Abstract
Hydrogen is being considered as an important option to contribute to energy system decarbonization. However, currently its production from renewables is expensive compared with the methods that utilize fossil fuels. This paper proposes a comprehensive optimization-based techno-economic assessment of a hybrid renewable electricity-hydrogen virtual power plant (VPP) that boosts its business case by co-optimizing across multiple markets and contractual services to maximize its profits and eventually deliver hydrogen at a lower net cost. Additionally, multiple possible investment options are considered. Case studies of VPP placement in a renewable-rich, congested area of the Australian network and based on real market data and relevant sensitivities show that multi-market participation can significantly boost the business case for cleaner hydrogen. This highlights the importance of value stacking for driving down the cost of cleaner hydrogen. Due to the participation in multiple markets, all VPP configurations considered are found to be economically viable for a hydrogen price of 3 AUD$/kg (2.25 USD$/kg), which has been identified as a threshold value for Australia to export hydrogen at a competitive price. Additionally, if the high price volatility that has been seen in gas prices in 2022 (and by extension electricity prices) continues, the flexibility of hybrid VPPs will further improve their business cases.
OVER the past decades, the focus on reducing the carbon footprint of electrical networks has been addressed by integrating renewable energy sources (RESs) and distributed energy resources (DERs) into the electrical networks. The rapid increase in RESs in turn has led to an increase in curtailed renewable energy. This has led to the emergence of hydrogen generated using renewable energy as a possible energy vector to help phase out fossil fuels. This hydrogen is created by using energy from RES to power an electrolyser. This process releases virtually zero emissions, and the hydrogen can then be used as a fuel source for vehicles [
The rise in RES has also caused a decrease in scheduled thermal generation, which traditionally would supply network services. Therefore, there is an opportunity for new players in the network to fill this gap and provide technical services and collect the associated revenues. Reference [
Reference [
Whilst it has been shown that electrolysers and other hydrogen-based resources can provide network services and participate in markets, previous works on business cases and economic feasibility of integrating hydrogen resources into the electrical networks overlook this. The economic benefits of integrating electrolysers into the electrical network have been studied, often operating in conjunction with RES such as wind farms [
It is clear therefore, that to create a high accuracy business case assessment for investing in electrolysers and other hydrogen-based resources, it is important to consider all possible revenue streams these resources could tap into in the electrical network, as well as possible limitations and revenue streams related to the electrical network. It has been shown in previous works that hydrogen resources can provide network services and participate in markets, but the impact of this on the business cases for hydrogen resources and net cost of producing hydrogen is currently lacking in the literature. The references are not aware of any work that provides a business case utilizing a comprehensive techno-economic assessment of hydrogen costs from an electrolyser coordinated in a VPP, that considers the full range of possible markets and services, as well as modeling the electrical network and utilizing current real-world prices. This could help inform the understanding of the true economic potential of hydrogen in the electrical network, which is key to accelerating the advent of the hydrogen economy.
This work considers a number of investment options for hydrogen-based resources with various market/service portfolios and conducts a comprehensive techno-economic assess-ment of the economic viability of fixed price hydrogen contracts within the context of the Australian energy network. To do this, a deterministic scheduling optimization considering multiple energy vectors is utilized. It is noted that, as there is no spot market for hydrogen in Australia, such as there is for natural gas, the electrolyser operator would need to enter into a sale and purchase agreement (SPA) with a third party who is buying the hydrogen. In this way, the fixed price hydrogen contracts considered in this work are the most likely method of buying and selling large quantities of hydrogen for the foreseeable future.
The optimization used in this work schedules and dispatches a multi-energy VPP participating in markets/contracts for:
1) An intra-day wholesale energy spot market.
2) A fixed-price hydrogen contract/SPA.
3) Six intra-day frequency control ancillary service (FCAS) markets to respond to contingency events to raise or lower network frequency.
4) A voltage control ancillary service (VCAS) in the form of upstream reactive power support.
5) A system restart ancillary service (SRAS) contract (if a resource is invested in which can provide the service).
6) A fast frequency response (FFR) contract with new renewable generators in the network to provide FFR on their behalf.
7) Contractual arrangements with local renewables to buy renewable energy curtailed due to congestion.
The main contributions of this work are:
1) Investigation of the economic viability of possible combinations of an electrolyser, hydrogen storage, fuel cells, and hydrogen-powered open cycle gas turbine (OCGT), for a fixed-price hydrogen contract/SPA compared with battery energy storage (as a current popular choice for utility energy storage which can also provide network services).
2) Presentation of comprehensive business cases assessing investment options for hydrogen-based resources considering the full range of markets and services, and the impacts of electrical network constraints.
3) A robust business case for the hydrogen-electricity VPP informed by sensitivity analysis of multiple market prices and magnitude of contractual services.
The rest of the paper is structured as follows. Section II presents the methodology used for the techno-economic assessments in this work. Section III outlines the case study to which the methodology is applied, based on the introduction of an electricity-hydrogen VPP into a renewable-rich area of the network. Section IV provides the results from the case study, and Section V summarizes the findings of this work.
The VPP modelling conducts a unit commitment utilizing multi-service, multi-energy optimal power flow studies for maximizing the profit of a VPP participating in multiple markets [
The centralized approach to the optimization proposed in this work with the VPP acting as a price-taker internalizes the real-world two-stage interaction between the VPP (and other commercial entities) and the market operator by having the VPP operator anticipate the network constraints and respond accordingly. This allows the commercial planning problem to be solved without modelling the full bidding process, but still factoring network constraints and market prices into the VPP operation.
