苗磊, 李擎, 蒋原, 崔家瑞, 王义轩. 深度学习在电力系统预测中的应用[J]. 工程科学学报, 2023, 45(4): 663-672. doi: 10.13374/j.issn2095-9389.2021.12.21.006 引用本文: 苗磊, 李擎, 蒋原, 崔家瑞, 王义轩. 深度学习在电力系统预测中的应用[J]. 工程科学学报, 2023, 45(4): 663-672. doi: 10.13374/j.issn2095-9389.2021.12.21.006 MIAO Lei, LI Qing, JIANG Yuan, CUI Jia-rui, WANG Yi-xuan. A survey of power system prediction based on deep learning[J]. Chinese Journal of Engineering, 2023, 45(4): 663-672. doi: 10.13374/j.issn2095-9389.2021.12.21.006 Citation: MIAO Lei, LI Qing, JIANG Yuan, CUI Jia-rui, WANG Yi-xuan. A survey of power system prediction based on deep learning[J]. Chinese Journal of Engineering , 2023, 45(4): 663-672. doi: 10.13374/j.issn2095-9389.2021.12.21.006 苗磊, 李擎, 蒋原, 崔家瑞, 王义轩. 深度学习在电力系统预测中的应用[J]. 工程科学学报, 2023, 45(4): 663-672. doi: 10.13374/j.issn2095-9389.2021.12.21.006 引用本文: 苗磊, 李擎, 蒋原, 崔家瑞, 王义轩. 深度学习在电力系统预测中的应用[J]. 工程科学学报, 2023, 45(4): 663-672. doi: 10.13374/j.issn2095-9389.2021.12.21.006 MIAO Lei, LI Qing, JIANG Yuan, CUI Jia-rui, WANG Yi-xuan. A survey of power system prediction based on deep learning[J]. Chinese Journal of Engineering, 2023, 45(4): 663-672. doi: 10.13374/j.issn2095-9389.2021.12.21.006 Citation: MIAO Lei, LI Qing, JIANG Yuan, CUI Jia-rui, WANG Yi-xuan. A survey of power system prediction based on deep learning[J]. Chinese Journal of Engineering , 2023, 45(4): 663-672. doi: 10.13374/j.issn2095-9389.2021.12.21.006 电力系统预测主要包括负荷预测、出力预测以及健康状态预测等。通过负荷预测，可以优化电力生产规划，从而更好地实现电能的精细化分配；通过出力预测，可以有效提升新能源电力消纳能力，实现电能的充分及合理利用；通过电力设备健康状态预测，可以及时发现设备运行隐患，从而进一步保障电力系统平稳安全运行。深度学习凭借其卓越的特征分析和预测能力，被广泛应用于电力系统运行及维护。本文首先归纳介绍了电力系统预测深度学习模型的特点、适用场景；其次，梳理了深度学习在面向民用及工业场景负荷预测、光伏及风电出力预测、机械及非机械设备健康状态预测中的应用前沿；最后，对深度学习在电力系统预测中所面临的关键问题、发展趋势进行了总结和展望。

