以倒傳遞類神經網路預測氣象站之風速與風向資料

PREDICTION OF WIND SPEED AND DIRECTION FROM METEOROLOGICAL STATIONS BY USING BACK-PROPAGATION NEURAL NETWORK

10.6652/JoCICHE.202309_35(5).0004

黎益肇(Yi-Chao Li)；陳瑞華(Rwey-Hua Cherng)；林柏宇(Bo-Yu Lin)；吳政陽(Cheng-Yang Wu)

倒傳遞類神經網路 ； 風速預測 ； 氣象資料 ； back-propagation neural network ； wind speed prediction ； meteorological Data

中國土木水利工程學刊

35卷5期（2023 / 09 / 01）

473 - 482

繁體中文;英文

過往在進行風環境分析時，常以季風為主要考量，蒐集鄰近氣象測站多年之風速風向資料進行統計分析，可能因時間過或儀器損壞導致統計代表性不足。本研究利用倒傳遞類神經網路（BPN），透過歷史氣象資料，調整網路輸入層參數、輸入的測站數量、隱藏層轉換函數並建立BPN模型。本研究以季風預測為主要目標，採土城測站鄰近氣象測站之2016～2020期間氣象參數作為訓練資料。比較了8組氣象參數組合、9種輸入測站組合、6種轉換函數組合、逐次比較誤差指標後選定最佳的輸入組合。結果顯示在比較目標測站風速歷時後，整體趨勢觀察頗為吻合，相關係數為0.769，平均風速偏差為0.11 m/s，適用於季風之風環境統計分析。而對於變異性較大的風速歷時，因局部極值風速則易被低估，使得平均誤差值26.4%偏高，故本預測模式無法合理反應極值風速，需進一步尋求適當模式予以修正。

Generally, when conducting wind environment analysis, monsoon winds were the primary consideration. However, wind speed and direction data from nearby meteorological stations were collected for years for statistical analysis, which might not be statistically representative due to time lapse or instrument damage. This study uses the backward-passing neural network (BPN) to adjust the input layer parameters, the number of input stations, the hidden layer conversion function, and the BPN model through historical meteorological data. The main objective of this study is monsoon prediction, and the meteorological parameters of the neighboring meteorological stations of the target stations for the period of 2016 ~ 2020 are used as training data. Eight combinations of meteorological parameters, nine combinations of input stations, and six combinations of conversion functions are compared. The best input combination is selected after comparing the error indicators one by one. The results show that after comparing the wind speed time-history data of the target stations, the overall trend is relatively consistent with the correlation coefficient of 0.769 and the average wind speed deviation of 0.11 m/s, which is suitable for statistical analysis of the monsoon wind environment. However, for the wind speed time-history data with large variability, the local extreme wind speed is easily underestimated, making the average error value 26.4% higher. Therefore, the prediction model cannot reasonably reflect the extreme wind speed, and we need to find a suitable model to correct it further.

1. Cadenas, E.,Rivera, W.(2009).Short term wind speed forecasting in La Venta, Oaxaca, México, using artificial neural networks.Renewable Energy,34(1),274-278.
2. Fadare, D.(2010).The application of artificial neural networks to mapping of wind speed profile for energy application in Nigeria.Applied Energy,87(3),934-942.
3. Filik, Ü. B.,Filik, T.(2017).Wind speed prediction using artificial neural networks based on multiple local measurements in Eskisehir.Energy Procedia,107,264-269.
4. Goodfellow, I.,Bengio, Y.,Courville, A.(2016).Deep learning.MIT press.
5. Hiester, T.,Pennell, W.(1979).,Richland, WA (USA):.
6. Khosravi, A.,Koury, R.,Machado, L.,Pabon, J.(2018).Energy, exergy and economic analysis of a hybrid renewable energy with hydrogen storage system.Energy,1087-1102.
7. Kisi, O.,Kerem Cigizoglu, H.(2007).Comparison of different ANN techniques in river flow prediction.Civil Engineering and Environmental Systems,24(3),211-231.
8. Kohonen, T.(1988).An introduction to neural computing.Neural Networks,1(1),3-16.
9. Koo, J.,Han, G. D.,Choi, H. J.,Shim, J. H.(2015).Wind-speed prediction and analysis based on geological and distance variables using an artificial neural network: A case study in South Korea.Energy,93,1296-1302.
10. Li, G.,Shi, J.(2010).On comparing three artificial neural networks for wind speed forecasting.Applied Energy,87(7),2313-2320.
11. Mabel, M. C.,Fernandez, E.(2008).Analysis of wind power generation and prediction using ANN: A case study.Renewable Energy,33(5),986-992.
12. Philippopoulos, K.,Deligiorgi, D.(2012).Application of artificial neural networks for the spatial estimation of wind speed in a coastal region with complex topography.Renewable Energy,38(1),75-82.
13. Rumelhart, D. E.,Hinton, G. E.,Williams, R. J.(1986).Learning representations by back-propagating errors.Nature,323(6088),533-536.
14. Salcedo-Sanz, S.,Perez-Bellido, A. M.,Ortiz-García, E. G.,Portilla-Figueras, A.,Prieto, L.,Correoso, F.(2009).Accurate short-term wind speed prediction by exploiting diversity in input data using banks of artificial neural networks.Neurocomputing,72(4-6),1336-1341.
15. Samadianfard, S.,Hashemi, S.,Kargar, K.,Izadyar, M.,Mostafaeipour, A.,Mosavi, A.,Nabipour, N.,Shamshirband, S.(2020).Wind speed prediction using a hybrid model of the multi-layer perceptron and whale optimization algorithm.Energy Reports,6,1147-1159.
16. Sharifi, Y.,Lotfi, F.,Moghbeli, A.(2019).Compressive strength prediction using the ANN method for FRP confined rectangular concrete columns.Journal of Rehabilitation in Civil Engineering,7(4),134-153.
17. Song, J.,Wang, J.,Lu, H.(2018).A novel combined model based on advanced optimization algorithm for short-term wind speed forecasting.Applied Energy,215,643-658.
18. Velázquez, S.,Carta, J. A.,Matías, J.(2011).Comparison between ANNs and linear MCP algorithms in the long-term estimation of the cost per kW h produced by a wind turbine at a candidate site: A case study in the Canary Islands.Applied Energy,88(11),3869-3881.
19. Vui, C. S.,Soon, G. K.,On, C. K.,Alfred, R.,Anthony, P.(2013).A review of stock market prediction with Artificial neural network (ANN).2013 IEEE International Conference on Control System, Computing and Engineering
20. Zakaryazad, A.,Duman, E.(2016).A profit-driven Artificial Neural Network (ANN) with applications to fraud detection and direct marketing.Neurocomputing,175,121-131.
21. Zuluaga, C. D.,Alvarez, M. A.,Giraldo, E.(2015).Shortterm wind speed prediction based on robust Kalman filtering: An experimental comparison.Applied Energy,156,321-330.
22. 朱佳仁(2006).風工程概論.臺北:科技圖書.
23. 林柏宇(2022)。臺北，國立臺灣科技大學營建工程系。
24. 郭建源(2017)。內政部建築研究所自行研究報告內政部建築研究所自行研究報告,臺北:內政部建築研究所。
25. 陳若華,吳國昌,陳海曙,蔡明樹(2001)。內政部建築研究所委託研究報告內政部建築研究所委託研究報告,臺北:內政部建築研究所。
26. 黎益肇,陳瑞華,林柏宇,江桐(2021)。內政部建築研究所委託研究報告內政部建築研究所委託研究報告,臺北:內政部建築研究所。