Safety assessment method for power operation environment based on multi-source data fusion

Traditional safety evaluation methods often grapple with limitations such as narrow index selection coverage, suboptimal evaluation accuracy, and questionable reliability. To address these challenges, this study introduces a novel safety assessment methodology for power operation environments, leveraging the potential of multi-source data fusion. This approach encompasses multi-modal and multi-source data mining, in conjunction with fusion processing of the gathered data specific to power operation environments. A comprehensive analysis of safety risk factors within power operations facilitates the construction of an intricate fault tree, which is employed to compute security risk values and complete the safety assessment of the power operation environment. Experimental outcomes indicate that the proposed method boasts an average accuracy of 93.2%, demonstrating robust evaluation stability and high efficiency. This efficacy underscores its potential as a valuable tool in enhancing the safety of power operations.