Short-term Electricity Load Forecasting Based on Improved Seagull Algorithm Optimized Gated Recurrent Unit Neural Network

Mengfan Xu; Junyang Pan

doi:10.4108/ew.5282

Authors

Mengfan Xu Anhui University of Finance and Economics
Junyang Pan Anhui University of Finance and Economics

DOI:

https://doi.org/10.4108/ew.5282

Keywords:

short-term electricity load forecasting, gated cyclic unit, seagull optimization algorithm, complete ensemble empirical modal decomposition

Abstract

INTRODUCTION: The complexity of the power network, changes in weather conditions, diverse geographical locations, and holiday activities comprehensively affect the normal operation of power loads. Power load changes have characteristics such as non stationarity, randomness, seasonality, and high volatility. Therefore, how to construct accurate short-term power load forecasting models has become the key to the normal operation and maintenance of power.

OBJECTIVES: Accurate short-term power load forecasting helps to arrange power consumption planning, optimize power usage and largely reduce power system losses and operating costs.

METHODS: A hybrid decomposition-optimization-integration load forecasting method is proposed to address the problems of low accuracy of current short-term power load forecasting methods.

RESULTS: The original power load time series is decomposed using the complete ensemble empirical modal decomposition method, while the correlation of power load influencing factors is analysed using Pearson correlation coefficients. The seagull optimisation algorithm is overcome to fall into local optimality by using the random adaptive non-linear adjustment strategy of manipulated variables and the differential variational Levy flight strategy, which improves the search efficiency of the algorithm. Then, the The gated cyclic unit hidden layer parameters are optimised by the improved seagull optimisation algorithm to construct a short-term electricity load forecasting model.The effectiveness of the proposed method is verified by simulation experimental analysis. The results show that the proposed method has improved the accuracy of the forecasting model.

CONCLUSION: The CEEMD method is used to decompose the original load time series, which improves the accuracy of the measurement model. The GRU prediction model based on improved SOA optimization not only has better prediction accuracy than other prediction models, but also consumes the least amount of time compared to other prediction models.

Downloads

Download data is not yet available.

References

Zheng T, Dai Zemei, Yao Jiahao, et al. Integrated energysystem control freedom index and its impact on economicdispatch[J]. China Electric Power, 2021(3): 1-17.

Deng Y.S., Jiao F.S., Zhang R.F., et al. A review of researchon electric load forecasting methods in distribution networkplanning[J]. Electricity and Energy Efficiency ManagementTechnology, 2019(14): 1-7.

Le Haozhe. Research on short-term power load forecastingalgorithm based on neural network hybrid model[D].Nanchang: Nanchang University, 2021.

Lin Jialiang, Li Nuanqun, Wu Yongfeng. Correlationanalysis of short-term load forecasting accuracy in powersystems[J]. Automation Applications, 2017(12): 125-127.

Zhu JC, Yang ZL, Guo YJ, et al. A review on the applicationof deep learning in power load forecasting[J]. Journal ofZhengzhou University (Engineering Edition), 2019, 40(05):13-22.

Li B., Kong X. Y., Zhu S. J., et al. Short-term loadforecasting for regional grids considering demand responsein an interconnected energy environment[J]. Power SystemAutomation, 2021, 45(01): 71-78.

Xu Qing, Zhou Chao, Zhao Shuangshang, et al. Research onshort-term electric load forecasting methods based onmachine learning [J]. Electricity Measurement andInstrumentation, 2019, 56(23): 70-75.

Zhong Ting. Research on short-term power load forecastingmethod based on EEMD decomposition and its application[D]. Xiangtan: Xiangtan University, 2020.

Wang Yanling, Wu Mengkai. A medium- and long-term loadforecasting model based on partial least squares regressionunder the new economic normal[J]. Power AutomationEquipment, 2018, 38(03): 133-139.

Song Yanru. Research and application of multi-objectiveoptimization-based hybrid algorithm for load forecasting andmachine learning [D]. Lanzhou: Lanzhou University, 2020.

Song Jefei. Research and application of short-term loadforecasting methods for power plants [D]. Xi'an: Xi'anUniversity of Architecture and Technology, 2021.

Yinsheng S, Chunxiao L, Bao L, et al. Principal ComponentAnalysis of Short-term Electric Load Forecast Data Based onGrey Forecast[C]//Journal of Physics: Conference Series.IOP Publishing, 2020, 1486(6): 062031. DOI: https://doi.org/10.1088/1742-6596/1486/6/062031

Wang R, Zhou C X, Lu J. Short-term electricity loadforecasting based on a combined model[J]. ComputerSimulation, 2018, 35(12): 396-400.

Zhang Yanyu, Xiao Xi. Analysis of the current status ofresearch on load forecasting of power systems at home andabroad [J]. Shandong Industrial Technology, 2016(11): 215-208. DOI: https://doi.org/10.1016/j.biortech.2016.04.026

Yang Yun, Cai Qiu Na, Zhang Qiao Yu, et al. Research onshort-term load forecasting of new energy power systemsbased on combined forecasting models[J]. AutomationTechnology and Applications, 2019, 38(08): 102-106.

Yan Chongxi, Chen Hao. Research on optimizing LSSVMshort-term power load forecasting method based onimproved tennies whisker search algorithm[J]. ElectricalMeasurement and Instrumentation, 2020, 53(06): 6-11.

Liu Qianying, Ruan Yingjun, Shi Xiang, et al. Time-by-timeelectrical load prediction for office buildings based onKmeans clustering and BP neural network algorithm[J].Thermal Power Engineering, 2018, 33(03): 139-144.

Zhang Shuqing, Li Jun, Jiang Anqi, et al. A novel two-stageshort-term power load forecasting based on FPA-VMD and BiLSTM neural networks[J]. Power Grid Technology, 2022, 46(08): 3269-3279.

Deng Zaiyu, Li Jian, Zhang Zhenyuan, et al. Short-termpower load forecasting based on EEMD-GRU-MLR [J].Power Grid Technology, 2020, 44(02): 593-602.

Napoli C, Pappalardo G, Tina G M, et al. Cooperativestrategy for optimal management of smart grids by waveletRNNs and cloud computing [J]. IEEE Transactions onNeural Networks & Learning Systems, 2016, 27(8): 1672-1685. DOI: https://doi.org/10.1109/TNNLS.2015.2480709

Hao Q, Srinivasan D, Khosravi A. Short-term load and windpower forecasting using neural network-based predictionintervals [J]. IEEE Transactions on Neural Networks &Learning Systems, 2017, 25(2): 303-315. DOI: https://doi.org/10.1109/TNNLS.2013.2276053

Zhu L, Li C, Guo L, et al. Short-term spatial and temporalload forecasting method based on GCN-GRU [J]. Energy and Environmental Protection, 2022, 44(04): 211-215.

Jiang F, He JQ, Zeng ZG, et al. Electricity price forecastingbased on decomposition-optimization-integration learningmethod[J]. Chinese Science: Information Science, 2018, 48:1300-1315. DOI: https://doi.org/10.1360/N112018-00075

Li Shaojie. Research on short-term power load forecastingbased on FFT-CNN-LSTM [D]. Nanchang: NanchangUniversity, 2021.

Yang Y. Research on feature selection algorithm based onbinary seagull optimization algorithm [D]. Nanchang:Jiangxi University of Finance and Economics, 2020.

Xu Songjin, Long W. Improved grey wolf optimizationalgorithm based on stochastic convergence factor anddifferential variance[J]. Science, Technology andEngineering, 2018, 18(23): 252-256.