Outdoor power usage function

What is Power Usage Effectiveness (PUE)?

The energy consumption growth of the DC’s systems are expected to account for 3.2 % of global carbon emissions by 2025. This remarkable growth in the energy consumption of DCs represents today a major challenge for energy efficiency. To meet these challenges, Power Usage Effectiveness (PUE) can be used to measure and optimize DC energy efficiency.

What happens if the outdoor unit stops?

When the outdoor unit stops, it still consumes some power “Standby Power”. (e.g. compress crankcase heater, PC board). The tenants share the standby power with a fixed value based on tenant’s indoor unit size. Z: The rated energy consumption for crank case heater and outdoor unit PCB.

Does CRAC subsystem power consumption affect outdoor temperature?

Based on the correlation results, we observe that the CRAC subsystem power consumption has a strong correlation with the outdoor temperature, followed by outdoor RH. For the first case study (ITEA3 RISE Sics DC), cooling power is produced using cold-air through the CRAC subsystems, in addition to the air-side cooling technique “Free cooling”.

Is power consumption a part of tenant sub-metering?

Otherwise, it is generally assumed that the indoor unit power consumption (which is relatively insignificant unless you are applying high CFM ducted units) is captured as part of the tenant sub-metering for internal power consumption (i.e. lights, plugs, appliances.)

What metric is used to measure power consumption?

Hence, the main metric used has been the Power Usage Effectiveness (PUE), a standard developed by the industrial consortium Green Grid . This metric evaluates the total amount of energy consumed by a DC, versus the energy needed to run the IT subsystems (servers, routers, switches, Racks, etc.), and the best value should be close or equal to 1.

How can power-related data be used to predict energy consumption?

The power-related dataset is obtained by running different workloads on the top of power-systems. Another work in the literature using the LSTM-based model is proposed by to predict the energy consumed by a DC. The stacked-LSTM model is used to learn the characteristics of the data and improve prediction accuracy.