2016 North American Power Symposium (NAPS) 2016
DOI: 10.1109/naps.2016.7747859
|View full text |Cite
|
Sign up to set email alerts
|

Solar forecasting by K-Nearest Neighbors method with weather classification and physical model

Help me understand this report

Search citation statements

Order By: Relevance

Paper Sections

Select...
2
1
1
1

Citation Types

0
16
0

Year Published

2017
2017
2024
2024

Publication Types

Select...
6
2
1

Relationship

0
9

Authors

Journals

citations
Cited by 24 publications
(16 citation statements)
references
References 18 publications
0
16
0
Order By: Relevance
“…The proposed hybrid approach gives a mean absolute bias error (MABE) value of 42 W/m 2 and the RMSE value of 242 W/m 2 in the experimental results. In another study [26], researchers proposed a novel kNN‐based forecasting model considering NWP, solar irradiation input parameters, and physical model of PV units. This proposed model also implements weather condition classification to achieve more accurate forecasting results.…”
Section: Analysis and Resultsmentioning
confidence: 99%
“…The proposed hybrid approach gives a mean absolute bias error (MABE) value of 42 W/m 2 and the RMSE value of 242 W/m 2 in the experimental results. In another study [26], researchers proposed a novel kNN‐based forecasting model considering NWP, solar irradiation input parameters, and physical model of PV units. This proposed model also implements weather condition classification to achieve more accurate forecasting results.…”
Section: Analysis and Resultsmentioning
confidence: 99%
“…The performances of hourly day-ahead PVPG forecasting by different models are further compared in Subsection 5.5. In this work, the proposed AD-LSTM model is mainly compared with a commonly-used persistence model (i.e., a forecasting model in which observations of the last day are persisted forward as the forecasting results) [5], a statistical model (i.e., the autoregressive integrated moving average model -ARIMA [13]), a conventional machine learning model (i.e., the k-nearest neighbors -KNN [17]), and the OL-LSTM model [30][31][32]. It is noted that the same hyper-parameters and network structures are adopted for both OL-LSTM and AD-LSTM.…”
Section: Day-ahead Forecasting Of Pv Power Generationmentioning
confidence: 99%
“…Accurate PVPG forecasting is a crucial research arena, and a substantial body of research has been devoted to it. In general, PVPG forecasting methodologies illustrated in previous works can be divided into four major categories: the physical model [11,12]; the statistical model [13][14][15][16]; the machine learning (ML) model [17][18][19][20][21]; and the hybrid model [22][23][24].…”
Section: Introductionmentioning
confidence: 99%
“…Some methods were used artificial intelligent techniques such as artificial neural network as in [16][17][18][19][20]. Other methods used data mining techniques like support vector machine [21][22][23][24], K-nearest neighbor as in [25][26][27]. In addition, there are some optimization techniques such as genetic algorithm and particle swarm in were used to predict and improve the solar system depending on environmental factors such as the temperature, wind, and cloud [28][29][30][31].…”
Section: Fig 3 Pv and Wind Energy Project Distributionmentioning
confidence: 99%