Abstract Accurate short-term electrical load forecasting plays an important role in ensuring the safe and stable operation of the power grid, optimizing energy management, improving the utilization rate of power generation equipment, and reducing operating costs. Owing to the incapability of traditional time series methods in dealing with data’s nonlinear characteristics, this article first compared the applicability and forecasting performance of classical machine learning methods in short-term electrical load forecasting. The adopted modeling methods include: support vector regression, Gaussian process regression and forward neural network; the values of metrics proved that the machine learning methods can obtain good prediction accuracy and are suitable for processing short-term load data with strong nonlinear characteristics; on this basis, a short-term electrical load forecasting method based on temporal convolutional network (TCN) deep learning method was proposed. This model has the ability to extract features from large samples of time series and realize prediction, and its structure can effectively solve the degenerate problem of deep network learning; finally, forecasting experiment was conducted on the real load data. The results showed that the TCN can acquire much higher accuracy, and the deep learning method outperforms the classical machine learning methods in non-linear feature extraction.
[1] Ayub N, Irfan M, Awais M, et al.Big data analytics for short and medium-term electricity load forecasting using an AI techniques ensembler[J]. Energies, 2020, 13(19): 5193-5213. [2] Barak S, Sadegh S S.Forecasting energy consumption using ensemble ARIMA-ANFIS hybrid algorithm[J]. International Journal of Electrical Power and Energy Systems, 2016, 82: 92-104. [3] 刘畅, 卓建坤, 赵东明, 等. 利用储能系统实现可再生能源微电网灵活安全运行的研究综述[J]. 中国电机工程学报, 2020, 40(1): 1-18. Liu Chang, Zhuo Jiankun, Zhao Dongming, et al.A review of the utilization of energy storage system for the flexible and safe operation of renewable energy microgrids[J]. Proceedings of the CSEE, 2020, 40(1): 1-18. [4] 李卫国, 陈立铭, 张师, 等. 分时电价下考虑储能调度因素的短期负荷预测模型[J]. 电力系统保护与控制, 2020, 48(7): 133-140. Li Weiguo, Chen Liming, Zhang Shi, et al.Short-term load forecasting model considering energy storage scheduling factors under time-sharing price[J]. Power System Protection and Control, 2020, 48(7): 133-140. [5] 李培强, 丰云鹤, 李欣然, 等. 考虑超短期负荷预测的储能电池参与电网一次调频控制策略[J]. 电力系统自动化, 2019, 43(19): 87-95. Li Peiqiang, Feng Yunhe, Li Xinran, et al.Control strategy for energy storage battery participating in primary frequency regulation of power grid considering ultra-short-term load forecasting[J]. Automation of Electric Power System, 2019, 43(19): 87-95. [6] 张淑清, 段晓宁, 张立国, 等. Tsne降维可视化分析及飞蛾火焰优化ELM算法在电力负荷预测中应用[J]. 中国电机工程学报, 2021, 41(9): 3120-3130. Zhang Shuqing, Duan Xiaoning, Zhang Liguo, et al.Tsne dimension reduction visualization analysis and MFOELM algorithm applied in power load forecasting[J]. Proceedings of the CSEE, 2021, 41(9): 3120-3130. [7] Boroojeni K G, Amini M H, Bahrami S, et al.A novel multi-time-scale modeling for electric power demand forecasting: from short-term to medium-term horizon[J]. Electric Power Systems Research, 2017, 142: 58-73. [8] 康重庆, 夏清, 刘梅. 电力系统负荷预测 [M]. 2版. 北京: 中国电力出版社, 2017. [9] Panapakidis I P, Skiadopoulos N, Christoforidis G C.Combined forecasting system for short-term bus load forecasting based on clustering and neural networks[J]. IET Generation, Transmission & Distribution, 2020, 14(18): 3652-3664. [10] Dai Yeming, Zhao Pei.A hybrid load forecasting model based on support vector machine with intelligent methods for feature selection and parameter optimization[J]. Applied Energy, 2020, 279: 115332-115340. [11] Gao Xin, Li Xiaobing, Zhao Bing, et al.Short-term electricity load forecasting model based on EMD-GRU with feature selection[J]. Energies, 2019, 12(6): 1140-1157. [12] Malekizadeh M, Karami H, Karimi M, et al.Short-term load forecast using ensemble neuro-fuzzy model[J]. Energy, 2020, 196: 117127-117136. [13] Khwaja A S, Zhang X, Anpalagan A, et al.Boosted neural networks for improved short-term electric load forecasting[J]. Electric Power Systems Research, 2017, 143: 431-437. [14] 王增平, 赵兵, 纪维佳, 等. 基于GRU-NN模型的短期负荷预测方法[J]. 