[0001] The present disclosure relates generally to renewable energy power systems, and more particular to systems and methods for optimizing power production of renewable power systems based on weather forecast data.
BACKGROUND
[0002] Wind power is considered one of the cleanest, most environmentally friendly energy sources presently available, and wind turbines have gained increased attention in this regard. A modern wind turbine typically includes a tower, generator, gearbox, nacelle, and one or more rotor blades. The rotor blades capture kinetic energy of wind using known airfoil principles. For example, rotor blades typically have the cross-sectional profile of an airfoil such that, during operation, air flows over the blade producing a pressure difference between the sides. Consequently, a lift force, which is directed from a pressure side towards a suction side, acts on the blade. The lift force generates torque on the main rotor shaft, which is geared to a generator for producing electricity. In addition, a plurality of the wind turbines may be arranged in a predetermined geological location and electrically connected together to form a wind farm.
[0003] During operation, wind impacts the rotor blades of the wind turbine and the blades transform wind energy into a mechanical rotational torque that rotatably drives a low-speed shaft. The low-speed shaft is configured to drive the gearbox that subsequently steps up the low rotational speed of the low-speed shaft to drive a high-speed shaft at an increased rotational speed. The high-speed shaft is generally rotatably coupled to a generator so as to rotatably drive a generator rotor. As such, a rotating magnetic field may be induced by the generator rotor and a voltage may be induced within a generator stator that is magnetically coupled to the generator rotor. In certain configurations, the associated electrical power can be transmitted to a turbine transformer that is typically connected to a power grid via a grid breaker. Thus, the turbine transformer steps up the voltage amplitude of the electrical power

such that the transformed electrical power may be further transmitted to the power grid.
[0004] In many wind turbines, the generator rotor may be electrically coupled to a bi-directional power converter that includes a rotor side converter joined to a line side converter via a regulated DC link. More specifically, some wind turbines, such as wind-driven doubly-fed induction generator (DFIG) systems or full power conversion systems, may include a power converter with an AC-DC-AC topology. [0005] For renewable power sources, such as the wind turbine power system described above, forecasting weather patterns or power and load requirements is essential to providing stable power production, since the fuel supply (e.g. wind or sunlight) is not consistently available. In areas where constant power output from a renewable power source is required, renewable power sources can be dramatically curtailed to ensure that the power output is not impacted by potential variability of the renewable source. However, such curtailment results in a general reduction of the power output. In cases where the variability of the renewable power source is not well known, the curtailment of the site can be dramatic, thereby leading to a significant loss in revenue or higher risks of penalties imposed for violating power commitments.
[0006] In view of the foregoing, systems and methods for optimizing power production of renewable power systems based on weather forecast data would be welcomed in the art. Accordingly, the present disclosure is directed to systems and methods for estimating a confidence factor for wind forecast data and adjusting the power output based on the confidence factor.
BRIEF DESCRIPTION
[0007] Aspects and advantages of the invention will be set forth in part in the following description, or may be obvious from the description, or may be learned through practice of the invention.
[0008] In one aspect, the present disclosure is directed to a method for optimizing power production of a renewable power facility. The method includes receiving, by a control device having one or more processors and one or more memory devices, a forecast of the renewable power facility, such as a power forecast, a weather forecast,

or a pricing forecast. The method also includes dynamically determining, by the
control device, a confidence factor of the forecast as a function of, at least one, one or
more environmental factors of the renewable power facility. Further, the method
includes determining one or more power commands of the renewable power facility
based on the forecast and the confidence factor. Moreover, the method includes
controlling the renewable power facility based on the one or more power commands
so as to dynamically adjust an amount of power held in reserve at the renewable
power facility.
[0009] In an embodiment, the method may include determining the confidence
factor of the forecast as a function of at least one of historical performance of the
renewable power facility and a desired level of risk.
[0010] In one embodiment, the function may include at least one of a weighted
mathematical function setting forth a gain for each of the one or more environmental
factors or a machine learning algorithm that adjusts an effect of the one or more
environmental factors based on performance over an extended time period.
[0011] In another embodiment, the environmental factor(s) may include at least
one of time of day, season, cloud index, temperature, availability of the renewable
power facility, losses of the renewable power facility, uncertainty of the function,
wind speed, turbulence intensity, wind shear, wind direction, air density, wake, and
power output changes.
[0012] In further embodiments, the method may include evaluating the forecast
by identifying forecasted weather patterns within the forecast.
[0013] In additional embodiments, the method may include applying statistical
analysis to the forecast of the renewable power facility to obtain a forecast
distribution. In such embodiments, the statistical analysis may include, for example,
mean, median, standard deviation, probability distribution, or combinations thereof.
[0014] Thus, in an embodiment, determining the confidence factor of the forecast
may include defining upper and lower bounds of the forecast distribution and
determining a variance of the forecast distribution.
[0015] In yet another embodiment, controlling the renewable power facility based
on the one or more power commands so as to adjust the amount of power held in
reserve at the renewable power facility may include increasing a power level held in

