Consumer Household Energy Demands

Demands of consumer households for electricity, gas and other fuels such as forms of renewable energy are used to model the energy expenditure of individual households and identify how individual consumption changes relative to energy price and availability. Baker et al. (1989) provide theoretical models to construct a constant utility-based measure of household welfare to identify the welfare costs and potential benefits of energy price changes and subsidies paid to particular energy items. The models provided by Baker et al. (1989) in addition provide econometric specification of energy-demand equations. Two models to estimate and quantify consumer demand and energy expenditure are given by Baker et al. (1989). The first is the two-stage budgeting model, which implements a recursive structure that allocates household income between fuels and non-fuels and secondly disaggregate fuel expenditure. The second model is an empirical specification, which exploits variation in demographic and other characteristics across households. Oladokun and Odesola (2015) review how household energy consumption and carbon emission modeling have evolved compared to historic research and models. According to Oladokun and Odesola (2015), primary approaches to model energy consumption and carbon emissions are econometric, building physics and statistical models. Importantly, the limitations of these models are also highlighted and these limitations are evident in many eco-city modeling techniques, typically because of the complexity and multivariate nature of energy demand. Limitations such as lack of transparency in model algorithms, complex and interdependent environments, limited indications of occupant-dwelling interactions and lack of qualitative input data hinder many energy consumption and supply-demand interaction models. Interdisciplinary approaches are required to holistically quantify household energy demand and consumption leading to carbon emissions. Interaction among the disciplines of complex technology, society, economics, climatology and culture (Oladokun and Odesola 2015) attempts to overcome the restraints of simplified approaches. Motawa and Oladokun (2015) expand on the research methodology through problem articulation, model boundaries, dynamic hypothesis, causal loop diagrams and model formulation of complex household energy consumption. Motawa and Oladokun (2015) conclude that the judgement and reasoning of researchers and industry practitioners have proven more valuable in optimizing household energy consumption and carbon emissions compared to data collected on household demands on energy. Multivariate and interdisciplinary data on energy consumption lead to complex and scenario-specific models, which must be altered and optimized for variability of the immediate environment, social aspects and economic prosperity. Essentially, a complex model of energy dependency of a household requires model boundaries distributed along three core types of variables, namely

  • • variables exhibiting endogenous behavior, whose value is therefore dependent and determined by the current or previous states of additional variables within the enclosed system,
  • • exogenous variables, which are factors in causal models whose value is not dependent from the current or previous states of additional variables within the enclosed system, and
  • • excluded variables.

In household energy consumption modeling, for example, instances of endogenous variables included in model boundary definitions include the effects of insulation on energy efficiency, the effects on energy efficiency of standard appliances, total household carbon emissions and also the number of births, mortality and household size. These all determine the energy demand of the household. Motawa and Oladokun (2015) list approximately 50 additional endogenous dynamic variables taken into account during modeling of household energy demand. Examples of exogenous variables include, among the detailed list of parameters listed by Motawa and Oladokun (2015), total floor area, solar flux, external air temperature and relative humidity. Each variable is independent of the states of other variables in the system. Excluded variables may, for example, include parameters related to the behavior, culture and social preferences of occupants, certain physical parameters of dwellings, such as their orientation, as well as parameters correlated to the external environment, such as new technologies, political qualms and energy security. Household classification and related end uses in addition structure and categorize demand for energy in households based on socio-economic parameters. Household groupings into, for example, low-, middle- and high-income electrified and non-electrified datasets determine the primary consumption areas or end uses within a household. These end uses typically include, for electrified households, food refrigeration, preparing and cooking of food, inside and outside lighting, space heating of rooms and living areas, water heating for consumption or washing and other end uses such as technology, entertainment and irrigation. For non-electrified households, the primary uses of energy are lighting, cooking, space heating and water heating only.

Marszal-Pomianowska et al. (2016) present a high-resolution model of household electricity from a combination of measured and statistical data in a bottom-up approach. The effects of parameters such as the number of occupants in a household, as well as their attitude to energy use, are included in the model. The structure of the implemented model is presented by Marszal-Pomianowska et al. (2016); it highlights input data for each appliance in the household, which includes activity probability based on the time of day and the day of the week and also the frequency of use of each appliance, the number of occupants and their respective approaches to energy savings, as well as certain seasonal variations and social factors that influence the use of energy. Essentially, the model creates a list of all appliances installed in the household, determines usage for each appliance and usage of energy of the household and generates resultant load profiles of specified resolution. The model is validated by obtaining reliable data for demand on overall energy use, a more practical approach than accurate data on the demand of single appliances. Results are categorized into the annual electricity demands of the individual households, with variations in the number of occupants, seasonal demand, peak and average demand, as well as daily load profiles. Potential enhancements of this model include improvements on algorithms used for lighting, including socio-economic factors and extending the model for various building topologies.

The following section reviews modeling considerations of the industry, commerce and mining sector, a heavy carbon-emitting industry compared to household energy demands and the largest contributor to carbon pollution through the burning of fossil fuels.

 
Source
< Prev   CONTENTS   Source   Next >