电力系统自动化, 2019, 43(5): 53-62. Wang Zengping, Zhao Bing, Ji Weijia, et al.Short-term load forecasting method based on GRU-NN model[J]. Automation of Electric Power Systems, 2019, 43(5): 53-62. [15] Yang Lintao, Yang Honggeng.Analysis of different neural networks and a new architecture for short-term load forecasting[J]. Energies, 2019, 12(8): 1433-1455. [16] 储晨阳,秦川,鞠平,等. 基于优化稀疏编码的超短期负荷滚动多步预测[J]. 电工技术学报, 2021, 36(19): 4050-4059. Chu Chenyang, Qin Chuan, Ju Ping, et al.Multi-step rolling ultra-short-term load forecasting based on the optimized sparse coding[J]. Transactions of China Electrotechnical Society, 2021, 36(19): 4050-4059. [17] 邓带雨, 李坚, 张真源, 等. 基于EEMD-GRU-MLR的短期电力负荷预测[J]. 电网技术, 2020, 44(2): 593-602. Deng Daiyu, Li Jian, Zhang Zhenyuan, et al.Short-term electric load forecasting based on EEMD-GRU-MLR[J]. Power System Technology, 2020, 44(2): 593-602. [18] Liu Tongxiang, Jin Yu, Gao Yuyang.A new hybrid approach for short-term electric load forecasting applying support vector machine with ensemble empirical mode decomposition and whale optimization[J]. Energies, 2019, 12(8): 1520-1539. [19] 徐佳宁, 倪裕隆, 朱春波. 基于改进支持向量回归的锂电池剩余寿命预测[J]. 电工技术学报, 2021, 36(17): 3693-3704. Xu Jianing, Ni Yulong, Zhu Chunbo.Remaining useful life prediction for lithium-ion batteries based on improved support vector regression[J]. Transactions of China Electrotechnical Society, 2021, 36(17): 3693-3704. [20] Richardson R R, Osborne M A, Howey D A.Gaussian process regression for forecasting battery state of health[J]. Journal of Power Sources, 2017, 357: 209-219. [21] Borovykh A, Bohte S, Oosterlee C W. Conditional time series forecasting with convolutional neural networks[EB/OL].2017, https://arxiv.org/abs/1703. 04691. [22] Rodríguez F, Martín F, Fontán L, et al.Very short-term load forecaster based on a neural network technique for smart grid control[J]. Energies, 2020, 13(19): 5210-5228. [23] 苏学能, 刘天琪, 曹鸿谦, 等. 基于Hadoop架构的多重分布式BP神经网络的短期负荷预测方法[J]. 中国电机工程学报, 2017, 37(17): 4966-4973. Su Xueneng, Liu Tianqi, Cao Hongqian, et al.A multiple distributed BP neural networks approach for short-term load forecasting based on Hadoop framework[J]. Proceedings of the CSEE, 2017, 37(17): 4966-4973. [24] 吴润泽, 包正睿, 王文韬, 等. Hadoop架构下基于模式匹配的短期电力负荷预测方法[J]. 电工技术学报, 2018, 33(7): 1542-1551. Wu Runze, Bao Zhengrui, Wang Wentao, et al.Short-term power load forecasting method based on pattern matching in Hadoop framework[J]. Transactions of China Electrotechnical Society, 2018, 33(7): 1542-1551. [25] Shi Heng, Xu Minghao, Li Ran.Deep learning for household load forecasting—a novel pooling deep RNN[J]. IEEE Transactions on Smart Grid, 2018, 9(5): 5271-5280. [26] Kong Weicong, Dong Zhaoyang, Jia Youfei, et al.Short-term residential load forecasting based on LSTM recurrent neural network[J]. IEEE Transactions on Smart Grid, 2019, 10(1): 841-851. [27] 赵兵, 王增平, 纪维佳, 等. 基于注意力机制的CNN-GRU短期电力负荷预测方法[J]. 电网技术, 2019, 43(12): 4370-4376. Zhao Bing, Wang Zengping, Ji Weijia, et al.A short-term power load forecasting method based on attention mechanism of CNN-GRU[J]. Power System Technology, 2019, 43(12): 4370-4376. [28] Bai Shaojie, Kolter J Z, Koltun V. An empirical evaluation of generic convolutional and recurrent networks for sequence modeling [EB/OL].2018, https://arxiv.org/abs/1803.01271. [29] Zhou Danhua, Li Zhanying, Zhu Jiali, et al.State of health monitoring and remaining useful life prediction of lithium-ion batteries based on temporal convolutional network[J]. IEEE Access, 2020, 8: 53307-53320. [30] Wu Pin, Sun Junwu, Chang Xuting, et al.Data-driven reduced order model with temporal convolutional neural network[J]. Computer Methods in Applied Mechanics and Engineering, 2020, 360: 112766-112778. [31] Oord A V D, Dieleman S, Zen H, et al. WaveNet: a generative model for raw audio[EB/OL].2016, https://arxiv.org/abs/1609.03499v2. [32] Wang Yongjian, Yang Ke, Li Hongguang.Industrial time-series modeling via adapted receptive field temporal convolution networks integrating regularly updated multi-region operations based on PCA[J]. Chemical Engineering Science, 2020, 228: 115956-115971. [33] He Kaiming, Zhang Xiangyu, Ren Shaoqing, et al. Deep residual learning for image recognition[EB/OL].2015, https://arxiv.org/abs/1512.03385. [34] Hewage P, Behera A, Trovati M, et al.Temporal convolutional neural (TCN) network for an effective weather forecasting using time-series data from the local weather station[J]. Soft Computing, 2020, 24(21): 16453-16482.
2022, Vol. 37 
