Quantum Leap in Wind Power in Asia: Structured Consultation

Asia Clean Energy Forum – Pre-event Workshop

Asian Development Bank, 6 ADB Avenue, Metro Manila, Philippines, 21 June 2010
Prepared by Mr. Soren Krohn

Turbine & Grid

Off-Grid Small
Development

Wind -Diesel
Financing &

Technology
Overview of Key Issues

Regulation
Policy &

Project
for Group Discussions

Costs
Risk Identification and Risk Allocation
***
Knowledge of Wind Energy in Local Financial Sector,
Depth of Local Long-Term Capital Market ***
Current Wind Turbine Market
*** * ***
Project & Turbine Size Constraints
*** * **
Modern Turbine Capabilities - Wind Farms as Power
Plants ***
Technical Grid Challenges
*** * ***
Institutional / Administrative / Economic Obstacles to Grid
Integration * *** * ** ***
Institution Building & Capacity Building, First Project
*** ** **
Economic Barriers at Power Generation Planning Stage
** ***
Energy Tariff & PPA Issues
** *** * ***
Permitting & Licensing
* *** * *
Public Land Use policy
*** **
Wind Resource Assessment & Site Data
* * *** *
Environmental & Social Impacts
* ** ***
Public Spatial Planning (Zoning)
* ** ***
Experience/capacity for IPP/BOO WInd Projects
* ** *** ***
Critical Mass Issues, Local Participation, Local Content
** ***
Wind-Diesel System Integration Challenges
***
*** = Primary theme with subject note; ** = Secondary theme; * = Tertiary theme;

The views expressed in this presentation are those of the author and do not necessarily represent those of the Asian Development Bank.
Financing and Costs
This session addresses financing and cost issues related to large-scale wind projects. There is
considerable overlap with the other sessions, in particular policy & regulation and project
development. The table attached below lists typical barriers to financing projects. Each subject is
treated in more depth in the notes for the other sessions. The Session Chair may wish to discuss
typical project finance structure, depending on the interest of the audience.

Key Issues:

Financing and Costs

• Why does financing for wind projects fail or succeed?

• Risk identification and allocation:

- Developer qualifications
- Electricity offtaker qualifications
- wind resource risk
- construction risk
- land risk
- environmental & social surveys risk
- grid risk
- revenue risk (PPA)
- power curtailment risk
- regulatory risk
- availability and operations risk
- health & safety risk

• Inadequate legal and regulatory framework

• Little knowledge of wind in local financial sector

• Depth of local long-term capital market

 Examples of Major Risks for Wind Farm Financing
Barrier
1 Risk allocation between
contracting parties
2 Developer qualifications:
Insufficient experience,
inadequate capital base
3 Electricity offtaker with little
IPP experience & poor credit
rating
4 Wind resource uncertainty, lack
of long-term reference data,
poor site measurements
5 Construction risk
6 Land risk, property rights
poorly defined, lack of
government land use policy
Possible Solutions
Note: Most mitigation measures mentioned below are
treated in more depth in the policy & regulation section
and turbine & grid section.
Risk should be carried by the party able to control the risk
or to most cheaply mitigate the risk. (e.g developer takes
wind risk, government guarantees against regulatory
changes with major economic impact). Risks beyond the
control of either party (e.g. exchange rate risk, price
index risk) are generally most cheaply carried by the
electricity purchaser.
Form consortia with experienced companies
Contractual framework needs to be established in
accordance with best international practice. Lack of
regulatory framework can partially be remedied by
"regulation through contract".
PPA may need to be backstopped by government
guarantee and partial risk guarantee from credit insurance
company / MIGA (political risk, payment risk)
Governments and development agencies can finance
modern mesoscale wind atlas work based on satellite
data, meteorological modeling and meteorological
reanalysis data. Guidelines for ToRs available from
World Bank (ESMAP).
Due to poor long-term meteorology data in many
developing countries, a ground-based long-term
provincial measurement program is often needed to
obtain reference data, which is used to calibrate
measurements in order to find long-term mean wind
speeds. Good examples of such programs in Egypt and
Syria.
Bankable site measurements need to be done by certified
consultants. At least 12 months of measurement if good,
long-term reference data is available from nearby
locations, otherwise preferably longer time period for
measurements.
Form consortia with experienced companies with local
knowledge and experience
Property rights programs for land registration may be
needed, government land use policy needed
7 Environmental and social risk
8 Grid risk: Inadequate grid
studies, including dynamic
stability studies. Missing grid
code or requirements that do not
correspond to actual grid
strength/stability
9 Revenue risk: Poor
creditworthiness of offtaker,
firm PPAs of 20 (or 15) years
duration not available, tariff
subject to political uncertainty
10 Tariff inadequate for rate of
return and/or debt service
coverage
11 Power curtailment risk
12 Availability and operations risk
13 Health & safety risks
Prior screening of land use by government extremely
useful (good examples in Denmark, Germany). Such
screening may involve bird studies, mapping microwave
corridors for telecommunications, aerial marking
requirements close to airports, potential archeological
finds etc.
Environmental and social impact assessments are needed,
and mitigation measures may need to be implemented.
Information and participation schemes for local
communities, landowner compensation schemes.
Grid code for wind turbines and wind farms (preferably
modeled on grid codes from major markets are needed).
Grid studies for site area required to ensure technical
feasibility.
Firm PPAs of 20 (or at least 15) years with fixed prices
are necessary to obtain an acceptable electricity price.
Wind projects are very capital-intensive and require a
(mostly fixed) tariff to service debt. Only O&M costs
(max 15-20%) need indexation.
Contracts must be take-and-pay and compensate for
actual loss due to grid outages or grid maintenance.
(Actual loss calculated from meteorology mast(s)
measurements on site and an empirical wind farm power
curve for the wind farm).
Power purchaser should require quality turbines from
experienced manufacturers, which are IEC certified by an
accredited entity as fit for purpose in the site
environment. Preferably turbine models, which have
proved to have a high availability in previous large wind
farm projects in similar climatic conditions.
Sufficient manufacturer warranties and service contracts
with training of local staff and manufacturer service team
in region. Spare parts and consumable stocks needed near
site.
Experienced wind farm operator with appropriate training
program for local staff.
Adequate health & safety program from developer
required. Occupational safety requirements should be
state-of-the-art from developed markets (e.g. fall
protection, and lifts in large turbines).
14 Inadequate legal and regulatory In countries new to wind power, in particular developing
framework countries, there is often a legal and regulatory vacuum.
This makes it impossible to design projects so that they
can be shown to be compliant with existing regulations,
which may in turn make it impossible to finance projects.
Developers are attracted to markets where the framework
conditions are known, or where at least the government is
aware of the regulatory gaps and capable of plugging
them through contract.

