原文信息：
Day-ahead probabilistic forecasting at aco-located wind and solar power park in Sweden: Trading and forecastverification
原文链接：
https://www.sciencedirect.com/science/article/pii/S2666792422000385
Highlights
We study the effect of aggregation at aco-located wind and solar power park.
We assess the performance of probabilisticforecasts in terms of quality and value.
Co-location improves the accuracy of forecaststhe most in the spring, summer and fall.
A ratio of 50% - 60% wind power in the combinedpark improves the accuracy the most.
The improved forecasts reduce the regulationcosts in the day-ahead market.
Abstract
This paper presents a first step in the field ofprobabilistic forecasting of co-located wind and photovoltaic (PV) parks. Theeffect of aggregation is analyzed with respect to forecast accuracy and valueat a co-located park in Sweden using roughly three years of data. We use afixed modelling framework where we post-process numerical weather predictionsto calibrated probabilistic production forecasts, which is a prerequisite whenplacing optimal bids in the day-ahead market. The results show that aggregationimproves forecast accuracy in terms of continuous ranked probability score,interval score and quantile score when compared to wind or PV power forecastsalone. The optimal aggregation ratio is found to be 50%–60% wind power and theremainder PV power. This is explained by the aggregated time series beingsmoother, which improves the calibration and produces sharper predictivedistributions, especially during periods of high variability in both resources,i.e., most prominently in the summer, spring and fall. Furthermore, the dailyvariability of wind and PV power generation was found to be anti-correlatedwhich proved to be beneficial when forecasting the aggregated time series.Finally, we show that probabilistic forecasts of co-located production improvetrading in the day-ahead market, where the more accurate and sharper forecastsreduce balancing costs. In conclusion, the study indicates that co-locatingwind and PV power parks can improve probabilistic forecasts which, furthermore,carry over to electricity market trading. The results from the study should begenerally applicable to other co-located parks in similar climates.
Keywords
Forecast value
Quantile forecasts
PV power
Wind power
Hybrid power park
Probabilistic forecasting
Graphics
Fig. 1. Flowchart illustrating the input data,method steps and analysis in this study. The input data are illustrated asparallelograms.
Fig. 2. Measured median of diurnal variabilityof PV power (upper subplot) and wind power (lower subplot) for the differentseasons at the co-located wind and PV power park. The blue envelopes show theobserved 20–80% quantile and the black solid lines show the median.
Fig. 4. Two example days of normalized PV, windand aggregation of equal level of wind and PV power production forecasts at theco-located park. The columns represent the different power sources. The rowsrepresent the different issue times in UTC. The prediction intervals are shownas envelopes with nominal coverage range from 10% to 90% and the observationsas solid red lines.
Fig. 5. PIT histograms for all look-ahead timesfor PV power, wind power and 50% aggregation level at the studied site forshuffle 2. Consistency intervals are denoted by the horizontal dashed lines andthe red lines show a perfectly reliable forecast. Note the differences on they-axes.
Fig. 6. PIAW as a function of the nominalcoverage rate of the prediction intervals for different nominal capacity ofwind power to the nominal capacity of the entire park. The rows corresponds tothe shuffles and the columns to the seasons.
Fig. 7. The rows present the CRPS andcorresponding decomposition scores as well as the variance for different sharesof wind in the combined park for the different seasons and shuffled data sets,respectively. The columns corresponds to the seasons.
关于Applied Energy
本期小编：李辉 审核人：冯玉鹏
《Applied Energy》是世界能源领域著名学术期刊，在全球出版巨头爱思唯尔 (Elsevier) 旗下，1975年创刊，影响因子11.446，CiteScore 20.4，高被引论文ESI全球工程期刊排名第4，谷歌学术全球学术期刊第50，本刊旨在为清洁能源转换技术、能源过程和系统优化、能源效率、智慧能源、环境污染物及温室气体减排、能源与其他学科交叉融合、以及能源可持续发展等领域提供交流分享和合作的平台。开源（Open Access）姊妹新刊《Advances in Applied Energy》现已正式上线。在《Applied Energy》的成功经验基础上，致力于发表应用能源领域顶尖科研成果，并为广大科研人员提供一个快速权威的学术交流和发表平台，欢迎关注！
公众号团队小编招募长期开放，欢迎发送自我简介（含教育背景、研究方向等内容）至wechat@applied-energy.org
点击“阅读原文”
喜欢我们的内容？
点个“赞”或者“再看”支持下吧！