The electrical power flows are modelled using a decoupled linear power flow model [
The optimization proposed is deterministic. As part of this planning problem, it is assumed that the VPP operating close-to-real-time will have good information on which to base its operation. Previous works have shown how a VPP operating close-to-real-time can effectively respond to uncertainty through receding horizon control, minimizing the loss in VPP revenue due to uncertainty [
VPP operation is optimized across the time horizon , in steps of length . Throughout, subscript t denotes a time-varying parameter, whilst denotes a variable’s value at time . The subscript k denotes the corresponding instance of a parameter or decision variable for specific resource . The subscript i denotes the corresponding instance of a parameter or decision variable for specific node .
A unit commitment model with linearized power flow constraints and fixed generation cost is used in this work, as is typical in such studies [
The objective function (1) is a sum of the cost (2) over each time step, where hours and hour is the time step length of the optimization (resulting in a 24-hour horizon with 30-minute resolution).
(1) |
(2) |
The VPP cost in (2) comprises eight terms: terms I and II represent buying active and reactive power from the power grid and , respectively; term III represents the revenue (negative cost) from selling hydrogen, where is the hydrogen sold at each node; term IV represents the cost of operating the scheduled resources as encoded by (3); term V represents the cost of curtailment ; terms VI and VII represent the revenue from raise FCAS service bids and lower FCAS service bids , respectively; and term VIII is the cost of purchasing curtailed renewable generation , which allows the optimization to consider a link between the VPP and RES in congested locations. A well placed VPP can buy energy from an RES that would otherwise be curtailed due to upstream line congestion.
The resource operating cost is encoded by the constraint:
(3) |
where is the committed status of resources in the time step before the optimization. There is a cost associated with the power generated by each resource , the power demanded by each resource , as well as a turn-on and turn-off cost.
The optimization problem includes the following linearized power flow constraints for :
(4) |
(5) |
(6) |
(7) |
(8) |
(9) |
(10) |
(11) |
(12) |
(13) |
(14) |
(15) |
The active and reactive branch power flows, and , respectively, are constrained in (4)-(7) by a linear approximation of the non-linear power-flow relationship. This linearization includes neglecting network losses. This is shown in (8) and (9) where the power flow through a line in each direction has the same magnitude, but opposite polarity. The nodal voltage magnitude and angle limit are set in (10) and (11), respectively. In (12) and (13), the active and reactive power flows in each line are defined as a function of nodal voltages. Equations (
The resource modelling is encoded by the following additional constraints for :
(16) |
(17) |
(18) |
(19) |
(20) |
(21) |
(22) |
(23) |
(24) |
(25) |
(26) |
(27) |
(28) |
(29) |
(30) |
(31) |
(32) |
(33) |
(34) |
where , , , and are the initial values for the normalized level of stored electrical energy for each resource , the normalized level of stored hydrogen at each node , and the active power generated or demanded by each resource, respectively. The minimum “up time” and “down time” of devices are constrained in (16) and (17). Device active power generation and demand limits are constrained in (18) and (19), respectively.
A hydrogen network is not modelled in this optimization, as there is not one currently in place in Australia. It is noted that in Australia, if there exists a natural gas distribution network nearby, the hydrogen can be injected into this network. If the VPP were to be paid for the amount of hydrogen injected into the natural gas network, from a VPP perspective, the model proposed in this work could also be used. However, modelling the flow of hydrogen in the natural gas network is an area of ongoing research [
In addition to buying/selling electrical energy and selling hydrogen, the resources can also participate in other markets and provide services to the network. Their ability to do so is dependent on their ramping capability as well as their power and energy headroom/footroom. For example, for a fuel cell to increase its active power output to provide a network service, there must be sufficient hydrogen reserves to perform this operation.
The contribution of each resource to providing network services is constrained as follows. For ,
(35) |
(36) |
(37) |
(38) |
(39) |
(40) |
(41) |
(42) |
(43) |
(44) |
(45) |
(46) |
(47) |
where denotes the maximum element of the vector argument . Each resource’s network service response capability is constrained by its ramp rate in (35) and (36), and its maximum and minimum power in (37) and (38), respectively. If the resource has electrical storage, its response is also limited by its available energy storage headroom/footroom in (39)-(42). Note that the energy capacity requirements in (39)-(42) are restricted to the set of resources that accommodate energy storage. Likewise, if it is a hydrogen-based device, it is constrained by the headroom/footroom in the node’s hydrogen storage (43)-(46). Further, (47) ensures that the amount of FFR provided by the VPP meets the contracted amount. The sets of resources and denote resources that supply/demand FFR services, respectively.
To summarize, the problem is to minimize (1) over the quantities labelled “variables” in the nomenclature section, e.g., , with hours, hour, subject to (2)-(47). This is run for each day in a year to determine annual operational results.
The case study used in this paper is based on an area of the South Australian (SA) network that is weakly linked to the rest of the power grid. Currently it consists of three distribution substations, two thermal open cycle gas turbines (OCGTs) (50 MW and 23 MW) and two wind farms (~70 MW). For this case study, the network has been supplemented with additional RES in the form of two 30 MW PV solar farms which are under consideration for installation, as shown in

Fig. 1 Diagram of network and position of proposed VPP.
