|
|
基于WCGAN与自注意力机制的光伏场景生成方法
|
Abstract:
以风能、光伏等清洁能源为主体的可再生能源在电力系统中的渗透率不断提高。电力系统中的可再生能源出力呈现出高度间歇性、随机性和波动性特征，这些特性将对电力系统的稳定运行产生重大的影响。同时高渗透率下的可再生能源出力也将对电力系统的调度优化控制提出更高的要求。如何精准刻画可再生能源出力特性成为了当前研究的热点，本文将Wasserstein条件生成对抗网络(WCGAN)与自注意力机制技术(SA)相结合，利用WCGAN强大的生成能力与SA数据增强技术在小数据样本训练时的优秀性能，学习可再生能源历史数据未知分布，生成符合观测数据分布规律的新样本，准确表征可再生能源出力的不确定性。经过对比实验，本文提出的方法在小样本数据上具有更佳的生成质量。
The penetration rate of renewable energy, mainly composed of clean energy such as wind and photovoltaic, in the power system is constantly increasing. The renewable energy output in the power system exhibits highly intermittent, stochastic, and fluctuating characteristics, which will have a significant impact on the stable operation of the power system. At the same time, the output of renewable energy under high penetration rates will also pose higher requirements for the optimization and control of power system scheduling. How to accurately characterize the output characteristics of renewable energy has become a hot topic in current research. This article will combine Wasserstein Conditional Generative Adversarial Network (WCGAN) with self-attention mechanisms data augmentation technology, utilizing the powerful generation ability of WCGAN and the excellent performance of self-attention mechanisms data augmentation in small data training to learn the unknown distribution of renewable energy historical data and generate new samples that conform to the distribution rules of observed data, accurately characterize the uncertainty of renewable energy output. After comparative experiments, the method proposed in this article has better generation quality on small sample data.
| [1] | 国家发展和改革委员会. 《2022年我国可再生能源发展情况》[EB/OL]. https://www.ndrc.gov.cn/fggz/hjyzy/jnhnx/202302/t20230215_1348799_ext.html, 2023-09-23. |
| [2] | 雷鸣, 杜鹏程. 基于风电功率概率预测的无功优化方法研究[J]. 山东电力技术, 2019, 46(1): 6-9. |
| [3] | 李湃, 管晓宏, 吴江. 基于大气动力模型的多风电场出力场景生成方法[J]. 中国电机工程学报, 2015, 35(18): 4581-4590. |
| [4] | Li, P., Guan, X., Wu, J., et al. (2015) Modeling Dynamic Spatial Correlations of Geographically Distributed Wind Farms and Constructing Ellipsoidal Uncertainty Sets for Optimization-Based Generation Scheduling. IEEE Transactions on Sustainable Energy, 6, 1594-1605. https://doi.org/10.1109/TSTE.2015.2457917 |
| [5] | Bludszuweit, H., Dominguez-Navarro, J.A. and Llobart, A. (2008) Statisticasl Analysis of Wind Power Forecast Error. IEEE Transactions on Power System, 23, 983-991. https://doi.org/10.1109/TPWRS.2008.922526 |
| [6] | 丁明, 吴义纯, 张立军. 风电场风速概率分布参数计算方法的研究[J].中国电机工程学报, 2005, 25(10): 107-110. |
| [7] | Tang, C., Wang, Y., Xu, J., et al. (2018) Efficient Scenario Generation of Multiple Renewable Power Plants Considering Spatial and Temporal Correlations. Applied Energy, 221, 348-357. https://doi.org/10.1016/j.apenergy.2018.03.082 |
| [8] | Chen, Y., Li, P. and Zhang, B. (2018) Bayesian Renewables Scenario Generation via Deep Generative Networks. 2018 52nd Annual Conference on Information Sciences and Systems (CISS), Princeton, 21-23 March 2018, 1-6. https://doi.org/10.1109/CISS.2018.8362314 |
