Adversarial hierarchical TOPSIS with FAHP dynamic weighting: A safety risk assessment framework for live work in electrical system

Under the intelligent upgrading of new-generation power systems, traditional safety assessment methods struggle to address dynamic risk coupling in multi-task scenarios due to increasing electrical system complexity and live work demands. This paper proposes a safety risk assessment framework for live work in electrical systems based on the adversarial hierarchical extraction TOPSIS method. A multi-task assessment indicator model integrating mechanical reliability, environmental disturbances, and human-robot collaboration is established through environmental analysis and historical data. By combining the Fuzzy Analytic Hierarchy Process and Criteria Importance Through Intercriteria Correlation, the method achieves dynamic weight adaptation for complex conditions. Improved TOPSIS quantifies solution distances to ideal references, while adversarial interpretive structural models hierarchically extract risk propagation paths. Experiments confirm the method’s high accuracy and real-time responsiveness for live work safety risk assessment.