The local area is home to large export facilities for shipping goods to international destinations. This is a prime location for hydrogen generation for international distribution, as this minimizes domestic transportation requirements. For this case study, the modelling approach is utilized by a potential VPP owner, and the inclusion of network constraints in the model allows them to understand the possible impact that the electrical network may have on their operation when the market operator is dispatching resources. This includes identifying where there may be potential for agreements with renewables to purchase their energy that would otherwise be curtailed, or to understand if there may be existing network constraints that could limit their ability to be dispatched. Information on constraints in the national electricity market dispatch engine, as well as resource bids and dispatch instructions are publicly available from the Australian Energy Market Operator (AEMO). The case study will consider the techno-economic benefits of installing an electricity-hydrogen VPP in the local network, using an electrolyser to generate hydrogen to sell to a third party. To reduce the net cost of generating this hydrogen, the VPP will also participate in additional markets and services. Possible resources to be included in the VPP are a battery energy storage system (BESS), electrolyser, FC, hydrogen OCGT, and hydrogen storage.
The position of these resources in the network, as well as the general network configuration, is shown in
A. Cost of Resources in VPP
To be able to fully consider the profitability of the VPP, it is not sufficient to only consider the revenue that the VPP accrues from operating in markets. The investment over the lifetime of the devices including capital expenditure (CAPEX) and operational expenditure (OPEX) should be considered to determine how much revenue the VPP needs to obtain. In this case study, it is assumed that the VPP operator will own all of the resources in the VPP. The devices that may be included in the VPP are shown in
Device | Capital investment | Fixed operation & maintenance | Size | CAPEX (M$) | OPEX per year (k$) | Lifetime (year) | EAC (M$) |
---|---|---|---|---|---|---|---|
BESS [ | 813 $/kW+543 $/kWh | 10 $/(kW·year) | 30 MW/8 MWh | 28.734 | 300 | 20 | 3.012 |
Proton exchange membrane (PEM) electrolyser [ | 1400 $/kW | 54 $/(kW·year) | 30 MW | 42.000 | 1620 | 20 | 5.585 |
Hydrogen OCGT [ | 1250 $/kW | 12.6 $/(kW·year) | 10 MW | 12.500 | 126 | 40 | 1.064 |
Hydrogen FC [ | 2109 $/kW | 58 $/(kW·year) | 5 MW | 10.545 | 290 | 20 | 1.285 |
Hydrogen storage [ | 1032 $/(kg·day) | 22 $/(kg·day) | 11475 kg/day | 11.842 | 252 | 40 | 1.140 |
The EAC represents the annual cost of owning, operating, and maintaining an asset. It is especially useful for comparing costs of assets with differing lifespans. The EAC is calculated as:
(48) |
where is the discount rate set to be 7%; and is the asset lifetime.
For this case study, the hydrogen storage capability is sized to be able to store 11475 kg of hydrogen each day. This size is calculated based on a 30 MW electrolyser operating with a capacity factor of 85% [
Case | Resource | EAC (M$) |
---|---|---|
1 | BESS | 3.01 |
2 | storage | 6.73 |
3 | FC | 8.01 |
4 | 7.79 | |
5 | FC | 9.08 |
6 | 12.09 | |
7 | 11.02 |
B. Services and Markets
There are 28 case studies considered in this work, utilizing seven VPP configurations and four market/service portfolios. A VPP with Portfolio A would participate in wholesale energy, curtailed RES, and hydrogen markets. Portfolio B includes all markets of Portfolio A as well as contingency FCAS markets. Portfolio C includes all markets of Portfolio B as well as provision of FFR. Portfolio D includes all markets and services of Portfolio C, and additionally includes VCAS and SRAS.
Past wind generation profiles and wholesale energy and FCAS prices are provided by AEMO [
1) Wholesale Energy
The optimization problem in Section II relates to maximizing revenue from energy exchange at the grid connection point in
2) Curtailed Renewables
Due to the remote location of this area of the network, and the high potential of renewables, there are times when RES generation is curtailed due to line thermal limits being reached. In this work, the VPP can purchase this energy that would otherwise be curtailed from RES for a price of 30 $/MWh.
3) Hydrogen
In these case studies, hydrogen is sold as part of a fixed-price contract. Prices of 2-3 $/kg (excluding storage and transport) have been identified as the target price region for Australia to be able to compete with other exporting countries [
4) Contingency FCASs
Contingency FCASs are the services to help the network cope with a sudden change in network load or generation (a contingency event), and in Australia, are divided into two sets of services. Contingency raise FCAS services are used to raise the system frequency and are divided depending on the required response time into fast (6 s), slow (60 s) and delayed services (300 s). In addition, there are also fast, slow, and delayed lower FCAS services with the same response time requirements that are used to lower the system frequency after a contingency. There are then six contingency FCAS markets that the VPP can choose to participate in.
5) FFR
FFR in the Australian network is defined as “the delivery of a rapid active power increase or decrease by generation or load in a timeframe of two seconds or less” [
6) VCAS
AEMO maintains voltage levels across the transmission network within relevant limits. This can be done by absorbing or injecting reactive power into transmission network connection points [
7) SRAS
In the Australian system, the provision of black-start services falls into SRAS. Each subsection of the Australian system is assessed on the magnitude of SRAS that it requires. For the SA system, this is 330 MW [
C. Feasible Operating Region (FORs) of VPP
A VPP can participate in markets and provide services by utilizing its electrical flexibilitya measure of the capability of a VPP to deviate from a set dispatch point. Flexibility maps, created using the approach proposed in [

Fig. 2 Feasibility maps of each of seven VPP configurations in Q-P space and H2-P space. (a) Case 1. (b) Case 2. (c) Case 3. (d) Case 4. (e) Case 5. (f) Case 6. (g) Case 7.
