Colocating wind and solar generation with
battery energy storage is a concept garnering much attention lately. An
integrated wind, solar, and energy storage (IWSES) plant has a far better
generation profile than standalone wind or solar plants. It results in better
use of the transmission evacuation system, which, in turn, provides a lower
overall plant cost compared to standalone wind and solar plants of the same
generating capacity. IWSES plants are particularly suitable for regions that
have set high targets for wind and solar generation but have limited land
available for project development.
The Concept of an IWSES Plant
Separate wind and solar plants connected
to the same point of interconnection do not constitute an integrated wind and
solar plant. In an IWSES plant, wind turbines, photovoltaic (PV) solar arrays,
and a battery energy storage system (BESS) are integrated into a single plant
using state-of-the-art controls. It is noteworthy that the integration of wind
and PV solar plants, as well a BESS, can be performed at the wind turbine/PV
array level or the farm level. In farm-level integration,
balance-of-plant (BOP) equipment such as transformers
and switchgears, as well as the upstream transmission evacuation system, is
shared by the wind turbines, solar arrays, and BESS. In turbine-level
integration, each converter may be potentially shared by a wind turbine, solar
array, and BESS. Benefits of an IWSES Plant An IWSES plant may offer several
benefits over standalone wind or solar plants of the same capacity, some of
which can be readily monetized while others cannot. If we are to understand
such plants’ economic value, then these benefits must be quantified. Here, we
first present the benefits of integrating wind and solar generation; we then
discuss the benefits accrued from the addition of energy storage.
Benefits from Integrating Wind and Solar
Generation The key benefits of an integrated wind-solar plant relative to
standalone wind and solar plants with the same cumulative capacity are as
follows:
✔ Decrease in
project development costs: For wind and solar projects, project development
studies are required to determine project feasibility, socioeconomic impact,
and impacts on the market, the environment, and the grid itself. Wind and solar
projects also necessitate the fulfillment of legal and regulatory requirements,
such as securing permits and licenses, land-lease agreements, environmental
impact assessments, construction permits, grid interconnection agreements, and
power purchase agreements. A reduction in project development costs can be
achieved through efficiencies in project studies and filings with the statutory
bodies because they are the same for both types of generation.
✔ Better use of
available land: Colocating wind and solar power plants conserves space and
increases the energy density (i.e., the amount of energy produced per acre of
land). In many wind plants, the land between the turbines is often left unused.
To make better use of space, wind and solar plants can be colocated, provided
that conditions for both wind and solar power generation are favorable at the
same location. Alternatively, if wind and solar plants can be sited adjacent to
one another, they can share the same transmission evacuation infrastructure.
✔ Complementary generation profile: A further benefit
of integrating wind and solar generation stems from the complementary nature of
both the diurnal and seasonal patterns of their generation. • Diurnal: Figure 1 shows the average hourly
generation during selected months for a wind-solar plant made up of roughly
500-MW each of wind and solar generation capacity. As the figure makes clear,
in general, wind generation dips during the day and increases late in the
evening. On the other hand, solar generation follows the sun and is highest
during the middle of the day. Combining wind with solar reduces the difference
between the generation levels during daytime and nighttime periods.
• Seasonal:
Figure 1 also shows how wind generation at this site is highest during the
monsoon months (June– September); this is also the period during which solar
generation is at its lowest due to more incidences of cloudy days.
✔ Potential savings in transmission evacuation costs:
An integrated wind-solar plant also has the potential for savings in evacuation
and transmission upgrade costs. Typically, in a wind or solar plant, the
collector system is designed to carry the maximum output of the plant. For
example, standalone 100-MW wind or solar plants would each have 100 MW of
evacuation capability. However, an integrated plant can make do with an
evacuation capability lower than 200 MW because wind and solar generation will
not peak at the same time. Figure 2 shows a range of cumulative hourly
generation during selected months for a 1,000-MW wind-solar plant. Here, the blue
line indicates the average cumulative hourly generation. In addition, the red
and green lines in Figure 2 indicate the bounds within which the cumulative
generation for each hour would occur 95% of the time. It is, thus, possible to
design the evacuation system with a rating lower than the sum of the maximum
capacities of the wind and solar plants. In the rare event that the
cumulative
generation exceeds
the rating of the evacuation system, the output of the wind or solar plant
could be curtailed to keep the power flow from exceeding the transmission
limit. Similarly, integrating wind and solar plants can reduce the need for
upgrades to the transmission system.
