Power to gas: addressing renewable curtailment by converting to hydrogen

Abstract

Renewable energy is the key to meeting increasing electricity demand and the decarburization targets in the generation mix. However, due to constrained power network capacity, a large volume of renewable generation is curtailed particularly from wind power, which is a huge waste of resources. There are typically three approaches to addressing excessive renewable: direct curtailment, the reinforcement of networks to expand transfer capacity, and the conversion of excessive renewable into other energy types, such as hydrogen, to transport. The costs and benefits of the three approaches could vary significantly across location, time, and penetration of renewable energy. This paper conducts a cost-benefit analysis and comparison of the three techniques to address wind curtailment. It uses a reduced 16-busbar UK transmission network to analyze the performance of the three approaches. The UK 2020 generation mix is used to quantify the saved renewable energy and incurred costs. The payback time and net present value of the two investment techniques are compared. From demonstration, it is reasonable to conclude that converting excessive wind power into hydrogen to transport is an environmentally friendly and cost-effective way to address wind curtailment.

Keywords

blending hydrogen / cost-benefit analysis / electrolysis / wind curtailment

Introduction

Due to the transportation constraints of traditional energy networks and increasing demand, it is necessary to increase renewable energy resources in the generation mix. For example, in 2011, wind curtailment in America was 2621.5 GWh, accounting for 8.5% of the generation capacity, in 2012, it was 1038 GWh, accounting for 3.7% of the total production, and in 2013, it was 479 GWh, accounting for 1.6% of the total production. In Germany, the total installed wind production was 29075 MW and 74 GWh of wind energy were curtailed from 2004 to 2006 [¹]. In the UK, the renewable energy curtailment was 483 GWh in 2013, costing more than £10 million [²]. However, it is impossible to deliver all renewable energy to end users because the transmission system might not have enough capability to transport it at the peak renewable output and there is a mismatch between the time of peak output from renewable generation and peak demand [³]. Thus, a huge amount of wind power was curtailed during wind output peak periods. Therefore, it is necessary to find appropriate alternative strategies to deliver renewable generation to end users.

Several papers have been published to address this wind curtailment. Luo et al. [⁴] have proposed different energy storage strategies, but energy storage can only store a small amount of energy, which is unable to resolve the curtailment issues efficiently at a large-scale. Wang and Redfern [⁵] have introduced an HVDC for linking wind farms and load centers in order to resolve curtailed wind power delivery, but the associated cost is extremely high. From the grid side, Lori Bird, and Wang [⁶] have put forward two main techniques which are system integration and forecasting improvement to improve transportation of renewable energy. The UK government [⁷] have proposed a formation of virtual power plants to remotely and automatically control electrical system. Zhang et al. [⁸] have discussed the application of electric boilers and pumped hydro to mitigate wind curtailment, where the benefits from reducing wind curtailment are also evaluated. Waite and Modi [⁹] have investigated the approach to extending network capacity by constructing new overhead lines to deliver the curtailed energy, but it takes a long time to extend the network system.

Another strategy for resolving renewable curtailment is to convert excessive renewable into other energy types, such as hydrogen through electrolysis [¹⁰–¹⁴]. Ursua et al. [¹⁰] have reviewed the hydrogen production through water electrolysis, surveyed the characteristics, advantages, disadvantages, and challenges of the existing electrolysis methods, and discussed water electrolysis integration with renewable energies. Rozendal et al. [¹¹] have used a biocatalyzed electrolysis to produce hydrogen from dissolved organic materials in wastewaters. Ehteshami et al. [¹⁴] have investigated the production of pure hydrogen from renewable sources by the Ethanol electrolysis process. There exist three different electrolyzers in the market: alkaline, polymer electrolyte membrane, and solid oxide electrolyzers. Alkaline water electrolysis is a mature technology, which is reliable and safe and exhibits long lifetime. Polymer electrolyte membrane is limited by its production capacity, short lifetime, and comparatively high investment cost. The solid oxide electrolyzer is obstructed by its long-term stability of the electrolysis cells. Therefore, the alkaline water electrolysis technology is selected in this paper to convert renewable into hydrogen, whose structure is demonstrated in Fig. 1.

