The invention discloses a power grid operation state judgment method and system based on fuzzy clustering and RS-KNN model. The method comprises the following steps: Step 1: selecting a characteristicvariable for discriminating the operation state of the power grid; and Step 2: using a fuzzy clustering algorithm to divide the health state of the power grid operation into four types according to the health degree, which are respectively health, sub-health, general illness and serious illness. Before clustering, in order to eliminate the influence of the data dimension, a normalization method needs to be adopted to process the characteristic variable data of the power grid. Step 3: taking the characteristic variables obtained in the step 1 as input variables of a training set, taking data corresponding to the classified power grid operation states obtained in the step 2 is used as an input variable of a training set, and setting labels corresponding to the four power grid operation states, namely, the health state, the sub-health state, the general illness state and the serious illness state, output by the training set to be 1, 2, 3 and 4 respectively. Step 4: using the obtained four kinds of power grid operation state data to train RS-KNN model.