7 Case study : Energy community in Les Vergers

This section presents the results of the simulation of a CEL at Les Vergers. Prior to the financial flows between each stakeholders of the CEL, energetic flows are analyzed. First few comments are given about the performances of the neighborhood energy system.

7.1 Performances of the neighborhood energy system

Few main points are important to state before analyzing the energetic flows of the CEL.

The load curve and the PV production data generated from QBuildings are only estimations : These are synthetic curves that only estimate the real private load curve and PV production of each building. In this present case, the annual electricity demand is under-estimated by 30%. The annual PV production is over-estimated only by 4%.
The modeled heat pump size is underestimated compared to the size of the real heat pump at Les Vergers
The maximum annual power demand of the neighborhood varies in function of the tariff at which electricity is sold. Thus, this tariff also influence the sizing of some units and the energy flows inside the neighborhood.

7.1.1 Heat pump sizing

The capacity of the modeled heat pump is only 3 MWth and not 5 MWth. This is due to the fact that REHO uses clustered weather data when determining the thermal needs. At Les Vergers, thermal needs are covered by the heat pump during the entire year, meaning the heat pump is sized in order to fulfill the most extreme thermal needs that only occur for few hours during the year. However, the extreme hours extracted with the 10 typical days during the weather data clustering, are not extracted within a day and are thus not considered in the synthetic weather data time-series of the year (see section 5.2.1.1). These periods are only used for sizing units as it is considered that the energy demand would be maximum for these hours. Yet, it appeared that the maximum heat demand of the 34 buildings did not happen during the extreme cold hour of the year. The hour of the year for which the heat thermal demand is maximum may also not happen during one of the typical day extracted. As a consequence, the sizing of the heat pump, may be underestimated as it is based only on the demand of the 10 extracted days.

A 3 MWth heat pump is therefore considered to be is enough to fulfill all the thermal needs of the neighborhood with the synthetic energy demand time-series of the year but may not cover the needs of the real energy demand along the whole year. However, over the 8760 hours of the year, the number of hours for which the thermal energy demand is higher than the maximum thermal energy demand of the 242 hours of the synthetic energy demand, is expected to be rather small (few dozens).

Moreover, it has been shown in the feedback analysis on the eco-neighborhood of Les Vergers (Schneider, Brischoux, and Hollmuller, n.d.), that the district heat pump was actually over sized in this neighborhood. Hence, from what has been said, it was decided to keep this sizing of 3 MWth for the modeled heat pump.

7.1.2 Maximum annual power demand

The ‘behavior’ of the energy system of the neighborhood is sensitive to the tariff at which grid electricity is sold. Indeed if they are charges on the neighborhood power demand, as it is the case for professional tariffs, the neighborhood will try to minimize its power demand in order to lower its cost. Table 7.1 shows the maximum annual power demand (column CEL's Pmax [kW] in kW), the maximal power required for electricity exports generated from PV panels (column PV stress [kW] in kW) and the effect of stochacticity on the power demand : For instance, with the BT tariff and with no battery nor EV’s chargers, the maximum power of the CEL represents 93% (column % of CEL Pmax, in %) of the sum of the maximum power of every members of the neighborhood (column Sum of members' Pmax [kW] in kW) (1282 kW out of 1379 kW). In this table, index indicates the grid electricity tariff considered in the optimization and whether district units are considered or not : bat, ev and evbat mean respectively that the district battery, EV charging stations and both together are considered.

Table 7.1: Maximum power demand for the whole neighborhood.
Scenarii	CEL’s Pmax [kW]	PV stress [kW]	Sum of members’ Pmax [kW]	% of CEL Pmax
BT	1282	1297	1379	93
proBT	1026	1310	1527	67
MT	1029	1300	1595	65
BT_bat	1331	1214	1505	88
proBT_bat	989	1322	1709	58
MT_bat	989	1280	1700	58
BT_ev	2562	1027	2914	88
proBT_ev	1891	974	2858	66
MT_ev	1841	963	2914	63
BT_evbat	2625	0	2829	93
proBT_evbat	1373	1039	3274	42
MT_evbat	1374	1027	3334	41

As explain in section 5.4.1 the CEL allows for its members to benefit from the stochasticity of the demand : the maximum power of the neighborhood is lower than the sum of maximum power of every of its members (see tables 7.1). The billing for power charges is made internally for each members : the contribution share \(\epsilon_i\) of each member to the sum of the maximum power of every members is calculated and then the effective maximum power \(P_{max}^{eff}\) of each member is computed by multiplying this share by the CEL maximum power as shown in equation (7.1).

\[\begin{equation} P_{max,i}^{eff} = \epsilon_i \cdot P_{max, CEL} \tag{7.1} \end{equation}\]

Where :

\(P_{max,i}^{eff}\) is the effective maximum power of the ith CEL member
\(\epsilon_i = \frac{P_{max,i}}{\sum_i P_{max,i}}\) is the contribution share of the ith member to the sum of the maximum power of every members
\(P_{max,CEL}\) is the maximum power of the CEL

Figure 7.1 shows the effective maximum power of every members of the CEL. It also allows to estimate the contribution of each members to the neighborhood maximum power. For instance, for the scenario with the proBT tariff and with a district battery and EV charger, the 34 buildings account for 20% of the annual maximum power of the neighborhood, the heat pump 26%, the battery 23% and the EV chargers for 30%. Especially, The relative importance of the EV chargers in the total load of the CEL can be here visualize.

