Climatic Long Term Trend and Prediction of the Wind Energy Resource in the Gwadar Port

doi:10.13209/j.0479-8023.2017.012

Abstract

Abstract:

Based on the 36-year (1979-2014) ERA-Interim 10 m sea surface wind data from European Centre for Medium-Range Weather Forecasts (ECMWF), the climatic long term trend of the wind energy resource of the Gwadar Port of Pakistan was analyzed. Using two linear regression and artificial neural network (ANN) techniques, the wind energy resource in the long term was predicted. The results show that wind energy in summer is richer than that in winter. The stability in summer is better than that in winter. For the past 36 years, the wind power density, occurrence of effective wind speed and energy level occurrence have significant annual decreasing trends, of -0.78 W/(m²· a), -0.21%/a, -0.22%/a separately. These trends mainly exhibit in summer, while no significant variation in winter. The stability (coefficient of variation, monthly variability index and seasonal variability index) does not have a significant long term trend for the past 36 years. From the prediction value, the wind energy resource in 2015 is similar to the multi-year average value, while the wind energy resource in 2016 is richer than the multi-year average value. For the year 2015-2016, the prediction wind energy will be more unstable than the multi-year average status. The results can provide scientific reference for the 21st Maritime Silk Road construction, development of remote islands and ports in the China seas.

Key words: 21st Maritime Silk Road, Gwadar Port, wind energy resource, climatic long term trend, prediction

摘要：

利用欧洲中期天气预报中心的ERA-Interim海表10 m风场资料, 计算巴基斯坦瓜达尔港的风能资源近36年(1979—2014年)期间的历史变化趋势, 并利用线性回归和BP神经网络两种方法, 对该港的风能资源进行长期年度预测, 得到以下结果。1) 瓜达尔港夏季的风能资源比冬季丰富, 且夏季的稳定性明显好于冬季。2) 近36年期间, 风能密度、有效风速频率和100 W/m²以上能级频率分别以-0.78 W/(m²·a), -0.21%/a和-0.22%/a的速度逐年显著递减, 且该趋势主要体现在夏季, 冬季无显著变化趋势; 风能资源的稳定性(变异系数、月变化指数和季节变化指数)无显著变化趋势。3) 从预测值来看, 瓜达尔港的风能资源在2015年与多年平均状态持平, 2016年则趋于更丰富; 2015—2016年, 风能资源的稳定性比多年平均状态略差。研究结果可以为“21世纪海上丝绸之路”建设以及中国海域的岛礁和港口建设提供依据和参考。

关键词: 21世纪海上丝调之路, 瓜达尔港, 风能资源, 历史变化趋势, 预测

CLC Number:

P743

Chongwei ZHENG, Yue GAO, Xuan CHEN. Climatic Long Term Trend and Prediction of the Wind Energy Resource in the Gwadar Port[J]. Acta Scientiarum Naturalium Universitatis Pekinensis, 2017, 53(4): 617-626.

郑崇伟, 高悦, 陈璇. 巴基斯坦瓜达尔港风能资源的历史变化趋势及预测[J]. 北京大学学报自然科学版, 2017, 53(4): 617-626.

Figures/Tables 11

Fig. 1 Topology of BP neural network[16,17]

Fig. 2 BP neural network operation flow chart

Fig. 3 Monthly wind power density in January, July and annual value for the past 36 years

Fig. 4 Monthly occurrence of effective wind speed in January, July and annual value for the past 36 years

Fig. 5 Monthly occurrence of wind power density greater than 100 W/m2 in January, July and annual value for the past 36 years

Fig. 6 Monthly coefficient of variation in January,July and annual value for the past 36 years

Fig. 7 Monthly variability index (a) and seasonalvariability index (b) for the past 36 years

Table 1 Climatic long-term trend of the wind energy parameters of the Gwadar Port

风能要素	1月		7月		逐年
风能要素	原始值	滑动平均	原始值	滑动平均	原始值	滑动平均
风能密度/(W · m^-2 · a^-1)	-0.03	-0.05	-0.89^*	-0.93^**	-0.59^**	-0.78^**
有效风速频率/(% · a^-1)	-0.07	-0.08	-0.25^*	-0.25^**	-0.17^**	-0.21^**
能级频率/(% · a^-1)	-0.07	-0.09	-0.22^*	-0.22^**	-0.18^**	-0.22^**

Table 2 Prediction precision of the wind energy of the Gwadar Port

项目	风能密度		有效风速频率		能级频率		变异系数
项目	线性回归	神经网络	线性回归	神经网络	线性回归	神经网络	线性回归	神经网络
CC	0.6593^*	0.8608^*	0.6724^*	0.7331^*	0.6634^*	0.7002^*	0.6352^*	0.8484^*
Bias	5.7897	3.0560	1.7890	-0.2516	1.9375	-0.5617	0.1101	0.0955
MAE	33.4049	18.2077	9.9247	7.3124	9.6059	9.0033	0.2073	0.1494
NRMSE	0.3347	0.1811	0.2795	0.2318	0.3174	0.2856	0.2028	0.1458
RMSE	42.8393	23.1813	12.4186	10.0006	12.5472	11.7034	0.2452	0.1762
SI	0.3316	0.1811	0.2766	0.2617	0.3136	0.2768	0.1812	0.1225

