A　novel　effective　method　for　identifying　＇Dispersed,　Disordered,　and　Polluting＇　(DDP)　sites　was　proposed　for　the　purpose　of　promoting　the　modernization　of　ecological　and　environmental　governance　capabilities　and　building　a　big　data　application　platform　for　enterprises’　pollution　prevention.　This　paper　aggregated　data　from　the　electricity　consumption　information　collection　system　and　characteristic　sensing　terminals,　including　user　daily　total　electricity　consumption　records,　peak　and　valley　electricity　consumption,　and　other　information.　Firstly,　we　used　the　hierarchical　K-means　algorithm　to　cluster　the　time　series　of　user　electricity　consumption　data.　After　ranking　by　cluster　features,　the　electricity　consumption　time　series　of　the　selected　suspicious　users　were　encoded　into　Gramian　angular　field　(GAF)　images.　Finally,　we　adopted　the　perceptual　hash　algorithm　to　build　the　model　to　identify　＇Dispersed,　Disordered,　and　Polluting＇　sites.　The　case　analysis　results　verified　this　method’s　feasibility,　rationality,　and　effectiveness.　©　2023,　The　Author(s),　under　exclusive　license　to　Springer　Nature　Singapore　Pte　Ltd.