Phase-controlled switching technology can remove the fault at the optimal phase of the short-circuit current and reduce damage from the short-circuit current to power equipment. The primary challenge lies in achieving accurate and rapid prediction of short-circuit current zero-crossing points. To this end, this paper proposes a prediction method for short-circuit current zero-crossing points based on collaborative feature extraction and enhancement. First, a feature extraction and enhancement module is constructed by combining a convolutional neural network (CNN) with squeeze-and-excitation (SE) attention mechanism to highlight the local key features of short-circuit currents. Then, the enhanced features are fed into a gated recurrent unit (GRU) to establish a temporal model by utilizing its gating mechanism and memory capability. Subsequently, a multiple-input multiple-output (MIMO) strategy is adopted to realize multi-step prediction. Finally, the chaotic evolution optimization (CEO) algorithm is combined with hyperparameter optimization to construct the CEO algorithm-based CNN-SE-GRU composite prediction model. Simulation and experimental results demonstrate that the hybrid prediction model achieves a coefficient of determination (R²) of 93.46%, a root mean square error (RMSE) of 69.33, and a mean absolute error (MAE) of 47.14. The model reduces the sampling duration to 2 ms with an average prediction time of 15.78 ms, representing a 97.51% reduction in prediction time compared to long short-term memory (LSTM) network recursive prediction. The mean absolute error for zero-crossing point prediction in experimental waveforms is 0.18 ms, with an average relative error of 5.89% at extreme values. Under simulations of different operating conditions and noise environments, the average prediction error for zero points is less than 0.32 ms. Compared with traditional methods, the multi-step hybrid prediction model significantly reduces prediction time while maintaining prediction accuracy, providing a theoretical basis for precise phase-controlled interruption of short-circuit fault currents by circuit breakers.
KASIKCII. Short circuits in power systems: a practical guide to IEC 60909-0[M]. Hoboken: John Wiley & Sons, 2018:167-173.
[2]
GUANC, DINGJ G, YAOX F, et al .Study on short-circuit current interruption characteristics of double-break fast vacuum circuit breaker within the minimum arcing time[J].International Journal of Electrical Power & Energy Systems, 2023, 147:108865.
DUANX Y, HUANGZ H, LIAOM F, et al .Research of controlled fault interruption based on complete response method[J].High Voltage Engineering, 2014, 40(8):2264-2270.(in Chinese)
[5]
PANOIUC, CIULICAD, PANOIUM, et al. Research based on modeling and simulation of the transient regime in controlled switching with high power switches[J]. Machines,2021,9(5): 99.
MAF Y, WEIY, LIL Q, et al. Study on phase-controlled interruption of high DC component short-circuit fault currents based on Prony algorithm[J].Advanced Technology of Electrical Engineering and Energy, 2025, 44(4):91-99.(in Chinese)
[8]
POLTLA, FROHLICHK .A new algorithm enabling controlled short circuit interruption[J]. IEEE Transactions on Power Delivery, 2003, 18(3): 802-808.
[9]
POELTLA, FROHLICHK. Two new methods for very fast fault type detection by means of parameter fitting and artificial neural networks[J]. IEEE Transactions on Power Delivery, 1999, 14(4):1269-1275.
[10]
THOMASR P, SOLVERC E .A method for controlled fault interruption for use with HV SF6 circuit breakers[C]//2007 IEEE Lausanne Power Tech.July 1-5, 2007, Lausanne, Switzerland.IEEE, 2008:1135-1140.
YUANZ, LUOC J, FANGC E, et al .Design of control system for controlled fault interruption based on half-wave Fourier algorithm[J].High Voltage Engineering, 2013, 39(4):869-875.(in Chinese)
HUANGZ H, DUANX Y, ZOUJ Y, et al. Fault current parameter estimation based on adaptive RLS algorithm[J]. Proceedings of the CSEE, 2014, 34(15):2460-2469.(in Chinese)
HUANGZ H, ZHANGD B, ZOUJ Y, et al .Design of control system for controlled fault interruption based on adaptive RLS algorithm[J].High Voltage Engineering, 2016, 42(10):3214-3220.(in Chinese)
[17]
YANGX, LONGQ, LIH, et al. Controlled fault current interruption scheme for improved fault prediction accuracy[J]. Applied Sciences, 2025, 15(6): 3106.
RONGW S, CAIZ Y. Research on zero prediction method of synchronous breaking short circuit current of breaker[J]. Transactions of China Electrotechnical Society,2019,34(15):3167-3174. (in Chinese)
[20]
WANA P, CHANGQ, AL-BUKHAITIK, et al .Short-term power load forecasting for combined heat and power using CNN-LSTM enhanced by attention mechanism[J]. Energy, 2023, 282:128274.
