Energy demand - TIAM-UCL

From IAMC-Documentation
Jump to: navigation, search
Model Documentation - TIAM-UCL
Corresponding documentation
Model information
Institution University College London (UCL), UK, https://www.bartlett.ucl.ac.uk/energy.
Solution concept Partial equilibrium (price elastic demand)
Solution method Linear optimisation
Anticipation Perfect Foresight

(Stochastic and myopic runs are also possible)

Demand drivers (population, GDP, family units, etc.) are obtained externally, via other models or from other sources (e.g. UN statistics, World Bank, IEA). Energy-service demands and respective drivers in the TIAM-UCL are presented in Table 2-1. The demands for energy services are linked to the drivers' projections via elasticities, see bellow. These elasticities of demands are intended to reflect changing patterns in energy service demands in relation to socio-economic growth, such as saturation in some energy end-use demands, increased urbanization, or changes in consumption patterns once the basic needs are satisfied. The energy-service demands for future years are projected using the following relationship:

35815634.png

Where k is a constant equal to 1 for most of the energy services demand. The constant k is population and number of households when the driver is GDP per Person (GDPP) and GDP per Household (GDPPHOU) respectively.

Table 2-1: Energy-services demand and respective drivers

Code Description Unit Driver
ICH Chemicals PJ PCHEM
IIS Iron and Steel Mt PISNF
INF Non-ferrous metals Mt PISNF
INM Non Metals PJ POEI
ILP Pulp and Paper Mt POEI
IOI Other Industries PJ POI
I00 Other Industrial consumption PJ Constant
NEO Industrial and Other Non Energy Uses PJ GDP
ONO Other non-specified consumption PJ GDP
AGR Agricultural demand PJ PAGR
CC1 Commercial Cooling - Region 1 PJ PSER
CCK Commercial Cooking PJ PSER
CH1 Commercial Space Heat - Region 1 PJ PSER
CHW Commercial Hot Water PJ PSER
CLA Commercial Lighting PJ PSER
COE Commercial Office Equipment PJ PSER
CRF Commercial Refrigeration PJ PSER
RC1 Residential Cooling - Region 1 PJ HOU/GDPPHOU*
RCD Residential Clothes Drying PJ HOU/GDPPHOU*
RCW Residential Clothes Washing PJ HOU/GDPPHOU*
RDW Residential Dishwashing PJ HOU/GDPPHOU*
REA Residential Other Electric PJ HOU/GDPPHOU*
RH1 Residential Space Heat - Region 1 PJ HOU
RHW Residential Hot Water PJ POP
RK1 Residential Cooking - Region 1 PJ POP
RL1 Residential Lighting - Region 1 PJ GDPP
RRF Residential Refrigeration PJ HOU/GDPPHOU*
NEU Non Energy Uses PJ GDP
TAD Domestic Aviation PJ GDP
TAI International Aviation PJ GDP
TRB Road Bus Demand Bv-km POP
TRC Road Commercial Trucks Demand Bv-km GDP
TRE Road Three Wheels Demand Bv-km POP
TRH Road Heavy Trucks Demand Bv-km GDP
TRL Road Light Vehicle Demand Bv-km GDP
TRM Road Medium Trucks Demand Bv-km GDP
TRT Road Auto Demand Bv-km GDPP
TRW Road Two Wheels Demand Bv-km POP
TTF Rail-Freight PJ GDP
TTP Rail-Passengers PJ POP
TWD Domestic Internal Navigation PJ GDP
TWI International Navigation PJ GDP
  • The driver is GDPPHOU for AFR, CHI, CSA, EEU, FSU, IND, MEA, MEX, ODA and SKO
Driver Elasticity

Driver elasticities determine the sensitivity of changes in energy-service demand to changes in the underlying driver. An elasticity of 1 means that a change of the underlying driver is exactly reflected in the energy-service demand. Energy-service demands with an elasticity below 1 are demand inelastic, while those with an elasticity of one or higher are demand elastic. In general it is assumed that energy-service demands grow slower than the underlying driver, such as GDP, GDP per capita or number of household. This decoupling of energy demand and economic growth is expected to increase during the 21st century so that all elasticities fall. Residential space heating (RH1), for example, has an elasticity of 0.8 in 2010, which drops to 0.5 in 2100. This means that initially the energy demand for space heating increases at 80% of household number growth, the specific underlying driver, and in the 2nd half of the century at only 50% of the household number growth rate.

