ZEMCH 2019 International Conference Proceedings April.2020 | Page 153
2. Methods
2.1 Disaggregation approach
In this study, the authors follow the previous disaggregation approach suggested by Kim et al.
(2019). A brief explanation is described here. South Korea has a temperate climate with four distinct
seasons. Winters (January and December) are typically cold and dry, while summers (July and August)
are hot and humid. Heating degree‐days (HDDs) and cooling degree days (CDDs) are approximately
2,191 and 769, respectively (Lee et al., 2013). Based on HDDs and CDDs, Figure 1 illustrates the energy
consumption pattern of most residential or commercial buildings. In the Figure, the Y‐axis indicates the
TEUI of the building (sum of gas, electricity, and district heating energy use), and the X‐axis represents
the month.
In the spring and autumn, for example May and October, the inflection points occur from heating
to cooling and from cooling to heating, respectively. If a certain building does not follow this pattern,
it may be regarded as an exception. In this pattern, the cooling‐related (Ec) and heating‐related (Eh)
energy usages can be disaggregated by subtracting the minimum value, regarded as the base load
energy (Eb), during the change in seasons from the monthly TEUI pattern.
The energy disaggregation utilizes the fact that the building energy consumption is lowest during
the change in seasons (spring and autumn). Accordingly, if a building includes large equipment (e.g.,
hospitals, banks, and data centers), then the base load energy can be large, but the heating and cooling
energy performance can still be fairly evaluated. The merits of this method are that there is no
assumption on linearity between heating (cooling) and outdoor temperature or heating (cooling)
degree‐days, and it is simple and inexpensive.
Figure 1. Monthly energy usage pattern in a temperate climate with four distinct seasons (source: Kim
et al., 2019)
2.2 Validation approach
To examine the adequacy of the aforementioned disaggregation method, the experiments are
designed as follows:
Experiment 1: the monthly TEUI is disaggregated into three indicators: namely Ec, Eh, and Eb.
Experiment 2: the monthly (1) electricity, (2) gas, and (3) other sources (e.g. district heating, LPG,
fuels, etc.) of energy are disaggregated into three indicators, respectively, and summed again into
to produce the final Ec, Eh, and Eb.
Experiment 3: the eight end‐uses of (1) cooling, (2) heating, (3) hot water supply, (4) lighting, (5)
ventilation (fan), (6) appliances, (7) indoor transportation, and (8) water supply, are disaggregated
into three indicators, respectively, and summed again to calculate the final Ec, Eh, and Eb.
The estimates (Ec, Eh, and Eb) from experiment 3 can be considered to be the best results of
experiments 1 and 2. If the results of experiment 1 or 2 are similar to that of experiment 3, the results
show the “practicality” of the disaggregation method by experiment 1 or 2, which only uses monthly
Disaggregation of weather-sensitive and -insensitive energy usage using monthly utility bill data
142