Several studies have been conducted on the cost-benefit analysis for wind curtailment. Reference [¹⁵] has proposed an active management strategy to address wind curtailment with voltage control and reactive power compensation. The optimal use of proposed method can be determined to accommodate wind energy in the distribution system by cost-benefit analysis. The proposed method is more profitable than network reinforcement. Reference [¹⁶] has applied cost-benefit analysis to determine the loss and profits from wind curtailment and network investment. The active control system can be an alternative method to accommodate curtailed wind. Reference [¹⁷] has converted wind curtailment to hydrogen as an energy storage system and evaluated its economic performance by cost-benefit analysis.

Excessive wind energy can be used to convert water into hydrogen and oxygen by electrolyzers, where hydrogen can be delivered to the customers by blending into natural gas pipelines. Reference [¹⁸] has discussed the possibility of blending hydrogen into the natural gas supply. Reference [¹⁹], focusing on electrolyzer modeling, has studied the operation performance under different conditions. If the volume of hydrogen is below 17%, the natural gas system will be unaffected. However, if the percentage of hydrogen is increased, the natural pipelines and end-users will suffer from the negative effects.

However, these methods mainly focus on the technical aspect without conducting the extensive cost-benefit analysis. In the worst scenario, the cost of some techniques might overwhelm the savings, making them infeasible in practice for reducing renewable curtailment. Therefore, it is necessary to conduct a cost-benefit analysis to identify the cost-benefit equilibrium of various techniques, and is equally important to compare the net savings of different techniques in order to find the most cost-effective one for reducing renewable curtailment.

Therefore, this paper conducts a cost-benefit analysis of three different strategies to tackle wind curtailment reduction by investing in the power system (Technique 1), by directly curtailing wind power (Technique 2), and by converting excessive wind energy into hydrogen to transport by natural gas systems (Technique 3). First, wind farms and electrolyzer are modeled according to their physical characteristics. Key indexes to reflect the costs and benefits of the three techniques are defined, where the costs are technique investment and the benefits are renewable energy savings. The reduced UK transmission network is used to analyze the effectiveness of the three techniques. Based on extensive analysis, the required investment and potential renewable energy savings are quantified. Payback periods and net present value (NPV) of the three techniques are calculated to find the most economical technique. This paper is of importance because it proposes converting wind energy to hydrogen which is a novel method to address wind curtailment; it proposes cost-benefit analysis to evaluate the impact of three different methods to address wind curtailment; it determines the best decision under different unit costs of electrolyzer and network; and it quantifies the net present value and payback time of different technologies which are the key criteria for decision makers.

Modeling of key elements

Wind generation modeling

Figure 2 [²⁰] shows the variation of wind power output with respect to wind speed. The cut-in speed is the minimum wind speed to drive the wind turbine and the minimum speed which could reach the rated output power is called rated output wind speed. The wind speed which would pose a risk to damage the rotor is called cut-out speed. If the wind speed is between the cut-in speed and the rated one, the output of a typical wind turbine, i.e., the actual mechanical power P_w is determined as [²¹]

(1){{\displaystyle P}}_{\mathrm{w}}={\displaystyle \frac{\mathrm{1}}{\mathrm{2}}}\cdot \rho \cdot A\cdot {{\displaystyle v}}_u^{\mathrm{3}}\cdot {{\displaystyle C}}_p,

(2)A=\pi \cdot r^{\mathrm{2}},

where A is swept area, r is blade length, ρ is air density, C_p is power coefficient, and v_u is wind speed.

With a given wind speed, the energy output of wind farms can be determined by Eq. (1).