reserve of the renewable power facility when the confidence factor is below a threshold and maintaining or decreasing the power level held in reserve of the renewable power facility when the confidence factor is above a threshold. [0016] In one embodiment, the control device may be a supervisory controller of the renewable power facility. In such embodiments, the method may include sending, via the supervisory controller, the confidence factor to a facility controller of the renewable power facility, determining, via the facility controller, the one or more power commands of the renewable power facility based on the forecast and the confidence factor, and sending the one or more power commands to an asset controller of the renewable power facility.
[0017] In several embodiments, the renewable power facility may include a wind farm having one or more wind turbines, a solar power plant having one or more solar power systems, a hydro power plant, an energy storage system, or combinations thereof as well as any other suitable renewable power facility. [0018] In another aspect, the present disclosure is directed to a system for optimizing power production of a renewable power facility. The system includes a control device comprising one or more processors and one or more memory devices. The control device is configured to perform a plurality of operations, including but not limited to receiving a forecast of the renewable power facility, dynamically determining a confidence factor of the forecast as a function of, at least one, one or more environmental factors of the renewable power facility, determining one or more power commands of the renewable power facility based on the forecast and the confidence factor, and controlling the renewable power facility based on the one or more power commands so as to dynamically adjust an amount of power held in reserve at the renewable power facility.
[0019] In one aspect, the present disclosure is directed to a method for forecasting power production of a renewable power facility to improve power production of the renewable power facility. The method includes receiving a power forecast of the renewable power facility, determining a confidence factor of the power forecast as a function of, at least one, one or more environmental factors of the renewable power facility, and adjusting a power reserve of the renewable power facility based on the power forecast and the confidence factor.

[0020] Variations and modifications can be made to these example embodiments of the present disclosure.
[0021] These and other features, aspects and advantages of the present invention will become better understood with reference to the following description and appended claims. The accompanying drawings, which are incorporated in and constitute a part of this specification, illustrate embodiments of the invention and, together with the description, serve to explain the principles of the invention.
BRIEF DESCRIPTION OF THE DRAWINGS
[0022] A full and enabling disclosure of the present invention, including the best
mode thereof, directed to one of ordinary skill in the art, is set forth in the
specification, which makes reference to the appended figures, in which:
[0023] FIG. 1 illustrates a schematic diagram of one embodiment of a wind
turbine system according to the present disclosure;
[0024] FIG. 2 illustrates a schematic diagram of one embodiment of a wind farm
having a plurality of wind turbines according to the present disclosure;
[0025] FIG. 3 illustrates a schematic diagram of one embodiment of a control
device of a wind turbine according to the present disclosure;
[0026] FIG. 4 illustrates a flow diagram of one embodiment of a method for
optimizing power production of a renewable power facility according to the present
disclosure;
[0027] FIG. 5 illustrates a block diagram of one embodiment of a system for
optimizing power production of a renewable power facility according to the present
disclosure;
[0028] FIG. 6 illustrates a timeline of one embodiment of historical performance
of a weather service forecast according to the present disclosure;
[0029] FIG. 7 illustrates a timeline of one embodiment of bucketing forecast
performance into various environmental conditions according to the present
disclosure;
[0030] FIG. 8 illustrates a graph of one embodiment of the forecast distribution
for facility power according to the present disclosure;
[0031] FIG. 9 A illustrates a graph of the power forecast with respect to time

according to conventional construction;
[0032] FIG. 9B illustrates a graph of one embodiment of the power forecast with
respect to time and probability according to the present disclosure;
[0033] FIG. 10 illustrates a graph of one embodiment of the power with respect to
time according to the present disclosure, particularly illustrating the improvement in
AEP due to increased forecasting accuracy.
DETAILED DESCRIPTION
[0034] Reference now will be made in detail to embodiments of the invention, one or more examples of which are illustrated in the drawings. Each example is provided by way of explanation of the invention, not limitation of the invention. In fact, it will be apparent to those skilled in the art that various modifications and variations can be made in the present invention without departing from the scope or spirit of the invention. For instance, features illustrated or described as part of one embodiment can be used with another embodiment to yield a still further embodiment. Thus, it is intended that the present invention covers such modifications and variations as come within the scope of the appended claims and their equivalents. [0035] Generally, example aspects of the present disclosure are directed to systems and methods for optimizing power ramping of renewable power systems based on weather forecast data. More specifically, the present disclosure provides a system and method for generating a confidence factor for given weather forecast data for a power asset, such as a wind turbine power system. By using the current weather forecast data, provided by either an internal or external source and estimating how accurate that forecast is based on a number of performance metrics and a desired level or risk, the confidence factor can be used to adjust the amount of power held in reserve. For example, the higher the confidence factor, the less excessive the curtailment may be, thereby improving power production.
[0036] Referring now to the drawings, FIG. 1 illustrates one embodiment of a wind turbine system 100 according to the present disclosure. Example aspects of the present disclosure are discussed with reference to the wind turbine system 100 of FIG. 1 for purposes of illustration and discussion. Those of ordinary skill in the art, using the disclosures provided herein, should understand that example aspects of the present