See policy & regulation section, project development

15 CDM/JI and other carbon
finance is difficult to handle for
developers
16 Depth of local long-term capital
market insufficient
section and turbine & grid section
CDM/JI and other carbon finance is difficult to handle for
developers, and hence attributed little or no value, when
calculating their required tariff. Carbon finance is
consequently best handled by government.
If it is not possible to obtain long-term finance in local
capital markets contracts will have to be in hard currency
- or indexed against hard currency.

June 18, 2010 • Soren Krohn • sk@skpower.net

Turbine and Grid Technology
This session addresses wind turbines and the realities of grids in developing countries in Asia,
where spinning reserve often may be a luxury, and rolling blackouts the rule, and where
frequency and voltage stability may be questionable. Many participants come from countries with
little or no wind power thus it may be useful to discuss that the challenges can be handled and are
handled even with high wind penetration above 20% by energy. The notes in the annex focus on
(1) main trends in the wind turbine market and (2) how the grid challenges are actually met in
countries with a large wind power penetration. This session will not discuss small turbines or
mini-grids, which are treated in another working group. The main topics that may likely be raised
on grid issues are listed in the table below.

Key Subjects:

Turbine and Grid Technology

Wind Turbines

• Current Wind Turbine Market

• Project & Turbine Size Constraints

• Modern Turbine Capabilities - Wind Farms as Power Plants

Grid Integration Issues

• Technical Grid Challenges

• Institutional / Administrative / Economic Obstacles

 Typical Transmission Grid Issues in Developing Countries
Barrier
1 Little or no knowledge of wind power
characteristics, (e.g. assumption that
wind power is intermittent100%-0% in
seconds rather than several hours)
2 No existing standard grid code adapted
for wind turbines or wind farms
3 Grid studies, including dynamic grid
stability studies for project unavailable
4 Weak grids and long radials to reach
(often remote) windy areas require grid
reinforcement / grid extension. Wind
developer demands for grid connection
in remote areas may be costly to meet.
5 No clear responsibility for transmission
systems operator to provide
interconnection for IPP wind farms
6 Transmission master planning not
adapted to IPPs: Long transmission
project queue, often 3 years of wait or
more. Master plan revision slow.
7 Wind projects are often required to bear
the cost of grid reinforcement / grid
extension, even if a stronger local grid
or a grid extension to remote areas also
benefits local consumers and electrical
utility.
8 First projects in a remote, high-wind
area cannot bear the cost of grid
extension, but additional projects could.
This »chicken and egg« problem
prevents wind development to take off
in potentially promising high-wind
zones.
9 Autogeneration wind projects face
problems of negotiating interconnection
fees, wheeling and banking rights, and
agreements on cost of balancing power.
Possible solutions
Capacity building for TSO staff required,
including basic wind power technology, power
quality and grid support properties of modern
wind turbines, meteorology, use of short-term
wind forecasting in dispatch. Simulation of power
generation based on historical local meteorology
data and historical hourly load curve.
Establish grid code for wind turbines and wind
farms based on mainstream large international
markets, but adapted to local grid conditions. If
first project is an IPP/BOO project, define and
subsequently recycle interconnection
requirements as a general grid code
For larger wind farms it is necessary to include
such studies in the transmission project related to
the wind farm.
It is extremely useful to prepare a generic grid
reinforcement cost study for each (wind-relevant)
section of the transmission grid (to be updated,
say after 3-5 years). This study will complement a
national mesoscale wind resource map in order to
search for economically suitable sites and begin
local wind measurements.
Separate transmission queue is needed for IPP
project pipeline, with clearly defined
responsibilities for transmission system operator.
The transmission grid should be considered a
public good, to be financed through a »postage
stamp« transmission tariff.
Central planning can be useful to make wind en-
ergy development take off: Governments and
development banks can help finance grid
extension to windy areas, where pre-assigned
sites can be tendered as a pipeline of IPP projects.
Example: Egyptian Red Sea Coast, where 3,500
MW of IPP & government-owned projects will be
built 300 km away from the main transmission
grid. Government, World Bank, African
Development Bank, EIB and KfW are financing
grid and EIA for the whole area.
Replicable models have been developed in e.g.
India and Morocco.
10 Grid maintenance planning not adapted
to wind IPPs: TSO may demand right to
interrupt grid, say 1% of the hours of
the year.
11 Connection requirements for small
wind farms (connected at distribution
voltage level) are sometimes as
demanding as for large wind farms
connected to the transmission grid.
12 Long »gate closure times« in electricity
market / least cost dispatch planning
makes it difficult for wind supplies to
be scheduled efficiently
13 Large concentrations of wind farms in
remote areas put additional demand on
grid management functions to ensure
grid stability.
14 Capacity credit - if part of tariff system
- may discriminate against wind by
assigning it zero capacity value
June 18, 2010 • Soren Krohn • sk@skpower.net
IPP contracts need to be take-or-pay contracts,
with damages to IPP equal to actual lost
production in case of any grid interruption, since
the maintenance event is controlled by the TSO
and can be planned for the low-wind season.
Otherwise developer may require some (1% /
capacity factor) in risk premium!
No technical need to apply transmission codes at
distribution voltage (MV) level.
Gate closure times can be shorter, i.e. limited only
by technical requirements for dispatch.
Dispatch center needs to run a short-term wind
generation forecast model, if there is a high level
of wind penetration in the grid control area.
Larger IPP wind farms should be required to
supply real-time wind data from on-site
meteorology masts and generation and
availability data from SCADA systems for the
wind forecasting model.
Grid codes should provide for remote control of
wind farms or clusters of wind farms by dispatch
center, e.g. for variable reactive power
compensation, and in emergency situations
possibility of energy curtailment from 0-100%.
Wind does have a capacity value in the grid,
which can be determined by simulation models
including historical data for wind and electricity
demand, and observing a given loss-of-load
probability. These analyses indicate that for
moderate amounts of wind in the grid, say, up to
20% by energy, the capacity value is about equal
to the average capacity factor for wind power.
Annex 1, Turbine and Grid Technology

Main Trends in Wind Turbine Market

and Impact on Developing Countries
• General tendency towards large turbines of 1.5-3.6 MW (70-120 m rotor diameter) in
large mainstream markets, though the onshore market seems to have reached a plateau in
terms of turbine size since from about 2006.

