Jenkins N., Strbac G., Ekanayake J. Distributed Generation

Подождите немного. Документ загружается.

4 x 0.4 MW = 1.6 MW

(4 x 0.1 MW = 0.4 MW)

50 MW

(12.5 MW)

10 MW

(2.5 MW)

33 kV

11 kV

CHP

5MW

(15 MW)

Wind farm

0.2 MW

(

1MW

)

132 kV

4 x 0.4 MW = 1.6 MW

(4 x 0.1 MW = 0.4 MW)

Figure 6.5 132/33 kV radial distribution system

A distribution network is typically designed to cope with the expected max-

imum loading condition, which is likely to occur at a time of maximum demand

with minimum local generat ion. With distributed generation, another extreme

condition needs to be considered, i.e. the condition where the generators produce

maximum output and demand is minimum. Therefore, in this example, these two

critical loading conditions are considered. The first number (without brackets) is

the loading or generation during maximum demand–minimum generation (this

would typically be associated with winter daytime), and the second number (in

brackets) is the loading or generation during minimum demand–maximum gen-

eration (this would typically be associated with summer night time).

The design of the distribution network should take into account the contribu-

tion of distributed generation to network capacity. This was discussed in Chapter 5.

For the sake of this illustrative example, an effective contribution of 5 MW (30% of

installed capacity) is allocated to the CHP generator, and 200 kW (20% of installed

capacity) to the wind turbine. In the context of DUoS pricing, it could be inter-

preted simply that the CHP and wind turbine are capable of substituting 5 MW and

200 kW respectively, of a distribution circuit capacity.

Minimum demand is assumed to be 25% of the maximum demand.

Pricing of distribution networks with distributed generation 157

The flows in both loading conditions can be obtained by simple inspection.

Critical flows are summarised in Figure 6.6. The arrows show the direction of the

flows. The critical loading of items of plant can be seen to be the largest power

flows of the two loading periods.

It can be observed that critical loading for the 11 kV feeder, 33 kV/11 kV and

132 kV/33 kV transformers is driven by maximum demand, while critical loading of

the 33 kV circuits is driven by maximum generation, and these occur at different

periods (time of use). The assets whose capacity is driven by maximum demand–

minimum generation condition (winter daytime) such as 11 kV feeders, 33 kV/11 kV

1.6 MW

(0.4 MW)

1.6 MW

(0.4 MW)

50 MW

(12.5 MW)

10 MW

(2.5 MW)

33 kV

11 kV

CHP

5MW

(15 MW)

Wind farm

0.2 MW

(

1MW

)

132 kV

Critical flows

58 MW

12.7 MW

13 MW

3MW

Min. demand

Max. Generation

Max. demand

Min. Generation

58 MW

8MW

13 MW

3MW

0.2 MW

12.7 MW

2.3 MW

0.6 MW

Figure 6.6 Computation of critical flows

Note that the critical flows determine the reference (optimal) ratings of the associated plant. The

reference rating of 11 kV and 33 kV circuits are 2  3 MW and 2  12.7 MW respectively. Due to the

topology of 11 kV circuits, one feeder must cope with all 11 kV loads when one of the 11 kV feeders

loses supply from the 33 kV/11 kV substation and the normally open point is closed. The optimal rating

of the 132 kV/33 kV substation is 2  58 MW. These reference ratings of the individual network

components (transformers and lines at various voltage levels) can be compared with the plant ratings of

the existing network.

158 Distributed generation

and 132 kV/33 kV transformers, are then classified as demand dominated (DD),

while the capacity of 33 kV feeders are driven by minimum demand–maximum

generation condition (summer night time) and are hence classified as generator

dominated (GD).

Given the direction of the critical flows and knowing the direction of demand-

and generation-driven flows (Figure 6.6), it is easy to see how demand and gen-

eration at various voltage levels will pay or get paid for the use of individual

network circuits. For example, an incremental increase in load of the demand

connected to an 11 kV feeder, durin g the maximum demand periods, will increase

the loading on the 11 kV feeder, 33 kV/11 kV and 132 kV/33 kV transformers.

Therefore, this demand will be charged for the use of these circuits and the total

charge will be based on maximum demand of 3.2 MW.

