Skip to content

Heat network

Jump to bottom Edit New page
Chael Kruip edited this page Oct 6, 2016 · 62 revisions

Table of Contents

Introduction

The heat network is currently still in Beta and therefore only available on the Beta server.

Heat networks are an interesting alternative for providing space heating and hot water to households as they can use excess heat from industry or other facilities such as swimming pools and ice skating rinks that would otherwise be wasted. Moreover, it is possible in some cases to supply heat networks with renewable heat, e.g. by a geothermal plant. Heat networks work best in energy systems that have been specifically designed to incorporate them. The best example of extensive implementation of heat networks can be found in Denmark.

User interface

Modelling a heat network in ETMoses starts in the Energy Transition Model and involves several steps that are described below.

  • First the user has to set which percentage of space heating and hot water is provided by a heat-network (a.k.a. district heating). Below is a screenshot showing how to set this percentage for space heating. A similar slider is available for hot water.

  • Next, the user can select the sources that feed the local heat network, which is shown in the image below. The options are:
  • Gas CHP
  • Biomass CHP
  • Biogas CHP
  • Geothermal
  • Central heat network (excess heat from agriculture, industry and central production and central large-scale dispatchable heaters)

The first four options are imported into ETMoses with their financial and physical attributes inherited from the ETM. The 'Central heat network' (CHN) is actually a very heterogeneous group of sources and is therefore simplified into two parts: a dispatchable and a must-run part. The relative importance of the parts is based on the amount of heat delivered in the ETM by dispatchable versus must-run sources contributing to the CHN. The financial attributes for these parts are based on research from Liandon and will be described in more detail below in the costs section.

NOTE: It is not necessary to include the agriculture- or industry sector in order to use the 'Central heat network'.

  • When the user opens the scaled ETM scenario in ETMoses, the percentage of space heating and hot water supplied by heat networks is converted into the number of households connected to the heat network by multiplying these percentages with the number of households in the LES (the number of base-load profiles in the technology matrix). These connections show up in the technology table as Households water heater district heating and Households space heater district heating. Note that the number of connections for space heating and hot water is not necessarily the same, as they are set by separate sliders in the ETM.
  • The different heat sources feeding the heat network are also imported from ETM and shown in a table. This table can be found when editing the LES in the tab Heat source list and is shown in the image below. The difference between must-run and dispatchable sources is explained below at calculation.

  • The user can edit several parameters:

    • Plant type (several pre-defined types exist, as mentioned above)
    • number of units: how many of this plant are available. Fractional units are allowed.
    • Capacity per unit [kW]: how big is the plant in energy terms. Relevant for dispatchables only.
    • Total production [kWh/y]: relevant for must-run only. Gives the total heat production in kWh per year.
    • Stakeholder: the actor that pays for the production asset
    • Profiles: Describes the production as a function of time in 15 minute values. Relevant for the must-run sources only.
    • Investment costs [EUR/kW] or [EUR/kWh]: The initial investment for a plant. For dispatchables it is relevant to express investment per kW installed capacity. For must-run it is relevant to express the investment per kWh of heat delivered per year.
    • Technical lifetime [y]: how long the plant is expected to last.
    • Fixed O&M costs [EUR/kW/y] or [EUR/kWh/y]: the expected costs per year for operation and maintenance. For dispatchables it is relevant to express this per kW installed capacity. For must-run it is relevant to express this per kWh of heat delivered per year.
    • Marginal costs [EUR/kWh]: the costs of producing one kWh of heat (including fuel costs, variable O&M costs and, possibly, costs for emitting CO2). NOTE: Marginal costs are not taken into account in the business case automatically. Of course, the user can manually use these marginal costs as fixed tariffs for heat production in Market models.
    • Distance to source [km]: how far the source is located from the feed-in point of the local heat network.
    • Order (for dispatchables): the order/priority in which sources will be dispatched is determined by the order in which the sources are dragged by the user.

  • The default parameters are based on research by Quintel and Liandon and can be found by following the links below. Note that for the three types of CHPs all marginal costs will be assigned to the heat production as it is assumed that they cannot necessarily sell the electricity they produce.

  • ETMoses also determines the heat network in the LES based on default parameters as provided by Liandon. These parameters can be found in this spreadsheet. The heat asset list can be found under the Heat asset list tab and is shown in the image below. As with all input data in ETMoses, the parameters can be adjusted by the user.