✔ Sharing of operations and maintenance (O&M) expenses: Another key
benefit of an IWSES plant is the reduction in O&M expenses. Fewer
maintenance personnel are required to maintain an integrated wind-solar plant
compared with those needed to maintain separately located wind and solar
plants. The skill sets needed to maintain wind and solar plants are similar—in
particular, the skills required to maintain power conversion equipment and the switchyard.
On the operations side, savings could be achieved by sharing the control room,
communication infrastructure, and personnel necessary for plant operation. Benefits
from Integrating a BESS A BESS connected to the bulk power system can provide several
generation- and transmission-level services such as energy arbitrage, reserves
regulation, transmission congestion management, and transmission asset
deferral. With aggregation and suitable regulatory changes, a customer-sited
BESS can perform additional customer-level services as well as upstream services
such as distribution-level and bulk power system services. However, a BESS
integrated with a wind solar plant can also provide many plant-level services
mandated by grid codes or regulations or incentivized through markets. The
stacking of plant-level and system-level services can help derive more value
from energy storage. A BESS integrated with a wind-solar plant can provide many
of the same system-level services as those provided by a standalone BESS, such
as the following:
✔ Energy shifting: Energy storage is frequently used for energy arbitrage or
load shifting. In the United States, pumped storage hydro plants were installed
mainly to integrate nuclear base-load power plants, charging at night when
energy prices are low and generating during the day when prices are high.
Battery energy storage has been used to provide energy during peak load hours
in several niche scenarios. It will be possible to use battery energy storage
to a greater extent as costs decline in the future.
✔ Ancillary services: Frequency regulation (i.e., the rapid adjustment in
generation or load to balance the system in the seconds-to-minutes time frame)
is one of the few applications that can provide full cost recovery for a commercial
battery storage project. However, ancillary services such as regulation are not
explicitly procured and compensated for in all systems. BESSs have recently found
a niche in balancing ancillary services such as fastfrequency response. A BESS
can also be used to provide black-start capability to the system and support
voltage on the bulk power system in addition to balancing ancillary services,
as discussed previously.
✔ Transmission and distribution congestion management and upgrade deferral:
Depending on its location, a BESS can alleviate transmission congestion by
serving the demand in the transmission-constrained area. The BESS can be used
to defer the need for transmission or distribution system upgrades by
time-shifting energy from low- to peak-demand periods. A BESS integrated with a
wind-solar plant can provide additional plant-level services. These plant-level
services may help the plant comply with grid codes or operational requirements.
While these services can be provided by a BESS that is not integrated with the
plant, more control and coordination would likely be required to achieve the
same results.
✔ Compliance with technical standards: The grid codes in various countries
require wind and solar plants to limit their ramp rates and, in some cases,
also provide primary frequency response. It may be economical for an integrated
BESS to provide these services rather than curtail the generation of the wind
and solar plants for this purpose.
✔ Compliance with operational requirements: In addition to facilitating wind
and solar plants’ compliance with technical standards, a BESS can also make it
easier for a plant to meet operational requirements, for instance, in the
firming (reducing forecast deviations) of wind and solar generation. In some
countries, wind plants are penalized if generation deviates from the forecast
by more than a certain percentage. As it reduces forecast errors, a BESS can be
used to lower the penalties imposed on wind plants.
