Claims: 1. A forecasting system for forecasting electric power production in a wind-solar hybrid power generation plant, the system comprising:
a wind power forecasting unit configured to generate a forecast for electrical power production from one or more wind turbines corresponding to one or more forecast intervals based on data received from one or more sensors disposed on the one or more wind turbines;
a solar power forecasting unit configured to generate a forecast for electrical power production from one or more solar panels corresponding to at least one of the one or more forecast intervals based on data received from one or more sensors disposed on the one or more solar panels;
a balance of plant forecasting unit configured to generate a forecast for the electrical power production of the wind-solar hybrid power generation plant based on one or more of the forecast for electrical power production for the one or more wind turbines, the forecast for electrical power production for the one or more solar panels, operating conditions of one or more electrical subsystems operatively coupled to the one or more wind turbines and the one or more solar panels, operating conditions of the balance of plant of the wind-solar hybrid power generation plant or combinations thereof; and
a hybrid farm power forecasting unit configured to generate a forecast for electrical power production from the wind-solar hybrid power generation plant based on the electrical power production of the wind-solar hybrid power generation plant from one or more previous forecast intervals and one or more forecasting models.
2. The system of claim 1, further comprising:
the one or more sensors disposed on the one or more wind turbines and configured to measure wind conditions, wind turbine system operating conditions, or a combination thereof, and
the one or more sensors disposed on the one or more solar panels and configured to measure light conditions, solar panel operating conditions, or a combination thereof.
3. The system of claim 2, further comprising a shading prediction unit configured to:
determine an effective panel area of the one or more solar panels for the one or more forecast intervals;
obtain image data of cloud conditions over the one or more solar panels from one or more cameras, one or more image sensors, one or more satellite based sensors, or combinations thereof;
map the effective panel area of the one or more solar panels to the image data;
determine a first area of a first set of areas corresponding to one or more shadows due to the cloud conditions over the effective panel area of the one or more solar panels at a first time interval of the one or more forecast intervals;
determine subsequent areas of the first set of areas corresponding to the one or more shadows due to the cloud conditions over the effective panel area of the one or more solar panels corresponding to subsequent time intervals of the one or more forecast intervals based on one or more prediction models;
determine a first area of a second set of areas corresponding to one or more shadows due to the one or more wind turbines on the effective panel area of the one or more solar panels corresponding to the first time interval of the one or more forecast intervals;
determine subsequent areas of the second set of areas corresponding to the one or more shadows due to the one or more wind turbines on the effective panel area of the one or more solar panels corresponding to subsequent time intervals of the one or more forecast intervals; and
combine the first set of areas with the second set of areas corresponding to the one or more forecast intervals to generate an aggregated area under shadow of the effective panel area of the one or more solar panels.
4. The system of claim 3, wherein the one or more prediction models comprise an artificial neural network (ANN), a multi-variable autoregressive integrated moving average (ARIMA) model, edge detection algorithms, position prediction algorithms, or combinations thereof.
5. The system of claim 3, wherein the solar power forecasting unit is further configured to:
determine an irradiance per unit area of the one or more solar panels based on the effective panel area of the one or more solar panels and the aggregated area under shadow of the effective panel area of the one or more solar panels for the one or more forecast intervals;
compute the electrical power generated by the one or more solar panels corresponding to the one or more forecast intervals based on the irradiance per unit area of the one or more solar panels; and
adjust the computation of the electrical power generated by the one or more solar panels corresponding to the one or more forecast intervals based on one or more degradation models of the one or more solar panels.
6. The system of claim 3, wherein the solar power forecasting unit is further configured to:
obtain the electrical power production for the one or more solar panels corresponding to one or more previous forecast intervals; and
determine an electrical power forecast for the one or more solar panels corresponding to the one or more forecast intervals, based on the electrical power production of the one or more solar panels corresponding to the one or more previous forecast intervals, one or more forecasting models and on the aggregated area under shadow of the effective panel area of the one or more solar panels for the one or more forecast intervals, or combinations thereof.
7. The system of claim 6, wherein the one or more forecasting models comprise an AR (autoregressive) model, an autoregressive integrated moving average (ARIMA) model, an autoregressive conditional heteroskedasticity (ARCH) model, a generalized autoregressive conditional heteroskedasticity (GARCH) model, a moving-average (MA) model, or combinations thereof.
8. The system of claim 1, wherein the wind power forecasting unit is further configured to:
select one or more wind forecasting models from a plurality of wind forecasting models based on performance metrics of the one or more wind forecasting models for prevailing wind conditions and wind turbine operating conditions; and
determine a forecast for electrical power production for the one or more wind turbines for the one or more forecast intervals based on the one or more wind forecasting models.
9. The system of claim 8, wherein the one or more wind forecasting models comprise an AR (autoregressive) model, an autoregressive integrated moving average (ARIMA) model, an autoregressive conditional heteroskedasticity (ARCH) model, a generalized autoregressive conditional heteroskedasticity (GARCH) model, a moving-average (MA) model, a neural network, a search-based model, or combinations thereof.
10. The system of claim 1, wherein the balance of plant forecasting unit is further configured to
adjust the forecast for power generation by the wind-solar hybrid power generation plant based on the forecast for electrical power generation corresponding to the one or more wind turbines, the forecast for electrical power generation corresponding to the one or more solar panels, one or more constraints of the one or more electrical subsystems, one or more constraints of the balance of plant of the wind-solar hybrid power generation plant, a renewable regulatory fund (RRF) metric, a regulatory metric, or combinations thereof.
11. A wind-solar hybrid power generation system, comprising:
one or more wind turbines configured to generate electrical power;
one or more solar panels co-located with the one or more wind turbines and configured to generate electrical power;
a forecasting subsystem configured to:
generate a forecast for electrical power production from the one or more wind turbines corresponding to one or more forecast intervals based on turbine operational data, weather data, or a combination thereof;
generate a forecast for electrical power production from the one or more solar panels corresponding to the one or more forecast intervals based on solar panel operational data, weather data, or a combination thereof;
generate a forecast for electrical power production of the wind-solar hybrid power generation plant corresponding to the one or more forecast intervals based on the forecast for electrical power production for the one or more wind turbines, the forecast for electrical power production for the one or more solar panels, operating conditions of one or more electrical subsystems operatively coupled to the one or more wind turbines and the one or more solar panels, operating conditions of the balance of plant of the wind-solar hybrid power generation plant or combinations thereof; and
a controller operatively coupled to one or more of the one or more wind turbines, the one or more solar panels, and the forecasting subsystem, wherein the controller is configured to control operation of the one or more wind turbines and the one or more solar panels based on the forecast for electrical power production of the wind-solar hybrid power generation plant.
12. The system of claim 11, further comprising:
one or more sensors disposed on the one or more wind turbines and configured to measure wind conditions, wind turbine system operating conditions, or a combination thereof, and
one or more sensors disposed on the one or more solar panels and configured to measure light conditions, solar panel operating conditions, or a combination thereof.
13. The system of claim 11, wherein the one or more electrical subsystems comprise a rotor converter, a grid converter, a direct current link, a boost converter, or combinations thereof.
14. The system of claim 11, wherein the forecasting subsystem further comprises one or more of a shading prediction unit, a solar power forecasting unit, a wind power forecasting unit, a balance of plant forecasting unit, and a hybrid farm power forecasting unit.
15. A method for forecasting electrical power production for a wind-solar hybrid power generation plant, the method comprising:
generating a forecast for electrical power production from one or more wind turbines for one or more forecast intervals;
generating a forecast for electrical power production from one or more solar panels for the one or more forecast intervals; and
generating a forecast for electrical power production of the wind-solar hybrid power generation plant for the one or more forecast intervals based on the forecast for electrical power production for the one or more wind turbines, the forecast for electrical power production for the one or more solar panels, operating conditions of one or more electrical sub-systems of the wind-solar hybrid power generation plant, operating conditions of the balance of plant of the wind-solar hybrid power generation plant or combinations thereof.