disclosure are also applicable in other power systems, such as synchronous, asynchronous, permanent magnet, and full-power conversion wind turbines, solar, hydro, an energy storage system, gas turbine, or other suitable power generation systems.
[0037] Furthermore, as shown, the wind turbine system 100 includes a rotor 106 includes a plurality of rotor blades 108 coupled to a rotating hub 110. The rotor 106 is coupled to an optional gearbox 118, which is, in turn, coupled to a generator 120. In accordance with aspects of the present disclosure, the generator 120 is a doubly fed induction generator (DFIG) 120.
[0038] The DFIG 120 can include a rotor and a stator. Further, as shown, the DFIG 120 is typically coupled to a stator bus 154 and a power converter 162 via a rotor bus 156. The stator bus 154 provides an output multiphase power (e.g. three-phase power) from a stator of the DFIG 120 and the rotor bus 156 provides an output multiphase power (e.g. three-phase power) of a rotor of the DFIG 120. Referring to the power converter 162, the DFIG 120 is coupled via the rotor bus 156 to a rotor side converter 166. The rotor side converter 166 is coupled to a line side converter 168 which in turn is coupled to a line side bus 188.
[0039] In example configurations, the rotor side converter 166 and the line side converter 168 are configured for normal operating mode in a three-phase, pulse width modulation (PWM) arrangement using insulated gate bipolar transistor (IGBT) or similar switching elements. The rotor side converter 166 and the line side converter 168 can be coupled via a DC link 136 across which is the DC link capacitor 138. In an embodiment, a transformer 178, such as a three-winding transformer, can be coupled to the line bus 188, the stator bus 154, and a system bus 160. The transformer 178 can convert the voltage of power from the line bus 188 and the stator bus 154 to a voltage suitable for providing to an electrical grid 184 via system bus 160. [0040] The power conversion system 162 can be coupled to a control device 174 to control the operation of the rotor side converter 166 and the line side converter 168. It should be noted that the control device 174, in typical embodiments, is configured as an interface between the power conversion system 162 and a turbine/asset controller 176. In one implementation, the control device 174 can include a processing device (e.g. microprocessor, microcontroller, etc.) executing computer-

readable instructions stored in a computer-readable medium. The instructions when executed by the processing device can cause the processing device to perform operations, including providing control commands (e.g. pulse width modulation commands) to the switching elements of the power converter 162 and other aspects of the wind turbine system 100.
[0041] In operation, alternating current power generated at the DFIG 120 by rotation of the rotor 106 is provided via a dual path to electrical grid 184. The dual paths are defined by the stator bus 154 and the rotor bus 156. On the rotor bus side 156, sinusoidal multi-phase (e.g. three-phase) alternating current (AC) power is provided to the power converter 162. The rotor side power converter 166 converts the AC power provided from the rotor bus 156 into direct current (DC) power and provides the DC power to the DC link 136. Switching elements (e.g. IGBTs) used in bridge circuits of the rotor side power converter 166 can be modulated to convert the AC power provided from the rotor bus 156 into DC power suitable for the DC link 136.
[0042] The line side converter 168 converts the DC power on the DC link 136 into AC output power suitable for the electrical grid 184, such as AC power synchronous to the electrical grid 184, which can be transformed by the transformer 178 before being provided to the electrical grid 184. In particular, switching elements (e.g. IGBTs) used in bridge circuits of the line side power converter 168 can be modulated to convert the DC power on the DC link 136 into AC power on the line side bus 188. The AC power from the power converter 162 can be combined with the power from the stator of DFIG 120 to provide multi-phase power (e.g. three-phase power) having a frequency maintained substantially at the frequency of the electrical grid 184 (e.g. 50 Hz/60 Hz).
[0043] The power converter 162 can receive control signals from, for instance, the system 174. The control signals can be based, among other things, on sensed conditions or operating characteristics of the wind turbine system 100. Typically, the control signals provide for control of the operation of the power converter 162. For example, feedback in the form of sensed speed of the DFIG 120 can be used to control the conversion of the output power from the rotor bus 156 to maintain a proper and balanced multi-phase (e.g. three-phase) power supply. Other feedback

