The invention relates to a method for evaluating the sortie ability of a shipboard aircraft according to an RBF based on a fast learning rate. The method comprises the steps of, according to an indexsystem established by a shipboard aircraft sortie ability index system model generator, generating a specified number of to-be-evaluated samples for training an RBF neural network based on the fast learning rate; adjusting the number of hidden layer neurons after judgment by an RBF hidden layer neuron increase and decrease judger; realizing robust adjustment of weights and adaptive fast adjustmentof the learning rate of the neural network; and finally realizing fast evaluation of the sortie ability of the shipboard aircraft. The non-linear mapping capability of the RBF neural network based onthe fast learning rate is used for evaluating the complex non-linear sortie ability of the shipboard aircraft, so that the problems that a traditional evaluation method is relatively high in subjectivity and complex in evaluation process can be avoided. The number of the hidden layer neurons is increased or decreased by comparing an evaluation value with a threshold value, so that the rationalityis higher; and the fast learning rate and a robust algorithm guarantee faster learning rate and stability of the neural network. According to the method, the effectiveness of the evaluation method isverified through a simulation experiment.