Zanesljivost napovedi proizvodnje energije kombiniranih vetrnih in sončnih elektrarn

Senica, Alenka (author), Žagar, Nedjeljka (mentor), Žagar, Mark (co-mentor)

Za enostavnejši prehod iz fosilnih goriv na obnovljive vire energije in s tem nizkoogljično družbo potrebujemo zanesljive napovedi o hitrosti in moči vetra, obsevanosti oziroma moči sončnega sevanja. Te napovedi omogočajo načrtovanje obratovanja vetrnih, sončnih ali kombiniranih elektrarn in integracijo energije proizvedene iz obnovljivih virov v električno omrežje. Za pripravo kakovostnih napovedi vremena je pomembna tudi njihova verifikacija, ki pripomore k razumevanju pomanjkljivosti prognostičnih modelov in razvoju postopkov za njihove izboljšave. Cilj magistrskega dela je preveriti, ali prognostični modeli bolje napovedujejo proizvodnjo energije kombinirane elektrarne kot proizvodnjo le vetrne ali le sončne elektrarne. Izbranih je bilo pet lokacij vetrnih parkov in merilnih postaj obsevanosti v severnem delu Nemčije. Tri lokacije se nahajajo v neposredni bližini morja, dve sta pomaknjeni bolj v notranjost severnega dela Nemčije. Z različnimi legami elektrarn smo želeli preveriti, kako bližina morja vpliva na količino proizvedene energije vetrnih parkov, saj so hitrosti vetra ob obalnih predelih višje. Na napovedljivost hitrosti in posledično moči vetra vpliva model, ki čim bolje opisuje procese v mejni plasti. Verifikacija hitrosti in moči vetra, obsevanosti in moči sončnega sevanja napovedi modela ECMWF ponazarja, da je napovedljivost obsevanosti oziroma moči sončnega sevanja boljša od napovedljivosti moči vetra. Primerjali smo tudi analize višje ločljivosti modela WRF z meritvami hitrosti vetra ter obsevanosti. Napake analiz modela WRF so višje od napak napovedi modela ECMWF, saj model WRF nima lastne asimilacije podatkov. Pomembno je, da se podatki asimilirajo v modelu visoke ločljivosti. Temeljita primerjava meritev in napovedi oziroma analiz modela je bila opravljena le za eno lokacijo. Na podlagi rezultatov primerjave kombiniranih elektrarn na različnih lokacijah vidimo, da je napovedljivost proizvodnje energije kombinirane elektrarne za štiri od petih obravnavanih lokacij boljša od napovedljivosti proizvodnje energije vetrne elektrarne. Pri verifikaciji napovedi proizvodnje vetrne in sončne energije opazimo dnevni cikel napake. Napake napovedi kombinirane energije so manjše, saj sta sončna in vetrna energija časovno medsebojno odvisni.

• Language: Slovenian
• Keywords: mejna plast, vetrovi, sevalnost, obsevanost, obnovljivi viri energije, vetrna energija, sončna energija, verifikacija vremenskih napovedi
• Work type: Master's thesis/paper
• Typology: 2.09 - Master's Thesis
• Organization: FMF - Faculty of Mathematics and Physics
• Year: 2019
• PID: 20.500.12556/RUL-109540
• COBISS.SI-ID: 383913
• Publication date in RUL: 05.09.2019

Secondary language

• Language: English
• Title: Forecast verification of the production of energy of the combined wind and solar power plants

Abstract

For a smooth transition from fossil fuels to renewable energy sources and consequently a low-carbon society, we need reliable forecasts of wind speed and wind power, solar irradiation or solar energy. The weather forecasts enable us to plan the operation of wind, solar or combined power plants and integrate energy generated from renewable energy sources into the power grid. For the preparation of quality weather forecasts, their verification is crucial to help us understand the shortcomings of prognostic models and develop procedures for their improvements. The aim of the master's thesis is to verify whether prognostic models better predict the production of energy of the combined power plants or energy generated only from wind farms, or only from solar power plants. In northern Germany, five sites of wind parks and weather stations with pyranometers were selected. Three sites are located in the vicinity of the sea, two in the interior of northern Germany. With different sites of combined power plants, we investigated how the proximity of the sea affected the amount of wind energy produced because wind speed on coastal areas is higher. The predictability of wind speed and wind power is affected by how well a model describes processes in the boundary layer. The forecast verification of wind speed and wind power, solar irradiation and solar energy of the ECMWF model showed that the predictability of solar irradiation or solar energy was better than the predictability of wind power. We also compared the WRF model downscaling analyses with wind speed and solar irradiation measurements. The WRF model downscaling analyses errors were higher than the ECMWF model forecasts errors since the WRF model does not have its own data assimilation. It is essential that data are assimilated in a high-resolution model. A thorough comparison of the measurements and forecasts or downscaling analyses was carried out for only one location. Based on the results of combined power plants which are located in different sites, we saw that the predictability of energy production of the combined power plant for four out of five sites considered was better than the predictability of wind power generation. When we verified the forecasts of wind and solar energy production, we observed a daily cycle of error. The predicted errors of combined energy were lower because solar and wind energy are interdependent over time.

• Keywords: boundary layer structure and processes, winds and their effects, radiative processes, solar radiation, energy analysis, renewable energy targets, wind energy, solar energy