The invention discloses a high-pressure heater state data analysis method. The method comprises the following steps: processing high-pressure heater state original data to obtain a training sample ofa high-pressure heater state monitoring model, clustering the training samples of the high-pressure heater state monitoring model by using an FCM clustering method based on a genetic simulated annealing algorithm according to the optimal clustering number to obtain a membership matrix, and obtaining the high-pressure heater state monitoring model according to the membership matrix and the trainingsamples; The method comprises the following steps: firstly, processing original data; obtaining a training sample of the detection model; clustering the training samples of the high-pressure heater state monitoring model by using an FCM clustering method based on a genetic simulated annealing algorithm; According to the FCM clustering method based on the genetic simulated annealing algorithm, thedefect that the FCM clustering method is prone to falling into a local optimal solution can be overcome, the global search capability is enhanced, the uncertainty and inaccuracy of data in equipmentmonitoring can be effectively processed, and therefore the real condition of equipment can be well reflected.