A question of accuracy

A temporary system monitoring is sufficient for consumption estimation?

Often a simple consumption estimation is used as a basis for energy audits, cost centre calculations or design calculations for the consumption of electrical systems. For this purpose, a temporary energy data acquisition is usually carried out at the respective measuring point with a retrofitted electricity meter.

 The consumption quantities recorded in this way can then be extrapolated to an annual consumption, whereby a certain error in estimation is unavoidable. How large this error actually is depends primarily on the actual consumption fluctuation at the measuring point and the duration of the sample measurement.


24H measurement compared to one month at the measuring point Transfer meter measurement

Day 1: 2,562 kWh extrapolation Month: 2,562 kWh*30 = 76,860 kWh Actual monthly consumption: 51,846 kWh

Deviation of the extrapolation = +48%.


The duration of the sample measurement may be limited to 24 hours. Energy data from one day are used for extrapolation over a whole year. In most industrial sectors, however, it is advisable to measure at least one day during the week and one weekend day, as there are often large differences in consumption. Since many industrial companies also have a strong weekly cycle, a measurement over an entire week will further approximate real consumption. The last step before a permanent measurement is made is a monthly measurement. With a good knowledge of seasonal and climatic effects, it is possible to estimate consumption with the slightest error, but the error quickly turns out to be in the high double-digit percentage range. Especially not to underestimate is the difficulty to estimate the error, if no actual consumption profile is known.

„The allocation of consumption to cost centers or individual plants is always a horror for us. Every month, three of us walk through the factory with clipboards and note down the meter readings. We have also carried out temporary measurements for individual plants and extrapolated them to one year. But the resulting interpretations were then rather a gamble because of course the necessary accuracy is missing. Only the correct dimensioning finally makes the whole thing economical“
(Energy representative of a South German mechanical engineering company)

7-day measurement compared to one month at the measuring point

Compressor measurement KW47: 321.96 kWh, extrapolation month: 1,407.84 kWh

Measurement KW48: 427.4 kWh, extrapolation month: 1,829.6 kWh

Actual monthly consumption: 1,597.76 kWh

Deviation extrapolation KW47 = -11.8 %

Deviation extrapolation KW48 = +14 %


„As of 1 January 2016, companies in the manufacturing sector that make use of tax relief in accordance with the Electricity Tax Act §9a or the Energy Tax Act §51 must record the relevant quantities of energy with meters. Thus it is no longer sufficient for the responsible main customs office if the quantities of energy eligible for relief are indicated by an estimate. Who seizes thus durably, does not only receive transparency but has in the case immediately all necessary data to the hand all necessary data
(Christian Schumacher, energy system analysis – ENIT of system)


The ENIT Agent can be installed at your premises in two simple steps. You can then test the product for three months with no obligation.

Test now