Figure 7.1: Effective maximum power of the CEL and its different members

Figure 7.1 and table 7.1 show that the influence of the tariff on the maximum power demand is more important when the sum of the maximum power of every consumers inside the CEL is the biggest : going form the BT tariff to a professional tariff only allows to reduce the maximum power by 20% when there is no battery nor EV chargers (from 1282 kW to 1026 kW in proBT). However, this reduction is maximum in presence of EV chargers as they represent a significant additional load for the neighborhood : with no battery, this reduction is around 26% (from 2562 kW to 1890 kW in proBT).

The battery also allows to reduce the maximum power demand when charges on the maximum power demand are included in the electricity selling price. But once again, its influence is the biggest when the sum of the maximum power of every consumers inside the CEL is the biggest, which means when EV chargers are considered. Indeed it’s influence is actually very limited when EV chargers are not considered : Adding a battery reduces the power from 1029 kW to 989 kW under the proBT tariff (a 4% reduction). Yet, as the battery also import electricity from the grid in this scenario, it also contribute to the neighborhood maximum power which is not the case when the neighborhood is subject to the BT tariff. In this later scenario, th battery solely consumes the excesses of electricity from the PV panels (see section 7.2.1). In the the proBT scenario, the contribution fractions of the buildings and of the heat pump are thus reduce due to the fact that the battery fraction contribution to the neighborhood maximum power is not 0. Therefore, even if the maximum power of the neighborhood does not reduces significantly due to the district battery, it allows nonetheless to reduce the effective maximum power of the buildings and of the heat pump.

The influence of the battery is bigger when EV chargers are consider : the maximum power demand goes from 1840 kW to 1373 kW with the EV chargers (a 25% reduction) under the proBT tariff. However, it is important to note that the battery does not reduce the maximum power if power demand is not charged as it is the case for the BT tariff.

7.2 Energy flows

All the energy fluxes hereafter presented are obtained for a scenario where all buildings and the heat pump can exchange energy like in a CEL or in a RCP with a microgird. To get the energy flows in a scenario where exchanges are restricted, it is sufficient to consider that the amount of energy exchange is simply import from the grid.

In the following figures, the x-axis indicates the grid electricity tariff considered in the optimization and whether district units are considered or not : bat, ev and evbat mean respectively that the district battery, EV charging stations and both together are considered.

Figures 7.2 and 7.3 show respectively the monthly electrical consumption of the 34 buildings and the monthly electrical consumption of the heat pump when electricity tariff is the proBT tariff. The exchanges between buildings inside the neighborhood are very little compare to the total monthly buildings consumption. In fact, the annual exchanges between buildings represent only 2% (67 MWh) of the total annual buildings consumption (3311 MWh). On the other hand, the amount of electricity the heat pump annually import from buildings PV production represent 20% (564 MWh) of its total annual consumption (2808 MWh). See Appendix 12 for the consumed energy volumes.

Note that similar results are found with the BT and MT tariff, that is why only the results for the proBT tariff are presented as it is the tariff the more prompted to be used for a CEL grouping.

Figure 7.2: monthly electrical consumption of the 34 buildings of the neighborhood

Figure 7.3: Monthly electrical consumption of the heat pump

Figures 7.4 and 7.5 respectively present the self-sufficiency and the self-consumption of the neighborhood. The contribution of each type of consumption inside the neighborhood is showed. Exchanges between buildings but mostly exchanges between the heat pump and buildings (where PV panels are installed) allowed to raise the annual self-sufficiency from 23% to 34%, and the annual self-consumption from 50% to 72% compred to a scenario where only buildings can consume their own PV production (see appendix 12) : SS_CEL and SC_CEL correspond to the SS and SC of the neighborhood when taking into account inter-members exchanges while SS_Bding and SC_Bding correspond to the SS and SC of the neighborhood when taking into account only buildings private PV consumption). In total, 2068 MWh of electricity are produced and consumed locally.

Figure 7.4: Monthly self-sufficiency of the neighborhood

Figure 7.5: Monthly self-consumption of the neighborhood

Figure 7.6 shows the monthly PV production. As expected, this production peaks in summer. In this scenario where exchanges between entities are allowed, exports to the grid are lowered. This could reduce curtailment needed to satisfy grid constraints.