电力系统 /  深度学习 /  负荷预测 /  出力预测 /  健康状态预测 Abstract: Power system is one of the largest and complex artificial engineering in the modern society. With the development of intelligence, digitization and long-distance technology, a large number of multi-source, multi-state and heterogeneous operational data have emerged. As a new trend direction of machine learning, deep learning has shown potential in data feature extraction and pattern recognition. Because of its excellent ability in data analysis and prediction, it is widely used in power system, which has a significant impact on optimizing power production planning, improving power production efficiency and energy utilization, and ensuring the smooth operation of the system influence. Based on massive quantities of data and by means of deep learning, it can better fit the nonlinear relationship between the factors affecting the subsequent operational state of the system, so as to further improve the prediction accuracy. Power system prediction includes load forecasting, new energy power prediction and state-of-health prediction. Power production planning can be optimized using load forecasting; thus, electrical energy can be finely dispatched. The capacity of new energy power consumption is improved through power prediction to reasonably use electrical energy. Potential equipment hazards can be timely found using power equipment health state prediction, thereby ensuring safe and smooth operation. First, in this paper, the characteristics and applicable scenarios of typical deep learning models are introduced, among them, deep belief network and stacked auto encoder belong to stack structure, so the structure is flexible and easy to expand, which is suitable for the modeling and feature extraction of unrelated data type; convolutional neural network shares convolution kernel internally to reduce the number of network parameters and is good at processing high-dimensional data type; recurrent neural network has feedforward and feedback connections, so it is suitable for processing sequence data with pre and post dependence. Second, the application frontiers of predictive power systems based on deep learning are reviewed, which include civil and industrial scenarios, photovoltaic and wind power, mechanical and non-mechanical equipment health state prediction. Finally, facing the challenges of power system in energy efficient allocation, high proportion of new energy power consumption, highly stable operation of power equipment and so on, the key problems and future development trends are presented.

Key words: power system /  deep learning /  load forecasting /  new energy power prediction /  health state prediction  ModelMain featureApply data typeTypical application scenario DBNUnsupervised learning
No need for large number of label data, and the training difficulty is lowSequence data without correlation before and after
(Time series data)Load forecasting, power prediction, equipment health state predictionCNNSupervised learning
Random initial value, sample data without preprocessingSequence data without correlation before and after,
(Multidimensional data)Load forecasting under multi energy spatiotemporal coupling, power prediction considering spatiotemporal correlation, health state forecastingRNNSupervised learning
Both feedforward and feedback connections are includedSequence data correlated before and afterLoad forecasting and power prediction under the scenario of severe power fluctuationSAEUnsupervised learning
Asymmetric connection, simple structure, easy to expandSequence data without correlation before and afterPower prediction, equipment health state prediction Deep learningCivil scenario
load forecastingIndustrial scenario
load forecastingProportion/% DBN[ 8 ][ 25 – 27 ]21.1CNN[ 13 , 28 – 30 ]–21.1RNN[ 31 – 37 ][ 17 , 18 , 38 ]52.6SAE[ 39 ]–5.2 Deep learningMechanical equipment
health state predictionNon-mechanical equipment
health state predictionProportion/
% DBN[ 9 , 68 ][ 69 ]16.7CNN[ 70 – 71 ][ 11 ]16.7RNN[ 72 ][ 73 – 75 ]22.2SAE[ 67 , 76 – 80 ][ 21 , 81 ]44.5 Shi J Q, Tan T, Guo J, et al. Multi-task learning based on deep architecture for various types of load forecasting in regional energy system integration. Power Syst Technol , 2018, 42(3): 698 doi: 10.13335/j.1000-3673.pst.2017.2368

史佳琪, 谭涛, 郭经, 等. 基于深度结构多任务学习的园区型综合能源系统多元负荷预测. 电网技术, 2018, 42(3):698 doi: 10.13335/j.1000-3673.pst.2017.2368 Tan M, Liu Y, Meng B M, et al. Multinodal forecasting of industrial power load using participation factor and ensemble learning // 2020 IEEE 4 th Conference on Energy Internet and Energy System Integration . Wuhan, 2020: 745 Zhao Y, Wang H M, Kang L, et al. Temporal convolution network-based short-term electrical load forecasting. Trans China Electrotech Soc , 2022, 37(5): 1242

赵洋, 王瀚墨, 康丽, 等. 基于时间卷积网络的短期电力负荷预测. 电工技术学报, 2022, 37(5):1242 Xu J H, Wang X W, Yang J J. Short-term load density prediction based on CNN-QRLightGBM. Power Syst Technol , 2020, 44(9): 3409

