Disclosed is a double-layer optimization based evaluation method for the external power receiving capability of a receiving-end power grid. The double-layer optimization based evaluation method includes setting basic parameters of a particle swarm optimization algorithm; commanding that Iter=Iter+1; judging whether the Iter is equal to 1 or not; judging whether the Iter is larger than Itermax or not; according to line parameters, load, power source parameters and network topology of the receiving-end power grid and transmission power of each external power transmission channel, creating an optimal power flow model of the receiving-end power grid and calculating an optical power flow of the receiving-end power grid; judging whether 'N-1' verification is satisfied or not; judging whether transient stability verification is satisfied or not; according to the transmission power of each external power transmission channel of the receiving-end power grid, calculating the external power receiving capability of the receiving-end power grid and updating a swarm extremum, an individual extremum of each particle and the like. The double-layer optimization based evaluation method has the advantages that the external power receiving capability of the receiving-end power grid can be evaluated scientifically and accurately, and guidance is provided for external power receiving capability improvement and grid structure optimization of the receiving-end power grid.