16. The method of claim 15, wherein generating a forecast for electrical power production from the one or more solar panels for the one or more forecast intervals comprises:
determining an aggregated area under shadow due to cloud conditions, one or more wind turbines, and combinations thereof or both on an effective panel area of the one or more solar panels;
computing an irradiance per unit area of the one or more solar panels, based on the effective panel area of the one or more solar panels and the aggregated area under shadow of the effective panel area of the one or more solar panels; and
determining the electrical power generated by the one or more solar panels based on the irradiance per unit area of the one or more solar panels.
17. The method of claim 16, further comprising performing one of:
adjusting the computation of the electrical power generated by the one or more solar panels based on one or more degradation models of the one or more solar panels; and
determining the forecast for electrical power generated by the one or more solar panels for the one or more forecast intervals based on electrical power production of the one or more solar panels corresponding to one or more previous forecast intervals, one or more forecasting models, and the aggregated area under shadow of the one or more solar panels corresponding to the one or more forecast intervals.

18. The method of claim 16, wherein determining the aggregated area under shadow of the effective area of the one or more solar panels comprises:
retrieving image data of the cloud conditions over the one or more solar panels;
mapping the effective area of the one or more solar panels to the image data;
determining a first area of a first set of areas corresponding to one or more shadows due to cloud conditions over the effective panel area of the one or more solar panels corresponding to a first time interval of the one or more forecast intervals;
determining subsequent areas of the first set of areas corresponding to the one or more shadows due to the cloud conditions over the effective panel area of the one or more solar panels corresponding to subsequent time intervals of the one or more forecast intervals, based on one or more prediction models;
determining a first area of a second set of areas corresponding to one or more shadows due to the one or more wind turbines on the effective panel area of the one or more solar panels corresponding to the first time interval of the one or more forecast intervals;
determining subsequent areas of the second set of areas corresponding to the one or more shadows due to the one or more wind turbines on the effective panel area of the one or more solar panels corresponding to the subsequent time intervals of the one or more forecast intervals; and
combining the first set of areas with the second set of areas to generate an aggregated area under shadow of the effective panel area of the one or more solar panels corresponding to the one or more forecast intervals.
.
19. The method of claim 15, wherein generating a forecast for electrical power production from the one or more wind turbines for the one or more forecast intervals further comprises:
obtaining weather data, wind turbine system data, or a combination thereof, from one or more sensors configured to sense wind conditions, wind turbine operating conditions or combinations thereof;
selecting one or more wind forecasting models from a plurality of wind forecasting models based on the performance metrics of the one or more forecasting models for the prevailing wind conditions and wind turbine operating conditions; and
determining the electrical power generated by the one or more wind turbines for the one or more forecast intervals, based on the one or more forecasting models.
, Description:BACKGROUND
[0001] Embodiments of the present specification relate generally to wind-solar hybrid power generation plants, farms or systems, and more specifically to systems and methods for enhancing operations of the wind-solar hybrid power generation plants.
[0002] Power generation from wind turbines in a wind farm varies based on prevailing wind conditions. For example, during peak wind conditions, the wind turbine may provide above-average electric power to the grid. Likewise, decreases in wind speed at the site may result in reduced electrical power provided to the grid. In recent times, regulatory policies require individuals and/or entities involved with wind power generation to guarantee a determined power evacuation to the grid for a given interval of time. Hence, it is becoming increasingly critical to provide accurate forecasts of available power for a given time interval and accordingly deliver the guaranteed power for that interval. Failure to deliver the guaranteed power may disadvantageously result in financial penalties on the individual and/or entities.
[0003] It is advantageous to provide an alternate power generation resource at the wind farm to compensate for any sudden, unforeseen short-term discrepancies from the regulatory requirement. Solar energy is a viable renewable energy alternative. Wind-solar hybrid power generation systems employ wind turbines that are co-located with solar panels for power generation at a site. Typically, wind farms need large parcels of land to be allocated for appropriate placement of wind turbines. Co-locating solar panels along with the wind turbines at the site affords better usage of the land. In addition, co-locating solar panels with wind turbines allows for sharing the balance of plant costs and infrastructure, including electrical subsystems and control systems, at considerable cost savings.
[0004] Unfortunately, in such hybrid power generation systems prevailing weather conditions have varying effects on the wind power generation system and the solar power generation system. For example, in many geographical regions such as India, peak wind conditions typically occur at night and hence preclude solar-based power generation. Also, during the day-time, solar power generation is subject to varying irradiation conditions due to environmental conditions such as cloud movement and shadowing that may be caused by wind turbines in such a hybrid configuration. In addition, the balance of plant also creates operational constraints, which in turn results in varying power patterns, hence posing a challenge in forecasting the power production.
BRIEF DESCRIPTION
[0005] In accordance with aspects of the present specification, a forecasting system for forecasting electrical power production in a wind-solar hybrid power generation plant is presented. The system includes a wind power forecasting unit configured to generate a forecast for electrical power production from one or more wind turbines corresponding to one or more forecast intervals based on data received from one or more sensors disposed on the one or more wind turbines. Moreover, the system includes a solar power forecasting unit configured to generate a forecast for electrical power production from one or more solar panels corresponding to at least one of the one or more forecast intervals based on data received from one or more sensors disposed on the one or more solar panels. In addition, the system includes a balance of plant forecasting unit configured to generate a forecast for the electrical power production of the wind-solar hybrid power generation plant based on one or more of the forecast for electrical power production for the one or more wind turbines, the forecast for electrical power production for the one or more solar panels, operating conditions of one or more electrical subsystems operatively coupled to the one or more wind turbines and the one or more solar panels, operating conditions of the balance of plant of the wind-solar hybrid power generation plant, or combinations thereof. Furthermore, the system includes a hybrid farm power forecasting unit configured to generate a forecast for electrical power production from the wind-solar hybrid power generation plant based on the electrical power production of the wind-solar hybrid power generation plant from one or more previous forecast intervals and one or more forecasting models.
[0006] In accordance with another aspect of the present specification, a wind-solar hybrid power generation system is presented. The system includes one or more wind turbines configured to generate electrical power. Furthermore, the system includes one or more solar panels co-located with the one or more wind turbines and configured to generate electrical power. In addition, the system includes a forecasting subsystem configured to generate a forecast for electrical power production from the one or more wind turbines corresponding to one or more forecast intervals based on turbine operational data, weather data, or a combination thereof, generate a forecast for electrical power production from the one or more solar panels corresponding to the one or more forecast intervals based on solar panel operational data, weather data, or a combination thereof, generate a forecast for electrical power production of the wind-solar hybrid power generation plant corresponding to the one or more forecast intervals based on the forecast for electrical power production for the one or more wind turbines, the forecast for electrical power production for the one or more solar panels, operating conditions of one or more electrical subsystems operatively coupled to the one or more wind turbines and the one or more solar panels, operating conditions of the balance of plant of the wind-solar hybrid power generation plant or combinations thereof. The system also includes a controller operatively coupled to one or more of the one or more wind turbines, the one or more solar panels, and the forecasting subsystem, where the controller is configured to control operation of the one or more wind turbines and the one or more solar panels based on the forecast for the electrical power production of the wind-solar hybrid power generation plant.
