**1. Introduction**

The European Union (EU) aims to reduce CO2 emissions by 80% compared to 1990 levels by the year 2050 [1]. The EU recognizes electrification of heating through heat pumps as one key step towards decarbonization [2]. Since buildings are responsible for 40% of EU's energy demand and emissions, the Energy Performance of Buildings Directive (EPBD) was created to reduce emissions caused by the operation of future buildings [3]. However, despite tightening regulation of new buildings, the existing energy-inefficient building stock that has mostly been built before modern regulations remains the biggest source of building-based emissions. To tackle this issue, the EU has called for national retrofit strategies to effect a positive change in the existing buildings as well [4]. Finnish building code also requires the consideration of energy efficiency whenever renovation tasks are performed on buildings [5]. This is important, as 79% of Finnish buildings have been built before the year 2000 [6] and certain mandatory renovation work provides a chance for lower cost energy efficiency improvements as well.

The influence of building retrofits on energy demand has been examined in many studies. The case of various Italian building types was examined in [7]. The results show the cost-e ffectiveness of di fferent levels of energy e fficiency, showing that energy consumption in single-family houses could go down by up to 77%. Similarly, the energy saving potential in Swedish detached houses was found to be 65–75% in [8]. Here the focus was on standardized buildings built between the years 1961 and 1980, increasing the usability of the suggested actions. Four reference buildings from di fferent regions were dynamically simulated in IDA-ICE, which is a multi-zone simulation tool for evaluating indoor thermal conditions and building energy consumption. Retrofit measures were added step-by-step based on their prevalence in real life, until the energy e fficiency matched passive houses. In Ireland, deep energy retrofits in semi-detached houses were only feasible with governmen<sup>t</sup> grants [9]. This study used six di fferent environmental indicators and included the environmental impact of the materials needed for retrofitting in addition to the operational impact. Thirty-five pre-determined retrofit packages were calculated using quasi-steady state equations in the DEAP software, a web-based tool for producing Building Energy Ratings. In the Finnish context, energy retrofits have been examined for apartment buildings [10,11], where both primary energy demand and CO2 emissions could be significantly reduced cost-e ffectively, especially using heat pumps, but also including improvements to the building envelope. In Finnish o ffice buildings CO2 emissions could be cost-e ffectively reduced by 50% while preserving thermal comfort [12]. In Finnish detached houses [13] life cycle costs and CO2 emissions could be most e ffectively lowered by deep energy retrofits in buildings with direct electric or oil-based heating. In these studies, dynamic simulation with IDA-ICE was combined with multi-objective optimization by a genetic algorithm, to go through hundreds of retrofit packages. Heat pumps in particular stand out as a good heating solution for the future [14]. As heating demand is increasingly met with electricity, short term fluctuations in electric power use can be expected to increase. However, all these studies have reported their results only on an annual level, leaving the seasonal changes unknown and providing no information about the peak power levels on short timescales.

The power levels in shorter time scales are important for the development of the energy generation infrastructure, especially as the share of undispatchable renewable energy increases. Investments into new power plants are based on capacity-based costs and the variable energy prices [15], which are again influenced by the instantaneous power demand and the available energy generation capacity. Increase of peak power demand may call for increased investments to power transmission lines [16] or to demand response services which are used to shift peak demand to lower load hours [17]. Strategies for reducing peak demand under uncertain loads are being developed such as in [18], which highlights the importance of energy flexibility in buildings. Important strategies include the price mechanism effect on occupant's behavior, centralized energy managemen<sup>t</sup> with demand response and HVAC peak load controls. A review on power system planning studies concluded that when modelling the power grid, smaller details such as heating systems should also be taken into account [19]. On the building side, a Norwegian study did report some monthly results on the deep retrofit of apartment buildings [20], though the main goal was to show the feasibility of evaluating retrofits using only the hottest and coldest months. A Spanish study showed the monthly renewable energy use and changes in peak demand after building retrofits [21]. Peak power demand reduction was also the focus of a study made for Dubai's cooling dominated climate [22]. Hourly power demand after building retrofits has been reported for Finnish apartment buildings [23], but a similar study on detached houses was not found. In many studies on building energy retrofits, heat pumps have been presented as the lowest emission heating solution. This is to be expected as there is a lot of low emission electricity generation in the Nordic countries, such wind power (Denmark and Sweden), nuclear power (Finland and Sweden), and hydro power (Finland, Norway, and Sweden) [24]. However, in a Sweden-based study on building retrofits, new heat pump systems were assumed to increase the total electricity use, thus forcing the use of high emission fossil fuel sources, under the assumption that all existing low emission generation is already in use [25]. This raises the question of whether the demand in other buildings could be reduced to make the existing supply of low emission electricity to stretch further or if the benefits of heat pumps have been exaggerated.

Many articles have been published on emission reduction and energy efficiency improvements in different building types. However, typically only annual changes to emission levels or energy consumption are reported, while few retrofit studies focus on the potential effects on the grid. In this paper, the changes in power levels of Finnish detached houses with different heating systems are examined individually and on the building stock level. The studied scenarios are based on previously optimized emission reducing configurations. The key questions are: What is the impact of deep energy retrofit on the seasonal and peak district heating and electric power demand of Finnish detached houses? How does the excess solar electricity generation of optimized building configurations compare to the hourly demand? What is the potential impact of large-scale building retrofits and electrification of heating on the electric power requirements of the whole detached house building stock?
