High voltage underground cable bonding optimisation to prevent cable termination faults in mixed high‐voltage lines

Akbal, Bahadir

doi:10.1049/iet-gtd.2019.1559

Cited by 6 publications

(1 citation statement)

References 25 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…These methods are used as objective function because these methods are widely used for the forecasting studies in electrical engineering [30][31][32][33][34][35][36]. In hybrid artificial neural networks, the weights of artificial neural network are updated by optimization method [37]. The prediction methods should be trained with the training data to predict MV, MC, and MHC values.…”

Section: Usage Of the Prediction Methods In Optimization Processmentioning

confidence: 99%

Optimum Cable Bonding With Pareto Optimal and Hybrid Neural Method to Prevent High Voltage Cable Insulation Faults in the Wind Energy Systems

AKBAL

2023

Preprint

View full text Add to dashboard Cite

Photovoltaic systems and wind energy systems are generally used in distributed system. However, zero-sequence and harmonic currents occur due to photovoltaic systems, and positive-sequence harmonic currents occur due to wind turbines. Thus, the sheath current in high voltage cables increases due to these currents and poor power quality in distributed generation plants. Metal parts of the cable are grounded to prevent the negative effects of sheath currents. When long high voltage cable lines are grounded with the method in the literature, high sheath current and voltage occur on the cable sheath. Therefore, the SSBLR method is designed as a new method for the distributed generation system. The SSBLR method has been optimized with prediction methods and multi-objective optimization methods to prevent negative effects of the sheath current. Pareto optimal method is used as multi-objective optimization method, and artificial neural network, hybrid artificial neural network and regression methods are used as prediction method. The most dominant result is searched using different estimation and optimization methods in pareto optimal method. When the artificial neural network-genetic algorithm hybrid method is used as a prediction method, and genetic algorithm used as optimization method, significant decreases are observed in sheath current and voltage.

show abstract