from other sensors can also be used by the controller 174 to control the power converter 162, including, for example, stator and rotor bus voltages and current feedbacks. Using the various forms of feedback information, switching control signals (e.g. gate timing commands for IGBTs), stator synchronizing control signals, and circuit breaker signals can be generated.
[0044] Various circuit breakers and switches, such as a line bus breaker 186, stator bus breaker 158, and grid breaker 182 can be included in the system 100 to connect or disconnect corresponding buses, for example, when current flow is excessive and can damage components of the wind turbine system 100 or for other operational considerations. Additional protection components can also be included in the wind turbine system 100.
[0045] Referring now to FIG. 2, the wind turbines 100 may be arranged together in a common geographical location known as a wind farm 200 and connected to the electrical grid 184. More specifically, as shown, each of the wind turbines 100 may be connected to the electrical grid 184 via their respective transformers 178. Further, as shown, clusters 206 of wind turbines 100 in the wind farm 200 may be connected to the electrical grid 184 via a substation transformer 202. Thus, as shown, the wind farm 200 may also include a farm/facility controller 204 for controlling operation thereof. Moreover, as shown, the facility controller 204 may also be communicatively coupled to a supervisory controller 208.
[0046] Referring now to FIG. 3, a block diagram of one embodiment of a control device 310 according to example embodiments of the present disclosure is illustrated. It should be understood that the control device/system 310 can be, for example, the converter controller 174, the asset/turbine controller 176, the facility/farm controller 204, and/or the supervisory controller 208. Thus, in some embodiments, the control device(s) 310 can include one or more processor(s) 312 and one or more memory device(s) 314. The processor(s) 312 and memory device(s) 314 can be distributed so that they are located at one more locales or with different devices. [0047] The processor(s) 312 and memory device(s) 314 can be configured to perform a variety of computer-implemented functions and/or instructions (e.g., performing the methods, steps, calculations and the like and storing relevant data as disclosed herein). The instructions when executed by the processor(s) 312 can cause

the processor(s) 312 to perform operations according to example aspects of the present disclosure. For instance, the instructions when executed by the processor(s) 312 can cause the processor(s) 312 to implement the methods discussed herein. [0048] Additionally, the control device 310 can include a communication interface 316 to facilitate communications between the control device 310 and various components of a wind turbine system, wind farm, or power system, including reactive power production requirements or sensed operating parameters as described herein. Further, the communication interface 318 can include a sensor interface 318 (e.g., one or more analog-to-digital converters) to permit signals transmitted from one or more sensors 320, 322 to be converted into signals that can be understood and processed by the processor(s) 312. It should be appreciated that the sensors (e.g. sensors 320, 322) can be communicatively coupled to the communications interface 318 using any suitable means, such as a wired or wireless connection. The signals can be communicated using any suitable communications protocol. The sensors (320, 322) can be, for example, voltage sensors, current sensors, power sensors, DFIG rotational speed sensors, temperature sensors, or any other sensor device described herein. [0049] As such, the processor(s) 312 can be configured to receive one or more signals from the sensors 320, 322. For instance, in some embodiments, the processor(s) 312 can receive signals indicative of a voltage or current from the sensor 320. In some embodiments, the processor(s) 312 can receive signals indicative of temperature (e.g. DFIG temperature, line side converter temperature) from sensor 322.
[0050] As used herein, the term "processor" refers not only to integrated circuits referred to in the art as being included in a computer, but also refers to a control device, a microcontrol device, a microcomputer, a programmable logic control device (PLC), an application specific integrated circuit, and other programmable circuits. Additionally, the memory device(s) 314 can generally include memory element(s) including, but not limited to, computer readable medium (e.g., random access memory (RAM)), computer readable non-volatile medium (e.g., a flash memory), a compact disc-read only memory (CD-ROM), a magneto-optical disk (MOD), a digital versatile disc (DVD) and/or other suitable memory elements. Such memory device(s) 314 can generally be configured to store suitable computer-readable instructions that, when

implemented by the processor(s) 312, configure the control device 310 to perform the various functions as described herein.
[0051] Referring now to FIGS. 4 and 5, a method 400 and system 500 for optimizing power production of a renewable power facility, such as the wind turbine power system 100, are illustrated. More particularly, FIG. 4 illustrates a flow diagram of one embodiment of a method 400 for optimizing power production of a renewable power facility, whereas illustrates FIG. 5 illustrates a schematic diagram of one embodiment of a system 500 for optimizing power production of a renewable power facility. The method 400 can be implemented by a control device, such as a control device 310. In addition, FIG. 4 depicts steps performed in a particular order for purposes of illustration and discussion. Those of ordinary skill in the art, using the disclosures provided herein, will understand that various steps of any of the methods disclosed herein can be adapted, omitted, rearranged, or expanded in various ways without deviating from the scope of the present disclosure.
[0052] As shown at (402), the method 400 includes receiving, by a control device, a forecast of the renewable power facility 100, such as a power forecast, a weather forecast, a pricing forecast, or similar or combinations thereof. For example, as shown in FIG. 5, the supervisory controller 208 receives the forecast 502 and current conditions 504, such as one or more environmental factors. For example, in certain embodiments, the environmental factor(s) may include at least one of time of day, season, cloud index, temperature, availability of the renewable power facility, losses of the renewable power facility, uncertainty of the function, wind speed, turbulence intensity, wind shear, wind direction, air density, wake, and power output changes. [0053] More particularly, the environmental factor(s) may include any factors that influence the confidence factor described herein. Therefore, such environmental factor(s) may be associated with a forecast in general and/or may also be associated with a specific type of power generate resource (e.g. wind, solar, hydro, energy storage, etc.). For example, environmental factor(s) that are associated with the forecast in general may include forecast look-ahead time or forecast resolutions. Forecast look-ahead time generally refers to the forward-looking horizon width for forecasts. In general, the uncertainty associated with forecasts that are further away from current time are less accurate compared to more immediate periods. The cause

