The invention relates to a pipeline monitoring and early warning method based on correlation analysis. The method comprises the following steps: step 1, monitoring a pipeline by a sensor unit to obtain pipeline strain data; step 2, acquiring pipeline strain data in an initial time window, and solving data correlation features and expectation E and variance sigma <2> of a correlation feature distribution function through the pipeline strain data in the initial time window; step 3, calculating support degree and validity of a monitoring value K-approach, thereby judging sensor reliability and pipeline damage situations. The method provided by the invention has no need of building a system parsing model, can correlatively analyze and extract feature information through the monitored pipelinestrain data, and can realize monitoring and early warning for pipeline strain states and evaluation on reliability of the sensor.