分布式发电的高覆盖率对电力系统设计和运行的影响分析

Aug. 2009

文章编号：1000-3673（2009）15-0037-10 中图分类号：TM7 文献标志码：A 学科代码：470·4051

分布式发电的高覆盖率

对电力系统设计和运行的影响分析

Bartosz Wojszczyk1，Omar Al-Juburi2，王靖3

（1．埃森哲公司，美国 罗利市 27601；2．埃森哲公司，美国 旧金山市 94105；

3．埃森哲公司，中国 上海市 200020）

Impact of High Penetration of Distributed Generation on System Design and Operations

Bartosz Wojszczyk1，Omar Al-Juburi2，Joy Wang3

（1．Accenture，Raleigh 27601，U.S.；2．Accenture，San Francisco 94105，U.S.；3．Accenture，Shanghai 200020，China） ABSTRACT: This paper addresses the topic of massive utility-oriented deployment of Distributed Generation (DG) in power systems. High penetration of DG presents significant challenges to design/engineering practices as well as to the reliable operation of the power system. This paper examines the impact of large-scale DER implementation on system design, reliable operation and performance and includes practical examples from utility demonstration projects. It also presents a vision for the utility of the future and describes DG technologies being implemented by utilities.

KEY WORDS: distributed energy resources；distributed generation；power system design and operation

摘要：文章探讨了电力公用事业大规模实施分布式发电的问题。分布式发电的日益普及将对电力系统的设计、工程实施以及运行的可靠性提出很高的要求。作者考察了分布能源(DER)的大规模应用对电力系统设计、安全运行与绩效的影响，并引用电力示范项目例举说明。同时，论文也展望了未来的电力公用事业发展远景，描述了应用于电力公用事业的分布式发电技术。

关键词：分布式能源；分布式发电；电力系统设计和运行

0 Introduction

Distributed generation (DG) or decentralized generation is not a new industry concept. In 1882, Thomas Edison built his first commercial electric plant— “Pearl Street”. This power station provided 110V DC electricity to 59 customers in lower Manhattan. In 1887, there were 121 Edison power stations in the United States delivering DC electricity to customers. These first power plants were run on water or coal. Centralized power generation became

possible when it was recognized that alternating current power could be transported at relatively low costs and reduce power losses across great distances by taking advantage of the ability to raise the voltage at the generation station and lower the voltage near customer loads. In addition, the concepts of improved system performance (system stability) and more effective generation asset utilization provided a platform for wide-area/global grid integration. In recent years, there has been a rapidly growing interest in wide deployment of DG. Commercially available technologies for DG are based on combustion engines, micro- and mini-gas turbines, wind turbines, fuel-cells, various photovoltaic (PV) solutions, low-head hydro units and geothermal systems.

Deregulation of the electric utility industry (in some countries), environmental concerns associated with traditional fossil fuel generation power plants, volatility of electric energy costs, Federal and State regulatory support of “green” energy and rapid technological developments all support the proliferation of DG units in electric utility systems. The growing rate of DG deployment suggests that alternative energy-based solutions play an increasingly important role in the smart grid and modern utility.

Large-scale implementation of DG can lead to situations in which the distribution/medium voltage network evolves from a “passive” (local/limited automation, monitoring and control) system to one that actively (global/integrated, self-monitoring, semi-automated) responds to the various dynamics of

38 Bartosz Wojszczyk等：分布式发电的高覆盖率对电力系统设计和运行的影响分析 Vol. 33 No. 15

the electric grid. This poses a challenge for design, operation and management of the power grid as the network no longer behaves as it once did. Consequently, the planning and operation of new systems must be approached somewhat differently with a greater amount of attention paid to global system challenges.

The principal goal of this paper is to address the topic of high penetration of distributed generation and its impact on grid design and operations. The following sections describe a vision for the modern utility, DG technology landscape, and DG design/engineering challenges and highlights some of the utility DG demonstration projects.

1 Vision for modern utilities 1.1 Centralized vs. distributed

The bulk of electric power used worldwide is

produced at central power plants, most of which utilize large fossil fuel combustion, hydro or nuclear

reactors. A majority of these central stations have an

output between 30MW (industrial plant) and 1.7GW.

This makes them relatively large in terms of both

physical size and facility requirements as compared

with DG alternatives. In contrast, DG is:

1）Installed at various locations (closer to the load)

throughout the power system and mostly operated by

independent power producers or consumers.