• Later generations of large turbines have advanced generator designs and modern power
electronics able to cope with demanding grid codes (voltage support, frequency support,
fault ride through, etc.)

• Large turbines may be less suited to new markets in developing countries because of
infrastructure limitations, in particular non-availability and cost of large cranes, port
facilities, road curvature, road instability for heavy loads and weak grids. Craning issue
resolution requires large projects, preferably at least 100-150 MW.

• Smaller market volume for medium-sized 750 kW - 1 MW turbines means large

manufacturers have less focus on product upgrades and product development in this
market segment.

• Fewer reliable manufacturers with a good track record are focusing on the medium-sized
turbine market, hence less competition.

• The wind turbine market was a seller's market in the three year preceding the global
financial crisis with high profit margins - and more expensive inputs upstream (steel,
generators, gearboxes, transformers etc.) There was a shortage of manufacturing capacity.
Severe shortages of supply of specialized components (bearings, flanges, castings) made
turbine manufacturers focus on existing costumers in well-established mainstream
markets in Europe and North America. Most projects below 50 MW did not attract
interest from large established manufacturers and developers. Lead times for orders was
up to 3 years from the large manufacturers, and a 25-30% down payment was necessary
to reserve a space in the production plan. There was only a scant interest from wind
consultants in working with developing country markets.

• Financing was relatively easy during the boom, with financiers, developers and equity
investors ready to take relatively high risks.

• The financial crisis has changed all this: Canceled orders due to lack of finance (much
stricter demands from lenders), overcapacity in most parts of the supply chain has made
for substantial price drops for wind turbines in the order of magnitude of 20-30%, meaning
that prices are somewhat close to the long-term tendency up to 2005. Lead times for
turbine orders are down to typically some 9 months between order and delivery on
site.

• Cancellation of projects in mainstream markets have made experienced international

developers more interested in markets in developing countries. New wind power
developers have come on the scene, particularly in Asia - mostly from China and Korea.
• The emergence of new manufacturers, particularly from China, may give some
downward pressure on wind turbine prices in the future, notably if these manufacturers
over time establish a good track record for availability. Wind farms with turbines from
the big established western manufacturers typically have availability rates around 97%,
so the performance bar is quite high.

• Financiers have become much more wary of projects in relation to risk taking. More
thorough project preparation is required, more focus on well-established low-risk turbines
from mainstream manufacturers. Equity more difficult to obtain for high-risk countries.

• The vast majority of projects in both developed and developing countries are IPP/BOO
wind projects, but the first pilot/demonstration project in developing countries are
sometimes EPC projects built by the national electrical utility.

• Type certification of wind turbines in accordance with the most recent IEC standards is a
must for all manufacturers in order that the client can obtain financing for his projects.
IEC norms, however, are mostly developed for typical European and North American
climates, whereas operation in major parts of Asia requires additional certification for
cold or hot climates, high humidity, dust and sand. Current norms and design basis were
not made for tropical cyclones / typhoons. Although turbines may possibly be designed to
safely survive tropical cyclones, there could be a cost advantage if an initiative were
taken to develop a better design basis under IEC for wind turbines for tropical cyclone
conditions [proposed by several international wind research institutes, which lack funding
for this work.]

• There is an increasing cost consciousness amongst major wind turbine manufacturers in

relation to building additional national or regional service organizations. Unless there is a
critical mass of projects in a country or region, manufacturers may not be willing to
supply turbines for the projects. This means that it may be necessary to do fairly large
projects 50-100 MW and up, and the former may only be workable if there are other
projects in neighboring countries. Another possibility is to bundle (consolidate) projects
on neighboring sites to a single project, so as to make the order attractive for a turbine
manufacturer.

June 18, 2010 • Soren Krohn • sk@skpower.net

Annex 2, Turbine and Grid Technology

Dealing With Grid Challenges in Countries with High

Wind Penetration
Historical Development
1. Until early 1990s: Wind turbines seen as marginal grid component. Apart from design of
protection of grid interface preferred action during grid disturbance was to disconnect
wind turbines.
2. Mid 1990s: Increasing plant size required interconnection studies and grid stability
studies.
3. 2000 onwards: Requirements to treat large wind farms more as power plants: Voltage
and frequency support, fault ride through, dynamic var support

Mainstream View of Challenges

1. Small penetration levels (say, 5% by energy): No major impacts, no need for heavy grid
code requirements. Load following & regulation impacts are small.
2. Issues of power systems control have to be seen at system level, e.g. balancing power is
required to deal with (load minus variable generation) - NOT to balance wind power
variability.

Variability and Uncertainty is Nothing New for Grid Operators

Variability and uncertainty prevail in all types of power generation: Power system control is all
about dealing with variability and uncertainty.

1. Variability:
1.1 Load varies by seconds, minutes, hours, day type and weather
1.2 Any supply resource may be unavailable or limited due to outages
1.3 Prices for power purchases and sales fluctuate

2. Uncertainty:
2.1 Operational plans are based on forecasts, some error is unavoidable
2.2 Supply resources available with some probability (usually high)

Wind Power Impact on Existing Variability and Uncertainty

1. Variability and predictability of wind is well characterized in different time domains:
Seconds, minutes to hourly (dispatch, ramping), diurnal (day/night) to seasonal (hydro
complementarity) scale.
2. Wind pattern variability (day/night, seasonal variation) and its correlation with load
varies with geography and can be analyzed in advance. Variability is reduced by
geographical spread / interconnection, since weather patterns move.
3. Generation from the individual wind turbine varies almost instantaneously with the
energy in the wind, but these second to second variations cancel out at wind farm level.
4. Regulation (fastest variations that are corrected by automatic action) are not the major
issue
5. Load following (10-minute domain) and wind forecast error reserves (hourly domain):
Best handled by deep, liquid real-time energy markets
 6. Ramping requirements may be reduced if wind is uncorrelated or negatively correlated
with load.
7. At increasing (10-20%) penetration levels, small but measurable increase in ramping
requirement can generally be met by existing generation with modest cost increase.
8. At higher (>30% levels) minimum load problems may appear:
- Large markets (energy, ancillary services, price responsive load)
- More flexible generation options
- Larger balancing areas and stronger interconnections
- Curtailment
- Energy storage (large hydro & pumped storage)

