[IEEE 2014 16th International Conference on Harmonics and Quality of Power (ICHQP) - Bucharest, Romania (2014.5.25-2014.5.28)] 2014 16th International Conference on Harmonics and Quality

978-1-4673-6487-4/14/$31.00 ©2014 IEEE

AC vs. HVDC Back to Back Interconnection cost benefit analysis

Oana Udrea, Gabriela Ungureanu CNTEE Transelectrica SA

Network Modelling Department Bucharest, Romania

[email protected]

George C. Lazaroiu, Mihnea Costoiu University Politehnica of Bucharest

Bucharest, Romania

Abstract — In this article was examined the technical and economic potential of connecting the high voltage electricity networks of three countries with a new high voltage transmission line. The analysis investigated options and technical recommendations for the optimal configuration a 750 kV transmission corridor. In doing so, it was recognized reconnection of the corridor holds the promise of improving energy security and increasing electricity trade between them by providing a new connection between the synchronous zone and the continental region of Europe. It was calculated the maximum regional generation cost savings resulting from the exchange of electricity via a new interconnecting line as described in each scenario.

Keywords — AC, Back to back, cost-benefit, OPF, PQ, wind farms.

I. INTRODUCTION

The investment scenarios are comprised of the following parameters:

• Voltage – 1x750kV, 1x400 kV, and 2x400 kV wereconsidered;

• Interconnection Mode – Synchronous orasynchronous interconnections via a high-voltage-back-to-back substation (HVDC B2B);

For all 750 kV investment scenarios, the economic indicators are not favorable. Although the power exchange levels afforded by the 750kV options are higher than the 400 kV options, the high cost of the 750 kV investments reduce the economic benefits to below that of all 400 kV options. In all cases, the 400 kV investment scenarios report higher economic benefit than any of the 750 kV scenarios.

The economic indicators reveal that connection through a HVDC B2B is superior to an AC synchronous connection in all investment scenarios. This is a result of HVDC technology reducing network constraints, allowing for greater power exchange levels in the corridor when the generation fleet is optimized. Even at lower levels of power exchange, the

savings to the region accruing from optimization are considerably higher with the HVDC connection than when compared to AC connection scenarios. Furthermore, HVDC B2B is the only solution providing system optimization savings when network constraints are at their maximum.

Directing the output of one of the generation units from one country to another via a 400 kV synchronous connection was considered as a potential intermediate step toward establishment transmission corridor. This would enable early electricity trade to commence between the countries prior to completion of the full corridor. Such a connection would provide with direct export/import capability to Europe, while potentially lowering total generation costs in the first country. The technical and economic benefits this scenario offers make it an attractive option for establishing the corridor in the region.

II. INPUT DATA ASSUMPTIONS

There were identified four scenarios in this paper from a total of 108 analyzed. Scenarios 22 and 34 are summarized in the following tables. Each scenario was analyzed under the assumption that a merchant line would be constructed and financed with a five year loan at 7% interest. In addition, the scenarios were analyzed assuming the line ownership and financing through a 20 year concessional loan at 4% interest.

TABLE I. SCENARIO 1 BASE CASE

Investment Scenario 22 Base Case (wind generation level in Country3 is 30%)

Corridor Country1 - Contry2 – Country3; 1x400kV AC line Substations Substation1 (Country1), Substation2 (Country2),

Substation3 (Country3) HVDC B2B 2x300 MW at Country2 Substation Line length 285 km

Inv Cost 388.845.000 USD Annual savings 44.383.621,18 USD

Financial assumptions

Interest rate: 7% Loan period: 5 years

Discount rate: 7%


Discount rate: 7%

72

Life time: 30 years Life time: 30 years B/C Ratio 1,13 1,12

NPV 61,5 Million USD 57,4 Million USD

This scenario examined the technical and economic benefits of establishing the all corridor between the Substation1 – Country1 and the Substation2 – Country 2 substation via a 400 kV AC line and a 2 x 300 MW high voltage back-to-back substation located at the Country2. The total length of this option is estimated to be 285 km. The sub-regional line construction investment cost is estimated at $389 million with the annual reduction in cost of generation an estimated $44.3 million over a 30 year life cycle of the project. Under the merchant line financing scenario, the Benefit to Cost ratio is 1.13. Under the TSO ownership with concessional financing scenario, the Benefit to Cost ratio is 1.12.

TABLE II. SCENARIO 3 BASE CASE

Investment Scenario 34 Base Case (wind generation level in Country3 is 30%)

Corridor Country1 - Contry2 – Country3; 1x400kV AC line Substations Substation1 (Country1), Substation2 (Country2),

Substation3 (Country3) HVDC B2B 2x300 MW at Substation3 (Country3) Line length 285 km

Inv Cost 388.845.000 USD Annual savings 43.952.664,95 USD

Financial assumptions


Discount rate: 7% Life time: 30 years


Discount rate: 7% Life time: 30 years

B/C Ratio 1,11 1,11 NPV 56,1 Million USD 52,1 Million USD

This scenario examined the technical and economic benefits of establishing the corridor between the Substation1 – Country1 and Substation3 – Country3 via a 400 kV AC line and 2 x 300 MW high voltage back – to – back substation in Country3.

