The invention discloses a clustering point wind power forecasting method based on tuple matching. The clustering point wind power forecasting method based on tuple matching comprises the following steps of inputting active power data of wind power clustering points; carrying out piecewise linearization processing on the wind power data of the wind power clustering points through a piecewise linearization method; expressing a result after the piecewise linearization processing in a tuple sequence mode; determining a tuple sequence S0 to be matched, and extracting formed tuple sequences to form a tuple sequence set Sset, wherein the tuple sequence S0 to be matched is a tuple sequence of the wind power output data within the time interval from T1-d to T1 one integral day d before the current time T1; obtaining a target tuple sequence S1, most matched with the tuple sequence S0 to be matched, in the tuple sequence set Sset through a tuple dynamic warping matching algorithm; calculating a transformation coefficient between the tuple sequence S0 to be matched and the target tuple sequence S1; operating historic data behind the target tuple sequence S1 by the utilization of the obtained transformation coefficient; outputting a wind power forecasting value of the clustering points. The clustering point wind power forecasting method based on tuple matching solves the problem that an existing wind power forecasting method cannot effectively forecast power of the wind power clustering points.