On the basis of reanalysis data from the European Center for Medium-Range Weather Forecasting (ERA-interim),124 stationsprecipitation data of the middle and lower reaches of the Yangtze River and the data of the 4 numerical forecast models, this paper took thelocal persistent storm process occurred in the middle and lower reaches from 30 June 2016 to 7 July 2016 as example, discussed the 4numerical models and one-dimensional moist vorticity (ODMV) and one-dimensional moist divergence (ODMD)’s diagnosis of storm location forMLRYRV. The results were summarized as follow: (1) No matter subjective analysis or spatial similarity with real precipitation, ODMV/ODMD isbetter than the 4 numerical models for the diagnosis of rain belt and strong precipitation center. (2) Japan-thin’s forecast result significantly weak,while T639-thin’s result is stronger than real precipitation, EC-thin, Grapes and T639-thin’s result also show the greater dispersion, onlyODMV/ODMD is the smallest mean error(ME) and root mean square error(RMSE) no matter for the rain belt or the precipitation center. (3) Withthe precipitation intensity increased, the Non-response rates of Japan-thin, EC-thin and Grapes increased significantly, the Threat Score (TS)reduced sharply; while T639-thin’s empty rate increased faster, the TS reduced quickly also; only ODMV/ODMD’s empty rate decreasedrapidly and Non-response rate increased slowly, the TS decreased slowly, to the highest standard (6h≥30 mm ), the TS still get 0.22.