Results

Heat load

The production and consumption of heat and hot water is shown at the Heat load tab on the main page of ETMoses; an example is shown in the image below.

In this chart, production is plotted as negative numbers equivalently to the load charts for gas and electricity. This chart shows both production and consumption separately rather than a net result. The user can check if supply and demand is balanced by looking for production deficits (plotted in red).

Business case

The heat costs and prices related to heat in the market model are taken into account in the total business case.
See the modelling approach for more information on the [cost calculation] (https://github.com/quintel/etmoses/wiki/Heat-network#costs).

Modelling approach

Infrastructure

The heat network is modelled as two main components which are kept quite abstract:

  • Primary (transmission) infrastructure. This contains one or more pipes from the source to the secondary network.
  • Secondary infrastructure. This contains the distribution pipes and the connections to the households including heat exchangers.

The reason to split up the costs of the heat network infrastructure in these two parts is that:

  • The cost of primary (transport) infrastructure, the main pipe leading from the source to the feed-in point of the network is mainly determined by the distance between source and feed-in point. Possible other factors include the type (thickness, insulation etc.) of pipe and if it has to be buried or not.
  • The cost of the secondary infrastructure will be mainly determined by the distribution and the type of the houses connected. ETMoses has eight pre-defined house types:
    • rural_detached
    • rural_terraced
    • village_detached
    • village_terraced
    • village_apartment
    • city_terraced
    • city_apartment
    • city_flat

The default parameters for both the primary and the secondary infrastructure have been researched by Alliander and can be found in this spreadsheet.

Sources

The sources that supply the local heat network provide heat in the following order:

  • heat network must-run. These sources supply the heat network with heat. The amount is defined by a yearly budget and the instantaneous production is described by a profile. The production profile for must-run sources is the profile scaled with the yearly production (similar to electricity demand profiles). The default profile is a constant profile. NOTE: The user can choose another profile.
  • heat network dispatchable sources. These also feed their heat into the network, but only if the heat demand surpasses the must-run supply. The order in which these sources are dispatched is determined by the order they appear in the heat source list (this can be adjusted by the user by dragging the sources of the list up or down).

If there is still remaining heat demand, optionally installed decentral heat sources (heat pumps, gas-fired heaters etc.), will try to supply demand which might be used to fill in the gaps if the heat network (both must-run and dispatchables) have insufficient production capacity to fulfil demand.

For both the dispatchable and must-run category, a 'generic' option is available which can be customised by the user to represent specific sources such as a local skating rink or swimming pool (must-run) or a utility sized heat pump (dispatchable).

NOTE: Electricity produced with CHP sources that supply the heat network is not used elsewhere in the LES. Therefore, the marginal costs of these CHPs assign all fuel costs to heat production and provide an upper limit for the realistic marginal costs.

Calculation

Energy

The heat network is modelled energetically (kW and kWh), meaning that temperature is not modelled explicitly.

The energy calculation works as follows for every 15 minute time-step:

  1. demand is set at the end-points of the electricity topology: defined by profiles, scaled with yearly demand and associated with buffers
  2. demand is satisfied by using the following options in order of preference:
  • the heat network
  • decentral production (heat pumps, gas-fired heaters etc.)
  1. if excess heat is present from must-run, it will be stored in the heat network buffer
  2. if more excess heat is present than can be accommodated in the buffer, it will be curtailed

Buffering in the heat network

The heat network has a buffering volume of 10 kWh per connection. This allows it to buffer heat when demand doesn't match the supply.

The buffer has two distinct volumes which are related to how the buffer is used:

  • 10 kWh (dispatchable): The buffer is filled up by dispatchable sources. This volume corresponds to a typical temperature of 60 degrees Celsius and a water volume of 191 liters. This can be seen as a typical 'thermostate'-approach where the heat sources work at a temperature where their COP is favourable.
  • 17.8 kWh (must-run): The must-run sources fill up the heat network buffer to a maximum temperature of 95 degrees Celsius. This approach maximises the (free) heat available even if it is produced at unfavourable COPs. Remaining heat that cannot be stored in the buffer and is not used to meet the heat demand during this time-step, is curtailed (thrown away).

See also this issue on Github for more background.