Figure 7.6: Monthly PV production

From these graphs, it can be concluded that the benefit of exchanges and thus of a CEL (or RCP microgird) is more significant in autumn and spring. Indeed it is the period where PV panels produce relatively lots of electricity while the thermal needs are still high : it is at that period that the exchange between the heat pump and the buildings are the biggest. On the other hand, exchanges between buildings are maximal in summer since the PV production is maximal and thus the excess of PV in certain buildings can be consumed by other buildings. However, these exchanges are very little compared to the total amount of energy needed within the neighborhood : inter-buildings exchanges represent only 3% of the CEL needs in summer, when these exchanges are maximum and when CEL needs are minimum (see figure 7.4). Annual inter-buildings exchanges represent only 1.1% of the annual CEL needs (6123 MWh). In a residential neighborhood where every buildings have PV panels installed on their roof, this is expected since every buildings already consume their own produced electricity. Annually, the amount of electricity consumed by the buildings coming from their own PV panels is around 1.5 GWh, 22.4 times more than the inter-buildings annual exchanges.

This highlights the fact that grouping structures, such as CEL, make sens in residential neighborhood only if district units are installed. Indeed, the exchanges between the buildings and the district heat pump, represent most of the neighborhood’s internal energy flows and enable to raise significantly the self-sufficiency together with the self-consumption as annual HP-buildings exchanges represent 9.2% of the annual CEL needs.

7.2.1 Influence of the district battery

As discussed in section 7.1.2 the district battery has a certain influence on the maximum power demand of the neighborhood. Indeed it represents a new electricity supplier, within the neighborhood. Therefore its presence changes the energy flows inside the neighborhood and alter the grid electricity demand of the neighborhood. As figure 7.1 illustrates, the usage of the battery is different depending on the tariff at which grid electricity is sold. Figures 7.7 and 7.8 show respectively the battery’s inflow and outflow when the BT tariff is applied and when the proBT tariff is applied. Inflows are negative while outflows positive. Note that MT tariff results are similar to the proBT tariff results. There are, hence, not presented.

Figure 7.7: Monthly inflows and outflows of the battery when the neighborhood is subject to the BT tariff

With the BT tariff, the battery is used to store the excess of electricity generated by the PV panels in summer. It is thus never fed by grid electricity. Hence, the battery stores annually 550 MWh of PV generated electricity that would normally be exported to the grid (see appendix 12). In summer the heat pump is already fed with PV panels generated electricity, thus it does not need electricity from the Battery. Therefore, the battery almost solely supplies the buildings in summer with its stored energy. It supplies annually 444 MWh to the buildings of the neighborhood. Note that the battery has a roun-trip efficiency of 0.9. As a consequence, the annual CEL’s self-sufficiency reaches therefore 37% and the annual self-consumption 92% (see table 7.2).

Figure 7.8: Monthly inflows and outflows of the battery when the neighborhood is subject to the proBT tariff

With the proBT tariff, the battery rather serves as an energy buffer provider. Indeed as charges on the power demand are included in the grid electricity tariff, the battery tries to minimize the maximum power of the neighborhood : it smooth out the heat pump’s grid electricity demand peaks in winter by supplying electricity to the heat pump at these specific moments while importing energy from the grid when the neighborhood’s grid electricity demand is lower. Since in this case the purpose of the battery is to limit the maximum annual power demand of the neighborhood, it does not have a significant impact on the neighborhood’s self-sufficiency nor on the neighborhood’s self-consumption (see table 7.2). Indeed it only supplies annually 42 MWh to the heat pump. More than 10 times less than the annual energy volume supplied by the battery when the neighborhood is subject to the BT tariff. Besides it only consumes 1.6 MWh of PV generated electricity in the entire year.

7.2.2 Influence of the EV chargers

The EV’s charger essentially represents an additional electric load for the neighborhood : it raises the annual electricity needs of the neighborhood from 6.1 GWh to 9.3 GWh, a 52% augmentation (see appendix 12). As a consequence, the neighborhood’s self-sufficiency falls to 27% but the neighborhood’s self-consumption reaches 89% (see table 7.2). Indeed, as a neighborhood electricity demand rises and the PV panels production remains the same, it is expected that the self-sufficiency drops and the self-consumption rises as EV’s chargers consume part of the electricity generated from the PV panels. Indeed EV charging stations consume 687 MWh of electricity coming from the PV panels over the entire year. This is more than HP-buildings exchanges. In fact, when EV chargers are present, the volume of energy exchanged between the HP and the buildings diminished to 350 MWh because the EV chargers capture a part of the PV generated electricity. Figure 7.9 shows the monthly electricity demand of the EV chargers. As results are similar for every tariffs, only proBT tariff results are presented.