of such inaccuracy is due to multiple factors. For example, in data-driven approaches, where time series based predictions are made, the autocorrelation indicates that the confidence level magnitude drops with increase in prediction horizon. In addition, in approaches that use exogenous inputs (other predicted inputs), typically weather information, the uncertainty of those inputs influences the power predictions. [0054] Forecast resolution generally refers to the averaging period involved in a forecast time block. For example, a 60-minute look-ahead forecast can be submitted at 15-minute resolutions (e.g. four forecast points). Given that the look-ahead period is held constant, higher forecast resolution will have a greater level of uncertainty associated with the predictions. The reason for such a result is due to the lack of data smoothing when averaging is performed over a longer window of time. [0055] Another environmental factor(s) associated with the forecast includes forecast service accuracy, which is illustrated in FIG. 6. For example, as shown, when using external sources for forecasting, this factor evaluates the historical performance of the service. For each outlook (e.g. +6 hours), the forecast is recorded. After 6 hours has passed, the current weather conditions experienced by the wind turbines are compared against the previously generated forecast and an accuracy metric is assigned to that timeslot (e.g. 6-hour forecast error). Accordingly, this factor can be recalculated for every potential outlook available from the weather service, allowing the facility 100 to provide varying confidences in hourly outlooks. [0056] Referring to FIG. 7, in further embodiments, the method 400 may include evaluating the forecast by identifying forecasted weather patterns within the forecast. In other words, the output of the forecasting performance evaluation from FIG. 6 can be further analyzed to bucket the performance into different forecasted weather patterns or other factors such as time of day, current season or environmental conditions that can be locally identified (e.g. thunderstorm, heavy rain, overcast, light rain, and/or clear skies as examples). Thus, in such embodiments, the weighted average of the error associated with each type of bucket can improve the estimate of the forecast confidence factor described herein. More specifically, when current conditions show high variation in wind speed and site power output, it can be expected that the volatility will continue into the near future. This uncertainty in the forecast can be included in the confidence factor calculation. In addition, there are

sites that experience daily variations in power output that can be predicted based on
historical records. The performance of this forecast and its output can also be
included in the projected forecast, and confidence factor calculation.
[0057] In addition to daily variations, the method 400 may include correlating the
forecast accuracy with environmental conditions to predict forecast accuracy in the
future. For example, in an embodiment, if a certain amount of wind shear is observed
as being associated with a forecast confidence of 10% and if a forecast predicts a
similar amount of wind shear, a future forecast confidence of 10% can be assigned to
that time period. Preparations can be made accordingly for that future time slot.
Thus, a profile can be built into the future that can be adjusted as more information is
collected.
[0058] Environmental factor(s) that may be associated with the solar power
facilities may include, for example, time of day, cloud index, ambient temperature,
availability, panel health, losses, and/or model uncertainty. In regard to time of day
and given the very nature of solar generation, there is a clear dependency of power
generation to daytime hours. Consequently, there is no uncertainty with power
predictions during off-solar hours. As for cloud index, typical solar forecasting relies
on a predicted global horizontal irradiance (GHI) information which does not account
for cloud cover (ex: clear sky model). The level of cloud cover (altitude) and the
extent of cloud cover impacts the final prediction. The uncertainty associated with
solar predictions must account for a range of "cloud indices" to characterize the
uncertainty.
[0059] Furthermore, depending on the temperature coefficient of the panel, which
changes by panel manufacturer, the power output and hence the forecasts will have an
associated uncertainty. A predicted temperature range can be used to bound the
uncertainty.
[0060] Availability of the solar power facility generally refers to an unplanned
outage that may occur due to faults or unwanted trips. Factoring such uncertainty is
challenging and can be attempted if failure probabilities and dependent factors are
known. For example, if probability of failure is higher at higher temperature, this
increase probability can be included.
[0061] Panel health is also an important factor to consider as panel degradation