许佳辉, 王向文, 杨俊杰. 基于CNN-QRLightGBM的短期负荷概率密度预测. 电网技术, 2020, 44(9):3409 Cheng L L, Zang H X, Xu Y, et al. Probabilistic residential load forecasting based on micrometeorological data and customer consumption pattern. IEEE Trans Power Syst , 2021, 36(4): 3762 doi: 10.1109/TPWRS.2021.3051684 Wei A, Mao D J, Han W L, et al. Short-term load forecasting based on EMD and long short-term memory neural networks. J Eng Therm Energy Power , 2020, 35(4): 203 doi: 10.16146/j.cnki.rndlgc.2020.04.028

魏骜, 茅大钧, 韩万里, 等. 基于EMD和长短期记忆网络的短期电力负荷预测研究. 热能动力工程, 2020, 35(4):203 doi: 10.16146/j.cnki.rndlgc.2020.04.028 Wang Z P, Zhao B, Ji W J, et al. Short-term load forecasting method based on GRU-NN model. Autom Electr Power Syst , 2019, 43(5): 53

王增平, 赵兵, 纪维佳, 等. 基于GRU-NN模型的短期负荷预测方法. 电力系统自动化, 2019, 43(5):53 Wang Z P, Zhao B, Jia X, et al. Short-term power load forecasting method based on difference decomposition and error compensation. Power Syst Technol , 2021, 45(7): 2560

王增平, 赵兵, 贾欣, 等. 基于差分分解和误差补偿的短期电力负荷预测方法. 电网技术, 2021, 45(7):2560 Chen H W, Wang S X, Wang S M, et al. Aggregated load forecasting method based on gated recurrent unit networks and model fusion. Autom Electr Power Syst , 2019, 43(1): 65

陈海文, 王守相, 王绍敏, 等. 基于门控循环单元网络与模型融合的负荷聚合体预测方法. 电力系统自动化, 2019, 43(1):65 Ma Y, Zhang Q, Ding J J, et al. Short term load forecasting based on iForest-LSTM // IEEE 14 th Conference on Industrial Electronics and Applications . Xi’an, 2019: 2278 Eskandari H, Imani M, Moghaddam M P. Convolutional and recurrent neural network based model for short-term load forecasting. Electr Power Syst Res , 2021, 195: 107173 doi: 10.1016/j.jpgr.2021.107173 Sun M Y, Zhang T Q, Wang Y, et al. Using Bayesian deep learning to capture uncertainty for residential net load forecasting. IEEE Trans Power Syst , 2020, 35(1): 188 doi: 10.1109/TPWRS.2019.2924294 Salakhutdinov R, Hinton G. Using deep belief nets to learn covariance kernels for Gaussian processes // Proceedings of the 20 th International Conference on Neural Information Processing Systems . New York, 2007: 1249 Peng W, Xu L W, Li C D, et al. Stacked autoencoders and extreme learning machine based hybrid model for electrical load prediction. J Intell Fuzzy Syst , 2019, 37(4): 5403 doi: 10.3233/JIFS-190548 Gal Y, Ghahramani Z. Dropout as a bayesian approximation: Representing model uncertainty in deep learning // Proceedings of the 33 rd International Conference on International Conference on Machine Learning . New York, 2016: 1050 Mitra P, Vittal V, Pourbeik P, et al. Load sensitivity studies in power systems with non-smooth load behavior. IEEE Trans Power Syst , 2017, 32(1): 705 doi: 10.1109/TPWRS.2016.2554398 Liu Y, Zhang H, Zhang A M. Improved load forecasting method based on load characteristics under demand-side response. Power Syst Prot Control , 2018, 46(13): 126 doi: 10.7667/PSPC170167

刘云, 张杭, 张爱民. 需求侧响应下基于负荷特性的改进短期负荷预测方法. 电力系统保护与控制, 2018, 46(13):126 doi: 10.7667/PSPC170167 Kong X Y, Li C, Zheng F, et al. Improved deep belief network for short-term load forecasting considering demand-side management. IEEE Trans Power Syst , 2020, 35(2): 1531 doi: 10.1109/TPWRS.2019.2943972 Tan X Y, Liu F, Ma J J, et al. Ultra-short-term pv power forecasting model based on dbn and t-s time-varying weight combination. Acta Energiae Solaris Sin , 2021, 42(10): 42 doi: 10.19912/j.0254-0096.tynxb.2019-1029

