Department of Engineering Sciences

Built Environment Energy Systems Group (BEESG)

In the Built Environment Energy Systems Group we study energy systems in buildings and energy systems in which buildings are components. How are district heating systems affected by energy efficiency improvements in buildings? What is the potential for on-site photovoltaics in the built environment? How can we construct smart grids with active buildings?

We approach this type of questions with methods for systems analysis and broad interdisciplinary collaborations. Our ongoing research includes optimisation studies of local and national power and district heating systems, variability assessment of solar power in the built environment, self-consumption of PV electricity in residential buildings, and development of simulation tools for solar energy systems

Based on our research we also develop stand-alone programs from our research that can be used for free for students and researchers. These programs can be found here.

Ongoing research areas

Read our publications

System integration of distributed solar power in the built environment

Over the last two decades, the installed capacity in photovoltaic (PV) systems worldwide has been increasing at an almost exponential rate. Consequently, different countries around the world are at different stages in integrating solar power into electricity markets and systems. This research field addresses various challenges and possible solutions associated with increasing the shares of solar power in the built environment and in the power system.

Examples of research questions are: How is solar power utilized most effectively in different types of buildings and communities? How can large shares of solar power be integrated in local distribution grids? How can the variability in large fleets of PV systems be quantified, predicted and handled?

Utilization of PV electricity in buildings and communities

Self-consumption of on-site PV electricity in buildings is gaining interest as a way to accommodate high PV penetrations in the power system. On markets where there is no substantial feed-in support for renewables, there is an economic incentive for PV self-consumption, as selling prices for PV electricity are normally lower than retail electricity prices.

Schematic outline of different options for increased PV self-consumption.

In the research project Small-scale solar electricity in buildings – Power for change in energy systems and everyday life, financed by the Swedish Energy Agency, we investigate the potential for different technological solutions to enhance the self-consumption of PV electricity in different types of buildings, communities and distribution grids, e.g. battery storage and control of HVAC systems in buildings. In collaboration with Linköping University we also study the behavioural impacts of PV installations.

Contact persons: Ph.D. student Rasmus Luthander and senior lecturer Joakim Widén.

Integration of large shares of solar power into electricity distribution grids

Experience and previous research on high-penetration PV in distribution grids have shown that high excess PV power can lead to overvoltage or overloading of grid components. Existing distribution grids have a certain hosting capacity for PV systems, below which relevant performance indices are kept within prescribed bounds. In order to accommodate higher shares than around 30-50% PV energy as a share of the annual demand, different solutions for increasing the hosting capacity have to be implemented. We investigate the potential for these together with the DSO Herrljunga Elektriska, in the project Evaluation of technological solutions for managing extensive connection of photovoltaic systems in electricity distribution grids, financed by the Swedish Energy Agency.

In the project Solar electricity in agriculture, financed by the Swedish Energy Agency, we have investigated the potential PV power generation on marginal lands and rooftops on agricultural properties, and quantified the hosting capacities of rural distribution grids. In collaboration with the Division of Industrial Management and Engineering, we also study demand response as a way to reach an efficient use of the distribution grids.

Contact person: Senior lecturer Joakim Widén.

Spatio-temporal variability in solar irradiance and solar power

With large amounts of solar power in power systems the need for balancing reserves on different time scales increases. When building-applied PV systems are integrated extensively into the built environment, cities may become big solar power producers. The solar power variability and availability will then depend on the geographical extent of the city and how the weather changes over it. To determine the need for increased system flexibility to handle this situation, new methods for quantifying and modeling the aggregate power output of an arbitrary set of distributed power plants are required.

Solar irradiance variability measurement
To quantify solar irradiance variability we use sensors based on a microcontroller with a photodiode, GPS och SD card (a) enclosed in a box with a teflon window (b) for easy mounting on buildings (c).

In the research project UppScaleSolar, financed by Energiforsk and the Swedish Energy Agency, we characterize aggregated solar power generation from extensive city-scale PV in future scenarios for the city of Uppsala. We develop models for geographically varying solar radiation and PV power generation on the city level and validate these with a monitoring network of solar irradiance sensors. In collaboration with the Division of Electricity, we also simulate large-scale solar, wind and wave power in future scenarios for the Nordic power system.

We have also developed a stand-alone programs for generating synthetic solar irradiancedata, which can be accessed here.

Contact persons: Ph.D. student David Lingfors and senior lecturer Joakim Widén.