[0007] In accordance with yet another aspect of the present specification, a method for forecasting electrical power generation in a wind-solar hybrid power generation plant is presented. The method includes generating a forecast for electrical power generation from one or more wind turbines for one or more forecast intervals. Also, the method includes generating a forecast for electrical power generation from one or more solar panels for the one or more forecast intervals. Furthermore, the method includes generating a forecast for electrical power generation in the wind-solar hybrid power generation plant for the one or more forecast intervals based on the forecast for electrical power production for the one or more wind turbines, the forecast for electrical power production for the one or more solar panels, operating conditions of one or more electrical sub-systems of the wind-solar hybrid power generation plant, operating conditions of the balance of plant of the wind-solar hybrid power generation plant, or combinations thereof.
DRAWINGS
[0008] These and other features and aspects of embodiments of the present specification will become better understood when the following detailed description is read with reference to the accompanying drawings in which like characters represent like parts throughout the drawings, wherein:
[0009] FIG. 1 is a schematic diagram of an exemplary wind-solar hybrid power generation plant including a forecasting subsystem for forecasting electrical power production of the wind-solar hybrid power generation plant, in accordance with aspects of the present specification;
[0010] FIG. 2 is a schematic diagram of one embodiment of the forecasting subsystem of FIG. 1, in accordance with aspects of the present specification;
[0011] FIG. 3 is a flow chart illustrating a method for forecasting electrical power production of the wind-solar hybrid power generation plant, in accordance with aspects of the present specification;
[0012] FIG. 4 is a flow chart illustrating a method for forecasting electrical power production of one or more solar panels in the wind-solar hybrid power generation plant, in accordance with aspects of the present specification;
[0013] FIG. 5 is a flow chart illustrating a method for forecasting electrical power production of one or more wind turbines in the wind-solar hybrid power generation plant, in accordance with aspects of the present specification; and
[0014] FIG. 6 is a flow chart illustrating a method for aggregating predicted electrical power production of one or more wind turbines with predicted electrical power production of one or more solar panels in the wind-solar hybrid power generation plant, in accordance with aspects of the present specification.
DETAILED DESCRIPTION
[0015] One or more specific embodiments of the present specification will be described below. In an effort to provide a concise description of these embodiments, all features of an actual implementation may not be described in the specification. It should be appreciated that in the development of any such actual implementation, as in any engineering or design project, numerous implementation-specific decisions must be made to achieve the developer’s specific goals such as compliance with system-related and business-related constraints.
[0016] When describing elements of the various embodiments of the present invention, the articles “a”, “an”, “the” and “said” are intended to mean that there are one or more of the elements. The terms “comprising”, “including” and “having” are intended to be inclusive and mean that there may be additional elements other than the listed elements.
[0017] The terms “system” and “plant” may be used interchangeably in the present specification in conjunction with the phrase “wind-solar hybrid power generation,” and are used to refer to the wind-solar hybrid power generation system as a whole, including the wind-solar hybrid farm and the balance of plant as a power source to an electrical grid. The term “wind farm” is used to refer to a collection of wind turbines or a collection of solar panels in the same location used to produce electrical power. Likewise, the term “solar farm” is intended to refer to a collection of solar panels in the same location to produce electrical power. In a similar fashion, the term “wind-solar hybrid farm” is used to refer to a collection of wind turbines and solar panels in the same location to produce electrical power.
[0018] It may be noted that in some jurisdictions, power forecasting is regulated via regulatory policies to occur, for example, at certain time periods. Some examples of these time periods include 15 minutes, 30 minutes, 1 hour, 3 hours, 8 hours, and the like. As will be described hereinafter, various embodiments of systems and methods for enhancing the operation of a wind-solar hybrid power generation plant by providing more accurate and timely power forecasts are presented. In particular, a forecasting system configured to provide more accurate and timely power forecasts of the wind-solar hybrid power generation plant is presented. The forecasting subsystem may use system operating data, weather conditions, and other exogenous conditions to forecast the electrical power production of one or more wind turbines and/or one or more solar panels in a wind-solar hybrid power generation plant for a given time period (forecast interval). This improved power production forecasting may enable more accurate trading of energy credits, enhanced pricing of energy, and lower insurance rates. In accordance with aspects of the present specification, the forecasting subsystem may further enhance the accuracy of the forecasts for electrical power generated from the wind turbines and solar panels by analyzing any effects of exogenous conditions, operating conditions of the machinery, and/or regulatory requirements on the individual forms of power generation in a wind-solar hybrid power generation plant.
[0019] In a wind power generation plant or a solar power generation plant, the term “balance of plant” refers to all the infrastructure and facilities of a wind or solar farm and their associated cost with the exception of the wind turbines or solar panels and all corresponding elements. The balance of plant therefore may include the civil infrastructure and costs, such as crane pads, hard standings, substation civil costs, farm network roads and the like. Also, the electrical components of the balance of plant may include transformers, Supervisory Control And Data Acquisition (SCADA) systems, electrical cabling, and substation electrical components such as switchgear, DC Links, rotor converters, medium voltage (MV) transformers, grid converters, boost converters, power protection systems, electric meters, and other components to facilitate the operational aspects of the wind farm. The balance of plant may also include components necessary to complete the connection to the distribution network such as switches, reinforcements, and protection and measuring elements.
[0020] In a wind-solar hybrid power generation plant, a significant portion of the civil and electrical infrastructure and costs of the balance of plant may be shared by the wind turbine systems and solar panels. In particular, the electrical substation/subsystem and other electrical components of the balance of plant may be shared by the wind turbine systems and solar panels, resulting in a considerable saving in installation and operational costs.
[0021] It may be noted that the terms “electrical substation” and “electrical subsystem” are interchangeable in the present specification and are used to refer to the electrical components to facilitate the operational aspects of the wind farm and complete the connection to the distribution network or electrical power grid.
[0022] Individual components of the electrical subsystem at the turbine level may have power rating constraints. Additionally, the balance of plant of the wind-solar hybrid power generation plant may have constraints on power evacuated to the grid, depending on the existing generating wind turbines and solar panels, pricing mechanisms, capacity for energy storage, demand management and other factors. In accordance with aspects of the present specification, the power constraints and operating conditions of the electrical subsystem and balance of plant are also accounted for in the generation of an accurate power forecast. The forecast data generated by the forecasting subsystem may be provided to a controller and/or system operator to control the operation of the wind turbines and the solar panels. Additionally or alternatively, the forecast data may be stored in a data repository for future analysis as historical power data.
[0023] In some situations, the forecast interval may be very short, for example, less than 10 minutes. In other situations, sensor data related to real-time weather conditions at the wind-solar hybrid farm site may not be available. Additionally or alternatively, there may be site-specific needs for advanced modeling. For example, complex terrain that creates high levels of turbulence intensity in the wind can pose challenges to using sensor information for predicting wind power. Similarly, sensor information for predicting solar power may be too cost prohibitive for an application that requires forecasts for short prediction intervals. In such situations, the forecasting subsystem is configured to generate a forecast for electrical power production from the wind-solar hybrid power generation plant for the forecast interval based on historical data corresponding to measured power or forecasted power at the farm level for previous forecasting intervals and statistical methods.
[0024] An exemplary wind-solar hybrid power generation system 100 is illustrated in FIG. 1. In a presently contemplated configuration, the wind-solar hybrid power generation system 100 includes a wind-solar hybrid farm 101, an electrical subsystem 120, a controller or control subsystem 138, and a forecasting subsystem 140.
[0025] The wind-solar hybrid farm 101 may include one or more wind turbines 102, one or more solar panels 104, and one or more sensors 106. The wind turbines 102 and the solar panels are configured to generate electrical power. For ease of description, power generated by the wind turbines 102 is generally referred to as wind power PWIND-K 108. In a similar fashion, power generated by the solar panels 104 is referred to as solar power PSOLAR-K 110.