Figure 7.9: Monthly electric vehicles consumption

7.2.3 Influence of the battery together with EV chargers

Here, both the district battery and the EV chargers are considered. As previously stated in 7.1.2, the battery’s influence is more significant in this case. Yet its usage is still dependent on which tariff is applied. Once again, MT tariff results are similar to the proBT results thus only the BT and the proBT results are compared. Figure 7.10 shows the monthly battery inflows and outflows when the BT tariff is applied. Again, inflows are negative while outflows positive. This profiles is similar to the one shown in figure 7.7 but here the battery supply not only the buildings but also the EV charging stations. Note that the battery still only reies on PV panels production as the battery is here only to store the excess production : it supplies in total 260 MWh annually to the buildings and to the EV chargers. On the other hand, EV chargers exchanges with the buildings still represent around 690 MWh annually.

Battery supplies and EV chargers exchanges with the buildings, together with the other internal exchanges make the annual neighborhood’s self-consumption to reach 100%. On the other hand, the annual neighborhood self-sufficiency drops to 29% (see tables 7.2).

Figure 7.10: Monthly inflows and outflows of the battery when the neighborhood is subject to the BT tariff and includes electic vehicles

Figure 7.11 shows the monthly battery inflows and outflows when the proBT tariff is applied. Again, this profiles is similar to the one shown in figure 7.7 but with exchanges between the battery and the EV’s chargers added. Note that the battery’s aim is still to reduce the neighborhood’s maximum power. That is why the exchanges between the battery and the EV chargers are not equal to the one presented in figure 7.10, even in summer. This is also why the energy provided by the battery to the EV chargers represents a bigger volume in winter than in summer, since at that period, the EV charging stations rely more on the grid than on PV panels electricity and hence, have a bigger power demand.

The presence of EV charging stations in the neighborhood has a direct influence on the optimal size of the district battery, since now it has a capacity of 4.4 GWh while it was only 1.3GWh in the proBT scenario with no EV chargers (see table 7.4). Indeed, the battery supplies, in this case, lots of energy to the EV chargers in order to reduce their maximum power. Annually it supplies 589 MWh from which 438 go to EV chargers, the rest goes to the HP as shown in figure 7.11. As the battery now supplies the EV charging stations continuously along the year, it can benefit from the excesses of electricity generated from PV panels in summer. As such, it consume up to 201 MWh of electricity from the buildings annually. As a consequence, the annual self-consumption equals 93% and the self-sufficiency 30%.

Figure 7.11: Monthly inflows and outflows of the battery when the neighborhood is subject to the proBT tariff and includes electic vehicles

7.2.4 Summary

The different tariffs as well as the presence of district units such as a district battery or EV charging stations modify the energy flows between the members of the neighborhood as well as the energy exchanges between the neighborhood and the grid. Tables 7.2 presents the annual self-sufficiency and the annual self-consumption of the neighborhood. SS_CEL and SC_CEL correspond to the SS and SC of the neighborhood when taking into account inter-members exchanges while SS_Bding and SC_Bding correspond to the SS and SC of the neighborhood when taking into account only buildings private PV consumption.

Table 7.2: Share of energy flows relative to each consumers’ needs
Scenarii	SS_CEL	SC_CEL	SS_Bding	SC_Bding
BT	34	73	23	50
BT_bat	38	92	21	50
BT_ev	27	88	16	51
BT_evbat	29	100	15	51
proBT	34	72	23	50
proBT_bat	34	72	23	50
proBT_ev	27	89	16	52
proBT_evbat	30	93	15	52
MT	33	71	23	49
MT_bat	34	72	23	50
MT_ev	27	88	16	52
MT_evbat	30	93	15	52

7.3 Financial flows

From the energy flows, financial flows are determined using the selected tariffs.

7.3.1 Tariffs and scenarii

Here some comments are made about the different considered tariffs.

7.3.1.1 Professional tariffs optimization

The annual power usage duration (DUP) for each building separately and for the whole neighborhood was calculated for the 2 different professional tariffs. This allows to determine the optimization tariff for both the 34 simple RCP (single building) and for the CEL. Tables 7.3 shows the CEL’s DUP and the number of buildings whose DUP is below 3000h for each scenarii. The CEL’s DUP is above the optimization threshold of 3600h for most of the scenarii, except in scenarii with the proBT tariff and when EV are considered. On the other hand, there are never more than 14 buildings who have a DUP bellow the 3000h optimization threshold. Thus, for sake of simplicity it has been decided that the tariff optimization for both proBT and MT tariffs would be the tariff optimization B for every end-users in every scenarii.

Table 7.3: Hypothesis on the DUP
Scenarii	CEL DUP [h]	# of buildings with a DUP<3000h
proBT	3952	10
proBT_bat	4123	10
proBT_ev	3594	11
proBT_evbat	4948	14
MT	3961	9
MT_bat	4109	8
MT_ev	3697	10
MT_evbat	4949	11

7.3.1.2 Battery tariff

From the \(N_{battery}\) different scenarii which include a district battery in the neighborhood’s energy system, the mean price \(p_{battery}^*\) of electricity sold by the battery owner has been deterimined thanks to equation (5.2). It is equal to 39.93 cts/kWh. Note that local distribution charges are added to this price to get the price at which customers will buy the electricity from the battery owner. Tables 7.4 shows the sizing of the battery in each scenario, its CAPEX and OPEX in MCHF per year and its CAPEX and OPEX per kWh in CHF/kWh.