Electrolyzer efficiency and hydrogen production

This paper proposes a simplified way to represent the efficiency of an electrolyzer system by the ratio of curtailed energy and the energy in daily produced hydrogen, which is represented by

(3){{\displaystyle \eta }}_{\mathrm{c}\mathrm{a}\mathrm{s}\mathrm{e}}={\displaystyle \frac{{{\displaystyle W}}_{\mathrm{H}}}{{{\displaystyle W}}_{\mathrm{D}}}},

where W_H is the daily produced hydrogen and W_D is the daily energy curtailment, which can be derived from

(4a){{\displaystyle W}}_{\mathrm{H}}={\displaystyle {\sum }_{t=\mathrm{1}}^{\mathrm{24}}{{\displaystyle \mathrm{H}\mathrm{o}}}_t}\cdot {{\displaystyle C}}_{\mathrm{1}},

(4b){{\displaystyle W}}_{\mathrm{H}}={\displaystyle {\sum }_{t=\mathrm{1}}^{\mathrm{24}}{{\displaystyle W}}_{d,t}},

where Ho_t is the hydrogen output at time t, W_d,t is the wind curtailment at time t, C₁ is the coefficient between N∙m³ of hydrogen and delivered energy, in kWh.

In Eq. (4), C₁ is determined by the ideal gas law (Eq. (5)) and the equations of hydrogen combustion (Eqs. (6-9)).

(5)P\cdot V=n\cdot R\cdot T,

(6)\mathrm{2}{{\displaystyle \mathrm{H}}}_{\mathrm{2}}\mathrm{O}+{\mathrm{2}\mathrm{e}}^{-}\to {\mathrm{H}}_{\mathrm{2}}+\mathrm{2}{\mathrm{O}\mathrm{H}}^{-},

(7){\mathrm{2}\mathrm{O}\mathrm{H}}^{-}\to {\displaystyle \frac{\mathrm{1}}{\mathrm{2}}}{{\displaystyle \mathrm{O}}}_{\mathrm{2}}+{\mathrm{2}\mathrm{e}}^{-},(\Delta \mathit{H}=\mathrm{286}\mathrm{k}\mathrm{J}\cdot {\mathrm{m}\mathrm{o}\mathrm{l}}^{-})

(8)\mathrm{3600}\mathrm{k}\mathrm{J}=\mathrm{1}\mathrm{k}\mathrm{W}\mathrm{h},

(9)n=m/M,

where V is the daily hydrogen output in N∙m³; P = 1 atm, T = 273 K, M = 2 for hydrogen; DH is the energy from per amount of material during electrolysis, where energy is measured in joules, and the amount of material is measured in moles; n is quantity in mole; m is the mass in grams; R = 0.082 atm∙L∙mol^–1; and hydrogen is 2 g/mol.

Therefore, C₁ is represented as

(10){{\displaystyle C}}_{\mathrm{1}}={\displaystyle \frac{\Delta H\cdot P\cdot V}{\mathrm{3600}\cdot R\cdot T}}.

The annual amount of hydrogen production (P_H) is

(11){{\displaystyle P}}_{\mathrm{H}}={\displaystyle {\sum }_{t=\mathrm{1}}^{\mathrm{8760}}{{\displaystyle \mathrm{H}\mathrm{o}}}_t}\cdot {{\displaystyle C}}_{\mathrm{2}},

whereC₂ is the coefficient between N∙m³ and kg, {{\displaystyle C}}_{{{\displaystyle H}}_z} is the price of hydrogen, t is the hours for one year.

C₂ can be determined by

(12){{\displaystyle C}}_{\mathrm{2}}={\displaystyle \frac{P\cdot M}{R\cdot T}}.

Methodologies for reducing wind curtailment

This section introduces the methodology used to compare the cost and benefits among three methods to address the curtailment by delivering the curtailed wind by reinforcing the existing networks, by abandoning the curtailed wind directly, and by converting the curtailed wind to hydrogen and then transport it by gas pipelines.

The flowchart in Fig. 3 depicts the cost benefits analysis of the three methods to address wind curtailment, which are network expanding, direct curtailing, and converting curtailed wind to hydrogen. The three main steps are as follows.

The first step is to determine the network structures, and load and demand information, which are the input elements. Then, the power flow on each branch, especially the overloaded branched resulting from wind curtailment, can be determined;

The second step is to quantify techniques to address the system overloading from each branch. The overloading of the system is checked branch by branch (from branch 1 to branch N) to ensure the clearance of the overloading in the system. Then, the overloading from each branch will be converted to the requested network capacity, the curtailed wind amount, and the produced hydrogen amount based on the equations in Subsection 3.1;

The third step is to assess the investment cost and benefits of these three techniques by implementing cost-benefit analysis. Then, the comparison of the effectiveness of the techniques is processed to find the optimal solution for wind curtailment.

