Integrated learning high-speed rotating bearing fault diagnosis method based on genetic programming algorithm

The invention relates to an ensemble learning high-speed rotating bearing fault diagnosis method based on a genetic programming algorithm. The method comprises the steps of 1, determining an input variable and an output variable of a high-speed rotating bearing fault diagnosis model; step 2, based on historical sample data of high-speed bearing operation, executing a tree-based genetic programming algorithm in computing nodes in cloud computing, and putting values of each group of input variables and output variables into one computing node, a plurality of optimal sub-models used for describing the health state and the fault severity of the high-speed rotating bearing are obtained through parallel calculation of a plurality of calculation nodes; and step 3, combining a plurality of optimal sub-models into an optimal total model through tree-based genetic programming based on an integrated learning modeling method of distributed calculation, and applying the optimal total model to fault diagnosis of the highly