The invention discloses a fault diagnosis method for a circuit breaker operating mechanism on the basis of CCA (Curvilinear Component Analysis)-SOM (Self-organizing Map). The fault diagnosis method comprises the following steps: firstly establishing a fault diagnosis data model of the circuit breaker operating mechanism; then performing dimension reduction on characteristic quantity extracted from circuit breaker opening and closing coil current by utilizing a curvilinear component analysis algorithm; subsequently classifying fault types by utilizing a self-organizing neural network algorithm to obtain self-organizing neural network algorithm training results; finally performing fault diagnosis on to-be-tested circuit breaker opening and closing coil current data by utilizing the self-organizing neural network algorithm training results. The fault diagnosis method for the circuit breaker operating mechanism on the basis of CCA-SOM, disclosed by the invention, has the benefit that a curvilinear component analysis and self-organizing neural network is applied to the fault diagnosis of the circuit breaker operating mechanism, so that the diagnosis speed is high, and the diagnosis result accuracy rate is high.