Short-term rapid prediction of stratospheric wind field based on PSO-BP neural network
-
摘要:
平流层风场环境对临近空间低速飞行器驻空飞行性能有重要影响。研究了基于PSO-BP神经网络的平流层区域风场建模与快速预测方法，根据历史风场数据，采用主成分分析法对数据进行降维处理，通过BP神经网络对风场进行预测建模，利用粒子群优化（PSO）算法对其进行优化，采用Biharmonic样条曲面插值方法构建区域预测风场。以南海地区5年历史风场为对象，对比分析了基于BP神经网络和基于PSO-BP神经网络的风场预测模型，结果表明：使用具有全局寻优特性的PSO算法改进BP神经网络，能够有效避免传统BP神经网络易陷入局部最优的缺点，提高预测精度；通过结合PSO-BP神经网络预测与Biharmonic样条曲面插值，可实现区域风场的预测。研究结果可为临近空间低速飞行器的轨迹规划与区域驻留等任务的高精度区域快速预报风场提供解决途径。
-
关键词:
- 平流层风场建模 /
- 临近空间低速飞行器 /
- BP神经网络 /
- 粒子群优化(PSO)算法 /
- Biharmonic样条曲面插值
Abstract:The stratospheric wind field environment has an important influence on the flight performance of near space low speed aircraft. In this paper, the modeling and prediction methods of stratospheric regional wind field are investigated based on PSO-BP neural network. Firstly, the principal component analysis method is implemented to reduce the dimensions of the historical wind field data. Then, the BP neural network, which is trained by the processed data to predict the wind field, is optimized with particle swarm optimization (PSO) algorithm. Finally, the Biharmonic spline surface interpolation method using multi-point prediction wind fields is studied to construct the regional prediction wind field. Taking the 5-year historical wind field data of a certain place, comparative study of the wind field prediction model based on BP neural network and PSO-BP neural network is conducted. The results show that PSO algorithm, which is characterized global optimization, can improve BP neural network by avoiding the disadvantage of easily falling into local optimization, and enhance the prediction accuracy. The integration method of PSO-BP neural network prediction and Biharmonic spline surface interpolation can provide the prediction of the regional wind field. The proposed method can provide high precision regional prediction wind field for trajectory planning and station keeping of near space low speed aircraft.
-
图 3 PSO算法的BP神经网络原理
Figure 3. Schematic diagram of BP neural network optimized by particle swarm optimization algorithm
图 4 海拔5~15 km东西方向风场数据降维后的第n阶相对模态能量及前n阶累积模态能量
Figure 4. The n-th order relative modal energy and the first n-order cumulative modal energy of east-west wind field data after dimensionality reduction between 5 km and 15 km
图 5 海拔15~30 km东西方向风场数据降维后的第n阶相对模态能量及前n阶累积模态能量
Figure 5. The n-th order relative modal energy and the first n-order cumulative modal energy of east-west wind field data after dimensionality reduction between 15 km and 30 km
图 6 东西方向上降维后的风场数据与实际风场数据对比
Figure 6. Comparison of east-west wind field data after dimensionality reduction and actual wind field data
图 7 海拔5~15 km南北方向风场数据降维后的第n阶相对模态能量及前n阶累积模态能量
Figure 7. The n-th order relative modal energy and the first n-order cumulative modal energy of north-south wind field data after dimensionality reduction between 5 km and 15 km
图 8 海拔15~30 km南北方向风场数据降维后的第n阶相对模态能量及前n阶累积模态能量
Figure 8. The n-th order relative modal energy and the first n-order cumulative modal energy of north-south wind field data after dimensionality reduction between 15 km and 30 km
图 9 南北方向上降维后的风场数据与实际风场数据对比
Figure 9. Comparison of north-south wind field data after dimensionality reduction and actual wind field data
图 10 东西方向上BP神经网络10次预测结果与实际风场对比
Figure 10. Comparison of 10 prediction results of BP neural network and actual wind field in east-west direction
图 11 东西方向上PSO-BP神经网络10次预测结果与实际风场对比
Figure 11. Comparison of 10 prediction results of PSO-BP neural network and actual wind field in east-west direction
图 12 东西方向上2种神经网络10次预测结果平均误差对比
Figure 12. Comparison of average error of 10 prediction results for two kinds of neural networks in east-west direction