Reducing Cost: Wind Plant Modeling

Historical modeling useful for (1) assessing generation sources affected by altered dispatch (2)
nodal forecast for managing congestion points in grid (3) interaction with hydro reservoir
management

1. Use historical measured wind speed data and/or meteorological modeling used for
weather forecasting to re-create the weather
2. Convert time series of wind speed data to generation using turbine power curves
3. Simulate smoothing effect of wind farm size and geographical spread (Nørgaard &
Holtinnen method)

Short-Term Wind Generation Forecasting

1. Short-term forecasting models are commercially available and reasonably accurate.

Regulation Conclusions: Cost Increases Small

For small penetration of wind power in the grid, 10-20% by energy, wind is manageable without
a major cost penalty. For larger penetration a moderate cost increase, but in cases where
electricity markets handle the balancing (Scandinavia or Pennsylvania-Maryland-New Jersey)
incremental costs are quite small, typically at or below 4-5 USD/MWh).

June 18, 2010 • Soren Krohn • sk@skpower.net

Policy and Regulation
This session addresses the legal and regulatory framework necessary to start large-scale grid-
connected wind energy development. Small-scale wind projects and technical grid and turbine
issues are covered in other sessions. Project development and financial issues related to the
transmission grid may have some overlap with this session, however.

Key Subjects:

Policy and Regulation

• Role of Government: Institution & Capacity Building

• Economic Barriers at Power Generation Planning Stage

• Energy Tariff & PPA Issues

• Permitting & Licensing Issues

• Public Land Use Policy (& Spatial Planning Issues)

The economic feasibility of developing large-scale wind energy depends primarily on having high
wind speeds on sites accessible by transmission and road, and on the economic cost of alternative
forms of power generation. The practical feasibility including the financing of projects depends
heavily on having an adequate legal and regulatory framework.

The tables below list typical barriers cited by developers and possible solutions. Some solutions
apply only to projects tendered on the basis of a tendered IPP/BOO project (on the basis of a
fixed kWh-price), whereas others apply to fixed-price feed-in tariffs (FIT); this is noted in the text
below.

Institution Building & Capacity Building,

First Pilot / Demonstration Project
Barrier
1 Many Government agencies involved
in regulation of wind area, little
effective coordination
2 Difficult to create comprehensive
legal and regulatory environment for
wind sector development
3 Government/utility pre-development
of tendered IPP/BOO projects is
necessary if they are tendered on
predetermined sites
4 It is critical that meteorology masts
be correctly installed in accordance
with the IEC standard and equipped
with MEASNET or equivalent
calibrated quality instruments
5 Technicians for turbine O&M are not
available
6 Procurement expertise in relation to
wind IPP/BOO projects is missing
Possible Solutions
In order to succeed building projects, wind power
needs to be high on the political agenda, and an
effective inter-agency task force needs to plug the
holes in the legal and regulatory framework. The task
force should consult with the wind industry for
clarification of technical issues.
Difficulty may to a certain extent be overcome by
tendering an IPP/BOO project internationally, and in
areas where regulation is missing, do »regulation by
contract«. If properly prepared, e.g. grid
interconnection requirements can be recycled as a
general grid code for wind turbines and wind farms.
Government/utility should ensure that a grid study
and a complete preliminary environmental and social
impact assessment have been completed before bids
are due. Final site measurements are most efficiently
done by pre-qualified developers (see Egyptian
tender model under wind resources)
First masts should be installed by certified
international wind measurement consultants, and
local staff be trained to erect and maintain masts (guy
wire tensioning, visual inspection, safe data
collection). Cellular phone mast erection contractors
often have staff that can be trained for this purpose.
First-rate turbine suppliers will train local staff to do
routine O&M work. Experienced operators of diesel
gensets or engineers managing ship's engines are
excellent candidates for this type of work.
Training program is required. Thermal projects are
technically and economically different from wind
projects (wind projects resemble small hydro, to a
certain extent), hence specific wind expertise is
required to assist writing RFPs and assess bids.
 Economic Barriers for Wind at Power Generation Planning Stage
Barrier
1 Market failure to include
externalities in the financial
analysis
2 Competition from subsidized
fuel for conventional thermal
power
3 Improper or no accounting for
fuel price risk in power systems
planning makes countries
choose short-term low cost
solutions without regard for
long-term risk
4 Dominance of conventional
thermal power allows fuel price
risk to be placed on clients - or
de facto absorbed by public
budgets.
5 Electricity markets not geared to
wind
Possible Solutions
Government should have utilities include externalities in
power generation planning
Economic analysis / compensation mechanism to utility
should include true opportunity cost of fuel savings (fuel
can be exported or import reduced, savings on fuel
subsidies)
Power system planning models such as WASP (not to be
confused with the wind resource analysis model, WAsP)
systematically choose minimum cost solutions with high
risk, even if alternative lower risk solutions are available
with a minimal cost increase. Sensitivity analyses do not
reveal the true risk.
Models for calculating the historical tradeoff between cost
and risk for power generation portfolios and for doing
portfolio optimization do exist. (World Bank, ESMAP
model for assessing Fuel Price Risk in Power Systems
Planning is in the public domain. More recent versions are
available at a cost, but such modeling requires additional
training.)
One key advantage of wind, hydro and geothermal projects
is that the fuel is free, and that electricity offtaker can do
20-year fixed price contract for electricity supplies.
Gate closure times in electricity markets (planning horizon
for power generation in number of hours) may be too long
to benefit from short-term wind energy forecasting. Should
be shortened to what is technically necessary for actual
dispatch. (See turbine & grid session).