Financial

Costs of heat production and consumption

It is possible to use the following heat-related 'measures' in Market models:

  • kWh of centrally-produced heat: measures the kWh produced by heat sources owned the selected stakeholder
  • kWh of heat consumed: measures the heat consumed from the network at end-points owned by the selected stakeholder
  • number of heat connections: measures the max number of district heating connections for the selected stakeholder (see section below for details)

Number of heat network connections

Connections to the heat network are, physically speaking, heat exchangers which need to be associated with buffers in ETMoses. In this case, the buffer-size should be reduced to a very small number, because houses with a connection to a heat network typically don't have heat buffers.

The number of heat network connections for space heating are extracted from the settings in the scaled ETM scenario. As with the other heating technologies of the households, the heat network connections are counted twice while in real life, the same connection to the heat network would suffice for both space heating and hot water. For connections to the heat network, this double counting is even more problematic than for the normal technologies because the number_of_heat_connections is one of the 'measures' in the market models. We decided to import the heat network connections in the same way as all other heating technologies, i.e. we separately take the following 'technologies' into account from the Energy Transition Model:

  • households_space_heater_district_heating_steam_hot_water
  • households_water_heater_district_heating_steam_hot_water

Although the above works fine for the technology matrix, the number_of_heat_connections measure in the market model transactions should not be the sum of households_space_heater_district_heating_steam_hot_water and households_water_heater_district_heating_steam_hot_water because houses typically only pay once for their connection to the heat network. Instead we take the maximum of heat/hot water connections as the number_of_heat_connections.

Costs of assets

The costs of assets are taken into account similarly to those of gas- and electricity assets by adding yearly fixed O&M costs to yearly depreciation costs for the stakeholder that owns the asset.

Sources

As described above, all relevant financial aspects of heat sources can be set in the heat source table:

Yearly costs for dispatchable sources

The yearly_costs (what is shown on the diagonal of the business case matrix) is defined as follows:

yearly_costs = initial_investment / technical_lifetime + om_costs_per_year

where the initial_investment is defined as:

initial_investment = total_initial_investment_per_mw * heat_capacity_per_unit * units

and

om_costs_per_year = om_costs_per_year_per_mw * heat_capacity_per_unit * units

Yearly costs for must-run sources

The yearly_costs (what is shown on the diagonal of the business case matrix) is again defined as follows:

yearly_costs = initial_investment / technical_lifetime + om_costs_per_year

where the initial_investment is defined as:

initial_investment = total_initial_investment_per_mwh * heat_production_in_mwh

and

om_costs_per_year = om_costs_per_year_per_mwh * heat_production_in_mwh

Primary infrastructure

The distance to the source(s) which can be set in the heat source list is used to calculate the costs of primary infrastructure: the transmission pipe between the source(s) and the secondary (distribution) network. The user can assign the following aspects of primary infrastructure:

  • Investment [EUR/km]
  • Fixed O&M costs [EUR/km]
  • Technical lifetime [Y]
  • Stakeholder

Yearly costs are calculated as

yearly_costs = initial_investment / technical_lifetime + om_costs_per_year

where the initial_investment is defined as:

initial_investment = total_initial_investment_per_km * distance_in_km

and

om_costs_per_year = om_costs_per_year_per_km * distance_in_km

These yearly costs are assigned to the selected stakeholder.

Secondary infrastructure

The cost of the secondary infrastructure (distribution) follows from the number of connections and some basic assumptions on the types of houses and the geography of the area. Depreciation costs for the heat network components and the heat technologies are calculated similarly to all other depreciation costs, see the results section for more details.

The costs for different combinations of types of houses and geography are summarised in the table below:

type Investment [EUR] Fixed O&M costs [EUR/y] Technical lifetime [y]
rural_detached 53300 566 40
rural_terraced 11800 146 40
village_detached 15800 191 40
village_terraced 8800 116 40
village_apartment 2200 44 40
city_terraced 6800 69 40
city_apartment 2200 44 40
city_flat 1800 36 40

Default parameters

Default parameters of the heat network are based on research by Liandon. This spreadsheet summarises the results of that research. Other default parameters follow from ETM.

The financial default attributes for the 'Central heat network' are based on reasearch by Liandon and summarised in the table below:

Part of the CHN Investment Fixed O&M costs Technical lifetime [y] Marginal costs [EUR/kWh]
Dispatchable 100 EUR/kW 2 EUR/kW/y 30 0.022
Must-run 0.5 EUR/kWh 0.02 EUR/kWh 30 0.025

For more background see issue #1097 on Github.

Add a custom footer

Clone this wiki locally