Recently, a techno-economic
feasibility study for IWSES plants was completed for two sites in India
(Ananthpur, Andhra Pradesh; and Kutch, Gujarat). This study was performed for
the Infrastructure Leasing & Financial Services Energy Development Company
Limited, a developer and financier of renewable projects in India under a
technical assistance grant from the U.S. Trade and Development Agency. The
deliverables of this study were as follows:
✔ developing the technical design of the IWSES plant, including the
transmission evacuation plan
✔ developing use cases for integrated energy storage appropriate for the
Indian system
✔ sizing energy storage to provide multiple plant and system-level services
✔ performing cost-benefit analyses of IWSES plants under existing and
proposed regulations
✔ preparing high-level environmental and social impact analysis guidelines
✔ outlining the financing plan for the projects using viability gap funding
and other incentives to support infrastructure projects that are economically
justified but fall short of financial viability
✔ recommending regulatory and policy changes required to integrate energy
storage in India
✔ conducting a reverse trade mission for Indian regulators and policy makers
to learn about the energy storage business in the United States. The steps
related to the technical design of the IWSES are summarized in the following
sections, and examples from the two project sites are also provided.
To accurately
assess the wind resource at a project site, it is important to obtain on-site
measurements of wind speed and other relevant meteorological parameters. The
ideal number of measurement locations is site specific. For the two project
sites, the monitoring mast data were quality checked, adjusted to the long-term
characteristics, and sheared up to the turbine hub height.
The wind frequency distribution across the
site was determined by combining these data with a wind resource grid file.
This information was used for turbine siting and energy production estimation. A
geographical information system (GIS)-based approach was used to identify the
buildable area suitable for siting turbines within the project, based on land
use and other constraints. Offsets were applied to various categories of land
use to define exclusion areas. Depending on the features within and surrounding
the project, the offset categories included property boundaries, residences, roads,
transmission lines, wetlands, streams, protected lands, high slopes, or
features that might affect the placement of turbines.
The wind turbines were then optimally placed
within the buildable area to maximize energy and minimize wake losses after
taking the necessary interturbine spacing into account. The optimization was
accomplished using the commercial wind farm design tool Openwind. Upon
completion of the optimization, some turbines were resited to improve project
construction and interconnection.
Good quality
measurement of irradiance data is required to perform a solar resource
assessment. If usable measurements are not available, it is common industry
practice to use high-quality satellite-modeled data sets. For the two project sites,
satellite-modeled data were used to achieve the lowest resource uncertainty.
The long-term period of record from the
SolarGIS database was used to estimate the long-term resource and create a
typical meteorological year (TMY). The TMY represents long-term resource,
temperature, and wind speed at the site on an hourly basis during a typical
year.
Based on available land area, setbacks, site
conditions, and industry-standard practices, optimal configurations were developed
for two PV solar technologies: crystalline and thin film. A number of
parameters were considered in the design, such as tilt angle, azimuth,
collector length, pitch, and dc–ac ratio. The main objective of the
optimization was to maximize the capacity factor and minimize shading losses,
while still achieving a cost-effective design that would meet the capacity
targets set for the sites.
The first step in the system design was to
optimize the tilt angle to the site latitude. The collector length was sized
based on module dimensions and industry-standard mounting approaches. The array
azimuth was set to true south to optimize energy production.
A pitch/shading analysis was conducted to determine
the optimal design of the PV configurations prior to energy simulation. The
result of this analysis was used to produce a plot of shading loss as a
function of pitch (the space between rows). The pitch/shading analysis resulted
in groundcover ratios dependent on the latitude. Next, the solar arrays were
laid out to optimize land use, minimize intrusion into exclusion zones, and
minimize the potential shading loss caused by proximity to the turbines. The solar
arrays were set back from all exclusion zones and boundaries with a 10-m
(33-ft) buffer zone. The buffer zones provide setbacks from sensitive areas and
a means of access to the solar equipment.
Roads of 10-m width were outlined through and around
the solar plants to provide access.