Barrier
1 20-year take-or-pay PPAs
unavailable
2 Feed-in tariff may be
modified politically at any
time
3 No sustainability of tariff
scheme
4 Creditworthiness of
electricity offtaker
inadequate
5 Fixed feed-in tariff (FIT) or
RE bonus per kWh
inadequate to ensure
profitability of projects
6 Wind energy cannot
compete financially due to
subsidies to fuels for power
generation
7 Economic incentives are
inadequate, (taxation etc.)
8 ....
Permitting and Licensing Issues
Barrier
1 Unexpected permitting
and licensing
requirements may wreck
an otherwise fully
developed wind project
Energy Tariff & PPA Issues
Possible Solutions
Wind energy is extremely capital intensive, so long-term
(basically) fixed-price PPAs are a necessity to obtain a
reasonable price per kWh, regardless of whether the tariff is
determined by a tender or by a feed-in scheme.
Changes to tariffs in feed-in based tariff systems should only
apply to new projects, where investments have not started.
Spread cost of wind energy on electricity tariff base, a levy on
transmission, RE fund, initiate compensation for true
opportunity cost of fuel saved from power generation
Projects may need to be backstopped by partial risk guarantees
(political risk and general payment risk) from MIGA or export
credit insurance organizations.
It is difficult to determine the appropriate level of a feed-in tariff
in a new market, i.e. a tariff, which is adequate, yet does not
give excessive profits on the best sites. Some mitigation can be
obtained by tariffs, which are differentiated by wind
resource or profitability (Danish, German or French models).
The best way to determine a commercially viable tariff is to
start with an IPP/BOO demonstration scheme and tender a few
wind farms on the basis of kWh price under such a scheme.
If fuel subsidies cannot be reduced, compensation mechanisms
to electricity offtaker for e.g. domestically produced fuel freed
for export or saved (rather than being given as subsidized fuel
for power generation)
Basically this is a question of determining an appropriate tariff
- by tendering or a fixed-price feed-in tariff (FIT). Special
incentive schemes may (politically usefully or not) serve to
make the pricing/tariff issue less transparent.
Possible Solutions
Permitting/licensing requirements should be built into
requirements for RFP/BOO projects, so that the winning bidder
will have demonstrated compliance with the requirements for
obtaining permits/licenses
For price-based tariff schemes (FIT) a »single window« approach,
i.e. a single government agency that coordinates all permitting
requirements is extremely useful. Has been done for offshore
wind with success in e.g. Denmark, and the model is being
implemented elsewhere, sometimes though national energy
agencies, sometimes national investment authorities.
 Public Land Use Policy
Barrier Possible Solutions
1 Exclusivity arrangements with In fixed-price feed-in tariff (FIT) systems,it is best to have
developers (often = land competition for land use (auction based on rent per MWh),
speculators) locks up valuable with annual land rent to be paid in any case, until project is
high-wind resource land, commissioned. Definitely a requirement to have a time limit
which remains undeveloped for land lease before commissioning must occur. Possible
model inspiration in Ireland's offshore wind territory lease
system.

For tendered BOO/IPP projects on sites to be found by

developers, non-exclusive letters of intent of Government to
do land leases can avoid the lock-up problem during the
bidding phase. (Good example by the Government of
Québec).

For tendered BOO/IPP projects on predetermined sites the

issue does not arise, (only one bidder will win the right to
use the site).

Land speculators may be unhelpful to the process. In that

case, limit access to public lands to pre-qualified, bona fide
developers, with sufficient technical and economic
qualifications. (Usually done in all tendered IPP/BOO
projects).
2 Land rent If any land rent must be collected, it is best (least risk for
developer) if based on actual energy production, i.e. an
amount per MWh.

In quantity-based systems (tendered IPP/BOO projects) land

rent will be reflected in the bid price, hence zero land rent
for public land may be optimal in price-tendered projects. It
is usually necessary to specify a minimum number of MW
for the particular land area to ensure the efficient use of
valuable high-wind resources. (The number of MW varies
with the terrain surface roughness, topography and wind
climate and requires expert advice).

June 18, 2010 • Soren Krohn • sk@skpower.net

Project Development
This session addresses project development issues for start large-scale grid-connected wind
energy development. Small-scale wind projects and technical grid and turbine issues are covered
in other sessions. There is some overlap with the sessions on policy & regulation, financial issues
and turbine & grid technology, however.

Key Subjects:

Project Development

• Wind Resources & Site Data - Wind Atlas & Long-Term Data

• Environmental & Social Impacts

• Public Spatial Planning (Zoning)

• Experience/Capacity for IPP/BOO Projects

• Critical Mass, Local Participation, Local Content

The economic feasibility of developing large-scale wind energy depends primarily on having high
wind speeds on sites accessible by transmission and road, and on the economic cost of alternative
forms of power generation. The practical feasibility including the financing of projects depends
heavily on having an adequate legal and regulatory framework.

The tables below list typical barriers cited by developers and possible solutions. Some solutions
apply only to projects tendered on the basis of a tendered IPP/BOO project (on the basis of kWh-
price), whereas others apply to fixed tariff systems (FIT), this is noted in the text below.

Wind Resource & On-Site Measurements

Barrier
1 Lack of knowledge of national wind
resource and probable generation costs
Possible Solutions
Mesoscale wind atlas based on satellite data,
weather model reanalysis data & meteorology
models. Preferably also national ground-based
meteorology mast measurement program to
verify this modeling. This can be used as a basis
for further exploration and measurements on
potential sites. World Bank (ESMAP) guidelines
for ToRs are available for mesoscale wind atlas
mapping.