Table 7.4: Battery’s characteristics
Scenarii	Size	CAPEX [kCHF/y]	CAPEX [CHF/kWh]	OPEX [kCHF/y]	OPEX [CHF/kWh]
BT_bat	7771	391	0.88	140	0.31
BT_evbat	8290	416	1.60	82	0.32
proBT_bat	1300	70	1.66	25	0.59
proBT_evbat	4430	225	0.47	167	0.35
MT_bat	1280	69	1.66	21	0.52
MT_evbat	4434	225	0.46	152	0.31

It is here clear that if the capital expenditure of the battery had been considered in the development of the energy sales price of the battery owner, this price would have been very high compared to the SIG tariffs and it wouldn’t have been beneficial for other consumers (the renters, the heat pump owner and the EV charging stations owner) to buy electricity from the battery owner. Indeed this price would have been over 1 \(CHF/kWh\) in most scenarii.

7.3.2 Results

In this section, financial results are shown for each stakeholder separately. Yearly balancing sheets are presented with error bars that account for the variation in the price of the local distribution charge. The base line scenario was done with the local distribution charge being equal the half of the distribution charge of the BT simple tariff, thus 5.25 cts/kWh. The variation goes from 0% (no local distribution charge) to 100% of the distribution charge of the BT simple tariff. These 2 case are extreme cases : the case where there is no local distribution charge is equivalent to a scenario with a RCP microgrid instead of the CEL scenario. Indeed, in a RCP microgrid scenario, members of the neighborhood can exchange energy through a private network : the DSO has no right to taxe exchanges. The case where the local distribution charge equals the distribution charge is also a extreme case as it won’t allow the PV producer to value its energy as it would get to expensive and it would break the principle of doing a reduced distribution charge for electricity that does not go through all the distribution infrastructure.

Note that the error bars allows to directly visualize the amount of money earn or spent due the implementation of the local distribution charge since the base its varation goes from 0% to 100% of the distribution charge of the BT simple tariff.

For each stakeholder yearly balancing sheets are given for relevant scenarii : for some of the stakeholders, balancing sheets are sometime alike between two scenarii. For sake of simplicity, only one scenario is presented in that case. This is the case for the PV panels owner, the renters and the heat pump owner : the scenarii with and without considering EV chargers in the neighborhood are similar under the grouping categories No grouping and RCP as in these scnearii no exchanges are allowed between the EV chargers owner and these 3 other stakeholders. Thus the scenarii with EV chargers and under the grouping category No grouping and RCP won’t be shown. For the Battery owner stakeholder, only the scenarii that includes a battery are presented. Similarly, for the EV charging stations owner, only the 12 scenarii that include EV chargers are shown.

Finally, as it was mentioned in section 5.4.1, some scenarii are not applicable in real life. Thus, additionally to the balancing sheet for every relevant scenarii, that allows to assess more easily the outcomes of being under one grouping category or tariff, a graph comparing only interesting applicable scenarii is given for every stakeholders. These scenarii are the the followings :

RCP scenario with proBT tariff
CEL scenario with proBT tariff
CEL scenario with a district battery and with proBT tariff
RCP scenario with EV charging stations and proBT tariff
CEL scenario with EV charging stations and proBT tariff
CEL scenario with a district battery and EV charging stations and with proBT tariff

The comparison is made by comparing each time one of the 4 CEL secnarii to one of RCP scenarii : if the CEL scenario includes EV chargers, it is compared to the corresponding RCP scenario with EV chargers. If not, it is compared to the RCP scenario with no EV chargers.

7.3.2.1 Energy supplier

Figures 7.12 and 7.13 show respectively the yearly balancing sheets of the energy supplier for the 12 scenarii without EV charging stations and the 12 others with EV charging stations.

Figure 7.12: Yearly balancing sheet of the energy supplier for the 12 scenarii with no electric vehicles

Figure 7.13: Yearly balancing sheet of the energy supplier for the 12 scenarii with electric vehicles

When comparing scenarii under a same grouping category but with different tariffs, it appears that the energy supplier makes more profit with the tariff BT. This is not surprising since this tariff is the most expensive for the end-user. Professional tariffs are less expensive for the end-user since these tariffs are available once a certain threshold of energy demand is reached. There is on the other hand no difference between proBT and MT tariffs. This is because the energy part of the tariff is the same in both pricing structure. Going from BT to proBT/MT makes the energy supplier lose between 60 and 70kCHF depending on the grouping category. This is relatively important comparing to the 390kCHF it earns yearly thanks to the BT tariff

When comparing scenarii under a same tariff but with different grouping category, it appears that the energy supplier makes little more profit in a CEL. The two figures indicate that it is because in CEL grouping category, the energy supplier spends less money remunerating the PV panels owner for electrical injection it sells at the same time less energy to its customers. Figure 7.14, which shows the comparison between the 6 selected applicable scenarii, illustrates that the energy supplier only save few thousands francs per year from going to the RCP proBT scenario to a CEL proBT scenario. These difference are negligible compared to its annual benefits of around 330 kCHF if EV are not included, and around 715 kCHF if EV are present : The losses due to selling less energy to the costumers is balanced out by the benefits made by less remunerating electricity injection.