谭小钰, 刘芳, 马俊杰, 等. 基于DBN与T-S时变权重组合的光伏功率超短期预测模型. 太阳能学报, 2021, 42(10):42 doi: 10.19912/j.0254-0096.tynxb.2019-1029 Lai C W, Li J H, Chen B, et al. Review of photovoltaic power output prediction technology. Trans China Electrotech Soc , 2019, 34(6): 1201 doi: 10.19595/j.cnki.1000-6753.tces.180326

(赖昌伟, 黎静华, 陈博, 等. 光伏发电出力预测技术研究综述. 电工技术学报, 2019, 34(6):1201 doi: 10.19595/j.cnki.1000-6753.tces.180326 Du J L, Liu Y Y, Liu Z J. Study of precipitation forecast based on deep belief networks. Algorithms , 2018, 11(9): 132 doi: 10.3390/a11090132 Zhang C Y, Chen C L P, Gan M, et al. Predictive deep boltzmann machine for multiperiod wind speed forecasting. IEEE Trans Sustain Energy , 2015, 6(4): 1416 doi: 10.1109/TSTE.2015.2434387 Khodayar M, Wang J H. Spatio-temporal graph deep neural network for short-term wind speed forecasting. IEEE Trans Sustain Energy , 2019, 10(2): 670 doi: 10.1109/TSTE.2018.2844102 Zhang Y C, Le J, Liao X B, et al. A novel combination forecasting model for wind power integrating least square support vector machine, deep belief network, singular spectrum analysis and locality-sensitive hashing. Energy , 2019, 168: 558 doi: 10.1016/j.energy.2018.11.128 Miao C X, Li H, Wang X, et al. Data-driven and deep-learning-based ultra-short-term wind power prediction. Autom Electr Power Syst , 2021, 45(14): 22 doi: 10.7500/AEPS20201127004

苗长新, 李昊, 王霞, 等. 基于数据驱动和深度学习的超短期风电功率预测. 电力系统自动化, 2021, 45(14):22 doi: 10.7500/AEPS20201127004 Zang H X, Cheng L L, Ding T, et al. A hybrid method for short-term photovoltaic power forecasting based on deep convolutional neural network. IET Gener Transm Distribution , 2018, 12(20): 4557 doi: 10.1049/iet-gtd.2018.5847 Wang H Z, Yi H Y, Peng J C, et al. Deterministic and probabilistic forecasting of photovoltaic power based on deep convolutional neural network. Energy Convers Manag , 2017, 153: 409 doi: 10.1016/j.enconman.2017.10.008 Sun Y C, Szűcs G, Brandt A R. Solar PV output prediction from video streams using convolutional neural networks. Energy Environ Sci , 2018, 11(7): 1811 doi: 10.1039/C7EE03420B Zang H X, Cheng L L, Ding T, et al. Day-ahead photovoltaic power forecasting approach based on deep convolutional neural networks and meta learning. Int J Electr Power Energy Syst , 2020, 118: 105790 doi: 10.1016/j.ijepes.2019.105790 Yan J C, Hu L, Zhen Z, et al. Frequency-domain decomposition and deep learning based solar PV power ultra-short-term forecasting model. IEEE Trans Ind Appl , 2021, 57(4): 3282 doi: 10.1109/TIA.2021.3073652 Hong Y Y, Rioflorido C L P P. A hybrid deep learning-based neural network for 24-h ahead wind power forecasting. Appl Energy , 2019, 250: 530 doi: 10.1016/j.apenergy.2019.05.044 Liu X, Yang L X, Zhang Z J. Short-term multi-step ahead wind power predictions based on A novel deep convolutional recurrent network method. IEEE Trans Sustain Energy , 2021, 12(3): 1820 doi: 10.1109/TSTE.2021.3067436 Wang F, Xuan Z M, Zhen Z, et al. A day-ahead PV power forecasting method based on LSTM-RNN model and time correlation modification under partial daily pattern prediction framework. Energy Convers Manag , 2020, 212: 112766 doi: 10.1016/j.enconman.2020.112766 Zhang Q, Ma Y, Li G L, et al. Applications of frequency domain decomposition and deep learning algorithms in short-term load and photovoltaic power forecasting. Proc CSEE , 2019, 39(8): 2221