can have a direct impact on the accuracy of converting a GHI estimate to power. Given that panel degradation is typically 1.5% to 2% in the first year (light induced degradation) and 0.5% to 0.6% per year afterwards, the uncertainty in the power prediction can be reduced by correcting for this degradation in the GHI-power transfer function.
[0062] Furthermore, in relation to losses of the solar power facility, forecasts are typically determined at the farm level. Therefore, a loss model is required to adjust the panel/inverter level forecasts to account for losses up to the substation (point of metering). This is an issue only when weather data is used to determine a forecast. Purely data driven methods use time series data at the farm level, which already accounts for losses within the facility.
[0063] Along those same lines, solar power facilities should consider model uncertainty. Model uncertainty generally refers to the analytical functions that are used to estimate the power and the uncertainty behind such models. From a practical perspective, reduced order models are a better tradeoff given limitations in maintaining a large parameter library and the sensor data needed to fully incorporate high fidelity models.
[0064] Environmental factor(s) that may be associated with the wind power facilities may include, for example, time of day, wind speed predictions, turbulence intensity and shear, air density, ambient temperature, wind direction, availability, power performance drift, and losses.
[0065] As for time of day, in most regions, wind speeds are higher during the evening period compared to daytime. Predictions of wind speed that are above rated have lesser uncertainty inherently due to the power regulation controls of wind turbines which regulate power to be at rated power. Further, the uncertainty impact of wind speed predictions (when various models are used) is seen along the variable speed region of wind turbines, with increasing uncertainty of power prediction up to the knee point.
[0066] In addition to the mean wind speed, the turbulence intensity (TI) and wind shear have a significant impact on power production. For example, higher TI values lead to lesser power production given the same mean wind speed. Shear similarly has an influence on power production, although TI and shear are negatively correlated, i.e.

during daytime TI is high while shear is low, and at nighttime TI is lower but with increased shear.
[0067] Furthermore, air density generally has a linear sensitivity to power prediction uncertainty and can in general be easy to correct for since it is correlated to ambient temperature and humidity. In addition, the impact of ambient temperature is related to how deration might be triggered due to converter limitations. At higher temperatures (e.g. greater than 30°C), the converter can activate a deration-based operation which can reduce the power generation potential of the connected wind turbine, hence deviating from what would have been expected based on wind speed alone. Therefore, the temperature prediction range can be leveraged to include such variation.
[0068] Moreover, given a wind speed prediction, there can be significant difference between expected and actual farm level power. This is primarily due to the impact of wake losses, which are a function of the wind direction in relation to the layout of the wind facility. Knowing a range of uncertainty in the wind direction (e.g. actual +-10 degrees) can help estimate the impact of wake losses at farm level for the given wind direction range.
[0069] Similar to solar power facilities, the availability of the wind turbine power facility can be difficult to estimate unless another statistical correlation is leveraged -probability of failure (trip) higher at higher loading level (wind speed, higher TI, ambient temperature, etc.). In addition, like solar power facilities, losses of wind turbine power facilities relates to power loss that occurs within the collector system up to the substation interconnection.
[0070] Still another factor effecting wind systems includes power performance drift. Such drift is related to gradual degradation of power performance (e.g. blade tip erosion, leading edge soiling, etc.) and only has a long-term impact. This can be corrected for by using "as measured" power curve instead of the published curve. [0071] Accordingly, and referring back to FIG. 4, as shown at (404), the method 400 includes dynamically determining, by the control device, a confidence factor of the forecast as a function of, at least one, the environmental factor(s) of the renewable power facility 100. For example, in an embodiment, the function F may be a simple weighted mathematical function that sets forth a gain for each of the environmental

factors. Alternatively, the function F may be a machine learning algorithm that adjusts an effect of the environmental factor(s) based on performance over an extended time period.
[0072] In particular embodiments, as shown in FIG. 5, the supervisory controller 208 may determine the confidence factor 510 as a function of the environmental factor(s), historical performance 508, and/or an operator's desired level of risk 506. Further, in an embodiment, the supervisory controller 208 may define the confidence factor 510 as set forth below:
Where xi, X2,.. .XN represent the environmental factors that contribute to the confidence interval.
[0073] Accordingly, as shown in FIG. 5, the supervisory controller 208 can then send the forecast 502 and the confidence factor 510 to the facility controller 204. Thus, referring back to FIG. 4, as shown at (406), the method 400 includes determining one or more power commands of the renewable power facility based on the forecast 502 and the confidence factor 510. More specifically, as shown in FIG. 5, the facility controller 204 can determine and send the power command(s) 512 to each of the asset controllers 176 of the facility 100.
[0074] For example, in an embodiment, the facility controller 204 may apply various statistical analysis to the forecast 502 of the renewable power facility 100 to obtain a forecast distribution, such as the example power forecast distribution 600 provided in FIG. 8. In other words, the forecast 502 can be defined as a distribution as opposed to a single deterministic value. In such embodiments, the statistical analysis may include, for example, mean, median, standard deviation, probability distribution, or combinations thereof. In particular embodiments, for example, the forecast 502 (F(t)) can be described by the expression set forth below:
F(t)={u(t),B+(t),B-(t),o(t)}
Where u(t) refers to the mean of the forecast distribution 600,
B+(t), B"(t) refer to the upper and lower bounds of the forecast distribution 600 (not
necessarily the maximum or minimum, examples of which are provided in FIG. 8),
and
o(t) is the variance of the forecast distribution 600.