2）Not centrally dispatched (although the development of “virtual” power plants, where many

decentralized DG units operate as one single unit, may

be an exception to this definition).

3）Defined by power rating in a wide range from

a few kW to tens of MW (in some countries MW

limitation is defined by standards, e.g. US, IEEE 1547

defines DG up to 10MW – either as a single unit or

aggregate capacity).

4）Connected to the distribution/medium voltage

network - which generally refers to the part of the

network that has an operating voltage of 600V up to

110kV (depends on the utility/country).

The main reasons why central, rather than distributed, generation still dominates current electricity production include economy of scale, fuel

cost and availability, and lifetime. Increasing the size

of a production unit decreases the cost per MW;

however, the advantage of economy of scale is decreasing—technological advances in fuel conversion have improved the economy of small units. Fuel cost and availability is still another reason to

keep building large power plants. Additionally, with a lifetime of 25~50 years, large power plants will continue to remain the prime source of electricity for many years to come [1].

The benefits of distributed generation include: higher efficiency; improved security of supply; improved demand-response capabilities; avoidance of overcapacity; better peak load management; reduction of grid losses; network infrastructure cost deferral; power quality support; reliability improvement; and environmental and aesthetic concerns (offers a wide range of alternatives to traditional power system design). DG offers extraordinary value because it provides a flexible range of combinations between

cost and reliability. In addition, DG may eventually

become a more desirable generation asset because it is “closer” to the customer and is more economical than central station generation and its associated

transmission infrastructure [2]. The disadvantages of DG are ownership and operation, fuel delivery (machine-based DG, remote locations), cost of connection, dispatchability and controllability (wind and solar). 1.2 Development of “smart grid” In recent years, there has been a rapidly growing interest in what is called “Smart Grid – Digitized Grid – Grid of the Future”. The main drivers behind this market trend are grid performance, technology

enhancement and stakeholders’ attention (Fig. 1). The main vision behind this market trend is the use of enhanced power equipment/technologies, monitoring devices (sensors), digital and fully integrated communications, and embedded digital processing to make the power grid observable (able to measure the states of critical grid elements), controllable (able to affect the state of any critical grid element), automated (able to adapt and self-heal), and user-friendly (bi-directional utility – customer interaction). The Smart Grid concept should be viewed through the modern utility perspective of remaining profitable (good value to shareholders), continuing to grow revenue streams, providing superior customer service,

investing in technologies, making product offerings cost effective and pain free for customers to participate and partnering with new players in the industry to provide more value to society. It is important to recognize that there is merit in the Smart Grid concept and should be viewed in light of it 第33卷 第15期

Market Trend Drivers Performance Areas Energy Efficiency Renewable Energy Power Quality/ Reliability Customer Choice Corporate Branding Maintainability and Cost of Ownership 电 网 技 术 39

Technologies Distributed Resources Stakeholders Federal/State Lawmakers/ RegulatorsEnvironmentalistsIn general, this market trend requires a new approach to system design, re-design and network integration and implementation needs. In addition, utilities will have to develop well-defined engineering and construction standards and operation and maintenance practices addressing high penetration levels of DG.

Communication Advances Protection and Control Advances Monitoring and Metering Advances MaterialsUtility Executives2 DG technology landscape

DG systems can utilize either well-established conventional power generation technologies such as low/high temperature fuel cells, diesel, combustion turbines, combined cycle turbines, low-head hydro or other rotating machines, renewable energy technologies including PV, concentrated PV (CPV), solar concentrators, thin-film, solar thermal and wind/mini-wind turbines (Tab. 1) or technologies that are emerging on the market (e.g. tidal/wave, etc.). Each of the DG technologies has its own advantages and disadvantages which need to be taken into consideration during the selection process (Tab. 2).

图1 智能电网市场发展驱动力 Fig. 1 Smart grid market trend drivers

bringing evolutionary rather than revolutionary changes in the industry.