Network investment (Technique 1)

This technique tackles wind curtailment by constructing parallel overhead lines when a branch is overloaded. For simplicity, an identical branch is assumed to be invested, which has the same parameter as that of the overloaded one. The investment cost of the network (C_N) is quantified by the length, curtailment capacity, and unit cost of the overhead lines which is expressed as Eq. (13) in Ref. [²²].

(13){{\displaystyle C}}_{\mathrm{N}}={{\displaystyle V}}_{\mathrm{C}}\cdot L\cdot {{\displaystyle C}}_{\mathrm{u}},

where V_C is the capacity of the overloaded branch, L is the length of transmission branch, and C_u is the unit cost of the network.

Direct wind curtailment (Technique 2)

In this technique, curtailed wind energy is abandoned directly. By comparing the branch flow with transmission capacity, curtailed wind energy is quantified. By reducing the generations that is near to the overloaded branches, the wind curtailment on each bus will be resolved and the overloaded branches will be brought back to the normal condition. According to the electricity price of the year, the capital loss can be calculated (C_w) for the direct curtailment.

(14){{\displaystyle C}}_{\mathrm{W}}={{\displaystyle W}}_{\mathrm{C}}\cdot {{\displaystyle C}}_{\mathrm{e}},

(15){{\displaystyle W}}_{\mathrm{C}}={\displaystyle {\sum }_{t=\mathrm{1}}^{\mathrm{8760}}{{\displaystyle W}}_{d,t}},

whereW_C is the sum of curtailed wind power of the year and C_e is the electricity price.

Direct wind curtailment (Technique 3)

This technique uses electrolyzers to convert curtailed wind into hydrogen and then to be transmitted by natural gas pipelines. The capacity of the electrolyzer system is determined by the maximum curtailed wind power which is specified in the day with the heaviest system overloading. With the efficiency of the electrolyzer in Eq. (3), the investment (C_ec) of electrolyzers can be determined by

(16){{\displaystyle C}}_{\mathrm{e}\mathrm{c}}={{\displaystyle W}}_{\mathrm{C}}\cdot {{\displaystyle C}}_{\mathrm{e}\mathrm{u}},

whereC_eu is the unit cost of the elecrolyzer.

With the annual hydrogen production in Eq. (10), the profit of generated hydrogen (C_ei) is

(17){{\displaystyle C}}_{\mathrm{e}\mathrm{i}}={{\displaystyle W}}_{\mathrm{C}}\cdot {{\displaystyle C}}_{\mathrm{e}\mathrm{u}},

Payback period and net present value

With investment costs and profits of all techniques, the payback periods and NPV could be used to compare their profitability.

The payback time is defined as the time required to recover the cost of an investment.

(18)\mathrm{Payback\; period}={\displaystyle \frac{{{\displaystyle c}}_{\mathrm{0}}}{{{\displaystyle c}}_t}},

where C₀ is the total initial investment costs and C_t is the annual cash inflows.

NPV is the difference between the present value of cash inflows and the present value of cash outflows, which is expressed as Eq. (19) in Ref. [²³].

(19)\mathrm{N}\mathrm{P}\mathrm{V}={\displaystyle {\sum }_{t=\mathrm{1}}^T\frac{{{\displaystyle c}}_t}{{(\mathrm{1}+r)}^t}}-{{\displaystyle C}}_{\mathrm{0}},

where r is the discount rate and t is the years of lifespan.

Case study

The proposed approach is demonstrated on a reduced UK transmission network described as 16-zone, 15-boundary radial network in Fig. 4 [²²]. Each node represents a GB zone, and each branch represents a boundary. Buses 1 to 14 represent the stretch of the UK network from north to south. The parameters such as the resistance, the reactance, and the conductance of the transmission branches, are obtained from Digest of UK Energy Statistics (DUKES).

