Data-driven, long-term prediction of building performance under climate change: Building energy demand and BIPV energy generation analysis across Turkey

Tamer, Tolga
Gürsel Dino, İpek
Akgül, Çağla
This research presents a methodological framework for lifetime energy demand and PV energy generation predictions for a given building considering the CC impacts through multivariate regression models. As a case study, a hypothetical office building in Turkey was selected. An existing linear morphing methodology was utilized to generate future weather files for all 81 cities in Turkey. For each year and city, corresponding weather metrics were calculated, and heating/cooling demand and PV energy generation values were computed through building energy simulations. Obtained data were used to develop two sets of multivariate regression models: (i) models to predict future weather metrics and (ii) models to predict future energy demand and generation. These models allowed lifetime energy demand and generation analysis (including associated GWP and cost) of the building considering CC impacts using only the current weather metrics of its location. For a lifetime of 60 years, considering CC impacts yielded substantially higher cooling (averaging at +0.5 MWh/m2 in the warmest region) and lower heating loads (averaging at − 0.4 MWh/m2 in the coldest region). For Turkey, the carbon intensity and the unit cost of cooling are much higher than those of heating. Therefore, the shift from heating to cooling has significant consequences in lifetime GWP and cost values (averaging +212 kg CO2-eq/m2 and +27 $/m2 , respectively, for the warmest region), emphasizing the importance of the decarbonization of the energy sector. The impact of CC on PV energy generation was limited (all-city average of +0.02 MWh/m2 for the building lifetime). Our regression-based approach can be further expanded to include not only various building parameters and types, but also supply-demand matching potentials.


Medium-Term Electricity Demand Forecasting Based on MARS
İlseven, Engin; Göl, Murat (2017-09-29)
The paper proposes use of multivariate adaptive regression splines (MARS) method to perform monthly electricity demand forecasting for medium-term. The model is developed based on specific example of Turkey; however is applicable to any other system. Performance of the proposed method is compared to that of multiple linear regression (MLR), generalized additive model (GAM), and artificial neural networks (ANN) methods. The validation process shows that the proposed model outperforms the other ones by test e...
A Policy design model for market formation of solar and wind electricity generation in Turkey
Erden Topal, Yelda; Erdil, Erkan; Department of Science and Technology Policy Studies (2016)
The aim of this study is to design technology policies for diffusion of solar and wind electricity generation technologies by analyzing the market formation dynamics from the perspective of key actors in the framework of technological innovation system approach. For this purpose, qualitative data is collected through 57 face-to-face semi-structured interviews with the key actors from private and non-private sectors who are engaged in economic activities of electricity generation, regulation of the market, c...
Cloud-based Optimal Energy Scheduling of Photovoltaics and Electric Vehicle-integrated Community Microgrids
Zehir, Mustafa Alparslan; Tufan Dogan, Osman; Merdanoglu, Hakan; Yakici, Ertan; Duran, Serhan; Can Akyildirim, Hayri (2022-01-01)
© 2022 IEEE.Community microgrid is one of the promising pathways to achieve higher levels of penetration of distributed generation from intermittent renewables and energy storage, further electrify heat and transport and enable active energy customers. Optimal energy scheduling of wide range and large number of flexible asset, using operational information from stakeholders (such dynamic pricing rates) and relying on customer preferences has been a processing power intensive major challenge. The inconsisten...
Assessing different sources of uncertainty in hydrological projections of high and low flows: case study for Omerli Basin, Istanbul, Turkey
Engin, Batuhan Eren; Yücel, İsmail; Yılmaz, Ayşen (2017-07-01)
This study investigates the assessment of uncertainty contribution in projected changes of high and low flows from parameterization of a hydrological model and inputs of ensemble regional climate models (RCM). An ensemble of climate projections including 15 global circulation model (GCM)/RCM combinations and two bias corrections (change factor (CF) and bias correction in mean (BC)) was used to generate streamflow series for a reference and future period using the Hydrologiska Byrans Vattenbalansavdelning (H...
Modeling electrical energy production in Northwestern Cyprus based on solar and wind measurements
Yenen, Mehmet; Fahrioğlu, Murat; Sustainable Environment and Energy Systems (2015-1)
This thesis presents the solar and wind energy assessment and aims to model the link between measurement and electrical energy production from wind and solar resources in Northwestern Cyprus. The measurement systems were installed and the measurements from these systems were analyzed thoroughly to meet the expectations of this thesis. Existing mathematical models were used to calculate electrical energy production figures for wind and solar energy. A circuit based Photovoltaic (PV) model from the literature...
Citation Formats
T. Tamer, İ. Gürsel Dino, and Ç. Akgül, “Data-driven, long-term prediction of building performance under climate change: Building energy demand and BIPV energy generation analysis across Turkey,” RENEWABLE AND SUSTAINABLE ENERGY REVIEWS, vol. 162, pp. 1–17, 2022, Accessed: 00, 2022. [Online]. Available: