This paper addresses the problem of high signal redundancy and mixed fault information in substation monitoring terminals by proposing an edge data reduction and traceability method for substation equipment based on association rule mining. First, high-dimensional monitoring signals are sparsely sampled and reconstructed through compressed sensing technology deployed on edge intelligent devices, significantly reducing data volume while ensuring signal feature integrity. This approach effectively alleviates the challenges of storage, processing, and transmission of high-dimensional monitoring data. Subsequently, an association rule mining algorithm is employed to construct an association rule database for substation equipment alarm information, revealing strong correlations between critical alarm signals and fault types. Combined with temporal feature analysis and dynamic updates of alarm propagation patterns, a comprehensive association rule database and traceability method are established. This forms a closed-loop system of perception, diagnosis, and maintenance, providing an effective solution for real-time monitoring and precise operation and maintenance in smart substations.
Wang et al. (Thu,) studied this question.