[0026] Further, in certain embodiments, the sensors 106 may be disposed on or in the vicinity of the wind turbines 102 and/or the solar panels 104. In certain embodiments, the sensors 106 may be configured to measure parameters related to operating conditions of the wind turbine machinery and/or the solar panel machinery. Some examples of the operating conditions of the wind turbine machinery and/or the solar panel machinery include, but are not limited to, pressure, temperature, blade revolutions per minute, vibration, torque, hours used, electrical power production, and the like. Additionally, the sensors 106 may also be configured to measure environmental conditions and/or weather related parameters, such as, but not limited to, wind direction, wind speed, atmospheric pressure, temperature, humidity, precipitation, cloud ceiling, visibility, light intensity, and the like. Moreover, the sensors 106 may include image sensors, cameras, and the like. In some embodiments, these sensors 106 may be configured to capture shadows of overhead clouds and/or neighboring wind turbine towers and blades on a solar panel 104. The sensors 106 may be communicatively coupled to the forecasting subsystem 140. In certain embodiments, the sensors 106 may be communicatively coupled to the forecasting subsystem 140 directly or indirectly via use of data repositories, data services, and the like. Moreover, the sensors 106 are configured to convey sensor data 116 to the forecasting subsystem 140. The sensor data 116 is generally representative of data gathered by the sensors 106.
[0027] The wind turbines 102 and the solar panels 104 may be operatively coupled to a controller or control subsystem 138. The control subsystem 138 is configured to control operations of the wind-solar hybrid power generation system 100. In one example, the control subsystem 138 may control the operations of the wind turbines 102 by controlling power inverter circuitry, electrical filtering circuitry, battery charging circuitry, and the like.
[0028] Moreover, the system 100 may include one or more electrical subsystems 120 that are operatively coupled to one or more of the wind turbines 102, the solar panels 104, the control subsystem 138, and the forecasting subsystem 140, as illustrated in FIG. 1. In some embodiments, the electrical subsystem 120 may include a rotor converter 122, a direct current (DC) link 124, a grid converter 126, and a medium voltage (MV) transformer 132. Additionally, the electrical subsystem 120 may include a boost converter 130 and charge controller 128. It may be noted that for ease of illustration, some components of the electrical subsystem 120 are depicted. However, the electrical subsystem 120 may include additional components or fewer components.
[0029] In the example of FIG. 1, the power PWIND-K 108 generated by the wind turbines 102 may be input to the electrical subsystem 120. Likewise, the power PSOLAR-K 110 generated by the solar panels may also be input to the electrical subsystem 120. In one embodiment, for certain kinds of wind turbines 102 such as a doubly fed induction generator (DFIG) based wind turbine, only a fraction of the wind power PWIND-K 108 (i.e., slip power) may be evacuated through the rotor converter 122. The remaining wind power PWIND-K 108 may flow through grid converter 126 and MV transformer 132. The DC link 124 may be configured to aggregate a fraction of the wind component of the power PWIND-K 108 and the solar component of the power PSOLAR-K 110 to generate electrical power at the farm level PFARM 134. In addition, the power PFARM 134 generated by the system 100 may be evacuated to a power grid at a switch 136.
[0030] As previously noted, the sensors 106 disposed on the wind turbines 102 and/or the solar panels 104 are configured to gather data and measurements regarding produced power, environmental conditions, and the like. Data related to the measured power produced by the wind turbines 102 may be generally denoted as PWIND-K data 112 and data related to the measured power produced by the solar panels 104 may be denoted as PSOLAR-K data 114. Additionally, the data corresponding to environmental conditions and operating conditions gathered by the sensors 106 is collectively denoted as sensor data 116. The sensors 106 are configured to convey the sensor data 116, the measured solar power data PSOLAR-K data 114, the measured wind power data PWIND-K data 112, or combinations thereof to the forecasting subsystem 140.
[0031] Further, as previously noted, the forecasting subsystem 140 is configured to provide more accurate and timely power forecasts of the wind-solar hybrid power generation plant 100. The forecasting subsystem 140 may use system operating data, weather conditions, and other exogenous conditions to forecast the electrical power production of one or more wind turbines 102 and/or one or more solar panels 104 in a wind-solar hybrid power generation plant 100 for a given time period (forecast interval) to enhance the efficiency of the system 100.
[0032] Accordingly, the forecasting subsystem 140 may be configured to obtain data from the wind turbines 102. In one example, wind turbine data may include environmental variables such as wind speed, wind direction, temperature, and the like. Additionally, the forecasting subsystem 140 may also be configured to obtain data related to the operating conditions of the wind turbines 102. Some examples of the operating conditions include availability of the wind turbines 102, the type of wind turbines in use, current wind power production, and the like. The data so acquired from the wind turbines 102 may be stored in a data repository.
[0033] In a similar manner, the forecasting subsystem 140 may be configured to acquire data from the solar panels 104. The solar panel data may include, for example, environmental variables such as light conditions, cloud formations over the solar panels 104, shadows due to turbine towers and blades, temperature, and the like. Also, the forecasting subsystem 140 may be configured to obtain data related to the operating conditions of the solar panels 104. Some examples of the operating conditions include data related to the degradation of the photo-voltaic cells of the solar panels 104, the type of solar panels in use, current solar power production, and the like. This data may be stored in the data repository. The forecasting subsystem 140 is configured to generate a more accurate and timely power forecast of the wind-solar hybrid power generation plant 100 based on the data acquired from the sensors 106, the wind turbines 102, the solar panels 104, or combinations thereof.
[0034] Additionally, the forecasting subsystem 140 may also be configured to enhance the efficiency of the forecast of power generation of the system 100 via use of a set of forecasting models, degradation models, edge detection algorithms, position prediction algorithms, and the like. In certain embodiments, these models may be maintained in the data repository. The forecasting subsystem 140 may use one or more of these forecasting models to predict the electrical power production from the wind turbines 102 corresponding to a forecast interval. The selection of the forecasting models may be based on an accuracy of the forecasting model for the prevailing weather conditions for the forecast interval and/or the performance/accuracy achieved by the various models in the past.
[0035] As will be appreciated, all portions/parts of the one or more solar panels 104 may not be uniformly exposed to the sun’s radiation. Hence, it may be desirable to determine an “effective panel area” of the solar panels 104. The term “effective panel area” of a solar panel 104 refers to an area of the solar panel 104 as seen from the sun. The effective panel area may be used to determine how much of the sun’s energy is impinging on the solar panel 104. Furthermore, the effective panel area of the solar panel 104 may vary based on the time of day.
[0036] Accordingly, the forecasting subsystem 140 may be configured to determine the effective panel area of one or more solar panels 104 in the wind-solar hybrid farm 101. More particularly, the forecasting subsystem 140 may be configured to determine the effective panel area of the solar panels 104 corresponding to different times of the day based on known solar trajectory and panel siting information. However, the effective panel area of the solar panel 104 may be further shadowed by clouds, turbine towers, and/or blades of the wind turbines 102 during the forecast interval.
[0037] In this example, the sensors 106 such as image sensors, cameras and the like may be configured to acquire/capture images of cloud conditions, turbine tower shadows, and/or turbine blade shadows. In accordance with aspects of the present specification, the forecasting subsystem 140 is configured to analyze the images of cloud conditions, turbine tower shadows, and/or turbine blade shadows to determine irradiance per unit area of the solar panel 104, while accounting for any shadows on the solar panel 104 during the forecast interval. The forecasting subsystem 140 may employ edge detection techniques, position prediction techniques, and the like to analyze the images in order to identify portions of the effective area of the solar panels 104 under shadow due to clouds, turbine towers and/or blades corresponding to the forecast interval. Moreover, the forecasting subsystem 140 may be configured to determine the irradiance of the solar panels 104 corresponding to the forecasting interval based on irradiance data associated with the effective area of the corresponding solar panels 104 and associated with the forecast interval. In addition, the forecasting subsystem 140 may also be configured to adjust the irradiance for the shadowed portion of the effective area of the solar panels 104 corresponding to the forecast interval. Further, the power generated by the solar panels 104 for the forecast interval may be determined based on the irradiance per unit area of the effective area of the solar panels 104.