Figure 7.14: Yearly difference between balancing sheets of the energy supplier for the 4 realistic CEL proBT scenarii compare to scneraio RCP proBT.

7.3.2.2 DSO

Figures 7.15 and 7.16 show respectively the yearly balancing sheet for the DSO for the 12 scenarii without EV charging stations and the 12 others with EV charging stations.

Figure 7.15: Yearly balancing sheet of the DSO for the 12 scenarii with no electric vehicles

Figure 7.16: Yearly balancing sheet of the DSO for the 12 scenarii with electric vehicles

Comparing scenarii under a same grouping category but with different tariffs shows that DSO’s revenues shrink when going form BT to proBT and from proBT to MT tariffs whatever the grouping category is. This is again because professional tariffs are less expensive for the end-user but revenues for the DSO are thus diminished. The two figures indicate that this is because way less energy is imported from the grid making the revenues from distribution charges to plunge. Power charges are not enough to compensate for this loss. This makes a huge difference especially when the grouping category is No grouping where annual benefits are diminished by almost half from BT to MT tariff. Remind that in MT, the DSO does not remunerate PV panels owner for their injection since end-users are connected to the MV network.

Comparing scenarii under a same tariff but with different grouping category shows that DOS’s revenues also diminish when going from No grouping to RCP and from RCP to CEL grouping category whatever the tariff is. It is also because less energy is import from the grid. Note that the difference between RCP and CEL is smaller compare to the difference between No grouping and RCP especially with the tariff BT simple : The DSO is already prompt to loose money through the implementation of simple RCPs.

Figure 7.17 shows the comparison between the 6 selected applicable scenarii. It shows that the DSO would loss between 72 and 150 kCHF per year if a CEL is implemented at Les Vergers compared to the RCP scenario. However it is to note that the DSO capture a certain amount of money thanks to the local distribution charge that mitigates its losses. If it could not receive this revenue, as it would be the case in a RCP with microgrid, the DSO would loose between 105 and 235 CHF. Indeed the implementation of the local distribution charge allows the DSO to taxes the energy flow within the neighborhood that uses its network. As a consequence, the DSO earns between 33 kCHF and 85 kCHF depending on weather a district battery or EV chargers are implemented or not (the revenue due to the implementation of the distribution charge can be visualized on the error bar). In a RCP with microgrid, where members can also exchange energy, the DSO cannot taxe these exchanges and would hence loose more more than in a CEL.

Figure 7.17 also indicates that if a battery is installed in the neighborhood, the DSO would loose more money, especially if EV charging stations are also installed as it has been seen that it is in that case that the battery usage is the more relevant. This is expected since the battery is a new energy supplier within the neighborhood. However it is also in that case where the DSO earns the more through the local distribution charge. Again, this is expected since the more district units are installed, the more local energy exchanges happen and the more it becomes a revenue source for the DSO thanks the local distribution charge.

The biggest source of reduction of the DSO’s revenue is the reduction of the CEL members’ maximum power as shown in figure 7.17. Indeed the DSO looses between 95 kCHF and 265 kCHF because each members of the CEL benefit from the maximum power charges reduction thanks to stochasticity. One way of mitigating this loss could be the creation of a new tariff that charges more the power component of the demand.

Figure 7.17: Yearly difference between balancing sheets of the DSO for the 4 realistic CEL proBT scenarii compare to scneraio RCP proBT.

7.3.2.3 PV panels owner

Figures 7.18 shows the yearly balancing sheets for the 18 relevant scenarii for the PV panels owner.

Figure 7.18: Yearly balancing sheet of the PV panels owner.

When comparing scenarii under a same grouping category but with different tariffs, the only difference that appears is with the tariff MT. This is because PV panels owner does not receive the remuneration for injection in the LV netwrok since end-users are connected to the MV network in MT. For the two other tariffs, the results are the same, except when the district battery is present. Indeed this is because the sizing and the inflows and outflows of the battery are different depending on which tariff is considered, as it has been seen. With the BT tariff, the battery will stored the excess PV panels production, making the volume of exchanges between the PV panels owner and the battery owner to rise. With professional tariffs, the battery will diminish the overall neighborhood maximum power and will more rely on grid electricity importations than on PV panels production.

