The invention discloses a method for mining state data association rules of extra-high voltage power transmission equipment. The method comprises the following steps of: establishing a one-to-one corresponding relationship between candidate rules and positions of particles on the basis of particle swarm optimization PSO, and respectively encoding each candidate rule into a particle; restricting anupper bound and a lower bound of the encoding according to a state data set of extra-high voltage power transmission equipment, and randomly initiating a particle swarm number and an iteration frequency; and setting a fitness function, carrying out parallel iteration in a maximum iteration frequency on the basis of a Spark distributed parallel processing framework, and determining the position ofthe particle corresponding to a globally optimal solution, namely, a state data association rule of the extra-high voltage power transmission equipment. According to the method, rules are quantitatively constructed as particles on the basis of the PSO, and discretization is not carried out in the association rule mining, so that the data correctness is ensured; and parallel design is carried outon the basis of Spark, so that the time consumption is decreased and a valuable solution is provided for solving the mining of association rules with numerical type attribute features under big data sizes.