The total length of this option is estimated to be 285 km. The line construction investment cost is estimated to be $389 million with the annual reduction in cost of generation estimated to be $44 million over the 30 year life cycle of the project.

III. COST – BENEFIT ANALISYS A. Wind farms in Country3

The base case study assumes that 30% of the installed wind power capacity in this country will be generated on a must run basis. As a sensitivity case, the calculations also looked at a case when 70% of the currently installed wind capacity is generated as must run (total installed capacity of the wind generators is increased such that their generation level under 30% capacity factor corresponds to 70% capacity factor of the installed capacity). This case significantly reduced calculated savings in the region to the extent that no scenario was economically feasible.

B. Wind Power Plant Generation Level in Country3 (Optimal Power Flow - OPF vs. 30% vs. 70%) To assess the effect of increasing wind generation in

Country3 from 30% capacity factor to the case of 70% capacity factor, the following scenarios were analyzed:

• 1x750 kV AC connection between Country1-Country3;

• Loading hour: Summer max; and

• Case_ VC-I_W30% vs. Case_ VC-I_W70%.

This investment scenario was the most dramatic, in the sense of power exchange reversal (north → south to south → north), and therefore preferred for discussion. The effects of wind generation levels on the total regional generation cost and optimum power exchanged are illustrated in Figure 1, for this investment scenario of 1x750 kV AC connection between Country1-Country3, as an example. In the 30% capacity factor must run wind case, the optimum power exchanged in terms of regional savings occurred when Country1 and Country2 exported 200 MW to Country3 and the total cost of electricity in the sub-region was approximately $1.404 million for the measured hour.

As seen in the figure, higher levels (in the 70% capacity factor case) of must-run wind generation in Country3 narrowed the room for the OPF algorithm to optimize the transmission corridor generation fleet. The total cost of generation in the sub-region at the optimum power exchange increased to $1.405 in this loading scenario and the direction of the power flow reversed to a south to north direction (Country3 exports to Country1 and Country2). Country3 then exported 450 MW to Country1 and Country2. The effect of wind generation increase on the total generation cost in Country3 was also deductible in Figure 1 (see the vertical shift in zero exchange cost).

Figure 1. An example of effect of increase in wind generation on optimum

exchange amount (Case_ VC-I_W30% vs. Case_ VC-I_W70%)

When the wind generation level in Country3 was increased to the 70% case, the OPF algorithm forced the high generation cost power plants in Country3 to reduce their generation. This continued until the reduction of generation in Country1 and Country2 became more cost effective than reducing power

73

generation in Country3. From this point on, Country3 started to export power to the north (see Figure 1-high wind scenario).

It is important to note that the OPF algorithm matched generation and demand while minimizing the total cost of generation. Because demand was fixed, the OPF algorithm adjusted the dispatch level of generators based on their production costs. The least efficient, most costly generators in the dispatch queue were the last to be dispatched and the first to have their generation levels reduced when demand decreased, more efficient generation was introduced to the queue, or when generation units were designated as must run for security of supply reasons or policy reasons.

In the 70% capacity factor case considered in this study, wind generation in the south region of Country3 was fixed at over 2,200 MW. This means the OPF algorithm was not permitted to optimize this generation in the queue because it was must run and was therefore dispatched in the first position. This 2,200 MW displaced more cost effective generation in the queue. As a result, the OPF algorithm began to reduce other generators in the queue, beginning with the least cost effective conventional generators in Country3, which were the highest cost generators in the region. The list of inefficient high cost generation plants in Country3 was soon exhausted. At that point, the OPF algorithm began reducing generation from Country1 and Country2. Accordingly, the power flow changed direction.

A similar effect was also observable in cost/benefit analyses. That is, the cost/benefit performance indicators were < 1 in all investment scenarios for the 70% capacity factor must-run wind parameter case.

C. Ignoring and Considering Voltage Constraints Voltage constraints were treated as local problems that can

be assumed to be managed by proper operations in the short term (e.g., proper selection of generator voltage set points) and relatively easy capacitor/reactor investments in the mid-term. For long term decision analysis it was more reasonable to work with OPF results performed and to ignore local voltage constraints.

In this section, substation voltages cases in which the voltage limits were ignored were investigated. These cases are as follows:

• Case_ VC-I_W30%; and

• Case_ VC-I_W70%.