When comparing scenarii under a same tariff but with different grouping category, it appears that the category CEL is the more beneficial for PV panels owner no matter the tariff. This is because there is a certain amount of electricity PV panels owner can sell directly to customers at a preferential tariff instead of selling it to the grid. The gab between No grouping and RCP is bigger than between RCP and CEL but the latter one is still significant. The presence of EV chargers, in particular, boost the PV panels owner revenue as shown in figure 7.19.

The variation on the local distribution charges has here no effect on the PV panels owner as it does not pay nor capture the revenue of this taxation.

Note that the CAPEX of the PV panels are not included here.

Figure 7.19: Yearly difference between balancing sheets of the PV panels owner for the 4 realistic CEL proBT scenarii compare to scneraio RCP proBT.

7.3.2.4 Renters

Figures 7.20 shows the yearly balancing sheets for the 18 relevant scenarii for the renters.

Figure 7.20: Yearly balancing sheet of the renters.

Comparing scenarii under a same grouping category but with different tariffs, shows that renters bill diminish when going form BT to proBT and from proBT to MT tariffs whatever the grouping category is. This is again because professional tariffs are less expensive for the end-user than the BT and because the MT tariff is even less expensive than the proBT tariff.

Comparing scenarii under a same tariff but with different grouping category shows that outcomes differ from one tariff to another :

With the BT tariff the RCP category and CEL category when no battery is installed are approximately as beneficial compared to the No grouping category, the CEL category being a little more advantageous : Indeed a little less energy is imported form the grid and instead imported form other building at a price a little more advantageous as shown in figure ??. However these differences between RCP and CEL are negligible compare to the renters’ annual bill. However, as soon as a battery is installed the bill of the renters rises as the electricity coming from the battery owner is the most expensive. Indeed, with the BT tariff, the battery owner sells almost all its stored energy to the renters.
With professional tariffs conclusions are less obvious : Figure 7.20 shows that the RCP grouping category is more expensive for renters because this tariff allows them to make huge economies which are mitigated by the fact that in RCP they buy local electricity to PV panels owner at a more expensive price than the proBT or MT price. Under a CEL, the stochasticity on the maximum power makes the price of electricity bought on the grid even cheaper making this grouping category the more profitable even though in this configuration, renters also buy expensive electricity to PV panels owner. The presence of the battery is no more an issue as the battery owner, in that case, sells energy to the heat pump owner and the EV chargers owner.

The variation on the local distribution charges makes little variation on renters annual bill except in the BT CEL scenario with a district battery since in this scenario, exchanges between the battery and the buildings are maximum and these excahnges are subject to the local distribution charges.

The biggest interest for the renters to be part of the a CEL at Les Vergers is to benefit from a reduction of their maximum power as shown in figure 7.21. This allows to think that is a new tariff was created that charges more the power demand of the customers, the renters would still benefit from grouping in a CEL.

Figure 7.21: Yearly difference between balancing sheets of the renters for the 4 realistic CEL proBT scenarii compare to scneraio RCP proBT.

7.3.2.5 Heat pump owner

Figures 7.22 shows the yearly balancing sheets for the 18 relevant scenarii for the heat pump owner.

Figure 7.22: Yearly balancing sheet of the heat pump owner.

When comparing scenarii under a same grouping category but with different tariffs, the MT tariff is the most advantageous not surprisingly as it is the cheapest for the end-user.

When comparing scenarii under a same tariff but with different grouping category it appears that the both categories No gouping and RCP yields the same results. This is because the heat pump is not included in any RCP since all RCP are simple RCP thus restricted to buildings. There are actually little differences between each grouping category with a similar tariff. Indeed the electricity price generated from PV panels, together with the local distribution charge, almost equal the BT tariff (25.37 cts/kWh against 26.24 cts/kWh). For the proBT tariff, The power charge reduction made available by the CEL is balanced out by the high price of PV panels electricity compared to the proBT tariff. This is even better illustrated in figure 7.23 that compares the 4 CEL selected applicable scenarii against the RCP scenarii in proBT. For the MT tariff scenarii however, the price of PV panels electricity is too high compared to the MT tariff to make the CEL attractive for this stakeholder. Here, if a new tariff that charges more the power demand of the heat pump owner was imposed, it wouldn’t be beneficial for the heat pump owner to gather in a CEL. However, figure ?? also chows that this could be mitigated by having an electricity price from PV panels production less expansive.

Figure 7.23: Yearly difference between balancing sheets of the heat pump owner for the 4 realistic CEL proBT scenarii compare to scneraio RCP proBT.

Figure 7.23 illustrate well how the district battery allows to mitigate the power charge of the heat pump owner. Note also that the heat pump’s PV panels electricity consumption decreases when EV charging station are considered. This is because the EV chargers capture a part of the PV panels electricity that would go to this stakeholder if they were not considered.

The variation on the local distribution charges makes the heat pump owner’s bill to rise or decrease by a maximum of almost 30 kCHF. This represent around 5% of the 615 kCHF bills of the heat pump owner in the CEL proBT scenario, a rather small part of the the total bill thus, but figure 7.23 shows that it makes the CEL scenarii with no EV chargers interesting or not compared to the RCP scenarii. This variation decreases when EV chargers are taken into account since the heat pump’s PV panels electricity consumption decreases in that case.

