Neighbourhood building stock model for long-term dynamic analyses of energy demand and GHG emissions. General model description and case studies
MetadataShow full item record
How should sustainable neighbourhoods be designed to reduce greenhouse gas emissions towards zero? What kind of information does decision makers need to make solid future plans on the neighbourhood level? A detailed understanding of a building stock’s characteristics and development over time is an underlying premise for reliable long-term building stock energy analyses. On the neighbourhood level, the building stock can be studied in large detail. Interactions between buildings and the local energy system can be analysed considering energy need, supply, local generation and local storage. Hourly resolution is needed to estimate peak heat and electricity loads in the neighbourhood. Further, greenhouse-gas (GHG) emissions resulting from the energy use in the buildings in the neighbourhood can be estimated by use of carbon intensities for the various energy carriers used in the neighbourhood. This report is deliverable D1.2.2 and a part of FME ZEN Work Package 1 Analytic framework for design and planning of zero emission neighbourhoods (ZEN). The goal for WP 1 is to develop definitions, targets and benchmarking for ZEN, based on customized indicators and quantitative and qualitative data. Additionally, life cycle assessment methodology for energy and emissions at neighbourhood scale will be developed, as well as a citizen-centred architectural and urban toolbox for design and planning of ZEN. A dynamic building stock model has been developed for energy- and GHG-emission scenario analyses of neighbourhoods. The model is generic and flexible and can be used to model any neighbourhood where building stock data is available. It makes use of a description of the current stock, as well as plans for construction, demolition and renovation activities in the neighbourhood. If plans are not available, the model may simulate stock activities by use of probability distributions. The neighbourhood building stock is segmented by use of archetypes defined by the buildings’ age, renovation state and floor area classes. Examples are grouping the two floor area types single family houses (SFH) and terraced houses (TH) together into a detached dwellings floor area class or grouping primary schools and secondary schools into a floor area class called “school buildings”. Hourly energy demand is estimated using delivered energy intensity profiles given for different archetypes of buildings or empirical data. Any number of different energy carriers and purposes can be defined and monthly or yearly carbon emission intensities can be given for each individual carrier. This serves as a basis to estimate hourly, monthly or yearly delivered energy and GHG emissions for a given neighbourhood under study. Two cases are analysed in this report: i) a hypothetical case of an imaginary neighbourhood consisting of apartment block (AB) and SFH dwellings, and ii) the Gløshaugen campus of the Norwegian University of Science and Technology (NTNU). Gløshaugen campus is a neighbourhood that has a high complexity of floor area types and usage. The purpose of the two very different case studies is not to provide reliable case studies at present, but to demonstrate how the model is capable of long-term analyses of both homogenous and complex neighbourhoods in order to offer detailed understanding of possible future hourly energy use and GHG emissions. For the hypothetical case, the model describes how the energy-efficiency of the stock improves over time due to renovation and demolition of older buildings and construction of new buildings with low energy need. The baseline scenario estimated annual delivered energy decrease from 150 kWh/m2 per 4 year at present to 90 kWh/m2 per year in 2070. Estimated GHG emissions decrease by 46% from 37 kton CO2-eq/year at present day to about 20 kton CO2-eq/year in 2070. Additionally, an advanced renovation scenario assuming that buildings being renovated have a higher probability of reaching better energy standards shows that the estimated annual delivered energy and GHG emissions will decrease faster in this scenario than the baseline scenario. Estimated annual delivered energy is 2% lower in 2025, 4% lower in 2030 and 7% lower in 2040 in the advanced renovation scenario than in the baseline scenario. Looking at aggregated GHG-emissions for the whole period, an estimated reduction of 8% from present day to 2070 compared to the baseline scenario is observed. Annual GHG emission gains compared to the baseline scenario are peaking around 2050 with 12% annual reduction of GHG emissions before natural renovation in the baseline scenario starts to catch up with the advanced renovation scenario. This is due to the fact that buildings in the baseline scenario go through renovation for the second time and reaches the third renovation state between 2050 and 2070. Constant monthly carbon intensities per energy carrier are assumed in the analysis, but it is likely that future monthly carbon intensities will change over the years of the period. A decrease in carbon intensities would lead to a further decrease in annual emissions over time. The neighbourhood building stock at NTNU campus Gløshaugen has a highly complex composition with 46 existing buildings (in total 300 000 m2 heated floor area) providing a large variety of functions related to education and research. 17 different floor area types are identified and distributed to 7 floor area classes. The planned future expansion of the campus is represented through construction of 120 000 m2 heated floor area before 2030. Average delivered energy intensity profiles per floor area class are modelled based on empirical data by using the simulation tool IDA ICE. The simulated profiles are used as energy model input. There is only one available energy profile per floor area class, regardless of the construction year and renovation state. Hence, the model is not able to estimate reductions in energy demand due to energy-efficiency of the stock through renovation and demolition of existing inefficient buildings or construction of new energy-efficient buildings. Carbon intensities are estimated monthly for district heating and grid electricity. Hourly and monthly peak loads, delivered energy and GHG-emissions are estimated for the whole neighbourhood at present year. The estimated long-term development in delivered energy and GHG emissions for Gløshaugen follows the stock development closely. This shows the weakness of using average profiles that do not reflect the differences in energy-efficiency state for buildings that are constructed in different periods or in different renovation states. A more detailed database of delivered energy intensity profiles is needed to create a more reliable long-term analysis taking into account stock activities and changes in the building stock characteristics. By changing different input parameters in the building stock, energy and GHG-emission model, different scenarios of future pathways can be studied. Various possible energy-efficiency measures can be analysed and compared with each other. This flexibility is a strength of the model as it makes analysing complex neighbourhoods possible. The model allows for creating roadmaps that decision makers can use when planning future development of neighbourhoods with building stocks and energy supply systems. The hourly time resolution makes it useful for electricity and district heating companies when planning future grid capacity need. The ability of the model to estimate and compare long-term changes in neighbourhood GHG emissions between scenarios makes it useful for decision makers aiming for future emission reductions.
PublisherSINTEF akademisk forlag