Electrical flexibility can be considered as either upward flexibility (the ability of a VPP to increase its active power injection or consume less active power) or downward flexibility (the ability of a VPP to decrease its active power injection or consume more active power). The amount of upward and downward flexibility that is available to the VPP at a specific time is dependent upon its dispatch point. For example, for the VPP in Case 2 (the electrolyser) to provide 10 MW of upward flexibility (for example to provide FFR), it must be operating with an active power dispatch point no greater than -10 MW, forcing it to absorb power while providing the service. However, if the VPP in Case 5 (electrolyser, fuel cell, and hydrogen OCGT) were providing that same 10 MW of upward flexibility, it would only need to have an active power dispatch less than 5 MW (assuming reactive power dispatch is 0 Mvar and all resources are on). Therefore, in Case 5, the VPP still has the capability to either inject or absorb active power (to respond to market prices) while providing this level of FFR. The VPP’s flexibility is also dependent on the VPP’s reactive power dispatch point. This highlights the importance of optimizing VPP operation in all markets/services, considering both active and reactive power simultaneously. Making these decisions in a non-holistic manner may reduce the revenue accrued from multi-market participation.
It can be observed by looking at the flexibility maps for Cases 1, 5, and 6 that when resources are aggregated, the resulting FOR is of greater size than the sum of the FORs of each individual resource. This aggregating of resources also dramatically increases the VPP flexibility in the H2-P space. This gives a VPP the ability to vary its hydrogen output whilst maintaining its active power dispatch point. In general, a VPP would want to maximize its hydrogen output for a set active power dispatch, as hydrogen can be monetized. However, if there exist binding constraints on the hydrogen infrastructure (storage size, export limits, etc.), the VPP could utilize its internal flexibility (by modulating internal dispatch factors) to accommodate these constraints while minimizing the effect that this would have on the VPP’s electrical operation.
A. Benefits from Multi-market Operation
The VPP revenue changes for the 28 cases with hydrogen price of 2 $/kg or 3 $/kg and FFR contract set at 50% or 33% of

Fig. 3 VPP revenues in 2017 for 50% or 33% contractual FFR provision and hydrogen prices of 2 $/kg or 3 $/kg. (a) 50% FFR and hydrogen price of 2 $/kg. (b) 33% FFR and hydrogen price of 2 $/kg. (c) 50% FFR and hydrogen price of 3 $/kg. (d) 33% FFR and hydrogen price of 3 $/kg.
The largest impact of multi-market participation can be seen when the VPPs add participation in the six contingency FCAS markets to their portfolio. The effect of this can be seen most prominently in VPPs containing a BESS (a resource which derives most of its value from FCAS). Considering a hydrogen price of 3 $/kg in Case 2A, the electrolyser capacity factor is 20.3%. This is close to the capacity factor that would be expected by considering only the wholesale energy prices over the year. In 2017, only 13.4% of price intervals are below the price threshold necessary to sell hydrogen at a profit. The reason in Case 2A where the electrolyser capacity factor is higher than this is due to the contractual arrangement the electrolyser has with the wind farm on node 18 to buy the energy that would otherwise be curtailed. This contract leads to an additional 1.05 M$ of hydrogen revenue in 2017 with a hydrogen price of 3 $/kg. Participation in FCAS also leads to an increase in electrolyser capacity factor of 25.1% in 2017, more than doubling the amount of hydrogen generated and sold in 2017.
In summary, multi-market participation is key to boosting VPP economic viability and can lead to additional hydrogen generation.
B. Sensitivity to Magnitude of Contractual FFR
The only service that can cause a reduction in VPP revenue in these case studies is FFR. This is because this is a contractual arrangement rather than a market where a VPP can respond to price signals when deciding whether to participate. The VPP configuration that is worst affected by this is Case 2 (only an electrolyser). This is because an electrolyser providing FFR must act as a load at the required magnitude (so that it can be turned off if required to provide the net increase in power output). During these periods of FFR provision, the electrolyser is very limited in how it can respond to the wholesale energy market prices (as explained in
In summary, VPP operators should carefully consider contractual arrangements they enter into to ensure that they will be profitable, as contracts can require reserving flexibility which could otherwise be used to generate revenue.
C. Sensitivity to Hydrogen Prices
Examining the revenues for different VPP configurations when the hydrogen price is 2 $/kg, as shown in
When the hydrogen price is changed to 3 $/kg, there is a large change of hydrogen-based VPP revenues. Increasing the hydrogen price allocates more values to the hydrogen produced when the VPPs are providing FFR services. This is because, for an electrolyser to provide FFR, necessarily it must be on and consuming electricity (as explained in Section III-C), and therefore creating hydrogen. As the price of the fixed-price SPA increases, the value of the hydrogen that is being created as a biproduct of being available to provide FFR also increases.
To summarize, for a hydrogen SPA priced at 3 $/kg and FFR requirement of 33%, all the VPP configurations manage to generate profit from providing FFR when compensated at 1.506 M$/year.
D. Sensitivity to Energy and FCAS Prices
VPP revenue is highly dependent on wholesale energy and FCAS market prices. To further consider the effects of changing wholesale energy and FCAS prices, as well as trends within these markets, case studies are run utilizing the market prices during 2013-2020. Informed by the findings in

Fig. 4 VPP revenues for 2013-2020 in Cases 1D-7D with hydrogen price of 3 $/kg and FFR requirement of 33% of PV generation.
