The invention discloses an operation state prediction method and system based on a BP (Back Propagation) neural network algorithm. The method and the system are applied to an electric information system to predict the operation state of the to-be-predicted equipment of the electric information system. The operation state prediction method comprises the following specific steps: firstly, determining a plurality of equipment indexes of the to-be-predicted equipment of the electric information system; according to historical data, taking the values of the equipment indexes for multiple times; utilizing the values obtained in multiple times to train a preset BP neural network model so as to obtain an operation state prediction model; and according to the practical equipment index of the to-be-predicted equipment and the operation state prediction model, obtaining the prediction operation state of the to-be-predicted equipment. Operation and maintenance personnel can carry out overhaul in time when a situation that the to-be-predicted equipment needs to be overhauled is determined according to the prediction operation state, so that adverse effects on the normal operation of a power grid system can be avoided.