Solar power forecasting: Difference between revisions

Revision as of 14:50, 11 April 2021

Revision as of 22:48, 8 May 2021

Line 1:

{{short description|Power forecasting}}

{{Multiple issues|{{more footnotes|date=July 2012}}}}

{{cleanup|reason=unencyclopedic, incomprehensible and unverified information|date=October 2017}}

{{underlinked|date=January 2013}}

{{more footnotes|date=July 2012}}

}}

{{Sustainable energy}}

'''Solar power forecasting''' is the gathering and analysis of data in order to predict the optimal conditions for [[Solar energy|solar power generation.]]

'''Solar power forecasting''' involves knowledge of the [[Sun]]´s path, the [[atmosphere]]'s condition, the scattering processes and the characteristics of a [[solar energy]] plant which utilizes the Sun's energy to produce [[solar power]]. Solar [[photovoltaic system]]s transform solar energy into electric power. The power output depends on the incoming radiation and on the solar panel characteristics. Photovoltaic power production is increasing nowadays. Forecast information is essential for an efficient use, the management of the electricity grid and for solar energy trading. Common solar forecasting method include stochastic learning methods, local and remote sensing methods, and hybrid methods (Chu et al. 2016).

'''SolarThis powerinformation forecasting''' involves knowledge ofincludes the [[Sun]]´s path, the [[atmosphere]]'s condition, the scattering processes and the characteristics of athe [[solar energy]] plant which utilizes the Sun's energy to produce [[solar power]]. Solar [[photovoltaic system]]s transform solar energy into electric power. The power output depends on the incoming radiation and on the solar panel characteristics. PhotovoltaicSolar power production is increasing nowadays. Forecastforecast information is essentialused for an efficient use, the management of the electricity[[Electrical grid|electric grid]] and for solar energy trading. Common solar forecasting method include stochastic learning methods, local and remote sensing methods, and hybrid methods (Chu et al. 2016).

==Generation forecasting==

Common solar forecasting method include [[stochastic]] learning methods, local and remote sensing methods, and hybrid methods (Chu et al. 2016).

The energy generation forecasting problem is closely linked to the problem of ''weather variables forecasting''. Indeed, this problem is usually split into two parts, on one hand focusing on the forecasting of solar PV or any other meteorological variable and on the other hand estimating the amount of energy that a concrete power plant will produce with the estimated meteorological resource.

It is useful to classify these techniques depending on the forecasting horizon

In general, the way to deal with this difficult problem is usually related to the spatial and temporal scales we are interested in, which yields to different approaches that can be found in the literature. In this sense, it is useful to classify these techniques depending on the forecasting horizon, so it is possible to distinguish between ''now-casting'' (forecasting 3–4 hours ahead), ''short-term forecasting'' (up to seven days ahead) and ''long-term forecasting'' (months, years...)

Solar radiation closely follows the physical and biological development of the earth. Its spatial and sequential heterogeneity powerfully influence the forcing of environmental and hydrological organisms by manipulating air temperature, soil moisture and vapor transpiration, snow cover and many photochemical procedures. Therefore, solar radiation drives place efficiency and plant life allotment, organism a key feature in undeveloped and forestry sciences that be obliged to be known precisely. The quantity of solar radiation obtainable at the earth’ surface is at the outset controlled at worldwide balance, organism above all precious by the Sun Earth geometry and the atmosphere. On the other hand, a complete explanation of its freedom time unpredictability require deliberation of limited procedure which frequently turn out to be also applicable, as is the casing in mountainous region. Predominantly, limited territory adjust the inward bound solar radiation by shadow casts, slope of elevation, surface gradient and compass reading, as a result, precise spatial model of inward bound solar radiation be supposed to regard as the pressure of the terrain surface. In the final time, more than a few events to consist of the confined terrain special effects in the solar radiation countryside have been projected, such as the use of Geographical Information Systems (GIS), artificial intelligence or post dispensation of satellite stand technique. Solar radiation can be also evaluated using numerical weather forecast (NWP) models. Nevertheless, the space and time balance determined with them and the incomplete computational ability frequently avoid the deliberation of terrain connected property.