The significantly higher FCAS prices in 2020 lead to much higher revenue in all configurations. If wholesale energy prices fall in the coming years due to increased RES integration, and FCAS prices continue to rise, then the profits of hydrogen-based VPPs will continue to increase. VCAS and SRAS with the assumed pricing are not significant compared with the other revenues. VCAS revenue varies from $58000-$157000 per year. SRAS when it is provided is worth $103000 per year.
In summary, VPP revenue is highly dependent on market prices, but diversifying market participation helps protect revenue generation from unfavorable conditions in a single market.
To assess the amount of hydrogen that is sold each year, it is compared with the maximum possible sale (i.e., if the electrolyser is operated at the maximum power constantly over the whole year) which we will refer to as the VPP’s capacity factor.
Year | Capacity factor (%) | ||||||
---|---|---|---|---|---|---|---|
Business as usual | Case 2D | Case 3D | Case 4D | Case 5D | Case 6D | Case 7D | |
2013 | 57.8 | 67.9 | 65.2 | 66.9 | 64.8 | 62.9 | 63.0 |
2014 | 82.9 | 89.3 | 88.4 | 89.0 | 88.2 | 87.6 | 87.7 |
2015 | 86.7 | 90.0 | 89.0 | 89.2 | 88.7 | 88.3 | 88.4 |
2016 | 56.6 | 71.8 | 69.2 | 69.9 | 68.3 | 67.0 | 67.1 |
2017 | 13.4 | 55.2 | 51.2 | 48.6 | 46.8 | 44.9 | 45.4 |
2018 | 15.2 | 61.0 | 57.9 | 53.4 | 52.1 | 51.2 | 54.5 |
2019 | 22.2 | 42.9 | 38.8 | 38.5 | 36.2 | 34.6 | 35.3 |
2020 | 76.2 | 89.7 | 88.5 | 87.2 | 86.2 | 86.0 | 88.1 |
Average | 51.4 | 71.0 | 68.5 | 67.8 | 66.4 | 65.3 | 66.2 |
These results highlight that multi-market participation allows additional profitable hydrogen generation, as well as unlocking additional revenue streams.
E. Benefits to Wider Network
The integration of a VPP can have wider network benefits, especially if the network is operating close to design limits (i.e., a congested network due to high RES export). These benefits can only be assessed by using an optimization that models the electrical network, such as the one proposed in this work.
1) Curtailed Renewables
A benefit of the VPP in a congested area of the network is the ability of the VPP to utilize renewable energy that would previously have been curtailed. As an example, in the 2017 “business as usual” case, 17% of the 156820 MWh of energy generated by the wind farm at node 18 would be curtailed due to line thermal limits. In Case 6
2) Easing of Network Congestion
One additional benefit of having an electrolyser provide FFR is that it may act to relieve congestion in the wider network, allowing generators to export more energy and accrue higher revenue. In the case studies, this is predominately the OCGTs at nodes 13 and 15. The easing of congestion that the VPPs deliver when providing FFR is worth on average 2.3 M$/year from 2013 to 2020. In this work, there is no mechanism for the VPP to access any of this additional value. However, the agreements could be made to have part of these funds be used to supplement the FFR payment that the VPP receives from the PV generators, leading to additional profits.
F. Value Metrics
The annual revenues obtained from the detailed modelling conducted in this work can be used as an input for cost-benefit analysis to determine if the 3 $/kg fixed-priced contract to sell hydrogen is economically feasible. To assess the economic viability of the proposed VPP configurations and 3 $/kg hydrogen SPA, two value metrics are used, i.e., net present value (NPV) and discounted payback period (DPP).
1) NPV
NPV represents the difference between the expected revenues of a project (converted into “today’s money” by using a discount rate) and the amount in initial investment required. In this way, NPV captures the total value of the project. The NPV of each option is calculated using (49).
(49) |
where is the discount rate of 7%; and is the number of years considered; and is the yearly income. The VPP includes devices with 20-40 years’ lifetimes. So, a conservative estimate for VPP lifetime of 20 years is considered for this analysis.
2) DPP
Another indicator that can be used to consider the economic viability of investment in a project is the DPP of the investment, which can be determined by (50). The DPP determines how long it takes to recoup the initial investment cost of a project while also incorporating a discount rate to recognize the changing value of money over time.
(50) |
The NPV and DPP for each VPP configuration with Portfolio D are shown in
Case | Average yearly revenue ($) | NPV ($) | DPP (year) |
---|---|---|---|
1D | 10630463 | 80707072 | 3.20 |
2D | 8167473 | 12721257 | 13.55 |
3D | 9669348 | 15014878 | 13.60 |
4D | 9039243 | 8121955 | 15.98 |
5D | 10368671 | 8588670 | 16.26 |
6D | 19657388 | 75081266 | 8.40 |
7D | 19143685 | 83473934 | 7.34 |
The ability to generate hydrogen via electrolysis and predominantly from renewable electricity for 2-3 $/kg is essential for the development of the Australian hydrogen export industry. While previous analysis has determined this is not feasible presently, the more detailed modelling conducted in this work highlights how multi-market/service participation and aggregation into a VPP can allow the sale of hydrogen for 3 $/kg to be economically viable. Furthermore, participating in multiple markets acts to increase the maximum value of wholesale energy below which the VPP can profitably create hydrogen. This in turn allows the VPP to create more hydrogen at a lower cost.
However, even with this innovative value-stacking approach, the prices of wholesale energy in the Australian market and the initial investment costs of resources are too high to allow hydrogen to be sold at 2 $/kg. It is important to analyze potential operation over a number of years to determine an accurate value for yearly revenue, as market prices vary greatly between years. The largest hurdle to a VPP’s ability to sell hydrogen at 2 $/kg is the current investment cost of the technology. However, as these technologies mature, investment costs will reduce and future multi-energy VPPs will be well placed to generate this low-cost hydrogen, especially if there is a reduction in wholesale energy prices or increase in FCAS prices.