A dc–ac ratio of 1.3 was used for the
configurations. This dc–ac ratio represents an industry standard that is value optimized
for current equipment costs to increase energy production and minimize energy
cost. The oversized dc array minimizes the losses upstream of the inverter,
which allows for greater energy production at low irradiance and increases the
ac capacity factor.
This initial phase took the design to the
point where the mutual interaction of the wind and solar equipment required
further consideration.
When designing an
integrated plant, it is important to ensure that the interaction between the
wind and solar plants is captured and taken into consideration. There are two
principal types of interactions that must be modeled: 1) the effect of the
turbines on the PV arrays and 2) the effect of the PV arrays on the wind-flow
field. The Effect of PV Arrays on Wind Turbine Siting At both project locations,
the siting of the turbines was carried out first because the wind resource was
more variable across the site than the solar resource; therefore, the optimum configuration
of wind turbines is more sensitive to their siting than that of the solar
array.
The presence of the PV panels can impact the
surface roughness and affect the wind flow through the turbine array.
An array of solar panels can be similar to other
topographic or locational features (trees, buildings, etc.), in that it alters
the surface roughness and affects wind flow through the array. Increased
surface roughness changes the profile of the atmospheric boundary layer as it
flows across the array, increasing the shear effect.
The approach for roughness modeling, turbine
wakes, and the impact on wind flow is based on a theory advanced by Sten
Frandsen (see “For Further Reading”) that defines wind farm equivalent
roughness. Frandsen stipulates that an infinite array of wind turbines is
represented as a region of uniform high-surface roughness.
The roughness imposes drag on the atmosphere,
causing both a downstream change in the structure of the boundary layer and a
reduction in the free-stream wind speed at the turbine hub height. Once the
equivalent roughness is defined, it is possible to calculate the hub-height wind
speed deep within the array, where the boundary layer has reached equilibrium
with the array roughness. At the project sites, the effect was comparable to
that of the local vegetation. The wind-flow field model considered the local
roughness, assuming that the panels would closely resemble the roughness of
local vegetation to account for this effect.
The difference in the comparative hub height
of wind turbines and PV arrays generally yields a small roughness effect on the
wind turbines by the arrays. Solar panels may have a significant impact on
local roughness if larger foliage (e.g., tall trees) will likely be removed to
make room for solar arrays or if a significant area upwind of the turbines in
the primary wind direction is expected to be covered with solar arrays.
However, this type of extreme scenario did not
apply for this project. The Effect of Wind Turbines on PV Siting The most
obvious way that solar arrays can be impacted by the location of wind turbines
comes from the shadow that they cast.
This shading effect depends on the time of day
and the orientation of the turbines with respect to the panels. Shadowflicker
analysis was adapted to assess the level of shading impact at receptor
locations across the solar arrays.
Using this customized methodology, the shading
fraction at each receptor was calculated for every hour. The affected
equivalent array time was computed as a function of the distance between
turbines and panels, the wind direction affecting turbine yaw, and the sun’s
local zenith and azimuth angle.
The affected equivalent array time was
computed to give a turbine-shading loss estimate. Site-specific solar resource data
were used to estimate the solar energy loss. The turbineshading energy loss
also accounts for diffuse light that is still present during shading at any
time of the day, even though the direct component may be blocked by the turbine.
This means that turbine shading does not
result in a total loss but only a power reduction during array-affected times.
Due to the electrical effect of modules in
series and the large reach of a single wind turbine blade shadow, the analysis
assumed that a shaded reflector corresponded to the presence of diffuse
irradiance on only that portion of the array for the hourly fraction
represented by the model.
The analysis further assumed that the
reduction in incident irradiation on the solar array was proportional to the
energy loss expected for turbine shading. Energy modeling was then used to
calculate the impact of the turbine shading on the solar production.
Partly because the effect of the shadows was
most pronounced at times of relatively low solar production, it was found that
the shading effect of the turbines was lower than 1% of annual production.