[0038] Moreover, over time, the photo-voltaic (PV) cells of the solar panels 104 may age and degenerate, consequently leading to a lower power per unit for a given lighting condition. In accordance with aspects of the present specification, the forecasting subsystem 140 is configured to employ a degradation model that captures power degradation of the solar panel 104 as a function of the electric power generated by the solar panel 104 over time to account for the degradation of the PV cells. In particular, the forecasting subsystem 140 may be configured to adjust the calculated power for the irradiated area of the solar panel 104 based on the degradation model to derive an accurate forecast for electrical power production of the solar panel 104.
[0039] In accordance with further aspects of the present specification, the wind turbines 102, the solar panels 104, and/or the sensors 106 may be configured to convey to the forecasting subsystem 140 updates regarding current environmental conditions, wind turbine and/or solar panel operating conditions. Additionally, the forecasting subsystem 140 may be configured to receive updates regarding system specific data, for example, hardware/software configurations, availability, operating state, maintenance schedules, regulatory data, or combinations thereof. These updates may occur in real-time, in certain embodiments. However, in certain other embodiments, the updates may be performed in a periodic manner.
[0040] Implementing the system 100 and more particularly the forecasting subsystem 140 as described hereinabove aids in providing more accurate and timely forecasts of power generation of the wind-solar hybrid farm 101. Additionally, the efficiency of the forecasts is further enhanced by accounting for various factors such as system operating data, weather conditions, and other exogenous conditions.
[0041] Referring now to FIG. 2, a schematic diagram 200 of one embodiment of the forecasting subsystem 140 of FIG. 1 is presented. As previously noted with reference to FIG. 1, the forecasting subsystem 140 is communicatively coupled to one or more wind turbines 102 and one or more solar panels 104. Additionally, the sensors 106 may be disposed on each of the wind turbines 102 and/or the solar panels 104 and may be communicatively coupled to the forecasting subsystem 140. Furthermore, the forecasting system 140 is communicatively coupled to a controller or controller subsystem 138. FIG. 2 is described with reference to the components of FIG. 1.
[0042] As illustrated in FIG. 2, a data repository 208 is communicatively coupled to the forecasting subsystem 140. The data repository 208 is configured to store sensor data (SED) 116, measured power from the wind turbines 102 PWIND-K data (PW) 112, measured power from the solar panels 104 PSOLAR-K data (PS) 114, weather data (WED) 206, turbine operational data (TOD) 204, solar panel operational data (POD) 202, and balance of plant operating data (BOPD) 218. The BOPD 218 may include constraints relating to the existing wind turbines and solar panels, pricing mechanisms, capacity for energy storage, demand management, and other factors. The BOPD 218 may further include power constraints of one or more components of the electrical subsystem 120.
[0043] The data repository 208 may additionally receive and store historical power production data produced by the wind turbines 102 and/or solar panels 104 of the wind-solar hybrid farm 101, and historical power production data at the farm or plant level. The historical power production data for each of the wind turbines, solar panels and the farm or plant as a whole corresponding to one or more previous time or forecast intervals may be stored in the data repository . For example, electrical power production values may be stored for the previous year, the previous month, the previous day, the previous 12 hours, the previous hour, the previous 15 minutes and so on. Moreover, the data repository 208 may also receive and store system specific data, for example, hardware/software configurations, availability, operating state, maintenance schedules, a renewable regulatory fund (RRF) metric, a regulatory metric, or combinations thereof.
[0044] Additionally, a forecasting model repository 210 may be operatively coupled to the forecasting subsystem 140. The forecasting model repository 210 may include one or more forecasting models (FM) 212, one or more shadow prediction models (PM) 214, and one or more degradation models (DM) 216. The forecasting models 212 may in turn include turbine-level models. These turbine-level models are specific models that are associated with a corresponding wind turbine 102 and may be executed independently for each wind turbine 102 in the wind-solar hybrid farm 101. Also, the turbine-level models may be employed for forecasting power generated by each of the wind turbines 102 individually. Turbine-level models may include wind speed prediction models and/or wind power prediction models.
[0045] The forecasting models 212 may also include wind direction prediction models, availability forecasting models, actual power curve models, and farm level models. Additionally, the forecasting models 212 may include persistence models. These persistence models include data, for example, such as wind speeds seen at the last available time point, power, wind direction, availability that is projected forward to the future, and the like. The forecasting models 212 may additionally or alternatively include autoregressive (AR) and/or autoregressive integrated moving average (ARIMA) based models. These models may in turn include statistical models that utilize values and errors corresponding to previous time intervals (for example, historical data) to predict future values. The forecasting models 212 may also include autoregressive models for circular time series (ARCS), autoregressive conditional heteroskedasticity (ARCH) models, autoregressive moving average (ARMA) models, generalized autoregressive conditional heteroskedasticity (GARCH) models, and/or moving-average (MA) models. In addition, the forecasting models 212 may further include search based models. These search based models are configured to search historical data for patterns that are similar to the currently occurring pattern and use those patterns to generate forecasts. The forecasting models 212 may also include neural network models. The neural network models may be trained using online time series data and other relevant inputs such as air density, temperature time gradient, or combinations thereof.
[0046] As previously noted, over time the PV cells of the solar panels 104 may age and degenerate. The degradation models (DM) 216 in the forecasting model repository 210 may include a power-time curve corresponding to the solar panels 104. By way of example, in the power-time curve, the x-axis may represent the time for which the solar panel 104 is in use. The time is typically measured in time units, such as number of hours, number of years, number of days, and the like. Also, the y-axis of the power-time curve represents the measured power output of the solar panel 104 as a percentage of the rated power of the solar panel 104. In particular, the percentage of the rated power represents a percentage of the measured power output in comparison to the rated power output. Some examples of percentage of the rated power include 100%, 80%, and the like. The degradation models 216 may also include asset life models where measured or calculated electrical power is characterized as a function of one or more solar panel properties.
[0047] The forecasting subsystem 140 may also use one or more prediction models (PM) 214 to generate the forecast. The prediction models 214 may include edge detection techniques, position prediction techniques, and the like. These techniques are representative of methods of predicting the movement of an image across a designated area over one or more time intervals. Also, edge detection techniques denote a set of mathematical methods which aim at identifying points in a digital image at which the image brightness changes sharply, or more particularly, has discontinuities. The position prediction algorithm tracks the movement of the cloud edges, which are identified by edge detection algorithms, and provides a prediction of the position of the edge at a future point in time. This allows for predicting the shadow conditions of the solar panels.
[0048] The prediction models 214 may also include an irradiation-power curve. In the irradiation-power curve, the x-axis represents irradiation per unit area of the solar panel 104 and the y-axis represents power as a function of the irradiation per unit area of the solar panel 104. In another example, measured electrical power from the solar panel 104 may be used to generate an estimated power curve map corresponding to electrical power generated by the solar panel 104 as a function of normalized irradiation data and the measured solar panel electrical power. The power curve data may be compared to historical, theoretical, and/or simulated power curve data based on the normalized irradiation data for the solar panel 104.