It is worth noting that the competitiveness of hydrogen generated from RES may be increased by external factors, such as the high and volatile gas prices experienced worldwide in 2022. The highly flexible nature of hydrogen-based VPPs will allow them to improve their business cases in more volatile markets. This will act to make hydrogen generated from RES more price-competitive. Additionally, it will mean that using hydrogen to power OCGTs will also become more valuable, as using natural gas becomes more expensive.
Nomenclature
Symbol | —— | Definition |
---|---|---|
A. | —— | Sets |
—— | Set of lines (branches) of network | |
—— | Set of lines (branches) connected to node | |
—— | Set of resources in the network | |
—— | Set of resources at node | |
—— | Set of hydrogen-based resources in the network | |
—— | Set of hydrogen-based resources at node | |
—— | Set of energy storage resources in the network | |
—— | Set of nodes in the network | |
B. | —— | Parameters |
, | —— | Allowable curtailments in each time period and across the whole horizon for resource |
—— | Operating cost of fixed resource | |
—— | Optimization time step length | |
—— | Available generation (if positive) or demand power (if negative) of resource k | |
—— | Initial resource commitment status | |
, | —— | Generation and consumption efficiencies of resource |
, | —— | The maximum and minimum nodal voltage angles at node |
—— | Price of curtailing active power at time | |
—— | Price of hydrogen at time | |
, | —— | Start-up and shut-down costs of resource k |
, | —— | Prices of wholesale energy and reactive power at time |
, | —— | Vectors of raise and lower frequency control ancillary services (FCASs) market prices at time |
—— | Cost of purchasing previously curtailed renewable energy at time | |
—— | Energy storage losses of resource k | |
, | —— | The maximum and minimum ramp up capabilities of resource k |
, | —— | Vectors of raise and lower FCAS service call times |
, | —— | Vectors of raise and lower FCAS service durations |
, | —— | The maximum and minimum active/reactive dispatch power ratios of resource k |
—— | Polarity of (-1 if negative and 1 if positive) | |
—— | Generation to be covered by fast frequency response (FFR) of resource k | |
, | —— | Generation and consumption costs of resource k |
—— | Active power curtailment when virtual power plant (VPP) is not present of resource k | |
, | —— | Susceptance and conductance of branch |
—— | Resource storage capacity | |
—— | Hydrogen storage capacity at node | |
—— | Normalized stored hydrogen target for end of optimization at node | |
, | —— | The maximum and minimum active power generations of resource k |
, | —— | The maximum and minimum active power consumptions of resource k |
, | —— | The maximum and minimum reactive power of resource k |
—— | Power flow limit of branch | |
—— | Number of time steps in optimization | |
, | —— | The minimum “up time” and “down time” of resource k |
, | —— | The maximum and minimum nodal voltage magnitudes at node |
—— | Normalized stored energy target of resource for end of optimization | |
C. | —— | Variables |
—— | Commitment status (0 is off and 1 is on) of resource k at time | |
—— | Nodal voltage angle at node at time | |
—— | Active power curtailment of resource k at time | |
—— | Total cost of network operation at time | |
—— | Cost of operating resources scheduled at time | |
—— | Normalized level of stored hydrogen at node at time | |
—— | Amount of hydrogen exported at node i at time t | |
—— | Active power injection of resource k at time | |
—— | Active power flow of branch at time | |
—— | Active power imported from power grid at time | |
—— | FFR bid of resource at time | |
, | —— | Active power generation and consumption of resource at time |
—— | Vector of all lower FCAS bids of resource at time | |
—— | Vector of all raise FCAS bids of resource at time | |
—— | Reactive power injection of resource at time | |
—— | Reactive power flow of branch at time | |
—— | Reactive power imported from power grid at time | |
—— | Nodal voltage magnitude at node at time | |
—— | Normalized level of stored energy of resource at time |
References
S. Clegg, L. Zhang, and P. Mancarella, “The role of power-to-transport via hydrogen and natural gas vehicles in decarbonising the power and transportation sector,” in Proceedings of IEEE PES Innovative Smart Grid Technologies Conference Europe, Torino, Italy, Sept. 2017, pp. 1-6. [Baidu Scholar]
S. Clegg and P. Mancarella, “Integrated modeling and assessment of the operational impact of power-to-gas (P2G) on electrical and gas transmission networks,” IEEE Transactions on Sustainable Energy, vol. 6, no. 4, pp. 1234-1244, May 2015. [Baidu Scholar]
M. Yu, K. Wang, and H. Vredenburg, “Insights into low-carbon hydrogen production methods: green, blue and aqua hydrogen,” International Journal of Hydrogen Energy, vol. 46, no. 41, pp. 21261-21273, Jun. 2021. [Baidu Scholar]