The installed generation capacity on different busbars in 2020 are listed in Table 1 [²⁴]. There are 12 buses which have a wind capacity of 29.5 GW in 2020 and the majority of the wind generations are located in the north of the UK. It can be observed that the total generation capacity is 91.1 GW, in which the traditional generation occupies 60%, wind occupies 32.4%. The generation maximum output is 30.03 MW and the maximum load is expected as 36.08 MW in 2020. Bus 15 has less generation and heavy load due to its location in the London area. Bus 13, located in the area around Norfolk and Triton, has the largest wind capacity of 4.5 GW. Gas-fired and wind generation account for more than 60% of the generation capacity in 2020. Compared with 2009, the penetration of renewable energy increases significantly and the use of coal declines dramatically [²⁴].

By selecting the wind speed on a random day as displayed in Fig. 5 [²⁵], the wind power at each busbar can be calculated. From the wind speed data, it can be observed that the wind speed from different buses is irregular. Buses 10 and 14 have the highest wind speed of around 12.5 m/s due to the offshore wind farms. The average wind speed at bus 16 is the lowest on this day which is 4.4 m/s. The wind power output is obtained by using the typical data from Ref. [²⁶]: the blade length (r) is 52 m, the air density (ρ) is 1.23 kg/m³, and the power coefficient (C_p)is 0.4. The typical cut-in speed is 3.5 m/s, the rated output wind speed is 14 m/s, and cut-out speed is 25 m/s.

The typical input data of the electrolyzer system is listed in Table 2 [²⁷]. The output of hydrogen is 1000 N∙m³∙h⁻¹ and the maximum delivery pressure is from 1.6 MPa to 3.2 MPa, where the power consumption for supporting the output is 4500 kWh∙(N∙m³) ⁻¹ at 380 V/220 V.

Results

This section provides the costs and benefits of the three techniques and compares their effectiveness.

Network investment (Technique 1)

Since identical branches are constructed to tackle overloads, the investment of expanding network is determined by the overloaded transmission branch under the worst overloading condition. By applying the wind speeds described in Fig. 5 to the network, nine branches are overloaded under the worst condition, during 1:00 to 8:00. The overloading is listed in Table 3. The heavy overloading occurs in TB8, TB10, and TB13 because the majority of wind farms are located in the north while the majority of the demand is in the south. TB8 has the harvest overloading of 6072 MW, which is almost blocked because of the small transmission capacity of this branch and the large wind output at busbar 9. The heavy congestion on TB 9 and TB 13 is caused by the fact that they connect the harvest loading areas, with a loading level 16 MW, which contains busbars 14, 15, and 16. Although TB 13 has the largest capacity of 11724 MW, the overloading is still 5028.8 MW on this branch. Combining with Table 1, the transmission branches which are close to buses 8 and 11 have heavy overloading, which is caused by the fact that there is no generation on these two busbars.

Therefore, the sum of the overloaded capacity for transmission branches is 26.3 GW and the total length of overhead lines is 1442 km. To accommodate these overloading resulting from wind penetration, by constructing new branches, a branch capacity of 55.2 GW is required. Based on the unit cost [²²], the capital investment (C₀) of the transmission network is calculated as £10.6 billion. The investment cost to mitigate the congestions is shown in Fig. 6. Although approximately a capacity of 10 times is required for TB8, this is not the most expensive investment for curtailment mitigation. The highest investment cost is from TB13, which is £3.6 billion, due to the large capacity of the new branch.

Wind curtailment (Technique 2)

Figure 7 shows the sum of all overloading and the wind curtailment for the whole system over a heavy overloading day. The curtailment curve represents the abandoned wind generation in time-series to mitigate the amount of overloading. The wind generation only mitigates a part of overloading during 11:00 to 15:00, which means this period has a heavy load and another type of generators should be responsible for it. During the curtailment period, the peak curtailment level of the system is reduced from 26.3 GW to 10.3 GW, most which (about 3.5 GW) is curtailed by the wind generation during this period. During the valley period, from 5:00 to 11:00, the curtailment is reduced from 4.6 GW to 2.2 GW. From 20:00 to 23:00, the wind curtailment is only reduced from 6.1 GW to 4.3 GW.