* ''now-casting'' (forecasting 3–4 hours ahead),

Otherwise, exclamation technique agree to us to acquire spatially persistent database from data evidence at inaccessible station greater than wide region. Even though their dependability is powerfully needy on the opening coldness between position, they eventually rely on experiential statistics, which have a superior precision than extra method. Therefore, while an adequate footage spatial thickness is accessible, disturbance method are preferred. Conventionally, solar radiation has not been as densely example as additional variables as temperature or rainfall, therefore the ease of use of capacity is frequently in short supply. Though, the number of experimental system which record solar radiation has developed and interruption has been converted into an appropriate technique for solar radiation evaluation. Nevertheless, radiometric stations are frequently come together approximately farmland or occupied region, typically during basin and plane area, while mountains at rest require enough footage thickness. This truth is particularly applicable afford the tall spatial unpredictability of solar radiation in these province. As an outcome, particular interruption method that tolerate include outdoor foundation should be used to make clear this extra spatial unpredictability. Several diverse spatial interruption techniques can be established. On the other hand, data ease of use in mountainous region is often extremely restricted. As a result, it is hard to construct a precise solar radiation climatology in hilly area to be used in environmental science, climate change.

* ''short-term forecasting'' (up to seven days ahead) and

* ''long-term forecasting'' (months, years...)

Solar radiation is a hardly illustration changeable with reverence to supplementary ecological variables such as temperature or precipitation, in fraction payable to the high maintenance price of the necessary radiometric sensors. It is extremely sensitive to ecological feature on or after local to limited balance. Predominantly, terrain surface confronts the conventional interruption method while forecast through far above the ground spatial decision are required, particularly due to the lack of measurement stations in mountainous areas. Geo-statistics front a stochastic move toward to resolve the spatial forecast difficulty that stop dependence on before imagine deterministic models and permit us to consist of the consequence of outside in sequence foundation stand on investigation data-sets.

==Nowcasting==

The term "Nowcasting" in the context of solar power forecasting, generally refers to same spatial and temporal scales as [[Nowcasting (meteorology)|meteorological Nowcasting]], which focuses on forecast horizons from a few minutes ahead, out to 4–6 hours ahead. In general, the 'Nowcasting' forecast horizon refers to those not well-served by global numerical weather prediction models, which produce outputs at up to hourly resolutions and only update every 6 hours. [[Solar power]] nowcasting then refers to the prediction of solar power output (or energy generation) over time horizons of tens to hundreds of minutes ahead of time with up to 90% predictability.<ref>{{cite web |last1=Vorrath |first1=Sophie |title=New APVI solar tool shows daily, time-based forecast for each state |url=https://reneweconomy.com.au/new-apvi-solar-tool-shows-daily-time-based-forecast-for-each-state-52445/ |website=RenewEconomy |language=en-AU |date=31 May 2019}}</ref> It has historically been very important for [[Electrical grid|electrical grid operators]] in order to guarantee the matching of supply and demand on [[Energy market|energy markets]]. Such solarSolar power nowcasting services are usually related to temporal resolutions of 5 to 15 minutes, with updates as frequent as every 5 minutes. The regular updates and relatively high resolutions required from these methods require automatic weather data acquisition and processing techniques, which are chiefly accomplished by two primary means:<ref>{{Cite web|url= https://www.elsevier.com/books/solar-energy-forecasting-and-resource-assessment/kleissl/978-0-12-397177-7 |title=Solar Energy Forecasting and Resource Assessment - 1st Edition|website=www.elsevier.com |access-date=2019-05-08}}</ref>

The regular updates and relatively high resolutions required for nowcast require automatic weather data acquisition and processing techniques. These are accomplished by three primary means:<ref>{{Cite web|url= https://www.elsevier.com/books/solar-energy-forecasting-and-resource-assessment/kleissl/978-0-12-397177-7 |title=Solar Energy Forecasting and Resource Assessment - 1st Edition|website=www.elsevier.com |access-date=2019-05-08}}</ref>

#'''Statistical techniques.''' These are usually based on [[time series]] processing of measurement data, including [[Surface weather observation|meteorological observations]] and power output measurements from a solar power facility. What then follows is the creation of a [[Training, validation, and test sets|training dataset]] to tune the parameters of a model (I. Espino eta al, 2011), before evaluation of model performance against a separate testing dataset. This class of techniques includes the use of any kind of statistical approach, such as [[Autoregressive–moving-average model|autoregressive moving averages]] (ARMA, ARIMA, etc.), as well as machine learning techniques such as [[Artificial neural network|neural networks]], [[support vector machine]]s (etc.). These approaches are usually benchmarked to a persistence approach in order to evaluate their improvements. This persistence approach just assumes that any variable at time step t is the value it took in a previous time.