Fig. 8 Prediction experiment of wind power density and coefficient of variation of the wind energy

Fig. 9 Prediction values of wind power density, occurrence of effective wind speed, occurrence of wind power density greater than 100 W/m2 and coefficient of variation for the future 24 months ((a)-(d)), and the abnormal values between prediction and multi-year average value ((e)?(h))

References 27

[1]	郑崇伟, 李崇银, 杨艳, 等. 瓜达尔港的风能开发可行性论证. 厦门大学学报(自然科学版), 2016,55(2):210-215
[2]	Chang T J, Wu Y T, Hsu H Y, et al.Assessment of wind characteristics and wind turbine characteristics in Taiwan.Renewable Energy, 2003,28:851-871
[3]	陈飞, 班欣, 祈欣, 等. 连云港沿海地区及近海风能资源评估. 气象科学, 2008,28(增刊1):101-106
[4]	毛慧琴, 宋丽莉, 黄浩辉, 等. 广东省风能资源区划研究. 自然资源学报, 2005,20(5):679-683
[5]	周荣卫, 何晓风, 朱蓉, 等. 中国近海风能资源开发潜力数值模拟. 资源科学, 2010,32(8):1434- 1443
[6]	郑崇伟, 李崇银. 中国南海岛礁建设: 重点岛礁的风候、波候特征分析. 中国海洋大学学报(自然科学版), 2015,45(9):1-6
[7]	郑崇伟, 李崇银. 中国南海岛礁建设: 风力发电、海浪发电. 中国海洋大学学报(自然科学版), 2015,45(9):7-14
[8]	谷兴凯, 范高锋, 王晓蓉, 等. 风电功率预测技术综述. 电网技术, 2007,31(2):335-338
[9]	Dee D P, Uppala S M, Simmons A J, et al. The ERA-Interim reanalysis: configuration and performance of the data assimilation system. Q J R Meteorol Soc, 2011,137:553-597
[10]	Song L N, Liu Z L, Wang F.Comparison of wind data from ERA-Interim and buoys in the Yellow and East China Seas.Chinese Journal of Oceanology and Limnology, 2015,33(1):282-288
[11]	ECMWF.ERA-Interim wind data [EB/OL]. (2011‒ 01‒01) [2015‒06‒22].
[12]	Bao X H, Zhang F Q.Evaluation of NCEP-CFSR,NCEP-NCAR,ERA-Interim,ERA-40 reanalysis datasets against independent sounding observations over the Tibetan Plateau.Journal of Climate, 2013,26:206-214
[13]	马永锋, 卞林根. ERA-Interim再分析和NCEP FNL分析资料在东南极中山站至 Dome A 断面的适用性研究. 极地研究, 2014,26(4):469-480
[14]	王烨芳, 李成才, 石光明, 等. 江苏环港风电场风能资源特征分析. 北京大学学报(自然科学版), 2013,49(6):955-963
[15]	黄世成, 姜爱军, 刘聪, 等. 江苏省风能资源重新估算与分布研究. 气象科学, 2007,27(4):407-412
[16]	张吉豫, 刘先华, 梁堃, 等. 一种基于人工神经网络的基本块重排方法. 北京大学学报(自然科学版), 2011,47(1):9-16
[17]	陈克海, 解学通, 黄舟, 等. 人工神经网络在星载散射计海面风场反演建模中的应用. 北京大学学报(自然科学版), 2007,43(4):460-467
[18]	黄茂怡, 黄嘉佑, 王勇, 等. 近年来CCA在气候分析与气候预测中的应用. 北京大学学报(自然科学版), 2001,37(1):128-135
[19]	郁亚娟, 王真, 郭怀成, 等. 基于人工神经网络的城市拓展区可持续发展指数序列研究. 北京大学学报(自然科学版), 2006,42(4):496-502
[20]	段晓峰, 许学工, 陈满春, 等. 基于长期验潮数据的海平面预测方法与案例应用. 北京大学学报(自然科学版), 2014,50(6):1065-1070
[21]	陈辉, 李双成, 郑度. 基于人工神经网络的青藏公路铁路沿线生态系统风险研究. 北京大学学报(自然科学版), 2005,41(4):586-593
[22]	谢衷洁, 黄香, 叶伟彰, 等. 人工神经网络及其在金融预报中的应用. 北京大学学报(自然科学版), 2001,37(3):421-425
[23]	Miao W W, Jia H J, Wang D. Active power regulation of wind power systems through demand response.Science China Technology Science, 2012,55:1667-1676
[24]	Zheng C W, Pan J.Assessment of the global ocean wind energy resource.Renewable and Sustainable Energy Reviews, 2014,33:382-391
[25]	郑崇伟, 黎鑫, 陈璇, 等. 经略 21 世纪海上丝路: 海洋资源、相关国家的海洋开发状况. 海洋开发与管理, 2016,33(3):3-8
[26]	郭梦婷, 蔡旭辉, 宋宇. 全国低风速气象特征分析. 北京大学学报: 自然科学版, 2016,52(2):219-226
[27]	Cornett A M.A Global wave energy resource assessment // Proceedings of the Eighteenth(2008)International Offshore and Polar Engineering Con-ference. Vancouver, 2008: Paper No. ISOPE-2008-TPC-579