7.3.2.6 EV’s chargers owner

Figures 7.24 shows the yearly balancing sheets for the 12 relevant scenarii for the EV charging stations owner.

Figure 7.24: Yearly balancing sheet of the EV chargers owner.

Here again, comparing scenarii under a same grouping category but with different tariffs, the MT tariff is the most advantageous not surprisingly as it is the cheapest for the end-user.

As for the heat pump owner, comparing scenarii under a same tariff but with different grouping category it appears that the both categories No grouping and RCP yields the same results. This is because the EV chargers are not included in any RCP since all RCP are simple RCP thus restricted to buildings. There are actually little differences between each grouping category with a similar tariff and when the battery is not considered. However, here the energy provided by the battery to the EV chargers represent a bigger volume than the energy provided by the battery to the heat pump. It makes thus a difference if the battery is considered or not, since the sells price of battery owner is very expensive. Figure 7.25 shows that the EV charging stations owner would loss 30 kCHF in the CEL scenario with battery and with the proBT tariff compared to the RCP proBT scenario which is yet, only a small fraction of the 820 kCHF it already pay in the RCP proBT scenario. However, if a new tariff that charges more the power demand was imposed, it would make the CEL even less attractive. However, figure ?? also chows that this could be mitigated by having an electricity price from PV panels production and from the battery less expansive.

Figure 7.25: Yearly difference between balancing sheets of the EV chargers owner for the 4 realistic CEL proBT scenarii compare to scneraio RCP proBT.

Here also the variation on the local distribution charges makes the EV chargers owner’s bill to rise or decrease by a small fraction of the total bill but it makes the CEL scenarii more or less interesting compared to the RCP scenario as shown in figure 7.25. These variations are of 35 kCHF with no battery (4% of the 800 kCHF bill) but are of almost 60 kCHF with a battery (7% of the 860 kCHF bill).

7.3.2.7 Battery owner

Figures 7.26 shows the yearly balancing sheets for the 6 relevant scenarii for the district battery owner. No comparison graph are shown for this stake holder since the battery is not considered in any RCP scenarii.

Figure 7.26: Yearly balancing sheet of the battery owner.

Note that since the battery owner’s electricity sells price does not include the battery CAPEX, it is normal that the annual balance is around 0. It is not exactly 0 as the sells price is the mean sells price \(p_{battery}^*\) over the \(N_{battery}\) scenarii that include the battery. Figures 7.26 perfectly illustrate the fact the battery usage depends on the tariff in use, as previously mentioned. If considering only professional tariff, as it would be the case for a CEL, the Use of the battery appears to be more relevant in the presence of EV chargers.

7.4 Summary

All stakeholders do not benefit form grouping in the form of CEL. However when comparing the situation of the RCP and the CEL the outcomes are more in favor of the CEL. Indeed it is important to remind that RCPs are organizational schemes already allowed in Switzerland. And the stakeholder who looses money in a CEL grouping also looses money from going to no grouping to RCP. This is the case for the DSO, SIG in other words. Moreover, this study focuses on simple RCP, but the model of consumption of a microgird RCP is similar to the CEL model except that the DSO is not rewarded for the usage of its network since, in a microgird RCP the network becomes private. Hence, the DSO is in a sense already prompt to lose money through RCPs. It is important however to be aware that the losses of SIG could be balanced by a revenue from an augmentation of its other customers charges.

On the other hand, Renters and PV panels owner would benefit from being in a CEL. Indeed, tariff would be more advantageous for the former and the latter would sell electricity at a better price. Regarding only realizable scenarii, there is no scenario better for both the renters and the PV panels owner than being in a CEL. For the heat pump owner and EV charging stations owner, the benefits from being in a CEL are not as pronounced as the ones for the renters and for the PV panels owner. Yet with a 5.25 cts/kWh local distribution charge, being in a CEL is still advantageous for the heat pump owner especially when EV chargers are taken into account. The EV chargers owner is also winning in being in a CEL except when it is supplied by the district battery which electricity is sold at a too high price.

A new tariff proposed by SIG could mitigate the poor outcomes for the DSO. In particular it could be done by increasing the power component of the distribution charges. However, this would make the CEL gropuing less attractive for the heat pump owner and the EV chargers owner. Yet, to mitigate this issue, savings could be made by having an electricity price for the PV panels production. This would transfer revenues from the PV panels owner to the DSO. Since the PV panels owner makes comfortable benefits from being in a CEL, this could be a fair solution. The impact of increasing the power component of the distribution charges on the other component of the grid electricity pricing should yet also be studied.

References

Schneider, Stefan, Pauline Brischoux, and Pierre Hollmuller. n.d. “Retour d’expérience énergétique Sur Le Quartier Des Vergers à Meyrin (Genève).”