#[[File:Satellite Based Solar Nowcasting.gif|thumb|An example of satellite-based cloud cover nowcasting, which is used to generate predication of solar power outputs. Credit: [http://solcastglobal.com Solcast]]]'''Satellite based methods.''' These methods leverage the several [[Geostationary orbit|geostationary]] Earth observing [[weather satellite]]s (such as [[Meteosat|Meteosat Second Generation (MSG) fleet]]'')'' to detect, characterise, track and predict the future locations of cloud cover. These satellites make it possible to generate solar power forecasts over broad regions through the application of image processing and forecasting algorithms. Key forecasting algorithms include cloud motion vectors (CMVs).<ref>{{Cite web|url= http://glossary.ametsoc.org/wiki/Cloud_motion_vector |title=Cloud motion vector - AMS Glossary|website=glossary.ametsoc.org |access-date=2019-05-08}}</ref> Relevant methods for applying physical models based on satellite image processing techniques provide an estimation of future atmospheric values can be found in ''Alvarez et al.'', 2010.

=== Statistical techniques ===

#[[File:Diagram-of-the-UCSD-sky-imager-USI-and-related-solar-and-sky-geometries-th-0-is-the.jpg|alt=|thumb|An example of sky-imager used for detecting, tracking and predicting cloud cover conditions in the vicinity of a solar energy facility of interest. Most often, these devices are used to make estimates of solar irradiance from the images using local calibration by a pyranometer. The solar irradiance short-term forecasts are then fed into PV power modelling routines to generate a solar power forecast. <br />Credit: [https://topex.ucsd.edu/rs/presentations/cortes.pdf UC San Diego]]]'''Ground based techniques.''' These techniques are generally used to derive irradiance forecasts with much higher spatial and temporal resolution compared with the satellite-based forecasts. Local cloud information is acquired by one or several ground-based sky imagers at a high frequency (1 minute or less). The combination of these images and local weather measurement information are processed to simulate cloud motion vectors and optical depth to obtain forecasts up to 30 minutes ahead.

#'''StatisticalThese techniques.''' These are usually based on [[time series]] processing of measurement data, including [[Surface weather observation|meteorological observations]] and power output measurements from a solar power facility. What then follows is the creation of a [[Training, validation, and test sets|training dataset]] to tune the parameters of a model (I. Espino eta al, 2011), before evaluation of model performance against a separate testing dataset. This class of techniques includes the use of any kind of statistical approach, such as [[Autoregressive–moving-average model|autoregressive moving averages]] (ARMA, ARIMA, etc.), as well as machine learning techniques such as [[Artificial neural network|neural networks]], [[support vector machine]]s (etc.). These approaches are usually benchmarked to a persistence approach in order to evaluate their improvements. This persistence approach just assumes that any variable at time step t is the value it took in a previous time.

=== Satellite based methods ===

#[[File:Satellite Based Solar Nowcasting.gif|thumb|An example of satellite-based cloud cover nowcasting, which is used to generate predication of solar power outputs. Credit: [http://solcastglobal.com Solcast]]]'''Satellite based methods.''' These methods leverage the several [[Geostationary orbit|geostationary]] Earth observing [[weather satellite]]s (such as [[Meteosat|Meteosat Second Generation (MSG) fleet]]'')'' to detect, characterise, track and predict the future locations of [[cloud cover]]. These satellites make it possible to generate solar power forecasts over broad regions through the application of [[image processing]] and forecasting [[Algorithm|algorithms]]. Key forecasting algorithms include cloud motion vectors (CMVs).<ref>{{Cite web|url= http://glossary.ametsoc.org/wiki/Cloud_motion_vector |title=Cloud motion vector - AMS Glossary|website=glossary.ametsoc.org |access-date=2019-05-08}}</ref> Relevant[[File:Diagram-of-the-UCSD-sky-imager-USI-and-related-solar-and-sky-geometries-th-0-is-the.jpg|alt=|thumb|An methodsexample of sky-imager used for applyingdetecting, physicaltracking modelsand basedpredicting oncloud satellitecover imageconditions processingin techniquesthe providevicinity of a solar anenergy estimationfacility of futureinterest. atmosphericMost valuesoften, canthese bedevices foundare inused ''Alvarezto make estimates of solar irradiance from the images using local calibration by eta alpyranometer.'', 2010The solar irradiance short-term forecasts are then fed into PV power modelling routines to generate a solar power forecast. <br />Credit: [https://topex.ucsd.edu/rs/presentations/cortes.pdf UC San Diego]]]