S. Bruce. (2018, Dec.). National hydrogen roadmap, CSIRO, Newcastle, Australia. [Online]. Available: https://www.csiro.au/-/media/Do-Business/Files/Futures/180314_EN_NationalHydrogen Roadmap_WEB_180823.pdf [Baidu Scholar]
X. Li and M. Mulder, “Value of power-to-gas as a flexibility option in integrated electricity and hydrogen markets,” Applied Energy, vol. 304, p. 117863, Dec. 2021. [Baidu Scholar]
G. Chicco, S. Riaz, A. Mazza et al., “Flexibility from distributed multienergy systems,” Proceedings of the IEEE, vol. 108, no. 9, pp. 1496-1517, Apr. 2020. [Baidu Scholar]
E. Corsetti, S. Riaz, M. Riello et al., “Modelling and deploying multi-energy flexibility: the energy lattice framework,” Advances in Applied Energy, vol. 2, p. 100030, May 2021. [Baidu Scholar]
H. Khani, N. A. El-Taweel, and H. E. Z. Farag, “Supervisory scheduling of storage-based hydrogen fueling stations for transportation sector and distributed operating reserve in electricity markets,” IEEE Transactions on Industrial Informatics, vol. 16, no. 3, pp. 1529-1538, Mar. 2020. [Baidu Scholar]
N. A. El-Taweel, H. Khani, and H. E. Z. Farag, “Hydrogen storage optimal scheduling for fuel supply and capacity-based demand response program under dynamic hydrogen pricing,” IEEE Transactions on Smart Grid, vol. 10, no. 4, pp. 4531-4542, Jul. 2019. [Baidu Scholar]
L. Zhang, S. Clegg, and P. Mancarella, “Modeling of electrolyzers in hydrogen vehicle refueling stations for provision of ancillary services,” in Proceedings of Bulk Power Systems Dynamics and Control Symposium, Espinho, Portugal, Sept. 2015, pp. 1-6. [Baidu Scholar]
B. A. Robbins and A. D. Domínguez-García, “Optimal reactive power dispatch for voltage regulation in unbalanced distribution systems,” IEEE Transactions on Power Systems, vol. 31, no. 4, pp. 2903-2913, Jul. 2016. [Baidu Scholar]
B. Zhang, A. Y. S. Lam, A. D. Domínguez-García et al., “An optimal and distributed method for voltage regulation in power distribution systems,” IEEE Transactions on Power Systems, vol. 30, no. 4, pp. 1714-1726, Jul. 2015. [Baidu Scholar]
E. Dall’Anese, S. V. Dhople, and G. B. Giannakis, “Optimal dispatch of photovoltaic inverters in residential distribution systems,” IEEE Transactions on Sustainable Energy, vol. 5, no. 2, pp. 487-497, Jan. 2014. [Baidu Scholar]
N. Ruiz, I. Cobelo, and J. Oyarzabal, “A direct load control model for virtual power plant management,” IEEE Transactions on Power Systems, vol. 24, no. 2, pp. 959-966, May 2009. [Baidu Scholar]
C. Suazo-Martinez, E. Pereira-Bonvallet, R. Palma-Behnke et al., “Impacts of energy storage on short term operation planning under centralized spot markets,” IEEE Transactions on Smart Grid, vol. 5, no. 2, pp. 1110-1118, Mar. 2014. [Baidu Scholar]
M. Tostado-Véliz, P. Arévalo, and F. Jurado, “A comprehensive electrical-gas-hydrogen microgrid model for energy management applications,” Energy Conversion and Management, vol. 228, p. 113726, Jan. 2021. [Baidu Scholar]
M. A. Pellow, C. J. M. Emmott, C. J. Barnhart et al., “Hydrogen or batteries for grid storage? A net energy analysis,” Energy & Environmental Science, vol. 8, no. 7, pp. 1938-1952, Apr. 2015. [Baidu Scholar]
H. Fu, Z. Wu, X.-P. Zhang et al., “Contributing to DSO’s energy-reserve pool: a chance-constrained two-stage VPP bidding strategy,” IEEE Power and Energy Technology Systems Journal, vol. 4, no. 4, pp. 94-105, Dec. 2017. [Baidu Scholar]
J. Naughton, H. Wang, S. Riaz et al., “Optimization of multi-energy virtual power plants for providing multiple market and local network services,” Electric Power Systems Research, vol. 189, p. 106775, Dec. 2020. [Baidu Scholar]
E. A. M. Ceseña, N. Good, A. L. A. Syrri et al., “Techno-economic and business case assessment of multi-energy microgrids with co-optimization of energy, reserve and reliability services,” Applied Energy, vol. 210, pp. 896-913, Jan. 2018. [Baidu Scholar]
R. Moreno, R. Moreira, and G. Strbac, “A MILP model for optimising multi-service portfolios of distributed energy storage,” Applied Energy, vol. 137, pp. 554-566, Jan. 2015. [Baidu Scholar]
D. Kroniger and R. Madlener, “Hydrogen storage for wind parks: a real options evaluation for an optimal investment in more flexibility,” Applied Energy, vol. 136, pp. 931-946, Dec. 2014. [Baidu Scholar]
S. McDonagh, S. Ahmed, C. Desmond et al., “Hydrogen from offshore wind: Investor perspective on the profitability of a hybrid system including for curtailment,” Applied Energy, vol. 265, p. 114732, May 2020. [Baidu Scholar]
M. Eypasch, M Schimpe, A. Kanwar et al., “Model-based techno-economic evaluation of an electricity storage system based on liquid organic hydrogen carriers,” Applied Energy, vol. 185, pp. 320-330, Jan. 2017. [Baidu Scholar]
L. Weimann, P. Gabrielli, A. Boldrini et al., “Optimal hydrogen production in a wind-dominated zero-emission energy system,” Advances in Applied Energy, vol. 3, p. 100032, Aug. 2021. [Baidu Scholar]