张倩, 马愿, 李国丽, 等. 频域分解和深度学习算法在短期负荷及光伏功率预测中的应用. 中国电机工程学报, 2019, 39(8):2221 Song L Q, Xie Q Y, He Y K, et al. Ultra-short-term wind power combination forecasting model based on MEEMD-SAE-Elman // 2020 IEEE 4 th Information Technology, Networking, Electronic and Automation Control Conference . Chongqing, 2020: 1844 Hossain M A, Chakrabortty R K, Elsawah S, et al. Predicting wind power generation using hybrid deep learning with optimization. IEEE Trans Appl Supercond , 2021, 31(8): 1 Wang H Z, Wang G B, Li G Q, et al. Deep belief network based deterministic and probabilistic wind speed forecasting approach. Appl Energy , 2016, 182: 80 doi: 10.1016/j.apenergy.2016.08.108 Jing B, Tan L N, Qian Z, et al. An overview of research progress of short-term photovoltaic forecasts. Electr Measur Instrum , 2017, 54(12): 1 doi: 10.3969/j.issn.1001-1390.2017.12.001

荆博, 谭伦农, 钱政, 等. 光伏发电短期预测研究进展综述. 电测与仪表, 2017, 54(12):1 doi: 10.3969/j.issn.1001-1390.2017.12.001 Wang Y F, Fu Y C, Xue H. Ultra-short-term forecasting method of photovoltaic power generation based on Chaos-EEMD-PFBD decomposition and GA-BP neural networks. Acta Energiae Solaris Sin , 2020, 41(12): 55

王育飞, 付玉超, 薛花. 基于Chaos-EEMD-PFBD分解和GA-BP神经网络的光伏发电功率超短期预测法. 太阳能学报, 2020, 41(12):55 Yan J, Gao C Y, Yu G B. Comparative analysis of prediction errors based on offshore wind power characteristics. Comput Integr Manuf Syst , 2020, 26(3): 648

闫健, 高长元, 于广滨. 基于海上风电功率特性的预测误差对比分析. 计算机集成制造系统, 2020, 26(3):648 Wang K J, Qi X X, Liu H D, et al. Deep belief network based k -means cluster approach for short-term wind power forecasting. Energy , 2018, 165: 840 doi: 10.1016/j.energy.2018.09.118 Sun C, Ma M, Zhao Z B, et al. Deep transfer learning based on sparse autoencoder for remaining useful life prediction of tool in manufacturing. IEEE Trans Ind Inform , 2019, 15(4): 2416 doi: 10.1109/TII.2018.2881543 Su L C, Xing M L, Zhang H. Fault detection of key components of wind turbine based on combination prediction model. Acta Energiae Solaris Sin , 2021, 42(10): 220

苏连成, 邢美玲, 张慧. 基于组合预测模型的风电机组关键部位故障检测. 太阳能学报, 2021, 42(10):220 Qi B, Wang Y M, Zhang P, et al. Deep recurrent belief network model for trend prediction of transformer oil chromatography data. Power Syst Technol , 2019, 43(6): 1892 doi: 10.13335/j.1000-3673.pst.2019.0030