For the generation-dominated 33 kV circuit, the relevant critical period is

determined by the coincidence of maximum generation and minimum demand.

Hence, demand connected at 11 kV will be rewarded for the use of this 33 kV

circuit, based on the load during minimum demand of 0.8 MW and the corre-

sponding reward to demand will be obtained during the summer night periods.

Consider now charges for the wind farm. The wind farm will be rewarded for

the use of the 11 kV network and 33 kV/11 kV and 132 kV/33 kV transformers and

will be charged for the use of 33 kV circuits. The rewards for using these circuits

will be based on the generator effective contribution to network capacity, i.e. 0.2

MW and apply during winter daytime. On the other hand, the charges for the use of

33 kV circuit will be based on the maximum generation output (1 MW) and applied

during summer night periods.

In order to evaluate network charges for individual users, per unit annuitised

capacity costs (£/kW/year) are allocated to each item of plant in the network. For

illustrative purposes, the estimate annuitised typical costs of 132 kV circuits, 132

kV/33 kV transformers, 33 kV circuits and 33 kV/11 kV transformers for typical

urban network in the United Kingdom are used.

The system is presented again in Figure 6.7 with all critical loadings

highlighted.

The typical annuitised costs of individual circuits are shown next to the net-

work model in Figure 6.7. DUoS exit charges for demand customers connected at

various points in the network are also listed. The polarity of charges is adopted to

be positive for downstream and negative for upstream power flows respectively.

Consider now the 132 kV/33 kV transformer. This is demand-dominated plant

since the direction of the power flow is downstream. Hence, all downstream

demand and generation customers pay and are paid 5.2 £/kW/year respectively for

the use of this particular plant during maximum demand conditions, while charges

are zero during the minimum demand period.

The next plant to be considered is the 33 kV circuit. This is a generation-

dominated plant since the direction of the critical power flow is upstream. Hence,

all downstream generation and demand customers pay and are paid 6.7 £/kW/year

respectively for the use of this plant during maximum generation condition, while

zero is charged during the maximum demand period.

Pricing of distribution networks with distributed generation 159

As shown in Figure 6.7, the total DUoS exit charges for demand customers

connected to the 33 kV busbar of the 33 kV/11 kV transformer is 5.2 £/k W/year

applied during the maximum demand period (5.2 £/kW/year for the use of the

132 kV/33 kV transformer and 0 £/kW/year for the use of the 33 kV circuit), and

DUoS entry charges of 6.7 £/kW/year during minimum demand period (0 £/kW/year

for the use of the 132 kV/33 kV transformer and 6.7 £/kW/year for the use of the

33 kV circuit use).

The 33 kV/11 kV transformer is demand-dominated plant since the direction of

the critical power flow is downstream. Hence, all downstream demand customers

are charged and all downstream generation customers are paid 4.3 £/kW/year for

the use of this particular plant during maximum demand conditions, while the

charges are zero during the minimum demand period.

Therefore, the total entry charges for the generation connected to the 11 kV busbar

of the 33 kV/11 kV transformer are 9.5 £/kW/year during maximum demand period

(5.2 £/kW/year for the use of the 132 kV/33 kV transformer, 0 £/kW/year for the use

of the 33 kV circuit and 4.3 £/kW/year for the use of the 33 kV/11 kV transformer,

and 6.7 £/kW/year during minimum demand period (0 £/kW/year for the use of the

132 kV/33 kV transformer) 6.7 £/kW/year for the use of the 33 kV circuit and 0 £/kW/

year for the use of the 33 kV/11 kV transformer).

Max. demand

Min. generation

Max. demand

Min. generation

Yardstick

0 £/kW

5.2 £/kW

0 £/kW

⫺6.7 £/kW

9.5 £/kW

⫺6.7 £/kW20.5 £/kW

5.2 £/kW

6.7 £/kW

4.3 £/kW

11 £/kW

50 MW

10 MW

33 kV

11 kV

CHP

5MW

(15 MW)

Wind farm

0.2 MW

(

1MW

)