图 13 南北方向上BP神经网络10次预测结果与实际风场对比
Figure 13. Comparison of 10 prediction results of BP neural network and actual wind field in north-south direction
图 14 南北方向上PSO-BP神经网络10次预测结果与实际风场对比
Figure 14. Comparison of 10 prediction results of PSO-BP neural network and actual wind field in the north-south direction
图 15 南北方向上2种神经网络10次预测结果平均误差对比
Figure 15. Comparison of average error of 10 prediction results for two kinds of neural networks in north-south direction
图 16 东西方向上20 km高度的区域风速示意图
Figure 16. Schematic diagram of regional wind speed in east-west direction at 20 km
图 17 南北方向上20 km高度的区域风速示意图
Figure 17. Schematic diagram of regional wind speed in north-south direction at 20 km
-
[1] 赵达, 刘东旭, 孙康文, 等. 平流层飞艇研制现状、技术难点及发展趋势[J]. 航空学报, 2016, 37(1): 45-56. https://www.cnki.com.cn/Article/CJFDTOTAL-HKXB201601004.htmZHAO D, LIU D X, SUN K W, et al. Research status, technical difficulties and development trend of stratospheric airship[J]. Acta Aeronautica et Astronautica Sinica, 2016, 37(1): 45-56(in Chinese). https://www.cnki.com.cn/Article/CJFDTOTAL-HKXB201601004.htm [2] LI Y, SHENG Z, JING J R. Feature analysis of stratospheric wind and temperature fields over the Antigua site by rocket data[J]. Earth Planet Physics, 2019, 3(5): 414-424. doi: 10.26464/epp2019040 [3] WANG J, MENG X, LI C. Recovery trajectory optimization of the solar-powered stratospheric airship for the station-keeping mission[J]. Acta Astronautica, 2021, 178: 159-177. doi: 10.1016/j.actaastro.2020.08.016 [4] 顾逸东. 气球科学观测100年[J]. 现代物理知识, 2020, 32(2): 3-12. https://www.cnki.com.cn/Article/CJFDTOTAL-XDWZ202002001.htmGU Y D. 100 years of scientific observation of balloons[J]. Modern Physics, 2020, 32(2): 3-12(in Chinese). https://www.cnki.com.cn/Article/CJFDTOTAL-XDWZ202002001.htm [5] WEI J, MA R, HOU X, et al. Analysis of the thermodynamical property of super-pressure balloons[J]. Acta Mechanica, 2019, 230(4): 1355-1366. doi: 10.1007/s00707-017-2021-2 [6] 杨燕初, 张航悦, 赵荣. 零压式高空气球球形设计与参数敏感性分析[J]. 国防科技大学学报, 2019, 41(1): 58-64. https://www.cnki.com.cn/Article/CJFDTOTAL-GFKJ201901009.htmYANG Y C, ZHANG H Y, ZHAO R. Shape design of zero pre-ssure high altitude balloon and sensitivity analysis of key para-meters[J]. Journal of National University of Defense Technology, 2019, 41(1): 58-64(in Chinese). https://www.cnki.com.cn/Article/CJFDTOTAL-GFKJ201901009.htm [7] 聂春雨, 祝明, 郑泽伟, 等. 基于Q-Learning算法和神经网络的飞艇控制[J]. 北京航空航天大学学报, 2017, 43(12): 2431-2438. doi: 10.13700/j.bh.1001-5965.2016.0903NIE C Y, ZHU M, ZHENG Z W, et al. Airship control based on Q-Learning algorithm and neural network[J]. Journal of Beijing University of Aeronautics and Astronautics, 2017, 43(12): 2431-2438(in Chinese). doi: 10.13700/j.bh.1001-5965.2016.0903 [8] 郭建国, 周军. 临近空间低动态飞行器控制研究综述[J]. 航空学报, 2014, 35(2): 320-331. https://www.cnki.com.cn/Article/CJFDTOTAL-HKXB201402003.htmGUO J G, ZHOU J. Review of the control of low dynamic vehicles in near space[J]. Acta Aeronautica et Astronautica Sinica, 2014, 35(2): 320-331(in Chinese). https://www.cnki.com.cn/Article/CJFDTOTAL-HKXB201402003.htm [9] 史智广, 张小强, 李锦清, 等. 平流层浮空器保压指标对驻空性能的影响[J]. 航空学报, 2016, 37(6): 1833-1840. https://www.cnki.com.cn/Article/CJFDTOTAL-HKXB201606014.htmSHI Z G, ZHANG X Q, LI J Q, et al. Effect of ground pressure-maintenance index on stagnation performance of stratospheric aerostats[J]. Acta Aeronautica et Astronautica Sinica, 2016, 37(6): 1833-1840(in Chinese). https://www.cnki.com.cn/Article/CJFDTOTAL-HKXB201606014.htm [10] 程晨, 王晓亮. 平流层飞艇热敏感因素分析[J]. 