表1 DG 技术概况

Tab. 1 DG technology overview

Technology DG Size Range

Fuel

Reciprocating Engine Natural gas, diesel, landfill gas, digester gas

Microturbine

Combustion Turbine Natural gas, liquid fuels 20%~45%

Fuel Cells

PV/CPV/ Concentrators

Wind

5kW~6+MW(3) 25~500kW 0.5~30+MW(3)

Natural gas, hydrogen,

propane, diesel 20%~30% (Recuperated)

500W~10 MW 1kW~1+MW(3) 2kW~5MWNatural gas, landfill, digester gas, propane, hydrogen, fuel oil 36%~60% (up to 85% with

cogeneration)

Sunlight Wind 10%~35% 40%

Efficiency 25%~45% Estimated Equipment Cost/

($/kW)(1)

O&M/($/kW) 0.01 Environmental

Emission controls required for NOx and CO

500~700 700~1200 700~1100 2000~5000 2500~5000 600

0.005~0.016 0.004~0.010 0.002(2) 0.002~0.004 0.01

Very low when controls are used

Nearly zero emissions

No emissions No emissions

Low (< 950 ppm) NOx

Footprint/(sqft/kW) 0.28~0.37 0.15~0.35 0.02~0.61 0.9~1.1 150~200 11000~15000

Note: (1) Does not include any U.S. federal and state grants or subsidies; (2) Does not include costs for replacement of fuel cell stacks; (3) 6+, 30+ and 1+

represent values, which are a little more than 6, 30 and 1 respectively.

3 DR interconnection requirements

DR interconnection design and engineering

details depend on the specific installation size (kW vs. MW); however, the overall components of the installation should include the following:

1）DG prime mover (or prime energy source) and its power converter.

2）Interface/step-up transformer.

3）Grounding (when needed—grounding type depends on utility specific system requirements)．

4）Microprocessor protective relays for:

① Three-, single-phase fault detection and DG overload (50, 51, 51V, 51N, 59N, 27N, 67).

② Islanding and abnormal system conditions detection (81o/u, 81R, 27, 59).

③ Voltage and current unbalances detection (46, 47).

④ Undesirable reverse power detection (32). ⑤ Machine-based DG synchronization (25). 5）Disconnect switches and/or switchgear(s). 6）Metering, control and data logging equipment.

40 Bartosz Wojszczyk等：分布式发电的高覆盖率对电力系统设计和运行的影响分析

Vol. 33 No. 15

表2 DG 的优劣

Tab. 2 DG advantages and disadvantages

Technology Advantages

Disadvantages

Low capital cost, quick Noisy, frequent maintenance

Reciprocating startup, fuel flexibility, intervals engine

high reliability, low natural gas pressure required

Small number of moving Low fuel to electricity parts, compact size, light- efficiency, loss of power

Microturbine weight, can utilize waste output and efficiency with

fuels, long maintenance higher ambient temperatures intervals

and elevation

Readily available over a Reduced efficiencies at part wide range of power output, load, sensitive to ambient

Combustion capable of producing high- conditions (temperature, Turbine

temperature steam using altitude), small system cost and exhaust heat, high power- efficiency not as good as larger to-weight ratio, proven systems reliability and availability

Quiet, proven reliability, Low energy density(1), limited

(1)

(1)

Fuel Cells

self reforming, high field test experience, low energy density(1)

temperature waste heat may limit cogeneration potential(1)

Works well for remote Local weather patterns and sun

PV/CPV/Conlocations, requires little conditions directly affect the centrators

maintenance,

potential of photovoltaic systems, some locations will not be able to use solar power

Power generated from Variable power output due to wind farms can be the fluctuation in wind speed,

Wind

inexpensive, no fuel visual impact-aesthetic

required

problem of placing them in higher population density areas

Note: (1) refers that the situation applies only to specific types of fuel cells．

7）Communication link(s) for transfer trip and dispatch control functions (when needed).

Tab．3[3] summarizes the common DG interconnection requirements of utilities for various DG sizes (some details will vary based on utility specific design and engineering practices).

4 Impact of DR integration and “penetration” level

Integration of DG may have an impact on system performance. This impact can be assessed based on:

1）Size and type of DG design: power converter type, unit rating, unit impedance, protective relay functions, interface transformer, grounding, etc.

2）Type of DG prime mover: wind, PV, ICE, CT, etc.

3）Intended DG operating mode(s): load shaving, base-load CHP, power export market, Volt-Var control, etc.