齐波, 王一鸣, 张鹏, 等. 面向变压器油色谱趋势预测的深度递归信念网络. 电网技术, 2019, 43(6):1892 doi: 10.13335/j.1000-3673.pst.2019.0030 Guo L, Lei Y G, Li N P, et al. Machinery health indicator construction based on convolutional neural networks considering trend burr. Neurocomputing , 2018, 292: 142 doi: 10.1016/j.neucom.2018.02.083 Yang B Y, Liu R N, Zio E. Remaining useful life prediction based on a double-convolutional neural network architecture. IEEE Trans Ind Electron , 2019, 66(12): 9521 doi: 10.1109/TIE.2019.2924605 Lu H, Vahid B, Nemani V P, et al. GAN-LSTM predictor for failure prognostics of rolling element bearings // IEEE International Conference on Prognostics and Health Management . Detroit, 2021: 1 Dai J J, Song H, Sheng G H, et al. Prediction method for power transformer running state based on LSTM network. High Volt Eng , 2018, 44(4): 1099 doi: 10.13336/j.1003-6520.hve.20180329008

代杰杰, 宋辉, 盛戈皞, 等. 采用LSTM网络的电力变压器运行状态预测方法研究. 高电压技术, 2018, 44(4):1099 doi: 10.13336/j.1003-6520.hve.20180329008 Sun S G, Wen Z T, Du T H, et al. Remaining life prediction of conventional low-voltage circuit breaker contact system based on effective vibration signal segment detection and MCCAE-LSTM. IEEE Sens J , 2021, 21: 21862 doi: 10.1109/JSEN.2021.3104290 Ma X, Hu H, Shang Y Z. A new method for transformer fault prediction based on multifeature enhancement and refined long short-term memory. IEEE Trans Instrum Meas , 2021, 70: 1 Zhao H S, Liu H H, Hu W J, et al. Anomaly detection and fault analysis of wind turbine components based on deep learning network. Renew Energy , 2018, 127: 825 doi: 10.1016/j.renene.2018.05.024 Lu B L, Liu Z H, Wei H L, et al. A deep adversarial learning prognostics model for remaining useful life prediction of rolling bearing. IEEE Trans Artif Intell , 2021, 2(4): 329 doi: 10.1109/TAI.2021.3097311 Jiang G Q, Xie P, He H B, et al. Wind turbine fault detection using a denoising autoencoder with temporal information. IEEE/ASME Trans Mechatron , 2018, 23(1): 89 doi: 10.1109/TMECH.2017.2759301 Meng Z, Zhan X Y, Li J, et al. An enhancement denoising autoencoder for rolling bearing fault diagnosis. Measurement , 2018, 130: 448 doi: 10.1016/j.measurement.2018.08.010 Xia M, Li T, Shu T X, et al. A two-stage approach for the remaining useful life prediction of bearings using deep neural networks. IEEE Trans Ind Inform , 2019, 15(6): 3703 doi: 10.1109/TII.2018.2868687 Sun Z X, Sun H X. Stacked denoising autoencoder with density-grid based clustering method for detecting outlier of wind turbine components. IEEE Access , 2019, 7: 13078 doi: 10.1109/ACCESS.2019.2893206 Chen Z Q, Chen X D, Olivira J, et al. Application of deep learning in equipment prognostics and health management. Chin J Sci Instrum , 2019, 40(9): 206 doi: 10.19650/j.cnki.cjsi.J1905122

陈志强, 陈旭东, Olivira J, 等. 深度学习在设备故障预测与健康管理中的应用. 仪器仪表学报, 2019, 40(9):206 doi: 10.19650/j.cnki.cjsi.J1905122 Fang X L, Yang Q, Liu G F, et al. Power supply restoration strategy for offshore multi-platform interconnected power system with faults. Autom Electr Power Syst , 2021, 45(7): 53

方晓伦, 杨强, 刘国锋, 等. 海上多平台互联电力系统故障后的供电恢复策略. 电力系统自动化, 2021, 45(7):53