Developers or land speculators, who have

2 Lack of reliable long-term wind data
makes wind energy resource estimates
uncertain
measured before others (and frequently taken
out options for land lease) often consider this a
low priority, since they prefer to remain in a
situation where they have an effective
knowledge monopoly and can lock up the best
land with good wind resources.
Government-run long-term wind measurement
program for each relevant region may be
needed. Good examples: DANIDA & GtZ-
financed program in Egypt, subsequently in
operation for >15 years. GtZ financed program
in 12 regions of Syria > 5 years. (Poorly planned
& poorly managed wind measurement programs
abound in many countries on several continents.)
3 Government does little or no pre-
development work for potential sites,
government does little or no regulatory
work related to wind energy
4 Moral hazard problem, if electricity
offtaker (government or utility) has
measured wind on a predetermined site to
be tendered as an IPP/BOO project:
Incentive to exaggerate resource, and
quality of measurements may be
insufficient, i.e. too risky for financiers.
(China and Morocco problem)
5 Low quality of wind measurement and
resource modeling requirements in
tendered IPP/BOO contracts increase risks
in projects
6 Poor or no digital topographical high-
resolution maps are available for sites
being tendered as IPP/BOO projects. This
increases risk in resource modeling and
consequently bid prices.
7 No geotechnical sampling prior to
tendering sites increases risks for bidders
(foundation costs).
Pre-development work on promising sites with
good wind resources, grid access and good
accessibility makes sense only if sites are
tendered competitively (by bidding for a MWh
price). If the site is thus pre-selected it is
important that the government take all the risks
under its control out of the project in order to
minimize the risk for bidders and their MWh
price. Prior EIA screening is also necessary.
Regulatory framework needs not be complete
for the first projects, the problems may be
solvable by »regulation by contract«. (See »First
Project« section)
If developers are to find sites on their own, then
logically pre-development work is the
developer's responsibility - e.g. in a classical
feed-in tariff (FIT) system. In this case,
however, it is necessary that the legal and
regulatory framework has been properly
established, dealing with all the issues listed in
these tables (and more).
Bidding developers, not the electricity offtaker
should measure wind on sites tendered for
IPP/BOO projects, since the developer takes the
wind resource risk. An operational model for a
voluntary joint site measurement program for
pre-qualified bidders has been developed in
Egypt for its 2,500 MW IPP/BOO wind
program. This model is now also being copied in
Syria.
It is the quality requirements of the developers
and their financiers, which are the determining
factor in whether a project succeeds, hence it is
preferable if minimum quality standards
correspond to the requirements of the bankers.
This is the basis for the mandatory measurement
requirements in the Egyptian IPP/BOO tender
model mentioned above.
The Egyptian tender model includes advanced
aerial laser scanning of sites. There are
economies of scale in site mapping, however,
and governments could digitally map multiple
sites as part of their pre-development work for
tendered IPP/BOO projects. There is more trust
in the offtaker doing this type of measurement
work than in their wind measurements, since
topographical measurement quality can be
verified ex post, but wind resource assessments
can only be verified after the wind farm has been
built.
The Egyptian tender model mentioned above
includes geotechnical sampling.
 Environmental and Social Impacts
Barrier Possible Solutions
1 Wind development Recommendation: National program of pre-screening of relevant
competes with other land regions for environmental & social issues, e.g. birds,
use in province/region telecommunications, archeological sites, waterways, etc., i.e.
preliminary environmental impact and social assessment studies
(EISA). Map layers can be combined with (1) wind atlas (2)
generic grid reinforcement cost map to find suitable development
areas. Most successful examples from Denmark and Germany
have been copied elsewhere (e.g. Spain, USA offshore).
2 Project risk whether For IPP/BOO projects on predetermined sites, government should
environmental and social do preliminary environmental impact and social assessment
impact assessment will studies, ensuring a near-certain approval of project.
be positive or not
3 Private land use: Poorly Land registration program may be needed to determine property
defined property rights rights in the area of the site. Special problems when handling
or indigenous people's collectively owned land by farming communities or indigenous
rights people's rights. Extensive literature on the subject available from
e.g. the World Bank (+ upcoming WB publication on best
practice for handling environmental and social issues in relation
to wind farms)
4 Private land use: Guidelines for compensation of landowners in accordance with
Landowner resistance to national practice for similar types of projects. It is important that
project all landowners within wind farm perimeter receive some sort of
compensation per turbine on their land + compensation for access
roads. In some jurisdictions transmission mast compensation rules
can be used as a model. Even landowners without turbines or road
use should receive some (lower) compensation to avoid political
blockage of project from non-compensated landowners. Good
elaborated model guidelines issued in Ontario, Canada.
6 Laws or resistance Land need not be purchased for wind farms, but can be leased for
against land being taken the duration of the PPA, and legislation and regulations should
out of farming allow this. Close to 98% of the land area will remain arable after a
wind farm has been installed (only turbine & transformer platforms
and access roads are needed for wind farming)
7 Local resistance to Community income sharing schemes are known from other power
project due to lack of generation projects. Best practice for local information / hearing
information / practice are known, e.g. from ADB, IFC or World Bank safeguard
participation guidelines.
8 Concerns about safety All wind turbines installed in the country must be required to be
for neighbors and type certified for a technical lifetime at least equal to the duration
workers on site of the PPA by an accredited entity in accordance with the most
recent version of the IEC 61400 standards as fit for purpose in the
site environment.
9 Ornithological concerns Ornithological studies (1 year) may be required as part of
about bird or bat environmental impact assessment (EIA) in critical areas.
populations Mitigation measures may be needed (e.g. temporary stoppage
during high-density migration, if wind farm is placed in an
important bird migration path). The determination of whether an
area needs additional studies is best done in the environmental
screening phase (point 1 above), where zones may be labeled red«
(prohibition), or »yellow« (bird studies required), or »green« (no
bird studies required).
 Public Spatial Planning (Zoning)
Barrier
1 Poor site area planning leads to
interference (wind shading) between
wind farms, and increases risks and
required tariffs uselessly (common
problem in very high-wind zones with
densely packed wind farms)
2 Poor site planning in relation to noise
& shadow flicker may cause problems
with neighbors and cause sites to
require re-planning
3 Military or civil aviation authorities
may object to siting
4 Telecommunications authorities may
object to siting
5 Road authorities may object to siting
Possible Solutions
Wind farms generate turbulence downstream, and
turbulent energy cannot be used for power
production. It is essential that wind farms be carved
out so that they do not shade one another. An
upstream wind farm will reduce energy production
downstream by 10-20% depending on turbulence
intensity and terrain surface roughness. Poor site
planning also means that use of the wind resource
will not be optimized, possibly wasting 15-20% of
the energy - and with correspondingly higher
required tariffs. Adjacent wind farms should be
long slices in the prevailing wind direction, with the
borderline following the prevailing wind. Buffer
zones for wind to recover are required between
wind farms, notably downstream. A single large
wind farm optimized by a single owner will usually
exploit the land area best.
There is excellent software on the market (e.g.
WindPro or Wind Farmer), which may be used to
define appropriate distances between turbines and
residences to meet a regulatory requirement that the
theoretical noise level will not exceed, say,
typically 40 dB(A) (this maximum limit needs to be
defined in regulations). Likewise, shadow flicker is
only a real problem in a narrow strip SE of each
turbine (in the Northern hemisphere). Exclusion
zones can be mapped using this standard software.
(Again, acceptable maximum number of shadow
flicker hours needs to be defined in regulations).
Rules for aerial markings on tall wind turbines need
to be regulated. Good standard models for this are
available (red/white stripes on blades, and for very
tall turbines, night lighting). But concrete decision
may need to be taken in relation to topography
(mountain ridges) and proximity to air corridors
(e.g. end of runways in airports are off limits, but
no major problems elsewhere around airports)
Turbines should not be placed directly in
microwave transmission corridors. Standard rules
are available for this. Otherwise no major radio or
TV interference problems: Wind turbines often
have double use as towers for cell phone
communications.
Setback from roads usually regulated to be about
100 m.
6 Decommissioning requirements for
wind farm undefined, risk of ghost
wind farms
Decommissioning requirements should be defined
in the PPA, or regulated generally. Best practice is
to require that foundations are removed to 1 m
below grade and that land is restored to its original
state after PPA termination. Any turbine, which has
been out of service for a year must be removed
from the site and the terrain restored as when
decommissioning. A security/guarantee
arrangement for this (bond) is useful to include in
the RPF documents or the PPA.
 Experience/Capacity for IPP/BOO Projects
Barrier Possible Solutions
1 Disagreement about Best practice for all tendered IPP/BOO contracts is that the party
which party bears who effectively can control each risk or who most cheaply can cover
which risks in it, bears that risk. For risks, which are outside the control of either
IPP/BOO contracts party such as exchange rate risks, and prices that affect project
economics, they are usually most cheaply carried by the electricity
offtaker. It is in the interest of the electricity offtaker to reduce
project risks as much as possible in order to achieve a low electricity
price. The offtaker will have an interest in doing as much pre-
development as possible in the special case of IPP/BOO tendering
on predetermined sites in order to reduce risk.
2 BOO or BOOT Wind farms have a standard certified technical lifetime of 20 years.
contracts? PPAs should generally have a term of 15-25 years, a
decommissioning requirement and terminate thereafter. It is unwise
to insert any option for the bidder to continue projects thereafter,
since economic conditions may change substantially in the meantime
(giving windfall capital gains to project owner).
3 Price indexation of The primary economic advantage of wind energy is that the
PPA contracts electricity offtaker can do fixed-price electricity contracts for a
(applies to FIT duration of 20 years. Wind farm projects should therefore normally
contracts as well) be done as primarily fixed-tariff (energy only) take-or-pay contracts
for the duration of the PPA. Wind farm owners are safe with this
solution, since they will take out nominal, not real (price-indexed)
loans from their financiers. A small component limited to labor and
parts content in O&M (maximized to about 15-20% initially) could
be indexed.