| [9] | Chen, Y., Wang, X. and Zhang, B. (2018) An Unsupervised Deep Learning Approach for Scenario Forecasts. 2018 Power Systems Computation Conference (PSCC), Dublin, 11-15 June 2018, 1-7. https://doi.org/10.23919/PSCC.2018.8442500 |
| [10] | Dong, W., Chen, X. and Yang, Q. (2022) Data-Driven Scenario Generation of Renewable Energy Production Based on Controllable Generative Adversarial Networks with Interpretability. Applied Energy, 308, Article ID: 118387. https://doi.org/10.1016/j.apenergy.2021.118387 |
| [11] | Wu, C., Chen, L., Wang, G., et al. (2020) Spatiotemporal Scenario Generation of Traffic Flow Based on LSTM-GAN. IEEE Access, 8, 186191-186198. https://doi.org/10.1109/ACCESS.2020.3029230 |
| [12] | Chen, Y., Wang, Y., Kirschen, D., et al. (2018) Model-Free Renewable Scenario Generation Using Generative Adversarial Networks. IEEE Transactions on Power Systems, 33, 3265-3275. https://doi.org/10.1109/TPWRS.2018.2794541 |
| [13] | 廖文龙, 任翔, 杨哲, 等. 基于隐式最大似然估计的风电出力场景生成[J]. 电力自动化设备, 2022, 42(11): 56-63. |
| [14] | 王守相, 陈海文, 李小平, 等. 风电和光伏随机场景生成的条件变分自动编码器方法[J]. 电网技术, 2018, 42(6): 1860-1869. |
| [15] | 廖文龙, 杨德昌, 葛磊蛟, 等. 基于像素卷积生成网络的可再生能源场景生成[J]. 高电压技术, 2022, 48(4): 1320-1331. |
| [16] | 朱瑞金, 廖文龙, 王玥珑, 等. 基于生成矩匹配网络的光伏和风电随机场景生成[J]. 高电压技术, 2022, 48(1): 374-385. |
| [17] | Vagropoulos, S.I., Kardakos, E.G., Simoglou, C.K., et al. (2016) ANN-Based, Scenario, Generation, Methodology for Stochastic Variables of Electric Power Systems. Electric Power Systems Research, 134, 9-18. https://doi.org/10.1016/j.epsr.2015.12.020 |
| [18] | Goodfellow, I., Pouget-Abadie, J., Mirza, M., et al. (2020) Generative Adversarial Networks. Communications of the ACM, 63, 139-144. https://doi.org/10.1145/3422622 |
| [19] | Kingma, D.P. and Welling, M. (2013) Auto-Encoding Variational Bayes. arXiv: 1312.6114. |
| [20] | 刘辉, 杨照华, 吴云, 等. 基于全局注意力机制的单像素成像图像增强方法[J]. 空间控制技术与应用, 2023, 49(6): 68-76. |
| [21] | Du, W.B., Wang, Y.L. and Qiao, Y. (2018) Recurrent Spatial-Temporal Attention Network for Action Recognition in Videos. IEEE Transactions on Image Processing, 27, 1347-1360. https://doi.org/10.1109/TIP.2017.2778563 |
| [22] | Pan, T.Y., Chen, J.L., Ye, Z.S., et al. (2022) A Multihead Attention Network with Adaptive Metatransfer Learning for RUL Prediction of Rocket Engines. Reliability Engineering & System Safety, 225, Article ID: 108610. https://doi.org/10.1016/j.ress.2022.108610 |
| [23] | Liu, L., Song, X. and Zhou, Z.T. (2022) Aircraft Engine Remaining Useful Life Estimation via a Double Attention-Based Data-Driven Architecture. Reliability Engineering & System Safety, 221, Article ID: 108330. https://doi.org/10.1016/j.ress.2022.108330 |
| [24] | Liao, W., Ge, L., Bak-Jensen, B., et al. (2022) Scenario Prediction for Power Loads Using a Pixel Convolutional Neural Network and an Optimization Strategy. Energy Reports, 8, 6659-6671. https://doi.org/10.1016/j.egyr.2022.05.028 |
