You are here

Guidelines for a well energy balanced district

Energy Balance of a District

Jürgen Schnieders, Passivhaus Institut,

As part of the Sinfonia project, the Passive House Institute has developed a calculation programme to assess and optimise city districts in terms of their energy efficiency. districtPH calculates detailed energy balances for buildings within the neighbourhood. Heat or electricity production in the district, both centrally and in individual buildings, is considered in the total energy balance. It is also possible to account for public supply structures as well as public consumers.

The development was aiming at two major fields of interest:

  • The energy balance of the district, including heat and electricity generators and grids, at a given point in time. Questions such as 'What is required to make the district zero-energy?', 'What would be an appropriate size for a seasonal heat storage?', 'How much energy will be exported from the district in a specific situation?' can be addressed.
  • The interaction of current and future retrofits with supply structures. Possible projections include the total primary energy demand or the CO2 emissions over several decades, depending on different scenarios for e.g. retrofit subsidies or district heating network installations. The probability of a refurbishment to a certain efficiency level can be defined, depending on factors like component age, time, subsidies, or existing efficiency level. The difficulties arising from the probabilistic nature of refurbishment rates were solved by implementing a Monte Carlo Simulation method.

Figure 1: districtPH

Source: Passive House Institute

Basic structure of districtPH

For reasons of flexibility and transparency, districtPH was realised as an Excel spreadsheet. Building-related energy consumption and the effects of refurbishment measures play a central part. In addition, districtPH considers user-related energy consumptions in buildings. This is supplemented by street lighting and the energy consumption of trams and other electric vehicles. A district heating system and the electricity grid, including short and long-term storages, can be represented. Estimation methods for the energy production from renewable sources were integrated.

For some purposes algorithms from the [PHPP] could conveniently be integrated, many other methods were developed from scratch. We assumed that the data acquisition would not be as accurate as when planning a new building. Design drawings will not be available, neither will exact component qualities, numbers of inhabitants, etc. Once accepted, this fact allows for entering the buildings in the district by assigning them to certain pre-defined building types from a typology. For the current version, due to the inhomogeneous availability of data, it was decided not to develop an import filter for GIS data, e.g. from CityGML.

One of the major goals in the development of districtPH is the prediction of how the district's energy demand evolves over time. Since future developments will always depend on many currently unknown parameters, the calculation results will have an unavoidable inaccuracy, which in turn justifies time-saving simplifications of the calculation methods themselves.

The first step in setting up an energy balance is to enter the buildings in the district. Each building is assigned to one of up to 30 building types, which can be either user-defined or chosen from the Episcope database ([Tabula 2018]). The buildings with their type, their positions and square meters of floor area can be entered. The building types already contain efficiency levels for all building components and the mechanical systems.

Excel can now calculate the energy balance of each building type with regard to heating, cooling, hot water, and electricity, and report the sum totals of e.g. delivered energy, CO2 emissions, or source energy. The relevant results are saved, and the district moves on to the next year. Now, with a user-defined probability, a retrofit of the building components to a different efficiency level takes place, and the calculation process starts again.

In order to deal with the exponentially growing number of buildings from year to year, a Monte Carlo method was selected: The number of building types remains constant in every time step, with each building type having only one renovation status, determined by retrofit probabilities. The whole simulation is repeated several times, with different random numbers, until the average of all individual results for the required quantity has been determined with sufficient accuracy.

This core calculation process is supplemented by several additional tools:

  • an import filter from the PHPP, for defining a building type from existing PHPP input data
  • a variant management, allowing for a comparison of e.g. different supply structures
  • an economics calculator, suitable to determine economically optimal renovation measures
  • a climate data worksheet, where local climate data can be entere
  • a set of worksheets for an hourly analysis of electricity and district heating networks

For further information on districtPH, please read On track towards a climate neutral city?, PHI District Tool or contact

Example Application for a district in the North of Darmstadt

In this section, a simple example will be laid out in order to illustrate potential uses of districtPH. This example was also used for the presentation of districtPH on the 22nd International Passive House Conference in Munich [Schnieders 2018].

A small district in the north of Darmstadt, Germany, was selected to be used as an example (Figure 2). The district covers an area of about 350 x 350 m. The railway line and the ring road on the western side are separated by a landscaped noise barrier approximately 20 m high.