Exchange rate indexation is another issue treated under the finance

subject. Price indexation of bids between the time of bidding and
financial close or commissioning is a separate issue.
 Critical Mass Issues, Local Participation, Local Content
Barrier
1 High local content
requirement beyond balance-
of-plant (i.e. roads,
foundations and electrical
works) is not economic for
small projects
2 Tower manufacturing is most
amenable to localization for
larger projects
3 Nacelle assembly is
(mistakenly) seen as a
valuable means of
employment and technology
transfer
4 Rotor blade manufacturing
requires high, continuous
order volume
5 High local content is more
difficult to achieve in price-
based (FIT) system than in
quantity-based tariff systems
6 Small project size fails to
attract experienced
international bidders and
turbine manufacturers in
tenders
June 18, 2010 • Soren Krohn • sk@skpower.net
Possible Solutions
The most important prerequisite for local manufacturing is
to have a stable wind program with a time horizon of at least
5-10 years, and a credible continuity of national policy
despite changes in government. Otherwise it is too expensive
to do investment and training of local staff. Local
manufacturing - particularly of small volumes - may imply
significant cost and reliability penalties. Simple one-off
programs in a single year will only generate local work on
installation (balance-of-plant), but this may be quite
significant - 20-35% of project investment.
Transportation costs often mean that towers may be
manufactured locally economically for larger projects in any
case. Towers account for a relatively large share of the value
of a wind turbine, about 15-20%. Local manufacturing
requires ISO 9000-series certification of the supply chain
Nacelle assembly accounts for around 2% of the price of a
wind turbine or less, hence there is no economic gain, little
employment and high quality risk associated with local
manufacturing. The manufacturing process for wind turbines
in not substantially different from other forms of large
machinery manufacturing.
Rotor blade molds are expensive assets, which - like blades -
are difficult to transport, hence they need to be run with high
capacity utilization, i.e. the local market has to be relatively
large and continuous. Raw materials normally need to be
100% imported. Blades typically account for 12-15% of the
value of a wind turbine. Local manufacturing requires ISO
9000-series certification of the supply chain
Annual volume is politically uncontrollable in a FIT system
and in practical terms also uncontrollable in a green
certificate system. Volume and suppliers can be controlled
accurately in a pipeline of IPP/BOO tenders (or EPC tenders
by the national utility/government)
Large, experience international wind developers with good
access to finance focus on projects in the 100-250 MW
range, or on pipelines or bundles of projects from this size
and up.
Small projects below 50 MW may have difficulty getting
turbines, if they are the first in a region without an
established service network.
A possible way of obtaining a critical mass of MW is to
bundle several non-contiguous project sites into a single
tender, as is being done in the Philippines.
Off-Grid Small Scale Wind (Wind-Diesel) Applications This
session is fairly distinct from the rest of the sessions and addresses the issues related to small
wind, primarily in hybrid wind-diesel mini-grids. The session does not address the market
for small battery charge, i.e. typically turbines from 100W-1000W. The note and table attached
below lists the typical barriers cited by developers of small systems.

Key Subjects:

Small Wind-Diesel Systems

Institutional barriers:
• Client base, Quality of service standards
• Tariffs/subsidies
• Diesel subsidies distort economics

Turbine supply barriers

• Quality of small turbines
• Manufacturer experience
• Operation and maintenance, service network

Systems integration barriers

• System design costs, (expensive to design from scratch)
• Diesel suitability and operating strategies / rules?
• Available expertise?
• Reliable controller systems?
Introducing one or more wind turbines in a diesel grid can be an economically efficient fuel-
saving measure, even with fairly poor wind resources of a mean of, say, 5 m/s at hub height, since
fuel is expensive. Wind is cleaner than other technologies.