浙江大学学报(工学版), 2020, 54(1): 202-212. https://www.cnki.com.cn/Article/CJFDTOTAL-ZDZC202001024.htmCHENG C, WANG X L. Thermal sensitivity factors analysis of stratospheric airships[J]. Journal of Zhejiang University (Engineering Science), 2020, 54(1): 202-212(in Chinese). https://www.cnki.com.cn/Article/CJFDTOTAL-ZDZC202001024.htm [11] 姜兆宇, 贾庆山, 管晓宏. 多时空尺度的风力发电预测方法综述[J]. 自动化学报, 2019, 45(1): 51-71. https://www.cnki.com.cn/Article/CJFDTOTAL-MOTO201901005.htmJIANG Z Y, JIA Q S, GUAN X H. A review of multi-temporal-and-spatial-scale wind power forecasting method[J]. Acta Automatica Sinica, 2019, 45(1): 51-71(in Chinese). https://www.cnki.com.cn/Article/CJFDTOTAL-MOTO201901005.htm [12] 邓小龙, 杨希祥, 麻震宇, 等. 基于风场环境利用的平流层浮空器区域驻留关键问题研究进展[J]. 航空学报, 2019, 40(8): 23-36. https://www.cnki.com.cn/Article/CJFDTOTAL-HKXB201908002.htmDENG X L, YANG X X, MA Z Y, et al. Review of important technologies for station-keeping of stratospheric aerostats based on wind field utilization[J]. Acta Aeronautica et Astronautica Sinica, 2019, 40(8): 23-36(in Chinese). https://www.cnki.com.cn/Article/CJFDTOTAL-HKXB201908002.htm [13] RAMESH S S, LIM K M, LEE H P, et al. Reduced-order modeling of stratospheric winds and its application in high-altitude balloon trajectory simulations[J]. Journal of Applied Meteorology and Climatology, 2017, 56(6): 1753-1766. [14] 胡亮, 顾明, 李黎. 基于本征正交分解的谱表示法模拟风场的误差[J]. 振动与冲击, 2011, 30(4): 12-15. https://www.cnki.com.cn/Article/CJFDTOTAL-ZDCJ201104006.htmHU L, GU M, LI L. Errors produced with proper orthogonal decomposition-based spectral representation method in wind velocity field simulation[J]. Journal of Vibration and Shock, 2011, 30(4): 12-15(in Chinese). https://www.cnki.com.cn/Article/CJFDTOTAL-ZDCJ201104006.htm [15] RONEY J A. Statistical wind analysis for near-space applications[J]. Journal of Atmospheric and Solar-Terrestrial Physics, 2007, 69: 1485-1501. [16] 孙斌, 姚海涛, 刘婷. 基于高斯过程回归的短期风速预测[J]. 中国电机工程学报, 2012, 32(29): 104-109. https://www.cnki.com.cn/Article/CJFDTOTAL-ZGDC201229016.htmSUN B, YAO H T, LIU T. Short-term wind speed forecasting based on gaussian process regression model[J]. Proceedings of the CSEE, 2012, 32(29): 104-109(in Chinese). https://www.cnki.com.cn/Article/CJFDTOTAL-ZGDC201229016.htm [17] 梁智, 孙国强, 俞娜燕, 等. 基于高斯过程回归和粒子滤波的短期风速预测[J]. 太阳能学报, 2020, 41(3): 45-51. https://www.cnki.com.cn/Article/CJFDTOTAL-TYLX202003007.htmLIANG Z, SUN G Q, YU N Y, et al. Short-term wind speed forecasting based on Gaussian process regression and particle filter[J]. Acta Energiae Solaris Sinica, 2020, 41(3): 45-51(in Chinese). https://www.cnki.com.cn/Article/CJFDTOTAL-TYLX202003007.htm [18] 李魁, 邓小龙, 杨希祥, 等. 基于本征正交分解的平流层风场建模与预测[J]. 北京航空航天大学学报, 2018, 44(9): 2013-2020. doi: 10.13700/j.bh.1001-5965.2017.0685LI K, DENG X L, YANG X X, et al. Reduced order modeling and prediction of stratospheric winds based on the proper orthogonal decomposition method[J]. Journal of Beijing University of Aeronautics and Astronautics, 2018, 44(9): 2013-2020(in Chinese). doi: 10.13700/j.bh.1001-5965.2017.0685 [19] BELLEMARE M G, CANDIDO S, CASTRO P S, et al. Autonomous navigation of stratospheric balloons using reinforcement learning[J]. Nature, 2020, 588(7836): 77-82. [20] ZHOU W, ZHOU P, WANG Y, et al. Station-keeping control of an underactuated stratospheric airship[J]. International Journal of Fuzzy Systems, 2019, 21(3): 715-732. [21] 张泰华, 姜鲁华, 周江华. 放飞过程中平流层飞艇运动与受力分析[J]. 北京航空航天大学学报, 2018, 44(4): 691-699. doi: 10.13700/j.bh.1001-5965.2017.0227ZHANG T H, JIANG L H, ZHOU J H. Kinematic and mechanical analysis on launch process of stratospheric airship[J]. Journal of Beijing University of Aeronautics and Astronautics, 2018, 44(4): 691-699(in Chinese). doi: 10.13700/j.bh.1001-5965.2017.0227 -
下载:
点击查看大图
计量
- 文章访问数: 413
- HTML全文浏览量: 111
- PDF下载量: 49
- 被引次数: 0