[0049] Additionally, the forecasting subsystem 140 may include one or more other units. In a presently contemplated configuration, the forecasting subsystem 140 is depicted as including a wind power forecasting unit 226, a hybrid farm power forecasting unit 228, a balance of plant (BOP) forecasting unit 230, a solar power forecasting unit 232, and a shading prediction unit 234. These units may be collectively referred to as component forecasting units 224. In certain embodiments, the forecasting subsystem 140 and the units 226, 228, 230, 232 and 234 in particular may be implemented as software systems or computer instructions executable via one or more processing units 220 and stored in a memory unit 222. The units 226, 228, 230. 232 and 234 may additionally or alternatively be stored and executed by other computing devices such as a workstation, personal computer (PC), laptop, notebook, tablet, cell phone, and the like. Further, the units 226, 228, 230, 232 and 234 may be implemented as hardware systems, for example, via FPGAs, custom chips, integrated circuits (ICs), and the like. In certain other embodiments, the data repository 208 and the forecasting model repository 210 may be part of a single larger data repository.
[0050] Each of the component forecasting units 226, 228, 230, 232 and 234 is configured to provide a forecast for electrical power production corresponding to different components of the wind-solar hybrid farm 101. The working of the component forecasting units 226, 228, 230, 232 and 234 will be described in greater detail in FIGs. 3-6.
[0051] Turning now to FIG. 3, a flowchart 300 illustrating a method for determining a forecast for electrical power production of a wind-solar hybrid farm 101, in accordance with aspects of the present specification, is presented. The method 300 is described with reference to the components of FIGs. 1-2.
[0052] In one embodiment, the various units 220, 222, 226, 228, 230, 232 and 234 of the forecasting subsystem 140 may be employed to generate an accurate estimate of the forecast for electrical power production of the wind-solar hybrid power generation plant/system 100 shown in FIG. 1. In some embodiments, various steps of the method 300 of FIG. 3 may be performed by one or more of the processor unit 220 and the memory unit 222 in conjunction with the component forecasting units 226, 228, 230, 232 and 234 of the forecasting subsystem 140. It may be noted that flowchart 300 illustrates the main steps of the method to generate an accurate forecast for electrical power production, and additional inputs and steps will be described in greater detail in FIGs. 4-6.
[0053] Reference numeral 330 is generally representative of a hybrid farm level power forecast, denoted as “hybrid power forecast”. The hybrid power forecast 330 is generated based on steps 301, 302, 304, 306, 312, 318, 324 and 326. The hybrid power forecast 330 may alternatively be generated based on steps 301, 302, 304 and 328.
[0054] The method 300 starts at step 301, where the processing unit 220 obtains one or more of sensor data 116, turbine operational data 204, solar panel operational data 202, and a desired forecast interval. In one embodiment, the processing unit 220 may acquire the data from the data repository 208. At step 302, the processing unit 220 checks if the acquired data includes sensor data 116. Accordingly, at step 302, if it is determined that sensor data 116 is not available, control passes to step 328.
[0055] Referring now to step 328, in one embodiment, the processing unit 220 and the hybrid farm power forecasting unit 228 may be used to generate the hybrid power forecast 330. Moreover, in some embodiments, the hybrid power forecast 330 may be generated based on historical electrical power production at the farm level 327 and one or more forecasting models 212 from the forecasting model repository 210. As previously noted, the forecasting models 212 may include AR and/or ARIMA based models, which include statistical models utilizing the values and errors corresponding to previous time intervals (e.g., historical data stored in the data repository 208) to predict future values. The forecasting models 212 may also include ARCS, ARCH models, ARMA models, GARCH models, and/or MA models.
[0056] With returning reference to step 302, if it is determined that sensor data 116 is not available, control passes to step 304. At step 304, a check is performed to determine the duration of the forecast interval. In one example, the check may entail determining if the forecast interval is extremely short. In the example of FIG. 3, the forecast threshold is 10 minutes. Hence, at step 304, the acquired forecast interval may be compared with the forecast threshold. If at step 304, it is determined that the forecast interval is less than 10 minutes, control is passed to step 328. However, if at step 304, it is determined that the forecast interval is greater than 10 minutes, control is passed to step 306.
[0057] Turning now to step 306, the processing unit 220 retrieves the measured power output PW 112 corresponding to one or more wind turbines 102 and the measured power output PS 114 corresponding to one or more solar panels 104. In one example, the processing unit 220 may retrieve the measured power outputs PW 112 and PW 114 from the data repository 208.
[0058] Subsequent to step 306, steps 312 and 318 may be performed concurrently or in a sequential manner. At step 312, the wind power forecasting unit 226 may retrieve turbine operational data TOD 204 and weather data WED 206 from the data repository 208. Additionally, the wind power forecasting unit 226 may retrieve one or more forecasting models FM 212 from the forecasting model repository 210. The wind power forecasting unit 226 may be configured to generate an electrical power production forecast for one or more wind turbines 102 based on one or more of TOD 204, WED 206, and FM 212. The working of the wind power forecasting unit 226 will be described in greater detail in FIG. 5.
[0059] Referring now to step 318, the solar power forecasting unit 232 may retrieve the predicted shadow area 320 of the one or more solar panels 104 from the shading prediction unit 234 and solar panel operating data POD 202 from the data repository 208. Additionally, the solar power forecasting unit 232 may retrieve one or more degradation models DM 216 and/or one or more forecasting models FM 212 from the forecasting model repository 210. The solar power forecasting unit 232 may be configured to derive a forecast for electrical power production for one or more solar panels 104 based on one or more of the predicted shadow area 320, POD 202, FM 212 and DM 216. Additionally or alternatively, the solar power forecasting unit 232 may be configured to use historical solar power production data, depicted in FIG. 4. The working of the solar power forecasting unit 232 and the shadow prediction unit 234 is described in greater detail in FIG. 4.
[0060] Control may be passed from steps 312 and 318 to step 324. At step 324, the processing unit 220 aggregates the forecast for electrical power production corresponding to wind turbines 102 generated at step 312 and the forecast for electrical power production corresponding to the solar panels 104 generated at step 318.
[0061] Subsequently, at step 326, the BOP unit 230 may retrieve balance of plant operating data (BOPD) 218 from the data repository 208. Additionally, the BOP unit 230 may process the aggregated forecast based on certain constraints associated with electrical subsystem 120 and the balance of plant of the wind-solar hybrid power generation system 100 to generate the hybrid power forecast 330. Some examples of these considerations include the power rating of one or more components 124, 122, and 126 of the electrical subsystem 120, the energy storage capacity of the wind-solar hybrid power generation system 100, demand management, pricing information, and the like.
[0062] Consequent to the processing of step 326, the hybrid power forecast 330 is generated. The hybrid power forecast may be stored in the data repository 208. Alternatively, the hybrid power forecast 330 may be received by the control subsystem 138 and/or a system operator. Control may be passed back to step 300 where the processor unit 220 of the forecasting system 140 may iterate through steps 301-328 of the method 300. Optionally, steps 301-328 may be iterated a finite number of times for a given forecast interval. Alternatively, steps 301-328 may be iterated at specified times of the day or night.
[0063] As will be appreciated, many of the sites where solar panels are installed experience temporal and spatial variability of the solar resource due to cloud cover influenced by terrain and local circulations. System operators need foreknowledge of imminent changes in solar energy output and the potential for large-scale ramp events occurring on short time scales. One of the dominant factors causing variability in solar power production is the variability of solar irradiance. For example, steep power ramps may be caused by the shadows of relatively fast-moving, low-level cumulus or stratocumulus clouds. Temperature is also a factor for most types of solar generation technology. More particularly, the material compositions of different types of solar panels have different sensitivities to the direct and diffuse components of solar radiation.
