Energy demand/Data uncertainties limitations: Difference between revisions

From IMAGE
Jump to navigation Jump to search
mNo edit summary
No edit summary
 
Line 5: Line 5:
==Data, uncertainty and limitations==
==Data, uncertainty and limitations==
===Data===
===Data===
The energy demand module has been calibrated for the 1971–2007 period in order to reproduce historical trends in fuel and electricity use (see papers on individual model components, such as [[Van Ruijven et al., 2010a]]). Using the historical input data on population and value added and the calculated energy prices as given, other drivers and model parameters were varied systematically within the range of values derived from the literature, in order to improve the fit ([[Van Ruijven et al., 2010a]]; [[Van Ruijven et al., 2010b]]).  
The energy demand module has been calibrated for the 1971–2015 period in order to reproduce historical trends in fuel and electricity use. Using the historical input data on population and value added and the calculated energy prices as given, other drivers and model parameters were varied systematically within the range of values derived from the literature, in order to improve the fit ([[Van Ruijven et al., 2010a]]; [[Van Ruijven et al., 2010b]]).  


The primary data source on energy use was the International Energy Agency (IEA). These data were complemented with data from other sources, such as steel and cement demand and production, and transport data from as described in the references of the different model components. The residential submodule uses data from national statistical agencies and household surveys ([[Van Ruijven et al., 2010a]]).
The primary data source on energy use was the International Energy Agency (IEA). These data were complemented with data from other sources, such as total material demand and production, travel demand, etc., as described in the (key) references of the different model components.  


===Uncertainties===
===Uncertainties===
The main uncertainties in modelling energy demand relate to the interpretation of historical trends, for instance, on the role of structural change, autonomous energy efficiency increases and price-induced efficiency improvements and their projection for the future ([[Van Vuuren et al., 2008]]).
The main uncertainties in modelling energy demand relate to the interpretation of historical trends, such as the existence of saturation levels and the potential for efficiency increases. The representation in TIMER is based on the assumption that demand for energy services tends to become saturated at some point. This is based on physical considerations and historical trends in sectors, such as residential energy use. However, economic models assume that income and energy use remain coupled, often even at constant growth elasticities. Secondly, insufficient data is available to fully understand (historic) energy demand trends on which global energy models can be calibrated. Various methods have been developed to study the implications of different model calibrations, which have previously been applied to the transport and residential submodules ([[Van Ruijven et al., 2010a]]; [[Van Ruijven et al., 2010b]])
 
Two uncertainties are the existence of saturation levels and the potential for efficiency increases. The representation in TIMER is based on the assumption that demand for energy services tends to become saturated at some point. This is based on physical considerations and historical trends in sectors, such as residential energy use. However, economic models assume that income and energy use remain coupled, often even at constant growth elasticities. Evidence for a constant growth can also be found in some sectors, notably transport and services.
 
In deciding between these different dynamics, the extent to which historical trends would be the best guide for the future is also unclear. A similar issue concerns the role of energy efficiency. Many techno-economic analyses of efficiency potential suggest large possibilities at rather low payback times. However, from a historical perspective, investments in efficiency have been significantly lower than optimal for cost minimisation. Other factors must be assumed to play a role in the form of perceived transaction costs. A critical issue is whether this efficiency potential could be exploited in the future.
 
In the model calibration, there is a large degree of freedom in parameter setting so that results fit historical observations. A method has been developed to identify the implications of different outcomes of model calibrations and has been applied to the transport and residential submodules ([[Van Ruijven et al., 2010a]]; [[Van Ruijven et al., 2010b]]).
 
The starting point is that insufficient data are available to fully understand historic trends and calibrate global energy models. TIMER has room for different sets of parameter values that simulate historical energy use equally well, but reflect different historical interpretations and result in different future projections. The recent trend to replace some energy models by a description of end-use functions and applying physical considerations will reduce some uncertainties as this enables better estimation of reasonable saturation levels. However, this method suffers from the fact that new energy functions may be developed in the future that could increase energy demand.


===Limitations===
===Limitations===
The main limitations of the [[TIMER]] energy demand model are listed in the introduction to the model. A critical factor in modelling energy demand is the level of detail, given the large number of relevant technologies. TIMER uses an intermediate approach, in which some key technologies are modelled explicitly, and others are included implicitly. For more detailed estimates of the potential of energy efficiency, it would be more appropriate to use a different model.
The main limitations of the [[TIMER]] energy demand model are listed in the introduction to the model. A critical factor in modelling energy demand is the level of detail, given the large number of relevant technologies. TIMER uses an intermediate approach, in which some key technologies are modelled explicitly, and others are included implicitly. For more detailed estimates of the potential of energy efficiency, it would be more appropriate to use a different model. Similar for the complexity of supply chains: all demand sectors represent individual sectors that co-exist in parallel without exchange of information (other than fuel demand, fuel price), meaning that e.g. the residential and the cement sectors are not linked but are treated separately.
</div>
</div>

Latest revision as of 16:17, 22 May 2019

Link to framework components overview

Parts of Energy demand/Data uncertainties limitations

  1. Introduction page
  2. Model description
  3. Policy issues
  4. Data, uncertainty and limitations
  5. Overview of references
Component is implemented in:
Aggregated component:Components:
Related IMAGE components
Projects/Applications
Key publications
References
TIMER model, energy demand module
Some sectors are represented in a generic way as shown here, the sectors transport, residential and heavy industry are modelled in specific modules.

Data, uncertainty and limitations

Data

The energy demand module has been calibrated for the 1971–2015 period in order to reproduce historical trends in fuel and electricity use. Using the historical input data on population and value added and the calculated energy prices as given, other drivers and model parameters were varied systematically within the range of values derived from the literature, in order to improve the fit (Van Ruijven et al., 2010a; Van Ruijven et al., 2010b).

The primary data source on energy use was the International Energy Agency (IEA). These data were complemented with data from other sources, such as total material demand and production, travel demand, etc., as described in the (key) references of the different model components.

Uncertainties

The main uncertainties in modelling energy demand relate to the interpretation of historical trends, such as the existence of saturation levels and the potential for efficiency increases. The representation in TIMER is based on the assumption that demand for energy services tends to become saturated at some point. This is based on physical considerations and historical trends in sectors, such as residential energy use. However, economic models assume that income and energy use remain coupled, often even at constant growth elasticities. Secondly, insufficient data is available to fully understand (historic) energy demand trends on which global energy models can be calibrated. Various methods have been developed to study the implications of different model calibrations, which have previously been applied to the transport and residential submodules (Van Ruijven et al., 2010a; Van Ruijven et al., 2010b)

Limitations

The main limitations of the TIMER energy demand model are listed in the introduction to the model. A critical factor in modelling energy demand is the level of detail, given the large number of relevant technologies. TIMER uses an intermediate approach, in which some key technologies are modelled explicitly, and others are included implicitly. For more detailed estimates of the potential of energy efficiency, it would be more appropriate to use a different model. Similar for the complexity of supply chains: all demand sectors represent individual sectors that co-exist in parallel without exchange of information (other than fuel demand, fuel price), meaning that e.g. the residential and the cement sectors are not linked but are treated separately.

Retrieved from "https://models.pbl.nl/index.php?title=Energy_demand/Data_uncertainties_limitations&oldid=36050"