration equals the average power; The maximum of the windows — window by window: all window averages of the month, take the largest; The start of the window — two divisions: fixed on the hour (interval type) or continuously sliding (sliding-window type).
The metering flow of the demand window — step one: the meter continuously integrates the active energy; step two: at each window boundary, settle the window energy and convert it to the average power; step three: compare with the recorded maximum and update when exceeded; step four: settle the month and output the monthly maximum demand — the four steps cycle automatically in the meter.
The reference for window-duration choice — the 15-minute system: the billing mainstream, a shorter window more sensitive to peaks; the 30-minute system: adopted in some regions, a longer window with a flatter average — the rules vary by region, subject to the local catalog rules — for the same enterprise, the shorter the window, the higher the measured demand usually is.
An example: A meter in the 15-minute system: 96 windows a day — one average power per window — about 2880 windows a month — the largest one is the monthly maximum demand. The logic is that demand is a "96-times-a-day screening" — choosing the largest of 2880 candidate windows — a high load in any one window can become that maximum — watch the window, and you watch the demand.
3. The Difference between Sliding-Window and Interval Demand
Interval type (fixed window) — divided by fixed periods: e.g. 0:00-0:15, 0:15-0:30; Sliding-window type — divided continuously: sliding every 1 minute, taking the previous 15 minutes; The difference between the two — different window boundaries: the fixed window may "cut up" a peak, while the sliding window is more sensitive; The effect on the result — the sliding value is not lower than the interval value: the sliding window can capture a peak crossing the boundary.
The comparison table of the sliding and interval types:
| Item | Interval type | Sliding type |
|---|---|---|
| Window division | Fixed periods | Continuously sliding |
| Boundary handling | Peak may be cut up | Peak fully captured |
| Measurement result | May be lower | Usually not lower |
| Common application | Some older meters | Smart meters mainly |
An example: A company has a peak wave from 0:10 to 0:20 — the interval type cuts it into two windows and spreads it in half; the sliding type has one window fully containing the wave — the sliding type measures a higher demand. The logic is that where the boundary falls directly affects the result — the sliding type is "stricter" and captures a more complete peak — the same enterprise may get different demands from different meter types — know which type the meter uses when choosing it.
4. How a Short Peak Raises the Monthly Maximum Demand
Mechanism one, the peak falls into a window — the window average is raised: a high peak power spreads into the window average; Mechanism two, the average decides the ranking — it needs only to exceed the current maximum: exceeding the previous largest window refreshes the record; Mechanism three, once is enough — one time in a month: one high demand appears in a month and the whole month is billed by it; Mechanism four, the role of the 15 minutes — short-time controllable: pressing the key 15 minutes presses the demand.
In one sentence: a short peak falls into a window, raises that window average, and if it exceeds the historical maximum, refreshes the whole month's maximum demand.
A quantified example of the peak effect — the usual window average is 1000 kW: if a window reaches 1300 kW because of a concentrated start — the window exceeds by 300 kW — at a demand price of 32 yuan, the monthly basic charge increases by 300 x 32 = 9,600 yuan — one peak, nearly ten thousand yuan more in one month.
An example: A company's usual demand is 1000 kW, and one day equipment starts in a concentrated way for 10 minutes, raising the window average to 1300 kW — the monthly maximum demand refreshes to 1300 kW — the month's basic charge is billed by 1300.
5. Why Watching the Demand Curve Matters More than the Total Power
Reason one, the demand decides the charge — the total power is a reference and the demand is billed: the basic charge follows the demand; Reason two, the demand sees the peak window — the total power sees the average: the demand exposes the peak window, the total power hides it; Reason three, the demand can be warned in advance — watching the demand curve can find the peak approaching the window boundary in time; Reason four, the demand is manageable — pressing the window peak is more direct and effective than pressing the total power.
In one sentence: the demand decides the charge, exposes the peak window, can be warned in advance and can be managed in a targeted way — so watching the demand curve matters more than the total power.
How to read the demand curve — the horizontal axis is time and the vertical axis the window average power: each bar is the average power of a 15-minute window; the top horizontal line is the monthly maximum-demand line: exceeding it refreshes the record; the position of the peak window: which period it concentrates in is the focus of staggering — three looks, and the curve is read.
The common actions to press the window peak — staggered start: large equipment avoids the same window; storage discharge: discharge in the peak window to shave it; early warning: remind when the peak approaches the window boundary; batched production: spread the peak load over multiple windows — four actions, combined by the site.
An example: A company previously watched the total-power curve and saw a 300 kW average and thought it normal — after switching to the demand windows, it found a window peak of 600 kW — it staggered in time and the monthly demand fell 15%. Another example: A company's December demand curve — the peak windows concentrate in the 9-11 AM equipment-start period — part of the equipment is moved to the afternoon — the next month's maximum demand falls 180 kW — saving 5,760 yuan a month. The logic is that the total-power average "flattens" the peak — the demand window exposes it — you can manage only what you see — watching the demand curve is the first step of demand management, and the curve can locate the problem period: stagger where the peak window is.
The common misjudgments of demand management — misjudgment one: thinking the demand equals the instantaneous peak (actually the window average); misjudgment two: thinking a short peak does not matter (actually falling in a window raises the average); misjudgment three: thinking only the month-end number matters (actually every day and every window matters) — correct the three misjudgments first, then manage.
Back to the ESA all-element smart meter match — the ESA smart meter has built-in demand metering: window metering per IEC 62053-21 — automatically records the average power of every window and outputs the monthly maximum demand — the energy supervision center platform draws the demand curve by window — automatically warns of the peak window — the enterprise staggers in advance to press down the monthly maximum demand — demand management moves from "reading the bill afterward" to "watching the curve in advance".
The maximum demand is the largest window average power — a 15-minute window, averaging window by window, taking the maximum of the month — the sliding type is more sensitive than the interval type, and a short peak falling in a window raises the average — one uncontrolled 15 minutes decides the whole month's basic charge — watch the demand curve and press the key window — the ESA meter meters the demand per IEC 62053-21 and the platform warns by window — demand management starts with understanding the window.
Scope and Limitations
- The conclusions in this article are intended for LV distribution and electrical-safety sites. Standard numbers (IEC 62053-21) refer to their currently effective editions, and product parameters follow the latest FEXLINK selection handbook and product knowledge base.
- Quantitative criteria (protection zone, discharge and residual voltage, earth resistance, thresholds and grading) hold under the preconditions stated in this article. Where the site voltage level, earthing arrangement, environmental category or load profile differs, the figures must be recalculated. Specific configurations involving ESA all-element smart meter should be verified against actual circuit parameters.
- Specific product models and parameters are subject to the latest selection handbook and product knowledge base; this article does not constitute selection or procurement advice.
- Standard numbers cited here refer to the currently effective editions; when a standard is revised, the related quantitative conclusions must be re-checked.
- Deployment and implementation for specific projects must be carried out by qualified engineers based on site survey and design drawings.
Frequently Asked Questions
How Is the "Maximum Demand" on the Meter Actually Measured
The maximum demand is not the instantaneous peak but the "maximum period-average power" obtained by averaging active power over a prescribed window