132 kV

58 MW

12.7 MW

13 MW

3MW

8 x 0.4 =

3.2 MW

Figure 6.7 Evaluation of DUoS exit charges

160 Distributed generation

Finally, the 11 kV feeder is demand-dominated plant since the direction of the

critical power flow is downstream. Hence, all downstream generation customers

are paid 11 £/kW/year for the use of this particular plant during maximum demand

conditions, while the charge is zero during the minimum demand period. Therefore,

the total charge for generation customers connected to the 11 kV circuit is 20.5

£/kW/year during the on-peak period (5.2 £/kW/year for the use of the 132 kV/33

kV transformer, 0 £/kW/year for the use of the 33 kV circuit, 4.3 £/kW/year for

the use of the 33 kV/11 kV transformer and 11 £/kW/year for the use of 11 kV

circuit) and DUoS entry charges of 6.7 £/kW/year during minimum demand period

(0 £/kW/year for the use of the 132 kV/33 kV transformer, 6.7 £/kW/year for the

use of the 33 kV circuit, 0 £/kW/year for the use of the 33 kV/11 kV and zero for

transformer and for the use of 11 kV circuit).

The DUoS charges (assuming positive polarity for demand customers) and

revenues collected from various users during peak demand and off-peak demand

conditions are given in Tables 6.7 and 6.8. The connection point G corresponds to

the balancing point. Note that 58 MW is imported under peak demand conditions

while 0.2 MW is exported under minimum demand condition from the grid supply

point (point G).

During the peak-load condition, the annual revenue is collected for all demand-

dominated assets, while for the generation -dominated plant revenue is recovered

Table 6.7 On-peak demand DUoS prices and revenues from demand and

generation customer

Connection

point

Price

(£/kW)

Demand

(MW)

Generation

(MW)

R demand

(£)

R generation

(£)

Total

(£)

G0058 000

F 5.2 50 0 260000 0 260000

E 5.2 0 5 0 26000 26000

D 9.5 10 0 95000 0 95000

C 20.5 3.2 0.2 65600 4100 61500

Total 420600 30100 390500

Table 6.8 Off-peak demand (peak generation) DUoS prices and revenues from

demand and generation customers

Connection

point

Price

(£/kW)

Demand

(MW)

Generation

(MW)

R demand

(£)

R generation

(£)

Total

(£)

G 0 0 0.2 0 0 0

F 0 12.5 0 0 0 0

E 6.7 0 15 0 100500 100500

D 6.7 2.5 0 16750 0 16750

C 6.7 0.8 1 5360 6700 1340

Total 22110 107200 85090

Pricing of distribution networks with distributed generation 161

during peak generation periods. The costs of the individual plant items for the

reference rating are given in Table 6.9.

In this particular case, the total annual revenue received for the demand-

dominated plant is £390500/year, as shown in Table 6.7. (This is exactly equal to

the total costs of the individual plant items as shown in Table 6.9, i.e. £390500 =

£301600 þ 55900 þ 33000.) On the other hand, the total annual revenue received

from DUoS charges during the off-peak demand period is £85090, as shown in

Table 6.8. This is exactly equal to the total cost of generation-dominated circuit.

The on- and off-peak demand DUoS related expenditure of individual users is

presented in Table 6.10. The total annual DUoS revenue equals the total annuitised

cost of the reference network.

References

1. Curcic S., Strbac G., Zhang X.-P. ‘Effect of losses in design of distribution

circuits’. Generation, Transmission and Distribution, IEE Proceedings.

2001;148(4):343–349.

2. Boiteux M. ‘La tarification des demandes en pointe: applicationde la

theorir de la vente au cout marginal’. Revue General de Electricite. 1949;

58:321–340.

3. Farmer E.D., Cory B.J., Perera B.L.P.P. ‘Optimal pricing of transmission and

distribution services in electricity supply’. Generation, Transmission and

Distribution, IEE Proceedings. 1995;142(1):1–8.

Table 6.9 Annuitised cost of individual plant items

Plant Unit cost (£/kW) Max. flow (MW) Cost (£)

Transformer 132 kV/33 kV 5.2 58 301600

Circuit 33 kV 6.7 12.7 85090

Transformer 33 kV/11 kV 4.3 13 55900

Circuit 11 kV 11 3 33000

Total 475590

Table 6.10 Annual DUoS charges for individual network users

User On-peak charge (£) Off-peak charge (£) Total charge (£)

Demand at F 260000 0 260000

Generator at E 26000 100500 74500

Demand at D 95000 16750 78250

Demand at C 65000 5360 60240

Generator at C 4100 6700 2600

Total 475590

162 Distributed generation

4. Nelson J.R. Marginal Cost Pricing in Practice. Prentice-Hall; 1967.

5. Strbac G., Allan R.N. ‘Performance regulation of distribution systems using

reference networks’. Power Engineering Journal. 2001;15(6):295–303.