[0075] In such embodiments, the upper and lower bounds and the variance assist in defining the confidence level or uncertainty associated with a predicted power output. Thus, in an embodiment, such information can be utilized to take specific actions or to make particular decisions when the variance of a forecast distribution is high.
[0076] Given the definition of uncertainty, the method 400 can include deriving the bounds by predicting the forecast distribution (FIG. 9B), instead of the current practice of providing only a power prediction (FIG. 9A). Referring particularly to FIG. 9B, the probability indicates the nature of the power forecast distribution and the uncertainty surrounding the mean forecast. In practical situations, the distribution may not be a known analytic distribution (e.g. gaussian, beta, etc.) and must be determined through a combinatorial approach. Such an approach can combine all the influencing factors in combination to determine the spread of the forecast distribution, which is represented by the expression below: F(i) = {»(t),B + (t),B - (t),o(t)}
= f^\_XtneanrXa'\f [^meffn,irer],£Zmettre,Za:],...)
[0077] Accordingly, the power forecast distribution can be obtained by incorporating the uncertainty of all the contribution factors and simulating the power prediction model against that input data.
[0078] Accordingly, referring back to FIG. 4, as shown at (408), the method 400 includes controlling the renewable power facility 100 based on the power command(s) 512 so as to dynamically adjust an amount of power held in reserve at the renewable power facility 100. Further, as shown in FIG. 5, each of the individual asset controlleds) 176 generates the power outputs 514 for the grid 184 according to the power command(s) 512. In an embodiment, for example, controlling the renewable power facility 100 based on the power command(s) 512 may include increasing a power level held in reserve of the renewable power facility 100 when the confidence factor is below a threshold and maintaining or decreasing the power level held in reserve of the renewable power facility 100 when the confidence factor is above a threshold. [0079] Certain sites may require strict up and down power ramping. Thus, at such

sites, forecasting power production is essential to avoid penalties for missed power commitments. When forecasting is inaccurate, power reserves in the form of energy storage, or continuous curtailment of the farm output is required. Accordingly, the system and method described herein provides a process through which a confidence factor is assigned to a power forecast. When that forecast is decidedly accurate, a smaller power reserve can be retained, directly impacting the annual energy production of the facility 100. When a forecast is decided to be inaccurate, the reserve power can be increased, potentially avoiding impending penalties. Such benefits can be better understood with respect to FIG. 10. As shown, given a forecast that allows the controller 310 to estimate the power output for the next X hours, the controller 310 can generate a commitment to bid into the energy markets. This commitment includes a margin from the actual forecast (such as 5% or 10%), to accommodate any potential volatility in the renewable source. If the forecast is inaccurate, this margin must be large, decreasing the power output of the facility 100. Thus, improving the forecast accuracy can have an immediate impact on AEP (as shown via the shaded region 700 of FIG. 10) by allowing the facility 100 to operate with a smaller power reserve.
[0080] As described herein, the accuracy of the forecast depends on several factors. Thus, the present disclosure introduces the confidence factor described herein that accounts for past performance of the various inputs to the forecasting function and allows the controller 310 to decide how accurately it can predict the power output of the facility 100. Thus, instead of setting a constant margin, the confidence factor dynamically allocates the reserve margin. In periods of high confidence, the facility 100 operates closer to the actual forecasted power output, whereas, in periods of low confidence, the facility 100 maintains more power in reserve, thus avoiding significant penalties on violating a power commitment.
[0081] The technology discussed herein makes reference to computer-based systems and actions taken by and information sent to and from computer-based systems. One of ordinary skill in the art will recognize that the inherent flexibility of computer-based systems allows for a great variety of possible configurations, combinations, and divisions of tasks and functionality between and among components. For instance, processes discussed herein can be implemented using a

single computing device or multiple computing devices working in combination. Databases, memory, instructions, and applications can be implemented on a single system or distributed across multiple systems. Distributed components can operate sequentially or in parallel.
[0082] Although specific features of various embodiments may be shown in some drawings and not in others, this is for convenience only. In accordance with the principles of the present disclosure, any feature of a drawing may be referenced and/or claimed in combination with any feature of any other drawing. [0083] This written description uses examples to disclose the invention, including the best mode, and also to enable any person skilled in the art to practice the invention, including making and using any devices or systems and performing any incorporated methods. The patentable scope of the invention is defined by the claims, and may include other examples that occur to those skilled in the art. Such other examples are intended to be within the scope of the claims if they include structural elements that do not differ from the literal language of the claims, or if they include equivalent structural elements with insubstantial differences from the literal languages of the claims.