表3 电力企业对于DG接入的要求

Tab. 3 DG interconnection requirements of utilities

DG > DG Requirements

less than DG 10~ DG 100~ 1000kW10kW 100kW

1000kW

or > 20%feeder load Disconnect Switch yes

yes yes

yes

Protective Relays:

Islanding Prevention &

yes yes yes yes Synchronization Other Protective Relays (e.g. unbalance) Optional Optional yes yes Dedicated Transformer Optional

Optional

yes

yes

Grounding Impedance (due to ground fault no no Optional

Often

contribution current) Special Monitoring &Control Requirements no Optional yes yes Telecommunication & Transfer Trip

no Optional Optional

yes

4）Interaction with other DG(s) or load(s).

5）Location in the system and the characteristics of the grid such as:

① Network, auto-looped, radial, etc.

② System impedance at connection point. ③ Voltage control equipment types, locations and settings.

④ Grounding design.

⑤ Protection equipment types, locations, and settings.

⑥ And other.

DR system impact is also dependent on the “penetration” level of the DG connected to the grid. There are a number of factors that should be considered when evaluating the penetration level of DG in the system. Examples of DG penetration level factors include:

1）DG as a percent of feeder or local interconnection point peak load (varies with location on the feeder).

2）DG as a percent of substation peak load or substation capacity.

3）DG as a percent of voltage drop capacity at the interconnection point (varies with location on the feeder).

4）DG source fault current contribution as a percent of the utility source fault current (at various locations).

4.1 DG impact on voltage regulation

Voltage regulation, and in particular voltage rise effect, is a key factor that limits the amount

第33卷 第15期 电 网 技 术 41

(penetration level) of DG that can be connected to the system. Fig. 2 shows the first example of the network with a relatively large (MW size) DG interconnected at close proximity to the utility substation.

Substation

LTC CT DG Supports most of feeder load Line drop compensator LTC Controller Large DG(many MW) Heavy Load No DG Heavy Load with DG eLight Load No DGgatlVoANSI C84.1 Lower Limit (114 v) Distance End

图2 临近电力企业变电站的DG连接

Fig. 2 DG connection close to the utility substation

Careful investigation of the voltage profile indicates that during heavy load conditions, with connected DG, voltage levels may drop below the acceptable/permissible by standards. The reason for this condition is that relatively large DG reduces the circuit current value seen by the Load Tap Changer (LTC) in the substation (DG current contribution). Since the LTC sees “less” current (representing a light load) than the actual value, it will lower the tap setting to avoid a “light load, high voltage” condition. This action makes the actual “heavy load, low voltage” condition even worse. As a general rule, if the DG contributes less than 20% of the load current, then the DG current contribution effect will be minor and can probably be ignored in most cases.

Fig. 3 and Fig. 4[3] show the second example of the network with DG connected downstream from the bi-directional line voltage regulator (VR). During “normal” power flow conditions (Fig. 3), the VR detects the real power (P) flow condition from the source (substation) toward the end of the circuit. The VR will operate in “forward” mode (secondary

Voltage profile RegulationDirectionDGVR P (kW) Load center with DG Q (kvar)

图3 VR双流向模式(正向流向)

Fig. 3 VR bi-directional mode (normal flow)

RegulationSubstationDirection10%Voltage profileDGVRP (kW) Q (kvar) 图4 VR 双流向模式(逆向流向)

Fig. 4 VR bi-directional mode (reverse flow)

control). This operation is as planned, even though the “load center” has shifted toward the VR.

However, if the real power (P) flow direction reverses, toward the substation (Fig. 4), the VR will operate in the reverse mode (primary control). Since the voltage at the substation is a stronger source than the voltage at the DG (cannot be lowered by VR), the VR will increase the number of taps on the secondary side; therefore, voltage on the secondary side increases dramatically.

Bi-directional voltage regulators have several control modes for coordination with DG operation. Bi-directionality can be defined based on real (P) and/or reactive (Q) power flow. However, reactive power support (Q) from DG is generally prohibited by standards in many countries. Therefore, VR bi-directionality is set for co-generation modes (real current) (Tab. 4) [3].

Tab. 5 gives examples of voltage change due to various DG sizes.