In total, 63,000 m² of living area were included in the calculation. The district is primarily populated with terraced housing and 3 and 4 storey multi-family houses. The areas of the terraced house plots range from 200 to 400 m2. For the most part, the district was developed between 1995 and 2005, and any older properties are only found on the eastern edge. There is a supermarket in the south of the district and a retirement home in the north-east corner of the area selected. A natural gas network has been laid in the district, but district heating is not yet available. In the model, it was assumed that the heating and hot water requirements of all the buildings would be covered by natural gas in 2018, the first year of the simulation.

Figure 2: Satellite picture of the example district

Source: Schnieders 2018

1.    Different Retrofit Strategies

Using districtPH, the effects of average quality retrofits and of the possibly resulting lock-in-effect on the total CO2 emissions were investigated. Simultaneously, the importance of the retrofit rate was examined.

The following 4 variants were considered:

A)   Retrofitting the wall, roof, floor slab and windows to the current minimum legal standard required in Germany according to the EnEV (roof 0.24, basement ceiling 0.30, exterior wall 0.24, window 1.3 W/(m²K)), and only if the building component in question is being retrofitted anyway. Window ventilation as before.

B)   Identical to A) but with the modernisation rate roughly doubled until 2028. This is realized by a shortened service life. We have therefore taken a realistic approach and assumed that primarily older building components will be modernised.

C)   Retrofitting the wall, roof, floor slab and windows as cost effectively as possible. This generally means using components of Passive House quality. The retrofit however is carried out only at the end of the service life of the building component in question. This follows a step-by-step retrofit in accordance with an EnerPHit Retrofit Plan. Installing a mechanical ventilation system with a highly efficient heat recovery system and improvements to the airtightness to EnerPHit level, as well as replacing the windows.

D)   The same as B) for the first 10 years, and thereafter, the same as C).

Figure 3 and Figure 4 illustrate the resultant CO2 emissions for space heating and hot water production (the power consumption is initially not taken into consideration here). It should first be noted that even after 50 years, only moderate reductions in emissions were achieved in the variant A).

If the rate – but not the quality – of modernisation is increased in the variant B), this initially reduces the CO2 emissions to a significant degree. The economic cost of this would, however, probably be considerable as purely energetic modernisations, which are not incorporated in the regular maintenance cycle, tend not to be financially viable. Sufficient funding would therefore have to be available to bring forward retrofits that were not due until a later date. In addition, manufacturers and tradespeople would have to build up the required capacities (which would then have to be run down again). If, as assumed in the example, the funding were to come to an end after 10 years, then – as shown in the graph – basically nothing more would be done in the following 10 years because all the building components (apart from the insulation level) would then be in relatively good condition. After this, further improvements would have to take place, so that in the long term the emissions are approximately equal to those in the variant A).

In variant C), the drop in emissions is initially very slow. This is partly due to the relatively new fabric of the buildings in the district used in the study. Subsequently, however, a strong, continuous and sustainable improvement can be seen.

Variant D) shows the outcome if this course of action is not taken until a later date, for example after 10 years of widespread intensive funding of broadly average quality. As in variant B), there is a longer pause after the end of the funding before the emissions resume a course similar to the variant C). However, the variant D) will not attain the final result of the systematic EnerPHit retrofit C) by the end of the 50-year financing period, whereby many opportunities will have been missed.

Figure 3: Timeline of the annual CO2 emissions in 4 different scenarios

Figure 4: In view of the sum total over 50 years, the lowest CO2 emissions are achieved through consistent EnerPHit retrofitting as part of the maintenance circle

Source: Schnieders 2018

On the basis of these findings, the following conclusions may be drawn: Substantial improvements to the fabric of the building which bring it to a sustainable level are crucial for reducing energy consumption in the building sector in the long term. In contrast, funding for average quality retrofits only provides short term improvements, before having a negative impact on the starting situation for any additional measures required.

2.    District Heating for Deep Retrofits

A second topic was the extent to which district heating is still worthwhile if the district is, in the long term, retrofitted to the EnerPHit level.

In the calculation, the area will now be provided with a district heating grid at usual operating temperatures (110 °C for winter and 80 °C for summer) (Figure 5). The heat is predominantly generated by a gas-driven CHP plant located near the supermarket with an overall efficiency of 85% (55% thermal and 30% electrical). The CHP plant is heat-driven and is designed to operate at a thermal output of 3.3 MW so that it generates approximately 3,000 full-load hours in the initial state of the district.