WHAT IS CLAIMED IS:
1. A method for optimizing power production of a renewable power
facility, the method comprising:
receiving, by a control device having one or more processors and one or more memory devices, a forecast of the renewable power facility;
dynamically determining, by the control device, a confidence factor of the forecast as a function of, at least one, one or more environmental factors of the renewable power facility;
determining one or more power commands of the renewable power facility based on the forecast and the confidence factor; and
controlling the renewable power facility based on the one or more power commands so as to dynamically adjust an amount of power held in reserve at the renewable power facility.
2. The method of claim 1, further comprising determining the confidence factor of the forecast as a function of at least one of historical performance of the renewable power facility and a desired level of risk.
3. The method of claim 1, wherein the function comprises at least one of a weighted mathematical function setting forth a gain for each of the one or more environmental factors or a machine learning algorithm that adjusts an effect of the one or more environmental factors based on performance over an extended time period.
4. The method of claim 1, wherein the one or more environmental factors comprise at least one of time of day, season, cloud index, temperature, availability of the renewable power facility, losses of the renewable power facility, uncertainty of the function, wind speed, turbulence intensity, wind shear, wind direction, air density, wake, and power output changes.
5. The method of claim 1, wherein the forecast comprises at least one of a power forecast, a weather forecast, or a pricing forecast.
6. The method of claim 1, further comprising evaluating the forecast by identifying forecasted weather patterns within the forecast.
7. The method of claim 1, further comprising applying statistical analysis to the forecast of the renewable power facility to obtain a forecast distribution.
8. The method of claim 7, wherein the statistical analysis comprises at

least one of mean, median, standard deviation, probability distribution, or combinations thereof.
9. The method of claim 7, wherein determining the confidence factor of
the forecast further comprises:
defining upper and lower bounds of the forecast distribution; and, determining a variance of the forecast distribution.
10. The method of claim 1, wherein controlling the renewable power
facility based on the one or more power commands so as to adjust the amount of
power held in reserve at the renewable power facility further comprises:
increasing a power level held in reserve of the renewable power facility when the confidence factor is below a threshold; and,
maintaining or decreasing the power level held in reserve of the renewable power facility when the confidence factor is above a threshold.
11. The method of claim 1, wherein the control device is a supervisory
controller of the renewable power facility, the method further comprising:
sending, via the supervisory controller, the confidence factor to a facility controller of the renewable power facility;
determining, via the facility controller, the one or more power commands of the renewable power facility based on the forecast and the confidence factor; and,
sending the one or more power commands to an asset controller of the renewable power facility.
12. The method of claim 1, wherein the renewable power facility comprises at least one of a wind farm having one or more wind turbines, a solar power plant having one or more solar power systems, a hydro power plant, an energy storage system, or combinations thereof.
13. A system for optimizing power production of a renewable power facility, the system comprising:
a control device comprising one or more processors and one or more memory devices, the control device configured to perform a plurality of operations, the plurality of comprising:
receiving a forecast of the renewable power facility;
dynamically determining a confidence factor of the forecast as a

function of, at least one, one or more environmental factors of the renewable power facility;
determining one or more power commands of the renewable power facility based on the forecast and the confidence factor; and
controlling the renewable power facility based on the one or more power commands so as to dynamically adjust an amount of power held in reserve at the renewable power facility.
14. The system of claim 13, wherein the plurality of operations further comprise determining the confidence factor of the forecast as a function of at least one of historical performance of the renewable power facility and a desired level of risk.
15. The system of claim 13, wherein the function comprises at least one of a weighted mathematical function setting forth a gain for each of the one or more environmental factors, or a machine learning algorithm that adjusts an effect of the one or more environmental factors based on performance over an extended time period.
16. The system of claim 13, wherein the one or more environmental factors comprise at least one of time of day, season, cloud index, temperature, availability of the renewable power facility, losses of the renewable power facility, uncertainty of the function, wind speed, turbulence intensity, wind shear, wind direction, air density, wake, and power output changes.
17. The system of claim 13, wherein the plurality of operations further comprise evaluating the forecast by identifying forecasted weather patterns within the forecast.
18. The system of claim 13, wherein the plurality of operations further comprise applying statistical analysis to the forecast of the renewable power facility to obtain a forecast distribution, the statistical analysis comprising at least one of mean, median, standard deviation, probability distribution, or combinations thereof.
19. The system of claim 13, wherein the control device is a supervisory controller of the renewable power facility, the system further comprising a facility controller and an asset controller, the plurality of operations further comprising sending the confidence factor to the facility controller, the facility controller configured to determine the one or more power commands of the renewable power

facility based on the forecast and the confidence factor and send the one or more power commands to the asset controller.
20. A method for forecasting power production of a renewable power facility to improve power production of the renewable power facility, the method comprising:
receiving a power forecast of the renewable power facility;
determining a confidence factor of the power forecast as a function of, at least one, one or more environmental factors of the renewable power facility; and
adjusting a power reserve of the renewable power facility based on the power forecast and the confidence factor.