Forecasting solar energy, electricity use and net-demand

Low variability and high reliability characterises the output of traditional electricity generators such as fossil-fuel and nuclear power plants. For network electricity operators, this meant that the task of matching electricity consumption and generation was a fairly straightforward one, even though electricity consumption has become more variable over the past decades. However, due to the increasing share of variable renewable energy sources (RESs) in the electricity generation mix, the generation side of the balance has become increasingly variable. In order to efficiently integrate large shares of RESs, such as PV power, into the modern electricity mix and retain a stable net demand in the power system, accurate probabilistic forecasts can be a cost-efficient solution. By communicating the uncertainty in future PV power production, stakeholders can minimise risk and ensure reliable power supply throughout the system

In this project, which is financially supported by the Swedish Energy Agency, we develop models to produce probabilistic forecasts taking into consideration both time and space. In addition, we investigate key features of probabilistic forecasts to increase the general understanding, for instance when PV power profiles are aggregated or when the share of PV power increases.

From our latest paper, which has the caption: "Probabilistic forecasts of net demand of (a) the GPs during the entire month of April 2012, (b) the GPs during a selection of days in April, (c) the ARIMA during the entire month of April 2012 and (d) the ARIMA during a selection of days in April. The shaded areas represent the 80% prediction intervals."

Contact persons: Assistant Professor Joakim Munkhammar


  • R. Luthander, J. Widén, D. Nilsson, J. Palm, Photovoltaic self-consumption in buildings: A review, Applied Energy 142 (2015) 80-94.
  • J. Widén, E. Wäckelgård, J. Paatero, P. Lund, Impacts of distributed photovoltaics on network voltages: stochastic simulations of three Swedish low-voltage distribution grids, Electric Power Systems Research 80 (2010) 1562-1571.
  • J. Widén, N. Carpman, V. Castellucci, D. Lingfors, J. Olauson, F. Remouit, M. Bergkvist, M. Grabbe, R. Waters, Variability assessment and forecasting of renewables: A review for solar, wind, wave and tidal resources, Renewable and Sustainable Energy Reviews 44 (2015) 356-375.

Electricity use in buildings

An increased amount of distributed and variable power production – such as photovoltaic (PV) power production - presents a challenge for grid operation and control. In order to assess the grid impacts, it is important with realistic models for electricity consumption in buildings and how it interacts with distributed intermittent generation. The main goal of our research in this field is to improve modeling of energy load profiles for use in building simulations and load flow calculations.

Household electricity use

When modeling energy use profiles in buildings – for electricity, space heating and domestic hot water – the activities of the residents are an important factor. Both the activities and the resulting load profiles are stochastic, which should be reflected in load models. A high-performance load model should reproduce key properties of energy use such as diurnal and annual variations, short time-scale fluctuations, load coincidence and diversity between households. 

In our group we have developed a stochastic Markov-chain model for household electricity use, which has been applied to building simulations and load flow calculations for power grids and has been widely referred to in the literature. The stochastic model has also been extended to include EV charging. Recently, we have also begun developing models based on probability distributions fitted to high-resolution data.

Household electricity
A surface plot of the household electricity use as characterized by the probability distribution model developed in the group.

Contact persons: Senior lecturer Joakim Widén and researcher Joakim Munkhammar.


  • Widén, J., Nilsson, A. M., Wäckelgård, E., A combined Markov-chain and bottom-up approach to modelling of domestic lighting demand, Energy and Buildings 2009: 41; 1001-1012.
  • Widén, J., Wäckelgård, E., A high-resolution stochastic model of domestic activity patterns and electricity demand, Applied Energy, 2010: 41; 1880-1892.
  • Munkhammar, J., Rydén, J., Widén, J., Characterizing probability density distributions for household electricity load profiles from high-resolution electricity use data, Applied Energy 2014: 135; 382-390.

Electric transportation

The Electric Transport Group, affiliated with BEESG, investigates electric vehicle (EV) charging by means of developing and using mathematical and statistical models to quantify, predict and forecast charging patterns over time and space. Electric vehicle charging is also studied from the perspective of local electricity use and in combination with local power generation, typically solar power, and in studies of impacts on the electricity grid. The main goal of the research is to facilitate the electrification of transport with mathematical modelling.

We develop models for simulating EV charging based on different charging strategies and availability of charging stations. These models are also extended to include PV power production as intermittent power sources in order to determine the impact on the power grid and the potential for self-consumption. Additionally, real-life EV charging data is analyzed, which will be incorporated in future models for improved model accuracy.

Studies made in our group have found that the matching between power production from PV and uncontrolled electric vehicle home-charging is limited, and that the introduction of EV charging presents new challenges for the grid in terms of high intermittent power demand. Characterizing EV charging – at home or at other locations – is necessary for design and operation of future grids, in particular for large-scale deployment of charging stations. There are also challenges in identifying and evaluating smart-charging strategies.

Current research is focused on electric vehicle charging on city scale, in particular in combination with local solar power, which is part of the EU ERA-NET Smart Grids Plus project SolarCharge2020 [Länk till]. This project is made in collaboration between Uppsala University, Solelia Greentech AB, Uppsala Parkerings AB, Uppsala Municipality, Arctic University in Tromsø, and Tromsø fylkekommune.