[0064] Cloud prediction is also complicated by the fact that clouds occur on a wide range of spatial and temporal scales, at different levels of the atmosphere, and with variable water or ice content. For example, individual cumulus clouds can be hundreds of meters in size and may have a typical lifecycle of less than one hour. On the other hand, clouds associated with large-scale storm systems may extend over 1,000 km and have a lifecycle of several days. Smaller scale cloud features may be embedded within larger-scale systems. Therefore, the challenge associated with predicting solar irradiance at the surface of the earth depends on the space and time scales that are of interest for specific applications.
[0065] Additionally, in one embodiment of the wind-solar hybrid farm 101 of FIG. 1, solar panels 104 co-located with the wind turbines 102 may be shadowed by neighboring wind turbine towers at different times of the day. Furthermore, the moving blades of neighboring wind turbines 102 may cast rapidly repeating shadows on the solar panel 104 at regular shorter intervals of time. In accordance with aspects of the present specification, to accurately predict the effective area and irradiance of a solar panel 104 in a wind solar hybrid farm 101 for a given forecast interval, the shadow on the solar panels 104 due to clouds may be determined. Additionally, the shadow on the solar panels 104 due to neighboring wind turbine towers and moving blades may also be determined. Subsequently, the areas of shadows on the solar panels 104 due to the clouds, the neighboring wind turbine towers and/or blades corresponding to the forecast interval may be aggregated to determine a more accurate effective area and irradiance per unit area.
[0066] Turning now to FIG. 4, a flowchart 400 illustrating a method for determining a forecast for electrical power production for one or more solar panels of a wind-solar hybrid power generation plant is presented. The method 400 may be described with reference to the components of FIGs. 1-2.
[0067] In one embodiment, steps 412 through 422 of the method 400 of FIG. 4 may be performed via use of one or more of the shading prediction unit 234, the solar forecasting unit 232 and the processor unit 220 of the forecasting subsystem 140. In a similar manner, steps 410 and 424 through 436 may be performed by the solar power forecasting unit 232 and the processing unit 220 of the forecasting system 140.
[0068] The method starts at step 410, where data is gathered by the solar power forecasting unit 232. The data may include sensor data 116, weather data WED 206, and irradiation data 404. The sensor data 116 may include camera images capturing overhead cloud conditions, and/or visible/infra-red cloud imagery from satellite based sensors for the site of the solar panel 104. Moreover, the weather data WED 206 may include information regarding weather conditions currently prevailing at the wind-solar hybrid farm 101. Additionally, the irradiation data 404 may include data derived from sensors or estimated from measured performance of the solar panel 104 at the site. In some embodiments, the data 116, 206, 404 may be retrieved by the processing unit 220 from the data repository 208. Also, the sensor data 116, the weather data WED 206, and the irradiation data 404 may collectively be referred to as cloud formation data.
[0069] Once the data is retrieved, at step 412, a cloud shadow may be mapped to the solar panel 104 for a first time interval. Accordingly, the position of the solar panel 104 at the site is retrieved by the shading prediction unit 234 from the data repository 208. In addition, the effective panel area of the solar panel 104 may be determined based on siting information corresponding to the wind-solar hybrid farm 101 and the known solar trajectory for the forecast interval. Furthermore, at step 412, the relative position of the effective panel area of the solar panel 104 with respect to the light source location is mapped onto an image circle of the image obtained from cameras, image sensors, and/or weather data corresponding to the first time interval. As used herein, the term “image circle” refers to a cross-section of a cone of light transmitted by a lens or series of lenses. The image circle represents a circle of light that is formed when this light strikes a perpendicular target such as film or a digital camera sensor. This mapping determines the area of the effective panel area of the solar panel 104 that is covered by the cloud shadow.
[0070] Furthermore, at step 414, the transition of the cloud shadow on the effective panel area of the solar panel 104 corresponding to subsequent time intervals may be predicted. In some embodiments, one or more edge detection models and/or position prediction models may be retrieved from the set of prediction models PM 214 by the shading prediction unit 234 from the forecasting model repository 210. These models may be used to predict the relative position of the shadow of the cloud with respect to the effective panel area of the solar panel 104 at time intervals subsequent to the first time interval of the forecast interval. In another embodiment, one or more forecasting models FM 212 may be retrieved from the forecasting model repository 210. Some examples of the forecasting models 212 include ARIMA models, ARMA models, and/or MA models, or combinations thereof. These forecasting models FM 212 may be used to track the relative position of the shadow of the cloud with respect to the effective panel area of the solar panel 104 at the subsequent time intervals. Subsequently, at step 416, the one or more areas of the effective panel area of the solar panel 104 that are covered by shadow due to clouds in the corresponding one or more time intervals of the forecast interval are determined. These one or more areas constitute a first set of areas of the solar panels 104 that are under shadow due to clouds for the forecasting interval.
[0071] In addition, wind turbine tower and blade shadow data 420 for one or more solar panels 104 corresponding to the forecast interval may be obtained by the shading prediction unit 234, as indicated by step 418. In one embodiment, the occurrence of shadows due to the wind turbine towers and blades on a solar panel 104 and their relative location to the effective panel area may be determined from siting information and the solar trajectory during the forecast interval. In addition, repetition of the shadow occurrences over the solar panel 104 for the forecast interval may also be predicted. By way of example, repeated shadows cast on the solar panel 104 by one or more turning blades may be predicted. In this way, one or more areas of the effective panel area of the solar panel 104 that are covered by shadows due to wind turbine towers and blades in the forecast interval are determined. These one or more areas constitute a second set of areas of the solar panels 104 that are under shadow due to wind turbine towers and/or blades during the forecasting interval.
[0072] Furthermore, the first set of areas and the second set of areas are combined to obtain the aggregated area under shadow of the effective panel area of the one or more solar panels 104, as indicated in step 422. In one embodiment, the second set of areas of the solar panels 104 under shadow due to wind turbine tower and blade shadows for the forecast interval obtained at step 418 may be overlaid, super-imposed and/or otherwise combined with the first set of areas of the solar panels 104 under shadow due to clouds obtained at step 416 to determine the aggregated area under shadow. In certain embodiments, the shading prediction unit 234 and the solar power forecasting unit 232 may be employed to perform step 422.
[0073] Subsequently, at step 424, an irradiated area corresponding to the solar panels 104 may be computed. To that end, a reduced effective panel area of the solar panels 104 may be determined. This reduced effective panel area is representative of the effective panel area of the one or more solar panels 104 reduced by the aggregated area under shadow obtained at step 422. Further, irradiance per unit area of the reduced effective panel area of the one or more solar panels 104 is computed. In one embodiment, one or more prediction models PM 214 may be retrieved from the forecasting model repository 210 by the solar power forecasting unit 234. The prediction models PM 214 may include an irradiation-power curve. Consequently, the forecast for electrical power production of the one or more solar panels 104 may be determined based on the irradiation-power curve and the computed irradiance per unit area for the reduced effective panel area of the solar panels 104.
[0074] In addition, the solar power forecasting unit 232 may retrieve one or more degradation models DM 216 from the forecasting model repository 210 and solar panel operational data POD 202 from the data repository 208, as indicated by step 426. Furthermore, a check may be carried out to verify if one or more degradation models DM 216 are available, as indicated by step 428. If the degradation models 216 are available, then control passes to step 432. At step 432, the forecast for electrical power production of the one or more solar panels 104 obtained at step 424 may be mapped to one or more DM 216 and POD 202 to generate an adjusted power forecast. In one embodiment, the forecast for electrical power production of the one or more solar panels 104 is down-corrected based on the one or more degradation models 216 corresponding to the solar panels 104 to generate the adjusted power forecast.
[0075] Moreover, at step 438, the adjusted power forecast may be stored to the data repository 208. Alternatively or additionally, the adjusted power forecast may be communicated to the control subsystem 138 and/or a system operator.