6. Mutale J., Jayantilal A., Strbac G. ‘Framework for allocation of loss and

security driven network capital costs in distribution systems’. IEEE Power-

Tech International Conference on Electric Power Engineering; Budapest 29

Aug–2 Sept 1999.

7. Mutale J., et al. A Framework for Development of Tariffs for Distribution

Systems with Embedded Generation CIRED’99. 1999; NICE, France.

8. Mutale J., Strbac G. Business Models in a World Characterised by Dis-

tributed Generation. EC funded project number NNE5/2001/256 (April 2002

to March 2004).

Pricing of distribution networks with distributed generation 163

Chapter 7

Distributed generation and future network

architectures

Rhyl Flats Offshore Wind Farm (90 MW)

Situated five miles off the North Wales coast, in Liverpool Bay [RWE npower

renewables]

7.1 Introduction

Distributed generation will continue to increase in importance as many countries

move towards de-carbonising their energy systems. The location and size of many

renewable energy sources and CHP plants require that they are connected to dis-

tribution networks and that effective use is made of existing circuits. This can only

be done cost-effectively by integrating the operation of the distributed generation

closely into that of the power system and changing the operation of distribution

networks from passive to active [1–5].

Traditionally, power networks have been supplied from large, rotating syn-

chronous generators that act as constant frequency and voltage sources. Hence, the

entire operating philosophy of the power system has been developed around

maintaining nearly constant voltage and, in the event of a short circuit, providing

fault current to operate protective relays. A rather small number of these large

rotating generators (some hundred units in Great Bri tain) are fitted with controls to

maintain frequency, voltages and reactive power flows, ensure stability and provide

damping of the power system.

Much distributed generation, either static or rotating, is connected to the net-

work through power electronic converters via a DC link. In their simplest form, the

power electronic converters simply inject real power at unity power factor into the

network to maximise the output of the generator. They only provide close to load

current into a network fault and the DC link decouples the generators from the

network and so removes the effect of their spinning inertia.

The connection of generators via power electronic converters allows much greater

flexibility and controllability than if a direct connection of a generator to the network is

made. Hence these new forms of distributed generation, with their different operating

characteristics, offer the possibility of new generator control philosophies as well as

new system operating practices. The historical investment in the traditional power

system is so great that its basic operating philosophy of constant voltages and fre-

quency is most unlikely to change quickly, but already there is interest in using vari-

able speed generators to provide injections of power at times of system frequency

excursions using their stored kinetic energy as well as increased system damping [6].

The present limited penetration of distributed generation, operated to inject

power with little consideration of the state of the power system, is already causing

restrictions to the amount of renewable energy and CHP that may be connected. It is

also leading to difficulties in the generation systems and transmission networks in

some European countries, e.g. at times of high wind power and low load or during

network disturbances. Therefore the present fit-and-forget philosophy, where dis-

tributed generators are viewed as negative loads and the distribution system is

operated in the traditiona l manner, will be superseded by active integration of dis-

tributed generation through active network management. This will result in a blur-

ring of the traditional distinction between transmission and distribution networks as

distributed generation is controlled for the benefit of the power system by a dis-

tribution system operator. This gives particular challenges in Great Britain with its

clear regulatory distinction between the suppliers of electrical energy and distribu-

tion network operators. The most effective way to coordinate and control a very large

number of small generators and loads (perhaps up to 10

units) has yet to be deter-

mined although their aggregation into virtual power plants has been suggested.

The attractions of distributed generation to enable individuals to engage more

closely with their energy supply should not be overlooked [7]. This is a rather

fundamental question and may well become more pressing as climate change and

energy security assume greater significance. It has similarities to the debate as to

whether it is better for individual computers to hold their own software or to use it

from a central source as required.

166 Distributed generation