The heat supply is supplemented by a 2,000 m2 solar thermal collector which can be installed on the railway line side of the landscaped noise barrier without incurring a significant adverse effect on its functionality. The collector is oriented towards the west and is inclined at an angle of 45°. It is supplemented by a storage device which operates during the day and enables the yields to be fully utilised even in summer.

Figure 5: Simplified overview of the district heating grid. It was assumed that every residential unit in the rows of terraced housing would require its own connection to the district heating transmission line.

Source: Schnieders 2018

The refrigeration units in supermarkets consume large amounts of electricity. In winter, the ideal use for the waste heat generated is to directly heat the supermarket. In summer, however, the waste heat can be used in the district heating network, but it must be passed through a heat pump to bring it to the corresponding temperature. A usable waste heat output of 57 kW was set during opening hours and 33 kW for other times.

At the starting point, the district heating grid incurred heat losses of 17% of the heat fed in. In summer, the solar thermal collector and supermarket approximately cover the heat losses of the network. Setting up a district heating grid would reduce the total CO2 emissions of the district (for heating, hot water and all electricity consumers) from 5,100 to -2,100 t/a. The emissions are counted as negative due to the CO2 credit for the surplus power. In Germany, renewable electricity is not replaced in the overall electricity mix (power grid operators prioritise purchasing electricity from renewable sources), instead it replaces the electricity which is generated by inefficient, coal-fired medium-load power stations. The corresponding CO2 factor amounts to 1,008 g/kWh. The gas CHP plant emits just 833 g per kWh of electricity, thereby simultaneously covering another portion of the thermal demand in the district.

The CO2 emissions calculated in this way are clearly misleading, not only because higher losses from the district heating pipelines would further reduce the emissions in terms of numbers, but also because the displacement electricity mix will drastically change over the decades, the latter being the relevant timescale for buildings and building modernisations. Germany has undertaken to reduce greenhouse gas emissions by 80% to 95% compared with 1990 by as early as 2050. This can only be achieved if coal-fired power stations are largely phased out.

One possibility for a more meaningful basis of assessment, especially for long term developments, is the PER system (for details, please see [Feist 2014] and [PHPP]). It is based on a future energy supply generated entirely from renewable sources. It comprises a cost-effective mix of photovoltaic installations, wind energy and biomass, tailored to regional availability. Methane, which is generated in summer from renewable electricity and reconverted to electricity in combined cycle power stations in winter, is used to cover the winter gap. According to this system, electricity generated by photovoltaics or wind, for example, has a PER factor of 1. Methane generated from renewable energies (renewable power-to-gas) has a value of 1.75. Electricity generated by CHP plants is evaluated in the same way as electricity generated by combined cycle power stations. The PER demand of the buildings in the district can therefore be calculated – the more electricity the CHP plant supplies, the lower the PER factor of the district heating system.

The PER demand of all the buildings in the district amounts to 21,100 MWh/a in the initial state. The transition to district heating would reduce the PER demand, though only to 17,600 MWh/a.

If the complete fabric of the buildings in the district were brought up to the EnerPHit level, the PER demand would fall to 7,700 MWh/a if the gas boiler continued in operation, however it would only fall to 8,000 MWh/a if the district heating system were used. The district heating grid losses would increase to 38% of the heat fed in. The CHP plant would then be operated at only 1,200 full load hours per year.

It can therefore be seen that the installation of a conventional district heating grid for the district investigated is not advisable, neither economically nor in terms of the cost of the energy supply. The heat fed in by the large solar thermal collector (11% contribution in the EnerPHit variant) and the supermarket’s utilisation of waste heat (3%) does not affect this assessment. Heating by gas would also be questionable in a renewable energy system. As it takes considerable effort to produce gas from renewable electricity, it is more beneficial to produce heating and hot water using efficient heat pumps: the PER demand then falls to 4,300 MWh/a. This outcome however cannot be applied universally, as CHP plants with greater electrical efficiency and in more densely populated districts can generate different results.


[Feist 2014] Feist, Wolfgang: Passive House – the next decade. In: Feist, W. (Ed.): Proceedings on the 18th International Passive House Conference 2014 in Aachen, Germany. Passive House Institute, Darmstadt, 2014.

[PHPP] Passivhaus-Projektierungspaket (PHPP), Version 9. Passive House Institute, Darmstadt, 2014.

[Schnieders 2018] Schnieders, Jürgen: Energy balancing at district level, 22nd International Passive House Conference, 9/10 March 2018, Munich.

[Tabula 2018] Typology Approach for Building Stock Energy Assessment. (accessed 20 Apr 2018).