Diffuse irradiation was still present even
during times of heavy shading. Optimal Wind and Solar Configuration It possible
to arrive at an optimal configuration for both the turbine and solar array
layouts only after 1) all site constraints have been defined and met, 2) the
time-dependent mutual interactions of the turbines and panels have been built
into the energy calculations, and 3) the wind and solar resources have been
accurately captured and modeled.
The wind and solar capacity for each phase of
the project was based on the resource assessment.
An additional levelized cost analysis was
performed to determine the inverter configuration as well as the make and model
of the PV panels and wind turbines to be employed for each phase. Table 1 shows
the wind and solar capacity by project phase
Development of
Energy Storage Use Cases The first step in the BESS design process was to
identify the uses of energy storage. Based on a survey of the needs of the system
as well as the expected revenues in the near and long term, seven applications
were identified to be of value. These were then grouped into primary and
secondary applications. The primary applications are those for which there is
an immediate need for storage or a payment mechanism exists. The secondary
applications are those for which a BESS will add value to the system but this
value can only be realized based on changes to existing regulations. Primary
Applications
✔ Penalty charge management under the deviation
settlement mechanism (DSM): The first primary application of energy storage
addresses financial penalty reduction in the DSM. In India, each state has a
state load dispatch center (SLDC) responsible for planning the operation of
internal generation to meet the forecasted load. The SLDCs also coordinate the
scheduling of interstate generators (generators designated to meet the load of
more than one state) with the help of the regional load dispatch center. Any
deviations from day-ahead interstate schedules are determined on a 15-min
basis, and penalties and incentives associated with these deviations are
determined according to the system condition (frequency) prevailing at that
time of deviation. In general, when the frequency is below 50.05 Hz and if the
state underproduces or overdraws, there is a penalty. Conversely, when the
frequency is below 50.05 Hz and if the state overproduces or underdraws, there
is an incentive. Figure 3 shows how a BESS can be discharged during periods
when the frequency is below 50.5 Hz and the state is withdrawing more than the
scheduled energy, as well as how it can be charged during other periods to
reduce the DSM penalty.
✔ Shifting renewable energy to peak demand hours: Another primary
application of energy storage is shifting renewable energy to peak demand
hours. This application was identified as key for the state of Andhra Pradesh,
which has a high target (8,100 MW of wind and 4,081 MW of solar) for renewable
energy. During off-peak load hours (8:00 a.m.–2:00 p.m.), the BESS charges and
stores energy, which it discharges during peak load hours (6:00–10:00 p.m.). Thus,
the integrated plant supplies a consistent amount of energy to customers during
peak hours through a combination of wind, solar, and energy storage.
✔ Forecast deviation reduction for wind and solar plants: Another of the
primary applications energy storage can provide is forecast deviation reduction
for wind and solar plants. The concept is that energy storage can be dispatched
to reduce a renewable energy plant’s forecast errors and thus prevent lost
revenues. In India, there is a tiered penalty structure for deviations in wind
and solar generation. For example, if the wind generation is below the
scheduled value in a 15-min block by 25%, then a deviation charge applies: 110%
of the fixed rate for balance energy beyond 15% and up to 25% (i.e., there is
no deviation charge for the first 15%).
The secondary use
cases of the BESS include 1) ramp management of the integrated wind-solar
plant, 2) fast frequency response, 3) primary frequency response, and 4) ramp
management for the system during morning and afternoon ramps.
Sizing of the BESS The generic BESS design
process is depicted in Figures 4–6.
Figure 4 illustrates the first step in the
evaluation, during which application-specific battery power command is
developed. As an example, from a native wind generation profile and forecast, application-specific
controls would generate the storage power needed to bring the total (wind and
storage) generation closer to the forecast. By combining these profiles with
local market rules and prices, revenue streams (or avoided penalties) can be calculated
for the specific application and market.