LIW Y, YANGG C, WENM, et al .A short-term power load forecasting method based on spatiotemporal graph attention[J].Journal of Hunan University (Natural Sciences), 2024, 51(2):57-67.(in Chinese)
[23]
YER K, WANGH F, BASHIRT, et al .Fast and accurate method for short-circuit current calculation in distribution network with IIDGs[J].International Journal of Electrical Power & Energy Systems, 2024, 155:109622.
CHENJ J, MIAOX R. Short circuit current peak prediction based on two-dimensional cloud model[J]. Power System Protection and Control, 2018, 46(23): 94-101. (in Chinese)
HUANGX, HEH Y, LUOD S, et al .A novel prediction method of short-circuit current[J]. Proceedings of the CSU-EPSA, 2017, 29(1):24-29.(in Chinese)
[28]
ZHENGX, WANGH F, JIANGK, et al. A novel machine learning-based short-circuit current prediction method for active distribution networks[J]. Energies, 2019, 12(19):3793.
CHENJ M, ZHANGP P. Application of grey BP neural network model in short circuit current peak prediction of power system[J]. Journal of Light Industry,2018,33(4):79-85. (in Chinese)
[31]
ALJARRAHR, AL-OMARYM, ALSHABID, et al. Application of artificial neural network-based tool for short circuit currents estimation in power systems with high penetration of power electronics-based renewables[J].IEEE Access, 2023, 11: 20051-20062.
HUANGL C, ZHANGL Y, HUY Y, et al. Fast prediction method of short-circuit current zero based on an algorithm of long short term memory[J]. High Voltage Engineering,2023,49(12):5022-5031. (in Chinese)
[34]
VEERAMSETTYV, EDUDODLAB R, SALKUTIS R. Zero-crossing point detection of sinusoidal signal in presence of noise and harmonics using deep neural networks[J]. Algorithms, 2021, 14(11): 329.
LID H, ZHAOM K, GAOL. Fault diagnosis method of chiller sensor deviation based on CNN-GRU[J]. Journal of Hunan University (Natural Sciences),2022,49(2):74-82. (in Chinese)
LIS S, LIUH R, GANY D, et al. Image dehazing network based on residual dense block and attention mechanism[J]. Journal of Hunan University (Natural Sciences), 2021,48(6):112-118.(in Chinese)
WANGM M, ZOUB, SHIL J, et al .Research on multi-step prediction strategies of Arctic sea ice extent based on LSTM[J].Marine Forecasts, 2025, 42(1):11-22.(in Chinese)
[43]
TAIEB SBEN, BONTEMPIG, ATIYAA F, et al .A review and comparison of strategies for multi-step ahead time series forecasting based on the NN5 forecasting competition[J].Expert Systems with Applications, 2012, 39(8):7067-7083.
XUS C, YANGW A, YANX F. Accurate interpolation algorithm for nonlinear errors in post-processing of five-axis machine tools[J]. Machine Building & Automation,2024,53(6):47-52. (in Chinese)
[46]
DONGY C, ZHANGS H, ZHANGH L, et al . Chaotic evolution optimization: a novel metaheuristic algorithm inspired by chaotic dynamics[J]. Chaos, Solitons & Fractals, 2025, 192: 116049.
ZHANGD P, LANZ, DUY H. Short-term wind power forecasting based on PCHIP-VMD dataanalysis and SSA-LSTM model[J]. Journal of Electronic Measurement and Instrumentation, 2025, 39(5): 251-261.(in Chinese)
HUANGK, HAOR K, GUOY F .State-of-health estimation method for lithium-ion battery based on comprehensive feature evaluation and model optimization[J].Proceedings of the CSU-EPSA, 2025, 37(5):131-140.(in Chinese)
HUANGL C, YANGH T, CUIM T, et al .A method for predicting short circuit current zeros in LSTM networks based on self-learning[J].Advanced Technology of Electrical Engineering and Energy, 2024, 43(2):78-86.(in Chinese)
[53]
DE STEFANIJ, BONTEMPIG .Factor-based framework for multivariate and multi-step-ahead forecasting of large scale time series[J].Frontiers in Big Data, 2021, 4:690267.
[54]
TAOC X, XINGB S, LIT G .A fault location method for hybrid transmission lines based on empirical Fourier decomposition[J].Archives of Electrical Engineering, 2023:1035-1053.
基金资助
甘肃省高等学校产业支撑计划资助项目(2023CYZC-72)
Gansu Provincial Higher Education Industry Support Program(2023CYZC-72)