The invention provides an alarm fusion method based on data center anomaly monitoring. The alarm fusion method comprises the following steps: S1, alarm collection: collecting alarm information from each data source monitored by a whole operation and maintenance system in a series of time windows; S2, alarm preprocessing: carrying out redundancy elimination operation on the collected alarm information, and extracting key alarm information in the collected alarm information; S3, alarm association analysis: discovering association rules of the preprocessed alarm information according to a certainalgorithm; and S4, alarm integration and output: performing de-duplication and integration operation on the newly arrived alarm information within a certain time according to the association rule ofthe alarm information discovered in the step S3, and outputting the alarm information to a database. According to the invention, redundant information in the alarm information can be eliminated. The alarm information with a causal relationship is fused, so that the workload and the working difficulty of operation and maintenance personnel are reduced. The operation and maintenance cost is reduced.