The embodiment of the invention provides a rotary mechanical equipment start-stop judgment method and system, and the method comprises the steps: obtaining the three-axis vibration signal data of rotary mechanical equipment, and extracting the feature data of a three-axis vibration signal from the three-axis vibration signal data; inputting the feature data into a plurality of preset machine learning classification models, and obtaining recognition results of the plurality of machine learning classification models; and obtaining the current start-stop state of the rotary mechanical equipment by adopting a model fusion strategy according to the identification results of the plurality of machine learning classification models. According to the method and the system provided by the embodimentof the invention, the current state of the industrial equipment is intelligently identified by utilizing the vibration signal of the machine, so that the accuracy of judging the start-stop state of the rotary mechanical equipment is improved, manual participation is not needed, equipment monitoring efficiency is improved, and the equipment monitoring cost is reduced.