By accumulating the hourly curtailed wind capacity, the daily power curtailment is 135.13 GWh. Therefore, the sum of curtailed power for 2020 would be 49 TWh by assuming that this daily curtailment data are an average for the whole year. Since the electricity price in the UK is £50/MWh [²⁸], the capital profit of the whole system is around £7.4billion.

Electrolysis (Technique3)

There are two main types of cost in converting wind curtailment into hydrogen, which are electrolyzer systems and pipeline costs. Here, it is assumed that the existing natural gas pipelines can accommodate the hydrogen without any updating costs needed. Since the price for the sample electrolyzer unit is around 1400 £/kWh [²⁷], the total investment is approximately £14.35billion.

The efficiency of electrolyzers can be deduced by Eq. (3), which is roughly around 79%. Combining Eq. (11), the annuity hydrogen production is 2687 t which equals to 106.75 GWh. Therefore, with the predicted hydrogen price of5 £/kg in 2020 [²⁹], the profit of the whole hydrogen system is around £4.58 billion/a.

Cost-benefit analysis of three techniques

Since the curtailed energy is worth £2.45 billion annually which is the energy loss in Technique 2. For Technique 1, the cost of investing network is £10.6 billion and the investment will be broken even by 18.37 years. In terms of NPV, it is assumed that the lifespan of electrolyzers is 20 years [³⁰] and the discount rate is 2.5% [³¹], NPV is £41.5 billion. For Technique 3, the profit of hydrogen technique is around £4.58 billion and the investment is approximately £14.35 billion. Therefore, the payback period for hydrogen is 3.13 years. Since the lifespan is 20 years and the discount rate is 3.5% [³¹], NPV is £52.45 billion. The results are summarized in Table 4.

Considering the uncertainty of the unit price of the electrolyzer and the unit price of the network, sensitive analysis can be implemented to mitigate the uncertainty and find the optimal solution. Figure 8 shows that the NPVs of both techniques are declining with the decrease of unit cost. The NPV of the hydrogen project decreases from £35.9 billion to £32.8 billion and the NPV of network investment is reduced from £79.3 billion to £67.9 billion with the unit cost changing form £400/kW to £100/kW for the electrolyzer and £400/(MW∙km), to £100/(MW∙km), for the network.

The dotted line in Fig. 9 represents the fact that the NPV difference is zero, which means two techniques have no economic difference in practice. At the right of the dotted line, since the unit cost of electrolyzer is smaller than £420/kW and the unit cost of the network is bigger than £400/(MW∙km), the NPV for Technique 3 is higher which means it can generate more profits.

Conclusions

This paper has compared three techniques in addressing wind energy curtailment and conducted the cost-benefit analysis. The following are some key findings from the demonstration.

The capital loss of wind energy is considerably high. The cost to extend electricity network is higher than that in converting curtailed energy into hydrogen. Hydrogen technique has the advantage in payback time. Because the net present value (NPV) of hydrogen project is higher than network investment, hydrogen project is more valuable to invest.

To summarize, blending hydrogen from renewable is an effective way to address energy curtailment. This method is not only environmentally friendly but also profitable. In future work, the uncertainties in wind power prediction and the parameters of electrolyzers will be analyzed. In addition, the stochastic flow analysis will also be considered in wind power uncertainties.

References

[1] Jennifer Rogers S F, Porter K. Examples of wind energy curtailment practices. National Renewable Energy Laboratory, 2010

[2] National Grid. Winter Outlook 2014/15. National Grid, 2014

[3] Lew D, Bird L, Milligan M, Wind and solar curtailment. The National Renewable Energy Laboratory (NREL), 2013

[4] Luo X, Wang J, Dooner M, Clarke J. Overview of current development in electrical energy storage technologies and the application potential in power system operation. Applied Energy, 2015, 137(C): 511–536

[5] Wang H L, Redfern M A. The advantages and disadvantages of using HVDC to interconnect AC networks. In: 45th International Universities Power Engineering Conference. Cardiff, Wales, UK, 2010

[6] Lori Bird J C, Wang X. Wind and solar energy curtailment: experience and practices in the United States. 2017–10,