Table 2-2: Driver elasticities for the United Kingdom

Energy-service demand 2010 2020 2030 2040 2050 2100
AGR 0.8 0.8 0.8 0.8 0.8 0.6
CC1 0.8 0.8 0.8 0.8 0.7 0.4
CCK 0.5 0.5 0.5 0.5 0.5 0.4
CH1 0.5 0.5 0.5 0.5 0.5 0.3
CHW 0.5 0.5 0.5 0.5 0.5 0.4
CLA 0.5 0.5 0.5 0.5 0.5 0.4
COE 0.5 0.5 0.5 0.5 0.5 0.4
COT 0.5 0.5 0.5 0.5 0.5 0.4
CRF 0.5 0.5 0.5 0.5 0.5 0.4
I00 0.6 0.6 0.6 0.6 0.6 0.5
ICH 0.8 0.8 0.8 0.8 0.7 0.5
IIS 0.7 0.7 0.7 0.7 0.7 0.5
ILP 0.8 0.8 0.8 0.8 0.7 0.5
INF 0.8 0.8 0.8 0.8 0.7 0.5
INM 0.8 0.8 0.8 0.8 0.7 0.5
IOI 0.8 0.8 0.8 0.8 0.8 0.6
NEO 0.6 0.6 0.6 0.6 0.6 0.5
NEU 1 1 1 1 0.9 0.5
ONO 0.6 0.6 0.6 0.6 0.6 0.5
RCD 1 1 1 1 1 0.8
RCW 1 1 1 1 1 0.8
RDW 1 1 1 1 1 0.8
REA 1 1 1 1 1 0.8
RH1 0.8 0.8 0.8 0.8 0.8 0.5
RK1 0.7 0.7 0.7 0.7 0.7 0.5
RL1 1 1 1 1 0.9 0.7
ROT 1 1 1 1 1 0.8
RRF 1 1 1 1 1 0.8
RHW 1 1 1 1 1 0.8
TAD 1.2 1.2 1.1 1.1 0.9 0.1
TAI 1.2 1.2 1.1 1.1 0.9 0.1
TRB 0.7 0.7 0.7 0.7 0.7 0.8
TRC 0.7 0.7 0.7 0.7 0.7 0.4
TRE 0.7 0.7 0.7 0.7 0.7 0.7
TRH 0.7 0.7 0.7 0.7 0.7 0.4
TRL 0.7 0.7 0.7 0.7 0.7 0.4
TRM 0.7 0.7 0.7 0.7 0.7 0.4
TRT 1.2 1.2 1.2 1.2 1 0.5
TRW 0.7 0.7 0.7 0.7 0.7 0.7
TTF 1 1 1 0.8 0.6 0.1
TTP 0.8 0.8 0.8 0.8 0.8 0.7
TWD 0.8 0.8 0.8 0.6 0.5 0.1
TWI 0.8 0.8 0.8 0.6 0.5 0.1

Non-energy demands are not explicitly considered.

Regional GDP per capita is a driver for the model, but there are no income distribution within a given region. Access issues are not considered either.

Regions can be split to additional subregions for the demand level, thus allowing to model demand separately for, for example, urban and rural areas in the Residential sector. Currently, USA and CAN have four and three geographic regions, respectively, while AFR, CHI, IND, MEA and MEX each have two ?sub-regions?, corresponding to rural and urban areas.

Behavioural change

Behaviour and heterogeneous agents are mostly not explicitly considered but are represented via price mechanisms e.g. there is no modal shift in the transport sector. With the exceptions of technology and region specific discount rates and price responsive energy service demands i.e. see Residential sector. Diffusion constraints can be implemented to simulate behavioural inertia (among the other barriers that are not explicitly included in the model).