To　solve　the　problem　of　optimal　layout　of　seismic　monitoring　stations　in　a　water　supply　network,　a　sample　attribute　matrix　was　built　through　the　concept　of　coefficient　influence.　A　dynamic　classification　model　for　optimizing　seismic　monitoring　stations＇　placement　in　water　distribution　networks　was　proposed　combined　with　dynamic　classification　method　(DT).　The　iterative　calculation　process　for　the　dynamic　classification　model　was　also　offered.　Firstly,　the　hydraulic　simulation　model　for　a　leaky　pipe　network　caused　by　earthquake　was　established　based　on　the　node　pressure　method.　The　flow　values　of　each　pipe　under　different　working　conditions　were　calculated　by　hydraulic　simulation　model　and　EPANET　software;　then,　the　flow　influence　coefficient　matrix　was　built.　Secondly,　the　dynamic　classification　method　was　introduced　for　optimizing　the　grading　of　all　the　pipes,　and　the　key　earthquake　monitoring　stations　were　selected　for　optimal　layout　combined　with　the　actual　situation.　Finally,　a　sample　network　was　calculated　by　the　model　and　its　relevant　calculation　programs.　The　analysis　results　show　that　the　classification　of　pipes　in　the　water　supply　network　is　reasonable　and　scientific,　and　seismic　monitoring　stations　distribute　uniformly　in　the　water　supply　network.　Furthermore,　the　model　can　wear　off　anthropogenic　impact　to　a　certain　extent　and　enhance　the　monitoring　effect　in　earthquake　and　economics　of　the　network　building.　©　2014　American　Society　of　Civil　Engineers.