表 4 无功电压控制模式

Tab. 4 Voltage regulator control modes

Control Modes

Detection of power

flow direction

1Locked forwardNo common Most 2Locked reverse

No modes

3Reverse idle

No 4Neutral idle

Real currentCareful consideration5Bi-directionalReal current6Co-generationReal currentFor system with DG 7Reactive bidirectionalReactive currentinstallation

表5 因不同DG容量导致的电压变化示例

Tab. 5 Examples of voltage change due to various DG sizes

DG Size (kW)

Feeder Portion of ΔV Total Voltage(DG 2 miles away Substation ΔV Change from the substation) (vector sum)

1000.102%0.075% 0.157%

1 0001.020%0.750% 1.567%

5 0005.080%3.740% 7.840%10 000

10.200%

7.490% 15.7%

Notes: DG connected 2 miles from the substation on 12.47kV feederwith 336kcmil conductor; 12.47kV, 10MVA, 10% equivalent systemimpedance at the substation bus．

42 Bartosz Wojszczyk等：分布式发电的高覆盖率对电力系统设计和运行的影响分析 Vol. 33 No. 15

4.2 DG impact on power quality

Two aspects of power quality are usually considered to be important during evaluation of DG impact on system performance: voltage flicker conditions and harmonic distortion of the voltage. Depending on the particular circumstance, a DG can either decrease or increase the quality of the voltage received by other users of the distribution/medium voltage network. Power quality is an increasingly important issue and generation is generally subject to the same regulations as loads. The effect of increasing the grid fault current by adding generation often leads to improved power quality; however, it may also have a negative impact on other aspects of system performance (e.g. protection coordination). A notable exception is that a single large DG, or aggregate of small DG connected to a “weak” grid may lead to power quality problems during starting and stopping conditions or output fluctuations (both normal and abnormal). For certain types of DG, such as wind turbines or PV, current fluctuations are a routine part of operation due to varying wind or sunlight conditions (Fig. 5).

Partly cloudy conditionsClear conditions Wk120/tupt80uO re40woP−200

09-1109-1309-15Date09-1709-19September 11 through September 20, 1998

图5 100 kW PV 电厂的功率输出波动 Fig. 5 Power output fluctuation for the 100 kW PV plant

Other types of DG such as ICE or CT can also have fluctuations due to various factors (e.g. cylinders misfiring and pulsation torque - one misfiring in a 1800 rpm engine translates to 15 Hz pulsation frequency). Harmonics may cause interference with operation of some equipment including overheating/ de-rating of transformers, cables and motors leading to shorter life. In addition, they may interfere with some communication systems located in the close proximity of the grid. In extreme cases they can cause resonant over-voltages, blown fuses, failed equipment, etc. DG technologies have to comply with pre-specified by standards harmonic levels (Tab. 6). In order to mitigate harmonic impact in the 表6 IEEE 519 – 1992常见配电系统的电流畸变限值 Tab. 6 IEEE 519 – 1992, current distortion limits

for general distribution systems

Maximum Harmonic Current Distortion in% of IL Individual Harmonic Order (Odd Harmonics)

ISC/IL <11 11≤H<1717≤H<23 23≤H<35

H≤35 TDD

<20。

4.0 2.0 1.5 0.6 0.3 5.020<50 7.0 3.5 2.5 1.0 0.5 8.050<100 10.0 4.5 4.0 1.5 0.7 12.0100<1000

12.0 5.5 5.0 2.0 1.0 15.0

>1000 15.0 7.0 6.0 2.5 1.4 20.0Even harmonics are limited to 25% of the odd harmonic limits. TDD refers to Total Demand Distortion and is based on the average maximum demand current at the fundamental frequency, taken at the PCC.

*All power generation equipment is limited to these values of current distortion regardless of ISC IL.

ISC =maximum short circuit current at the PCC

IL =maximum demand load current (fundamental) at the PCC H=harmonic number

system the following can be implemented：

1）Use an interface transformer with a delta winding or ungrounded winding to minimize injection of triplen harmonics.

2）Use a grounding reactor in neutral to minimize triplen harmonic injection.

3）Specify rotating generator with 2/3 winding

pitch design.

4）Apply filters or use phase canceling

transformers. 5）For inverters: specify PWM inverters with high switching frequency. Avoid line commutated inverters or low switching frequency PWM – otherwise more filters may be needed.

6）Place DG at locations with high ratios of

utility short circuit current to DG rating.

A screening criterion to determine whether

detailed studies are required (stiffness factor) to assess DG impact on power quality can be performed based on the ratio between the available utility system fault current (ISC) at the point of DG connection and the DG’s full load rated output current (IDG) (Tab. 7).

