The invention provides a method and system for evaluating the operation state of an intelligent electric energy meter based on a Bayesian network and a genetic algorithm. The method comprises the following steps: firstly, discretizing attribute indexes with continuous values in the attribute indexes for evaluating the running state of the electric energy meter; based on historical data based then,for the discretized attribute index, obtaining a discretized attribute index; determining a set of optimal attribute indexes by adopting a genetic algorithm; based on historical data, training a plurality of Bayesian nets; adopting a genetic algorithm to select an optimal Bayesian network according to the fitness function; and finally, observing the attribute index value of the intelligent electric energy meter. the operation state of the intelligent electric energy meter is determined through a Bayesian network reasoning algorithm; According to the method and the system, the state evaluationof the intelligent electric energy meter is accurately realized, great convenience is provided for the work of safe operation, state monitoring, data reading, electricity price issuing and the like of a power grid, the huge cost of periodically recovering and detecting the electric energy meter is effectively saved, and a powerful technical guarantee is provided for replacing the intelligent electric energy meter.