=== Ground based techniques. ===

#[[File:Diagram-of-the-UCSD-sky-imager-USI-and-related-solar-and-sky-geometries-th-0-is-the.jpg|alt=|thumb|An example of sky-imager used for detecting, tracking and predicting cloud cover conditions in the vicinity of a solar energy facility of interest. Most often, these devices are used to make estimates of solar irradiance from the images using local calibration by a pyranometer. The solar irradiance short-term forecasts are then fed into PV power modelling routines to generate a solar power forecast. <br />Credit: [https://topex.ucsd.edu/rs/presentations/cortes.pdf UC San Diego]]]'''Ground based techniques.''' These techniques are generally used to derive [[irradiance]] forecasts with much higher spatial and temporal resolution compared with the satellite-based forecasts. Local cloud information is acquired by one or several ground-based sky imagers at a high frequency (1 minute or less). The combination of these images and local weather measurement information are processed to simulate cloud motion vectors and [[optical depth]] to obtain forecasts up to 30 minutes ahead.

==Short-term solar power forecasting==

''Short-term'' forecasting provides predictions up to 7seven days ahead. This kind of forecast is also valuable for grid operators in order to make decisions of grid operation, as well as, for electric market operators.<ref>{{cite journal|last1=Sanjari|first1=M.J.|last2=Gooi|first2=H.B.|title=Probabilistic Forecast of PV Power Generation based on Higher-order Markov Chain|journal=IEEE Transactions on Power Systems|date=2016|volume=32|issue=4|pages=2942–2952|doi=10.1109/TPWRS.2016.2616902}}</ref>

Under this perspective, the meteorological resources are estimated at a different temporal and spatial resolution. This implies that meteorologicalMeteorological variables and phenomena are looked from a more general perspective, not as local as nowcasting services do.

=== InNumerical thisweather sense,prediction most===
Most of the approachesshort maketerm useforecast ofapproaches differentuse numerical weather prediction models (NWP) that provide an initial estimation of weather variables. Currently,These severalincluded models are available for this purpose, such asthe [[Global Forecast System]] (GFS) or data provided by the European Center for Medium Range Weather Forecasting ([[ECMWF]]). These two models are considered the state of the art of global forecast models, which provide meteorological forecasts all over the world.

In order to increase spatial and temporal resolution of these models, other models have been developed which are generally called mesoscale models. Among others, [[HIRLAM]], [[Weather Research and Forecasting Model|WRF]] or [[MM5 (weather model)|MM5]] are the most representative of these models since they are widely used by different communities.

To run these models a wide expertise is needed in order to obtain accurate results, due to the wide variety of parameters that can be configured in the models. In addition, sophisticated techniques such as [[data assimilation]] might be used in order to produce more realistic simulations.

Finally, someSome communities argue for the use of post-processing techniques, once the models’ output is obtained, in order to obtain a [[Probability|probabilistic]] point of view of the accuracy of the output. This is usually done with ensemble techniques that mix different outputs of different models perturbed in strategic meteorological values and finally provide a better estimate of those variables and a degree of uncertainty, like in the model proposed by Bacher et al. (2009)

==Long-term solar power forecasting==

''Long-term'' forecasting usually refers to forecasting of the annual or monthly available resource. This is useful for energy producers and to negotiate contracts with financial entities or [[Public utility|utilities]] that distribute the generated energy.

In general, these long-term forecasting is usually done at a lower scale than any of the two previous approaches. Hence, most of these models are run with [[Mesoscale meteorology|mesoscale]] models fed with reanalysis data as input and whose output is postprocessed with [[Statistics|statistical]] approaches based on measured data.

==Energetic models==

Any output from anya model described above must then be converted to the electric energy that a particular solar PV plant will produce. This step is usually done with statistical approaches that try to correlate the amount of available resource with the metered power output. The main advantage of these methods is that the meteorological prediction error, which is the main component of the global error, might be reduced taking into account the uncertainty of the prediction.

As it was mentioned before and detailed in ''Heinemann et al.'', these statistical approaches comprises from ARMA models, neural networks, support vector machines, etc.

On the other hand, there also exist theoretical models that describe how a power plant converts the meteorological resource into electric energy, as described in Alonso et al. The main advantage of this type of models is that when they are fitted, they are really accurate, although they are too sensitive to the meteorological prediction error, which is usually amplified by these models.

Hybrid models, finally, are a combination of these two models and they seem to be a promising approach that can outperform each of them individually.