4.3 DG impact on ferroresonance Classic ferroresonance conditions can happen with or without interconnected DG (e.g. resonance between transformer magnetization reactance and underground cable capacitance on an open phase). However, by adding DG to the system we can increase the such as: DG connected rated power is higher than the rated power of the connected load, presence of large capacitor banks (30% to 400% of unit rating), during

DG formation on a non-grounded island.

第33卷 第15期

电 网 技 术 43

表7 电能质量筛选标准

Tab. 7 Power quality screening criterion

Stiffness Factor (SF)

Recommendations

Insignificant: no concern that flicker or voltage SF > 250

change will be an issue for any type of DG source

100 < SF≤250

Nearly Insignificant: very little concerns unless DG is started/stopped frequently

Minor Concerns: moderate concerns for wind and 50 < SF≤100

PV. Needs to asses rates of fluctuations and start/stop cycles

Significant Concerns: any DG needs analysis of

25 < SF≤ 50 planned start/stop cycles and output fluctuations. May require mitigation equipment

Very significant concerns: DG can cause serious

15 < SF≤25 voltage flicker and fluctuations. Requires mitigation equipment and/or system changes

Extreme concerns: voltage changes may be so SF < 15

severe that DG interconnection is not possible without extreme application of mitigation equipment of system upgrade

4.4 DG impact on system protection

Some DG will contribute current to a circuit current on the feeder. The current contribution will raise fault levels and in some cases may change fault current flow direction. The impact of DG fault current contributions on system protection coordination must be considered. The amount of current contribution, its duration and whether or not there are any protection coordination issues depends on:

1）Size and location of DG on the feeder.

2）Type of DG (inverter, synchronous machine, induction machine) and its impedance.

3）DG protection equipment settings (how fast it trips).

4）Impedance, protection and configuration of feeder.

5）Type of DG grounding and interface transformer.

Machine-based DG (IEC, CT, some microturbines and wind turbines) injects fault current levels of 4-10 times their rated current with time contribution between 1/3 cycle to several cycles depending on the machine. Inverters contribute about 1-2 times their rated current to faults and can trip-off very quickly--many in less than 1 cycle under ideal conditions. Generally, if fault current levels are changed less than 5% by the DG, then it is unlikely that fault current contribution will have an impact on the existing system/equipment operation. Utilities must also consider interrupting capability of the equipment, e.g. circuit breakers, reclosers and fuses

must have sufficient capacity to interrupt the combined DG and utility source fault levels. Examples of DG fault contribution on system operation and possible protection mis-coordination are shown in Fig. 6 & Fig. 7.

115 kV 13.2 kV Fault Contribution fromAdjacent FeederDG Might Trip The/utilityFeeder Breaker andRecloser FaultDGRecloser A DG Recloser B 图6 非计划的保护跳闸(后馈) Fig. 6 Undesirable protection trip (back-feeding)

The recloser has 13.2 kV tripped On its first instantaneous shot， Now the DG must trip before a fast reclose is Adjacent Feederattempted by the utilityRecloser A Islanded AreaDG Recloser B (Normally Open) 图7 非主动形成的 DG 孤岛 Fig. 7 Unintentional DG islanding

5 DG interconnection–utility demonstration project examples [3]

5.1 Utility 1: ground-fault current contribution for a synchronous DG and changing transformer configuration

Ground-fault current contribution for 100 kVA, 500kVA and 2MVA synchronous DG is being investigated on some rural feeders in the U.S (Fig. 8).

44 Bartosz Wojszczyk等：分布式发电的高覆盖率对电力系统设计和运行的影响分析 Vol. 33 No. 15

bus 1 bus 2 bus 3 bus 4 bus 5DGInterconnection utilitytransformer 10 MVA3 MVA 4 MVA3 MVA 13.8 kV0.95 pf lag 0.95 pf lag

0.90 pf lag

图8 带有同步 DG 的乡村馈线

Fig. 8 Rural feeder with synchronous DG

In addition, during investigation, the transformer configuration (Delta/Wye/Grounded Wye) on the DG side and utility side was changed.

The results of this investigation are presented in Tab. 8. The DG fault current contribution changes in a range from less than 1% (for Delta/Delta transformer configuration) to approx. 30% for 2MVA DG with Delta\\Grounded Wye transformer configuration (at bus 2). Slight changes of the fault current for the non-grounded utility-side are due to an increase in pre-fault voltage.