We are also part of the Swedish Electromobility Centre.

Daily mean power use from household activities and electric vehicle charging generated using the Widén household electricity use model and the Grahn-Munkhammar electric vehicle home-charging model (See references below).

We have also developed the Grahn-Munkhammar model as a stand-alone software, which can be accessed here.

Contact persons: Group leader Assistant Professor Joakim Munkhammar.


  • Widén, J., Nilsson, A. M., Wäckelgård, E., A combined Markov-chain and bottom-up approach to modelling of domestic lighting demand, Energy and Buildings 2009: 41; 1001-1012.

  • Widén, J., Wäckelgård, E., A high-resolution stochastic model of domestic activity patterns and electricity demand, Applied Energy, 2010: 41; 1880-1892.

  • Munkhammar, J., Bishop, J.D.K., Sarralde, J. J., Tian, W., Choudhary, R., Household electricity use, electric vehicle home-charging and distributed photovoltaic power production in the city of Westminster, Energy and Buildings 2014: 86; 439-448.

  • Munkhammar, J., Rydén, J., Widén, J., Characterizing probability density distributions for household electricity load profiles from high-resolution electricity use data, Applied Energy 2014:135;382-390.

  • Munkhammar, J., Rydén, J., Widén, J., On a probability distribution model combining household power consumption, electric vehicle home-charging and photovoltaic power production, Applied Energy 2015: 142; 135-143.

  • Grahn, P., Munkhammar, J., Widén, J., Alvehag, K., Söder, L., PHEV Home-Charging Model Based on Residential Activity Patterns, IEEE Transactions on Power Systems 2013: 28; 2507-2515.

  • Luthander R., Shepero M., Munkhammar J., Widén J., Photovoltaics and opportunistic electric vehicle charging in a Swedish distribution grid, 7th Internation Workshop on Integration of Solar into Power Systems in Berlin 2017.

Energy use in buildings and its impact on district heating production

In the EU, and most of the developed world, the building sector accounts for approximately 40% of the total energy use. The potential for energy savings is significant, and the recast of the Energy Performance of Buildings Directive (EPBD) is an important legislative measure to improve energy-efficiency in buildings. The EPBD states that all new buildings should be nearly zero energy buildings (NZEB) by 2020, and it emphasizes the importance to promote measures that transform existing buildings undergoing major renovations into nearly zero energy buildings. How can significant building energy use reductions be achieved? How can energy-efficient buildings be incorporated with surrounding energy systems, such as district heating and power systems? 

Energy efficiency in buildings

The common principle for nearly zero-energy buildings is that it should be a building with outstanding energy performance and a prominent part of its energy demand should be supplied by energy from renewable energy sources. However, the exact requirements are defined on a national level. In Sweden, the proposed definition is based on an energy performance parameter and the ability to count on-site and nearby renewable energy production towards the energy performance. One of the challenges with the concept of nearly zero-energy buildings is that it should be in line with the general EU targets of reduced greenhouse gas emissions, reduced energy use and increased share of renewable energy sources. An implication of this is that local variations in energy systems and temporal disparities in renewable energy supply influence the building’s coherence with the general EU targets. In our research we study the building energy balance, its role in the energy system, and possible trade-offs between the two.

Using a case study we have for example examined the combination of energy efficiency measures and transformation of the energy system for a community located outside the urban energy system. The community consists of semi-detached and detached houses and was constructed in the end of the 1960s. The houses are in need of renovations and the project examines the trade-offs between local combined heat and power (CHP), PV electricity, and energy efficiency measures in terms of self -sufficiency and primary energy use.

District heating adaption to building energy efficiency

92 % of the heat used for space heating and domestic hot water in Swedish multi-family residential buildings is produced and supplied in Sweden´s about 500 district heating systems. These systems are, however, not only used for production and distribution of heat. Production of district heating is unique in the sense that it is possible to combine with other processes, such as incineration of household waste, industrial production and thermal power generation. District heating also provides the possibility to utilize low exergy sources that would otherwise be wasted.

A reduced heat demand in district-heated buildings might, therefore, affect other systems and processes, such as waste treatment and the balance of the power system.  Also, the district heating sector in Sweden was affected by the deregulation reform of the Swedish power system in 1996. This reform changed the status of district heating from being generally considered a public utility for heat supply into being a heat supply technology in competition with other heat supply technologies, such as heat pumps and in-house boilers. The power system reform also opened up district heating business for private actors and pricing on district heat was liberalized. This can be considered a risky business since district heating systems are structured in many aspects as natural monopolies.

Contact person: Researcher Magnus Åberg and associate senior lecturer Annica Nilsson.