When the voltage limit constraints were ignored in the OPF analyses, it was observed that the voltages in certain nodes were beyond the operational limits defined in Table III. When examined in the HVDC investment scenarios, the voltages at these nodes were closer to permissible limits when compared to AC interconnection scenarios. Therefore, the voltages of the key nodes in the AC interconnection cases are the focus of this section of the study.

TABLE III. VOLTAGE LEEL LIMITS FOR ANALISYS

The voltages at key nodes on the corridor are illustrated in

Figure 2 for AC interconnection investment scenarios. Since the optimum power exchange level was less than thermal limits of the line in many investment scenarios, the voltages at the key nodes close to the corridor were higher than 1 pu (per unit) in almost all scenarios, as illustrated in Figure 2. Nevertheless, the maximum voltage deviation was observed to be around +20%, which could be resolved through proper compensation by adding reactors. Note that only the reactors that were available at the substations in Country3 and Country1 are modelled in the analysis, as was agreed to by the project working group members. This result demonstrated the importance of additional reactors.

Figure 2. The voltage levels after OPF solution at optimum exchange power

(Case_ VC-I_W30%)

D. 750 kV vs. 400 kV interconnection In term of benefit/cost ratio, the 400 kV connection as

superior to the 750 kV connection in all AC and DC scenarios. The reason was that the total investment cost of the 750 kV is option was significantly higher than that of 400 kV. Although the optimum power exchange level in the 750 kV scenarios was higher than the 400 kV scenarios as illustrated in Table IV, the additional savings due to the higher levels of power exchange were not sufficient to compensate for the additional investment cost of 750 kV option when compared to the 400 kV option.

TABLE IV. COMPARISON OF OPTIMUM POWER EXCHANGE AMOUNTS AND SAVINGS (750KV VS. 2*400KV) – CASE W-30%

E. AC vs. HVDC B2B Interconnection For the HVDC B2B was greater than that for the AC

connection in all investment scenarios. This was because the OPF enlarged the exchange levels through the HVDC B2B because this technology relaxed technical constraints to increased power flows. Even at low levels of exchange, the regional savings were considerably higher when compared to

74

AC connection scenarios. Furthermore, the HVDC B2B option was the only solution that allowed for optimization savings when technical constraints were at their maximum (e.g., winter peak loading conditions as illustrated in Table V).

TABLE V. COMPARISON OF OPTIMUM POWER EXCHANGE AMOUNTS – CASE W-30%

Load Flow

Scenario Substation Variant

Voltage Level

Variant

Optimum power exchange (MW)

AC HVDC B2B

1 Country1 – Country3

1 x 750kV 200 900

2 Country1-Country2-Country3

1 x 750kV 200 850


1 x 750kV 200 850


1 x 750kV 200 850


2 x 400kV 150 800


2 x 400kV 200 800


2 x 400kV 200 800


2 x 400kV 150 800


1 x 400kV 150 600


1 x 400kV 150 600


1 x 400kV 150 800


1 x 400kV 150 800


1 x 750kV 700 250


1 x 750kV 700 300


1 x 750kV 700 250


1 x 750kV 700 250


2 x 400kV 700 250


2 x 400kV 700 250


2 x 400kV 700 250


2 x 400kV 700 250


1 x 400kV 650 250


1 x 400kV 650 250


1 x 400kV 650 250


1 x 400kV 650 250


1 x 750kV 250 100


1 x 750kV 250 100


1 x 750kV 250 100


1 x 750kV 250 100


2 x 400kV 200 100


2 x 400kV 200 100


2 x 400kV 200 100


2 x 400kV 250 100


1 x 400kV 200 100


1 x 400kV 200 100


1 x 400kV 200 100


1 x 400kV 200 100

TABLE VI. COMPARISON OF OPTIMUM POWER EXCHANGE AMOUNTS AND SAVINGS (1X400KV AC VS. HVDC B2B) – CASE W-30%

Another aspect very important is the possibiliy of

reducing quality disturbances: the reactive power capability of DC connection using B2B can be used to control stress in the network, thus contributing improved quality. Furthermore, the faster the response caused by the growth switching frequency provides new levels of performance on the control power quality such as reducing harmonics, flicker, voltage dips, and so on.

IV. POWER QUALITY PROBLEMS Approximately 70% of all power quality related problems can be attributed to faulty connections or wiring [4]. Power frequency disturbances, electromagnetic interference, transients, harmonics and low power factor are the other categories of power quality (PQ) problems that are related to the source of supply and types of load [5]. Among these events, harmonics are the most dominant one. The effects of harmonics on PQ are specially described in [6]. According to the IEEE standard, harmonics in the power system should be limited by two different methods: one is the limit of harmonic current that a user can inject into the utility system at the point of common coupling (PCC) and the other is the limit of harmonic voltage that the utility can supply to any customer at the PCC. Details of these limits can be found in [7]. Again distributed generation interconnection standards are to be followed considering PQ, protection and stability issues [8].

