Research on a Data-Driven Safety Risk Assessment Model for Electrical Switching Operations in Substations

With the continuous development and increasing complexity of power systems, the safety of switching operations in substations has become an important issue for ensuring the stable operation of the power grid. In response to this issue, this paper proposes a safety risk prediction method for electrical switching operations based on the integration of an Improved Black Kite Algorithm (IBKA) and Extreme Learning Machine (ELM). The traditional Black Kite Optimization Algorithm (BKA) tends to fall into local optima when solving high-dimensional complex problems, which affects optimization performance. To address this, several strategies have been applied to improve the BKA, leading to the development of an Improved Black Kite Algorithm. These improvements include an adaptive search strategy, enhanced local optimization, and a hybrid intelligent mechanism, all of which effectively enhance the global search capability and local convergence of the IBKA, thus improving the optimization accuracy and efficiency. Experimental results show that the safety risk prediction model based on IBKA-ELM performs excellently in predicting the risks of switching operations, achieving an accuracy of 0.926, which significantly outperforms traditional machine learning models. Compared to traditional models, the IBKA-ELM model not only achieves higher prediction accuracy but also demonstrates stronger robustness. The IBKA-ELM model proposed in this paper provides an effective solution for predicting the safety risks of electrical switching operations in power systems, enabling the early identification of potential risks and offering decision support for the safe operation and management of power systems, with broad application prospects.