表8 故障电流的贡献

Tab. 8 Fault current contribution results

Interconnection Ground fault current (A)

Transformer Small DG Medium DG Large DG (100 kVA)

(500 kVA)

(2 MVA)

DG side

Utility side

Bus4 Bus3 Bus2 Bus4 Bus3 Bus2 Bus4Bus3Bus2

No DG & Trans. 9338 7271 6544 9338 7271 6544 933872716544Delta Delta 9340 7273 6546 9347 7281 6554 936473016575Gnd-Wye Delta 9340 7273 6546 9347 7281 6554 936473016575Delta

Wye

9340 7273 6546 9347 7281 6554 936473016575

Gnd-Wye Gnd-Wye 9379 7310 6582 9459 7386 6657 953574616732Delta Gnd-Wye 9446 7391 6662 9894 7795 7059 1104689208182

5.2 Utility 2: customer-based reliability enhancement on rural feeders – planned islanding application

The planned islanding application is being investigated on some rural feeders in Canada to improve the reliability of supply for rural communities, where the corresponding distribution substation is only supplied by a single high voltage (HV) line [4]. Customers on those feeders may experience sustained power outages for several hours a few times per year due to environmental and weather impacts that cause line break-downs and power outages. A local independent power producer (IPP) equipped with additional equipment is employed to supply the load downstream of the substation when the HV line is down or during maintenance periods of the substation. The IPP is paid an additional bonus if it can successfully serve the load during a power outage.

Fig. 9 shows a planned islanding example that consists of a 3MW run-of-river hydro power based IPP connected to a medium voltage feeder downstream of a 69kV/25kV distribution substation. The peak load of the feeder is about 1.5MW. Subsequent to scheduled substation disconnection (for maintenance), or accidental (forced) power outage due to a fault on the HV feeder, the feeder loads are supplied by the IPP. The hydro generation unit has load following and black-start capabilities to overcome possible transients due to load/generation mismatch at the time of disconnection to sustain the island. The IPP can serve the load on its connection feeder and in some cases, also on the adjacent feeder depending on the load demand and water-flow level at the time of the power outage. Both the substation and the IPP are equipped with necessary automatic synchronization equipment for reconnection back to the HV line without interrupting the load.

Utlity grid Substation63 kV 25 kV F2 F1 2.5km IPP 25kV/4.16kV 6.0kmHGHydro Generator1.5MW/0.8Mvar Load3MW.PF=0.85图9 规划的孤岛应用

Fig. 9 Planned islanding application

5.3 Utility 3: peaking generation and DG for demand reduction application

This application addresses a utility approach toward peak shaving and demand reduction which is attractive to those LDC that purchase electricity from larger utility companies based on a particular rate structure. The cost of electricity for LDC is normally calculated based on energy rate (MWh), the maximum demand (MW) and surcharges due to exceeding the agreed upon maximum demand. The peak-time cost of electricity can be as high as 10~20 times the regular rates.

In this case, LDC may install peaking units or have an agreement with specific customers in the LDC’s area that already have on-site backup generation units. The peaking units are operated only during peak-load time for a total of 100 to 200 hours

第33卷 第15期 电 网 技 术 45

per year based on 5~10 min dispatch commands. In return, the participating facilities are paid for the total power supplied during peak-demand periods at an agreed-upon rate that compensates for both for management of the load on this feeder since the energy storage unit can provide back-up generation in case of a sudden reduction in wind power production; therefore, it increases the load carrying capacity of the generation/maintenance costs and plant upgrading costs in order to respond to utility dispatch commands.

5.4 Utility 4: energy storage applications for firming up intermittency of distributed renewable generation (DRG)

Medium to high penetration of renewable energy resources (RES) can cause large power fluctuations due to the variable and dynamic nature of the primary energy source, such as wind and solar photovoltaic generation. Power fluctuations may cause reverse power flow toward the main grid, especially during light load conditions of the feeder. Furthermore, due to inherent intermittent resource characteristics, the firm capacity of a large RES-based DG may be very low and ultimately the utility grid will still be the main provider of the spinning reserve capacity and emergency backup generation in the area. Deployment of distributed energy storage units, when adequately sized and properly co-located with RES integration, has been explored by several utility companies in the U.S. to firm up power fluctuations of the high penetration of renewable energy (wind and solar) and to reduce adverse impacts on the main grid. Fig. 10 shows an application of energy storage that locally compensates the variation in power output of a large wind farm and averages out the power fluctuations. Hence, the power flow measured at the point of common coupling (PCC) can be controlled based on a pre-scheduled profile, below permissible demand capacity of the feeder.