Market
Grid-connected projects are developed by utilities or IPPs, while stand-alone renewable energy
systems are usually owned by consumers - or in a few cases by renewable energy service
companies. Mini-grid renewable energy systems generally require a significant degree of
community involvement. Utilities are usually not interested in off-grid areas and slow to respond
when there are technical problems with mini-grids because of the high transaction costs. Mini-
grid electrification uses different business models, and require specifically designed tariff
schemes coordinated with subsidies.

The supply side of the market is characterized by many small firms with very low production
volume. These firms have enough difficulties to survive in the marketplace, and usually have
little of no expertise to help clients develop a viable business model.

Market Regulation
Most mini-grid service providers are often not regulated or over-regulated. If regulated, quality of
service standards tend to be unrealistically high and expensive to comply with. Mini-grid
electrical systems for use in developing countries have in the past often suffered from over-design
and lack of repeatability. This means that engineering and designing each energy supply system
from the bottom up have made project planning almost as expensive as the system itself.

The rural electrification agency is inevitably the de facto regulator because of its administration
of subsidies to small grid operators. There is a “chicken and egg” problem for subsidies and
tariffs. Potential operators must know both tariff and subsidy levels before they can make
investment decisions. Government will usually decide on external subsidies, but a regulator can
nullify government granted subsidies with low tariffs.

Technology
This note refers to wind-diesel wind installations, typically with 5 kW to 330 kW turbines (5 m to
33 m rotor diameter). This market is distinct from:

1. The large grid-connected mainstream high-volume wind market for wind turbines,
typically 850 kW - 3.6 MW (50 m -120 m rotor diameter)
2. The very small 100-1000 W wind battery chargers

Wind-diesel integration has been most successful in medium-sized diesel grids in e.g. Australia,
China and on US naval bases, using conventional mainstream turbines of 225 kW and up.

In the late 1970s and early 1980s mainstream grid-connected turbines were typically 15-75 kW,
but this market segment has disappeared from the mainstream wind turbine market due to
economies of scale. Mainstream manufacturers now focus on wind turbines from 850 kW to 3.6
MW. This market is dominated by large manufacturers with billions of turnover with many years
of cumulative design and operating experience and large service networks.

Present-day manufacturers of small wind turbines have generally developed their own designs of
turbines, mostly in the 15-75 kW range, and some have licensed somewhat larger, older turbine
designs from mainstream manufacturers. Few of these manufacturers have a long track record or
a global service network. The vast majority of commercial small turbine designs from 25 kW and
up tend to be no less complex to build and maintain than large wind turbines.
Installing a wind turbine in a diesel environment makes the system more complex, requiring
experts for hybrid system integration. An electronic control system is required to distribute
generation between the different types of generators and to maintain voltage and frequency.

Such a system also includes a variable dump load, where excess electricity is wasted in a resistor
bank. Such a control system is fairly complex, since many diesel units cannot operate well below
load factors of around 50% without developing a coke-like substance within the engine. Suitable
diesel engines, which can operate at low loads, often with extra insulation to maintain temperature
and standard operating procedures to clean the engines may be necessary in small
grids. It is economically preferable to have one small diesel genset and a number of larger ones in
order to run a wind-diesel system efficiently (without having large diesels produce energy that
has to be wasted). That is a different optimization than what is preferred in pure diesel grids (with
a few large standard units).

The following table lists key barriers in this market and possible solutions:

Institutional Barriers
Barrier
1 Inexperienced client base, often co-
operatives. Often little or no assistance
from national utility.
2 Most mini-grid service providers are often
not regulated or over-regulated.
3 Quality of service standards tend to be
unrealistically high and expensive to
comply with. Project planning is often
almost as expensive as the system itself.
4 Worldwide, almost all rural electrification
programs involve some forms of subsidies.
It is often difficult to agree on adequate
tariffs and subsidies cover O&M costs and
allow for recovery of capital costs.
Regulators may often nullify pre-existing
payment schemes.
5 Subsidies for diesel do not enter current
financial project calculations, i.e. economic
analysis is distorted
Possible Solutions
Rural electrification agency should provide
support and capacity building
The rural electrification agency is inevitably the
de facto regulator. There are replicable, good
regulation models available.
Standards for service such as power quality
must be set realistically, affordable, easily
monitored and enforced.
Benchmarks should be used whenever possible
rather than actual costs for prices or subsidies.
Individual cost-of-service calculations are often
not workable. "Regulation by contract" is
needed: Regulators should not be able to
unilaterally change the tariff for mini-grid
operators during the contract period.
System costs should be evaluated on the basis
of true economic costs
 Wind Turbine Supply Related Barriers
Barrier Possible Solutions
1 Small volume, lack of replicable projects
2 Commercially unattractive niche, few experienced suppliers with
a long track record. Many small suppliers focused on debugging
product, and with limited commercial long-term viability. Many
upstarts focused on entering market for mainstream machines Create high market
after a brief learning period. Upstarts very dependent on volume, with multiple
government assistance for technology development. replicable projects
3 Market extremely dependent on development aid, few projects
have been sustainable and lasted 20 years.
4 Cost of service network similar to that of large mainstream
machines
5 Wind turbines are generally as complex to build and maintain as Project for additional
large turbines assessment and research
required by international
wind research institutes

Systems Integration Barriers

Barrier Possible Solutions
1 Conservative norms for grid quality require expensive non-
standard »gold-plated« power engineering designs, hence
high unit costs
Develop or allow use of
2 Conservative norms for diesel unit minimum load and
current standard solutions
operating strategy (barrier from utilities, and diesel
with acceptable quality of
manufacturers)
supply.
3 Generic problem with smallest diesel genset unit size and
required minimum diesel load (plant strategy is different
from conventionally optimized diesel grids)
4 Hybrid systems with diesel requires experienced integrators
on both controller and diesel side - very few available.
5 Few, tested, reliable controllers (w/ automatic dump loads).
Create high market volume,
Grid integration is difficult, either complex turbines with
with multiple replicable
synchronous generators are required and/or frequency
projects
converters.
6 Standardization and repeat orders fail to take off, hence
vicious circle for manufacturers and integrators

June 18, 2010 • Soren Krohn • sk@skpower.net