A. Rabiee, A. Keane, and A. Soroudi, “Technical barriers for harnessing the green hydrogen: a power system perspective,” Renewable Energy, vol. 163, pp. 1580-1587, Aug. 2021. [Baidu Scholar]
W. Sun and G. P. Harrison, “Active load management of hydrogen refuelling stations for increasing the grid integration of renewable generation,” IEEE Access, vol. 9, pp. 101681-101694, Jul. 2021. [Baidu Scholar]
J. Yang, Z. Ning, C. Kang et al., “A state-independent linear power flow model with accurate estimation of voltage magnitude,” IEEE Transactions on Power Systems, vol. 32, no. 5, pp. 3607-3617, Sept. 2017. [Baidu Scholar]
J. Naughton, H. Wang, M. Cantoni et al., “Co-optimizing virtual power plant services under uncertainty: a robust scheduling and receding horizon dispatch approach,” IEEE Transactions on Power Systems, vol. 36, no. 5, pp. 3960-3972, Sept. 2021. [Baidu Scholar]
S. Riaz, G. Verbič, and A. C. Chapman, “Computationally efficient market simulation tool for future grid scenario analysis,” IEEE Transactions on Smart Grid, vol. 10, no. 2, pp. 1405-1416, Mar. 2019. [Baidu Scholar]
A. de Corato, I. Saedi, S. Riaz et al., “Aggregated flexibility from multiple power-to-gas units in integrated electricity-gas-hydrogen distribution systems,” Electric Power Systems Research, vol. 212, p. 108409, Nov. 2022. [Baidu Scholar]
T. S. Brinsmead, P. Graham, J. Hayward et al. (2015, Sept.). Future energy trends: an assessment of the economic viability, potential uptake and impacts of electrical energy storage on the NEM 2015-2035. CSIRO. Newcastle, Australia. [Online]. Available: https://www.aemc.gov.au/sites/default/files/content/fa7a8ca4-5912-4fa9-8d51-2f291f7b962 [Baidu Scholar]
1/CSIRO-Future-Trends-Report.pdf [Baidu Scholar]
McKinsay & Company. (2021, Feb.). Hydrogen insights: a perspective on hydrogen investment, market development and cost competitiveness. [Online]. Available: https://hydrogencouncil.com/wp-content/uploads/2021/02/Hydrogen-Insights-2021.pdf [Baidu Scholar]
S. Moss. (2019, Dec.). 2019 costs and technical parameter review, Aurecon Australasia Pty Ltd. Brisbane, Australia. [Online]. Available: https://www.aemo.com.au/-/media/Files/Electricity/NEM/Planning_and_ Forecasting/Inputs-Assumptions-Methodologies/2019/Aurecon-2019-Cost-and-Technical-Parameters-Review-Draft-Report.PDF [Baidu Scholar]
Australian Energy Market Operator. (2021, Dec.). Market data: NEMWEB. Australian Energy Market Operator, Melbourne, Australia. [Online]. Available: http://www.nemweb.com.au [Baidu Scholar]
Australian Government: Bureau of Meterology. (2021, Dec.). About one minute solar data. [Online]. Available: http://www.bom.gov.au/climate/data/oneminsolar/about-IDCJAC0022.shtml [Baidu Scholar]
SA Power Networks. (2021, Dec.). Resource library. [Online]. Available: https://www.sapowernetworks.com.au/resource-library [Baidu Scholar]
Electricity Connection for Retail Customers, National Electricity Rules 5.3A.12(a), 2022. [Baidu Scholar]
Australian Energy Market Commission. (2020, Dec.). Frequency control rule changes, Sydney, Australia. [Online]. Available: https://www.aemc.gov.au/sites/default/files/2020-12/Frequency%2 0control%20rule%20changes%20-%20Directions%20paper%20-%20December%202020.pdf [Baidu Scholar]
C. Christiansen and N. Hillmann. (2017, Jun.). Feasibility of fast frequency response obligations of new generators, AECOM Australia Pty Ltd. Sydney, Australia. [Online]. Available: https://www.aemc.gov.au/sites/default/files/content/661d5402-3ce5-4775-bb8a-9965f6d93a94/AECOM-Report-Feasibility-of-FFR-Obligations-of-New-Generators.pdf [Baidu Scholar]
M. Gatt. (2022, May). Power system security guidlines. Australian Energy Market Operator. Melbourne, Australia. [Online]. Available: https://www.aemo.com.au/-/media/Files/Electricity/NEM/Security_and_ Reliability/Power_System_Ops/Procedures/SO_OP_3715---Power-System-Security-Guidelines.pdf [Baidu Scholar]
Australian Energy Market Operator. (2019, Dec.). 2019 network support and control ancillary services (NSCAS) report, Australian Energy Market Operator, Melbourne, Australia. [Online]. Available: https://aemo.com.au/-/media/files/electricity/nem/planning_and_ forecasting/isp/2019/2019_nscas_report.pdf?la=en [Baidu Scholar]
M. Gatt. (2021, Feb.). SRAS guideline: system restart ancillary services incorporating boundaries of electrical sub networks, Australian Energy Market Operator, Melbourne, Australia. [Online]. Available: https://aemo.com.au/-/media/files/electricity/nem/security_and_ reliability/ancillary_services/sras/sras-guideline-2021.pdf?la=en [Baidu Scholar]
Australian Energy Market Operator. (2021, Feb.). Non-market ancillary services (NMAS) cost and quantity report 2019-2020, Australian Energy Market Operator. Melbourne, Australia. [Online]. Available: https://aemo.com.au/-/media/files/electricity/nem/data/ancillary_services /2021/nmas-cost-and-quantities-report-2019-20.pdf?la=en [Baidu Scholar]