Next, as shown in Figure 5, energy storage
technologies are selected and sized by processing the battery power command
into a use-intensity map. Relying on a database of physics-based performance
models, the installed cost to meet the application requirements is calculated
for each technology, including storage and BOP.
Finally, as shown
in Figure 6, the asset life is determined from the duty profile, asset type,
and size by using asset-specific calendar and cycle-life models. Combining all
calculations, project economics can be estimated based on revenue, initial
cost, and life (number of replacements over the project life).
This analysis used a database of energy
storage technologies built at GE Global Research for the comparison and
evaluation of various energy storage assets available in today’s market.
The underlying asset database includes equivalent
circuit models of over 100 energy storage assets spanning a wide range of
technologies and options. These energy storage equivalent-circuit models were
developed from a combination of data sheets and follow-up discussions with
suppliers to confirm model results. For the primary and secondary use cases
described previously, the key input for determining the charge/discharge profile
of the BESS was the simulated 15-min wind and solar profiles and the corresponding
short-term forecasts for each phase of the project.
Other key model inputs include wind and
solar tariffs, the DSM penalty structure, and ancillary service tariffs. Based
on the technical design and financial analysis, a 10-MW, 15-MWh battery lithium
manganese cobalt-oxide BESS was recommended for the phase 1-a 41-MW integrated wind-solar
project. This BESS was designed primarily to perform the DSM penalty reduction
function but, when not doing so, could also be used for other primary and secondary
applications.
BESS sizing for the remaining phases of the
project will depend significantly on future energy storage-related regulations
and the cost of the BESS. For each of the remaining phases of the Ananthpur and
Kutch projects, the BESS ranged in size from 10–60 MW, with a capacity of
30–280 MWh. Design of the Collector System The collector system was designed
using a mixture of industry best practices for wind and PV collector systems.
It is common practice in India to use overhead conductors for standalone wind
plants and underground cables for PV solar plants. A main consideration in the
collector design was
whether underground cables
should be used for the collector system or a mixed-construction approach using
both underground cables and overhead conductors would be preferable. Designs
for both cases were compared to evaluate these options. For the
mixed-construction case, underground cable was used to feed the PV systems or
whenever the conductor path entered a wind turbine exclusion zone.
In this case, riser poles were used to
transition between underground and overhead construction.
All underground design was used for phase 1-a
of the project to account for the tight arrangement of PV arrays and wind
turbine exclusion zones, even though the cost was slightly higher than with the
mixed-construction approach. For phases with overhead transmission lines, the
routing was designed such that there was no loss of PV generation due to shadowing
from the lines.
To evacuate the power, a 33-KV/220-KV pooling
substation was designed for each phase of the project.
The BESS was sited at the pooling substation
and connected to the low-voltage side of the substation.
The BESS was sited at the pooling substation
and connected to the low-voltage side of the substation.
The final design
step was the design of the controls and communications system that enables the
BESS to communicate with the plant’s wind turbines and solar arrays and also
enables the SLDC and Renewable Energy Management Center to perform the various
applications.
The proposed plant control architecture
consists of three control device types:
✔ Plant energy management system: This system orchestrates all control and
communication actions and provides the interface to the utility and SLDC.
✔ Plant master control station: This station is a
hardened computer that hosts critical real-time control services. Its main
function is to compute the set point for the storage system’s power, i.e. the
command for the BESS charging or discharging cycles. It also computes the plant-level
commands for the wind and solar plants.
✔ Plant edge control station: The edge plant-level
controllers are hardened industrial controllers located at the BESS as well as
at each wind or solar plant. These units control the actual power flow at the
point of interconnection based on local measurements and commands received from
the master-level controller. The plant edge controllers are typically
multifunction controllers that provide the physical interface to the lower inverter-level
controllers and sensors.
The next step is
to develop the 41-MW IWSES plant as a grant-funded pilot project. The lessons
learned from the demonstration project can be used to develop the technical
design, commercial arrangements, and energy storagerelated policies and
regulations that will have a strong influence on future energy storage projects
in India.