[7] GOV.UK. Wind imbalance in UK Technical solutions for mitigation. 2017–12,

[8] Zhang N, Lu X, McElroy M B, Reducing curtailment of wind electricity in China by employing electric boilers for heat and pumped hydro for energy storage. Applied Energy, 2016, 184: 987–994

[9] Waite M, Modi V. Modeling wind power curtailment with increased capacity in a regional electricity grid supplying a dense urban demand. Applied Energy, 2016, 183: 299–317

[10] Ursua A, Gandia L M, Sanchis P. Hydrogen production from water electrolysis: current status and future trends. Proceedings of the IEEE, 2012, 100(2): 410–426

[11] Rozendal R A, Hamelers H V, Euverink G J, Metz S J, Buisman C J. Principle and perspectives of hydrogen production through biocatalyzed electrolysis. International Journal of Hydrogen Energy, 2006, 31(12): 1632–1640

[12] Kavadias K, Apostolou D, Kaldellis J. Modelling and optimisation of a hydrogen-based energy storage system in an autonomous electrical network. Applied Energy, 2017, 227: 574–586

[13] Cai W, Zhang Z, Ren G, Quorum sensing alters the microbial community of electrode-respiring bacteria and hydrogen scavengers toward improving hydrogen yield in microbial electrolysis cells. Applied Energy, 2016, 183: 1133–1141

[14] Ehteshami S M M, Vignesh S, Rasheed R K A, Chan S H. Numerical investigations on ethanol electrolysis for production of pure hydrogen from renewable sources. Applied Energy, 2016, 170: 388–393

[15] Nursebo S, Chen P, Carlson O, Tjernberg L B. Optimizing wind power hosting capacity of distribution systems using cost benefit analysis. IEEE Transactions on Power Delivery, 2014, 29(3): 1436–1445

[16] Hu Z, Li F. Cost-benefit analyses of active distribution network management, part II: investment reduction analysis. IEEE Transactions on Smart Grid, 2012, 3(3): 1075–1081

[17] Parissis O S, Zoulias E, Stamatakis E, Integration of wind and hydrogen technologies in the power system of Corvo island, Azores: a cost-benefit analysis. International Journal of Hydrogen Energy, 2011, 36(13): 8143–8151

[18] Schouten J A, Michels J P J, Janssen R. Mixtures of hydrogen and natural gas: thermodynamic and transportation properties. 2017–10,

[19] Sarrias-Mena R, Fernández-Ramírez L M, García-Vázquez C A, Jurado F. Electrolyzer models for hydrogen production from wind energy systems. International Journal of Hydrogen Energy, 2015, 40(7): 2927–2938

[20] Windpower Program. Wind turbine power output variation with steady wind speed. 2017–10,

[21] Manyonge A W, Ochieng R M, Onyango F N, Shichikha J M. Mathematical modelling of wind turbine in a wind energy conversion system: power coefficient analysis. Applied Mathematical Sciences, 2012, 6(6): 4527–4536

[22] Chaudry M, Jenkins N, Qadrdan M, Wu J. Combined gas and electricity network expansion planning. Applied Energy, 2014, 113(6): 1171–1187

[23] Investopedia. Net present value—NPV. 2017–11,

[24] Qadrdan M. Modelling of an integrated gas and electricity network with significant wind capacity. Dissertation for the Doctoral Degree. Welsh: Cardiff University, 2012

[25] MSW.Weather Stations. 2017–10,

[26] Npower. Wind turbine power calculations. 2017–10,

[27] KAPSOM. Water electrolysis hydrogen generator. 2017–11,

[28] NERA Economic Consulting. Energy supply margins: commentary on Ofgem’s SMI. 2017–09,

[29] ECUITY. A vision for the UK hydrogen economy. 2018–07,

[30] Ogden J M. Cost and performance sensitivity studies for solar photovoltaic/ electrolytic hydrogen systems. Solar Cells, 1991, 30(1–4): 515–523

[31] GOV.UK. Social impact bonds: discount rates and net present value. 2017–10,

RIGHTS & PERMISSIONS

Higher Education Press and Springer-Verlag GmbH Germany, part of Springer Nature