The controlled level of power flow at the PCC also drastically reduces the reserve capacity requirement

Energy Storage Time Utility Grid CHP ΔPG Wind farm LoadOutput Power ΔPW

Time 图10 为消除 DRG 功率波动的馈线潮流控制的储能单元

Fig. 10 Feeder flow control energy storage to

smooth out DRG power fluctuations

wind farm.

6 Conclusion

A growing number of electric utilities worldwide are seeking ways to provide excellent energy services and become more customer-focused, competitive, efficient, innovative, and environmentally responsible. Distributed Generation is becoming an important element of the electric utility’s Smart Grid portfolio in the 21st century. Present barriers to widespread implementation of DG are being reduced as technologies mature and financial incentives (including government- and- investor-supported funding) materialize. However, there are still technical challenges that need to be addressed and effectively overcome by utilities. Distributed Generation should become part of day-to-day planning, design and operation processes/practices. Special consideration should be given to the following:

1）Transmission and distribution substation designs that are able to handle significant penetration of DG

2）Equipment rating margins for fault level growth (due to added DG).

3）Protective relays, and settings that can provide reliable and secure operation of the system with interconnected DG (that can handle multiple sources, reverse flow, variable fault levels, etc.).

4）Feeder voltage regulation and voltage-drop design approaches that factor possible significant penetration of DG.

5）Service restoration practices that reduce chance of interference of DG in the process and even take advantage of DG to enhance reliability where possible.

6）Grounding practices and means to control DG induced ground fault over voltages.

References

[1] KEMA Consulting．Power Quality and Utilization Guide，Section 8

－Distributed Generation and Renewables[M/OL]．Leonardo

Energy，Cooper Development Association．http://www.Copperinfo.

co.uk．

[2] Willis H L，Scott W G．Distributed Power Generation：Planning and

Evaluation[M]．New York：Marcel Dekker，2000．

[3] Wojszczyk B，Katiraei F．Distributed Energy Resources－Control，

46 Bartosz Wojszczyk等：分布式发电的高覆盖率对电力系统设计和运行的影响分析

Vol. 33 No. 15

Operation，and Utility Interconnection[C]//Seminar for Various North American Utilities，2007&2008.

[4] Abbey C，Katiraei F，Brothers C，et al．Integration of distributed

generation and wind energy in Canada[C]//IEEE PES GM，Montreal，2006：7.

Received date：2009-06-18． Biographies：

Dr. Bartosz Wojszczyk is the Global MarketDirector for Accenture Intelligent Network Services.Bartosz leads demand generation efforts in smartgrid/systems operations/sales across the globe. He

provides thought leadership and advisory on strategic

Bartosz Wojszczyk Smart Grid/Smart City, Intelligent Network Solutions client engagements. Bartosz is atechnology and business expert in a number of areas including T&Ddesign, engineering & operations, smart grid, renewable and distributedenergy technologies (resources), advanced/integrated automation, widearea monitoring protection and control (WAMPAC), and energy storage.He holds a master and doctoral degree in Electrical Power Engineering.Located in Raleigh, USA, Bartosz is a member of the IEEE Power

Engineering Society and Standards Association (Secretary of IEEE Distributed Generation & Energy Storage Working Group).

Omar Al-Juburi is a senior executive with Accenture’s utilities industry group’s Intelligent Network Solutions (INS) practice. His expertise is in power systems operations, energy markets, complex program delivery, and smart grid planning. Mr Al-Juburi consults for

clients globally. He is located in San Francisco, USA.

Joy Wang is senior manager for Accenture Intelligent Network Solutions (INS). Ms. Wang has, for the last eight years, focused on the transmission and distribution sector of the utility industry. She has deep industry knowledge and provides management and technology consulting services to several China utilities. Currently Ms. Wang leads the intelligent network service practice, a practice that specializes in smart grids and smart metering/advanced metering infrastructure (AMI), of Accenture China. She is involved in several INS pilot projects in China including smart gridplanning and dispatching center optimization. Ms. Wang holds a master degree in Economics. She is located in Shanghai, China.

（实习编辑 董佳馨）