[0076] With returning reference to step 428, if the degradation models 216 are not available, control passes to step 430. At step 430, solar historical power production data 408 of the one or more solar panels 104 is retrieved from the data repository 208. One or more forecasting models 212 may be used to derive a forecast for electrical power produced by the one or more solar panels 104 for the forecast interval, based on the solar historical power production 408 of the one or more solar panels 104. Furthermore, at step 434, the forecast may further be down-corrected based on the predicted shadowed area 320, of the effective solar panel area of the one or more solar panels 104. Moreover, control may be passed to step 438, where the forecast may be stored to the data repository 208. Alternatively or additionally, the forecast may be communicated to the control subsystem 138 and/or a system operator.
[0077] FIG. 5 is a flowchart 500 illustrating a method for determining a forecast for electrical power produced by one or more wind turbines 102 of a wind-solar hybrid power generation plant. The method 500 is described with reference to the components of FIGs. 1-2. In one embodiment, steps 508 through 514 may be performed by the wind power forecasting unit 226 and the processor unit 220 of the forecasting subsystem 140.
[0078] At step 508, the wind power forecasting unit 226 may retrieve one or more sets of data from the data repository 208. These sets of data may include weather data WED 206, sensor data 116, turbine operational data 204, and wind historical power production data 506 corresponding to the one or more wind turbines 102. In accordance with aspects of the present specification, selection of a data set may include accounting for current meteorological events, such as a wind ramp event. During a wind ramp event, wind conditions may change from a period of low wind to a period of high wind. Accordingly, a subset of the weather data WED 206 and the sensor data 116 that primarily/predominantly includes wind ramping data may be selected to enhance accuracy of the power forecast. The selection of a subset of data may additionally include accounting for certain maintenance events. For example, if one or more of the wind turbines 102 are undergoing maintenance or are scheduled to undergo maintenance in the near future (within the forecast interval), then the forecast for electrical power production may be lower. Also, data related to the one or more specific wind turbines 102 that are undergoing maintenance may be excluded from use in determining the forecast. Likewise, when the one or more specific wind turbines 102 resume operation, then the data may be adjusted to incorporate turbine operating data 506 and historical power data 508 corresponding to the one or more specific wind turbines 102.
[0079] Referring now to step 510, the wind power forecasting unit 226 may select one or more forecasting models 212 from the forecasting model repository 210. In one embodiment, based on prevailing conditions for the forecast interval, the one or more forecasting models 212 with the best or most suited performance metric may be selected. More particularly, the selection of the one or more forecasting models 212 may be based on finding a match between the selected data subset from step 508 that reflects the prevailing weather conditions such as wind speed, wind direction, cloud conditions, and the like, and a specific forecasting model 212 that provides the best accuracy of forecast for the prevailing weather conditions characterized by the selected data subset.
[0080] Specifically, the wind power forecasting unit 226 may select one or more turbine-level models, such as wind speed prediction models, wind power prediction models, wind direction prediction models, and the like. The wind power forecasting unit 226 may also select one or more availability forecasting models, power curve models, and/or persistence models. The selected models may additionally AR and/or ARIMA based models, ARCS, ARCH models, ARMA models, GARCH models, and/or MA models. The selected models may further include search based models, neural network models, or a combination thereof.
[0081] The wind power forecasting unit 226 may use the selected data subsets from step 508 and the selected models from step 510 and derive a forecast for electrical power production of the one or more available wind turbines 102 for the forecast interval, as indicated by step 512. Further, at step 514, the forecast data may be stored in the data repository 208, communicated to the control subsystem 138 and/or a system operator.
[0082] Having obtained a wind power forecast and a solar power forecast, the forecasting subsystem 140 is configured to apply one or more constraints associated with one or more components of the electrical subsystem 120 and balance of plant. The forecasting subsystem 140 may also adjust the power forecasts for wind and solar components based on one or more regulatory or revenue based criteria. By way of example, if electrical power production from solar panels 104 is priced more profitably for the wind-solar hybrid power generation plant operator for a given time period, the forecast for the wind-solar hybrid power generation plant may be adjusted to provide proportionally more power from the one or more solar panels 104. In another example, in extremely varying wind conditions that may cause high power fluctuation in the electrical subsystem 120, the forecast may be adjusted to provide proportionally more power from the one or more solar panels 104.
[0083] FIG. 6 is a flowchart 600 illustrating a method for further enhancing a forecast of power generated by a wind-solar hybrid power generation plant. In one embodiment, the power forecast may be enhanced by applying balance of plant (BOP) considerations, electrical subsystem component constraints, energy storage capacity, demand management and other influencing regulatory metrics and exogenous conditions to forecasts of wind and solar power. In one embodiment, the method 600 may be performed by the balance of plant (BOP) forecasting unit 230 and the processing unit 220 of the forecasting subsystem 140. The method 600 may be described with reference to the components of FIGs. 1-5.
[0084] At step 603, wind forecast data 601 generated by the wind power forecasting unit 226 (see method 500 of FIG. 5) and solar forecast data 602 generated by the solar power forecasting unit 232 (see method 400 of FIG. 4) are retrieved by the BOP forecasting unit 230 from the data repository 208. Subsequently, at step 604, the balance of plant operating data BOPD 218 is retrieved from the data repository 208. As previously noted, the BOPD 218 may include constraints relating to the existing generating wind turbines and solar panels, pricing mechanisms, capacity for energy storage, demand management and other factors. The BOPD 218 may further include power constraints of one or more components of the electrical subsystem 120. Further, at step 606, these constraints/considerations are applied to down-correct the wind forecast data 601 and the solar forecast data 602. Additionally, based on regulatory and pricing metrics, the wind and solar forecasts 601, 602 may be re-adjusted to lower or higher values, as indicated by step 608.
[0085] Moreover, at step 610, a loss compensation factor to offset the loss of power occurring in one or more components of the balance of plant such as cabling, transformer losses, and combinations and variations thereof may be applied to the wind forecast and solar forecast to further adjust the forecast values. Consequent to the processing of steps 603-610, an aggregate forecast 612 of the wind forecast and the solar forecast may be generated. In addition, the BOP forecasting unit 230 may communicate the aggregated forecast 612 to the data repository 208, the control subsystem 138, and/or a system operator as indicated by step 614.
[0086] Various embodiments of a forecasting system for a hybrid wind-solar power generation plant that enhance forecasting for future electrical power production of the wind-solar hybrid power generation plant are presented. Moreover, the forecasting system is configured to factor in additional conditions such as the shared electrical unit component considerations, regulatory metrics, financial/economic constraints, prevailing weather conditions to produce an improved forecast. Further, the forecasting system is also configured to account for shadows caused by clouds and/or neighboring wind turbine towers and/or blades to enhance the accuracy of the power forecast.
[0087] It is to be understood that not necessarily all such objects or advantages described above may be achieved in accordance with any particular embodiment. Thus, for example, those skilled in the art will recognize that the systems and techniques described herein may be embodied or carried out in a manner that achieves or improves one advantage or group of advantages as taught herein without necessarily achieving other objects or advantages as may be taught or suggested herein.
[0088] While the technology has been described in detail in connection with only a limited number of embodiments, it should be readily understood that the specification is not limited to such disclosed embodiments. Rather, the technology can be modified to incorporate any number of variations, alterations, substitutions or equivalent arrangements not heretofore described, but which are commensurate with the spirit and scope of the claims. Additionally, while various embodiments of the technology have been described, it is to be understood that aspects of the specification may include only some of the described embodiments. Accordingly, the specification is not to be seen as limited by the foregoing description, but is only limited by the scope of the appended claims.