V. CONCLUSIONS The increase in wind generation in Country3 dramatically

limited the ability of the countries to optimize the regional generation fleet based on the cost of production. In some investment scenarios, the flow of power changed direction

75

from north → south to south → north when the wind power plant generation in Country3 increased from the 30% case to the 70% case and it was designated as must run. This occurs when the OPF algorithm forced inefficient high cost generators, first in Country3 and then in Country2 and Country1, to reduce their generation in favor of must run wind. This process continued until the reduction of generation in Country1 and Country2 became so much more cost effective than the reduction of generation in Country3 that the power flow changed direction. From this point onward, Country3 began exporting power in a northward direction to Country2 and Country1.

A similar conclusion was observed in the cost/benefit analyses. The cost/benefit performance indicators were significantly less favorable when wind generation was modeled at the 70% case capacity factors. In this scenario, the benefit/cost ratio was < 1 for all investment scenario cases.

It is important to note that for the investment scenario of a 400 kV connection passing through a HVDC B2B substation, the benefit/cost ratio was > 1, when Country3 must run wind generation was modeled with a 30% capacity factor.

The analyses results showed that, in case of high wind generation scenarios in Country3, the benefit of the new transmission corridor will depend on power export capability of Country3 to the rest of the network.

The 400 kV connection was superior to all 750 kV connection scenarios for both AC and DC options. This was because the total Investment cost of the 750 kV option was significantly higher than that of 400 kV. Although the optimum power exchange level for the 750 kV options as higher than that of 400 kV scenarios, the additional savings accruing to the region failed to make up for the increased investment cost associated with the 750 kV option.

Connection through the HVDC B2B was superior to connection through AC options in almost every investment scenario considered. This was because the HVDC connection reduced technical network constraints to increase power exchange, enlarging the scope for power flows in the sub-region.

In fact, HVDC B2B was the only investment solution which resulted in benefit/cost > 1 when considering the scenario of Country3 must run operating with a 30% capacity factor. Therefore, HVDC technology based interconnection of the countries seemed the most rational solution in the short/mid-term.

There was no significant difference revealed in the cost/benefit analyses for the different investment scenarios related to the configuration of the corridor, i.e., either directly from Country1 to Country3 or through Country2. If the

interconnection between the countries were realized in intermediate steps, (for example, if the connection between Country3 and Country2 were realized before all three countries are interconnected), energy trade between Country31 and Country2 could begin before the trading among all three countries by directing a generator in Country2 to operate synchronously with Country3 in island mode.

The results of this calculation should be considered as indicative for potential further studies which will take into account specific aspects of the wholesale electricity markets in the region with the purpose of more clearly quantifying benefits of investment in a new sub-regional transmission corridor.

The analyses were conducted using OPF. Accordingly, the results are bound by its limitations. These include the commitment of existing generation units not considered in the load flow analyses that were focused on three loading hours (winter and summer max and summer min). Complementary studies could consider this planning problem as a security constrained unit commitment (SCUC) problem (hourly based along a year) which, requires additional availability and pricing of spinning reserve requirements, locational marginal prices at substations nodes, etc.

ACKNOWLEDGMENT The work has been funded by the Project financed by the

European Social Fund, Priority Axis 1: “Education and training in support for growth and development of knowledge based society”, Key area of intervention 1.5 – “Doctoral and post-doctoral programs in support of research”.

REFERENCES [1] Lazaros P. Lazaridis, “Economic Comparison of HVAC and HVDC

Solutions for Large Offshore Wind Farms under Special Consideration of Reliability”, Master’s Thesis, Royal Institute of Technology Department of Electrical Engineering, Stockholm 2005.

[2] Dodds, B. Railing, K. Akman, B. Jacobson, T. Worzyk, “HVDC VSC (HVDC light) transmission – operating experiences,” CIGRE 2010

[3] Siemens PTI Documentation. [4] S.M Halpin, L. L. Grigsby The Electric Power Engineering Handbook,

CRC Press LLC (2001),pp15.4 [5] C.Sankaran, Power Quality, CRC Press (2002) [6] R D. Henderson, P J. Rose, “Harmonics: The Effects On Power Quality

And Transformers”, IEEE Trans Industry Appl, 1994, Vol 30(3) [7] S.M Halpin, L. L. Grigsby The Electric Power Engineering Handbook,

CRC Press LLC (2001),pp15.22-23 [8] IEEE 1547, IEEE Standard for Interconnecting Distributed Resources

with Electric Power Systems, 2003

76

Documents

[IEEE 2014 16th International Conference on Harmonics and Quality of Power (ICHQP) - Bucharest, Romania (2014.5.25-2014.5.28)] 2014 16th International Conference on Harmonics and Quality