The invention discloses a tightening technology online fault detector based on process data analysis and a method of the detector. The tightening technology online fault detector comprises a tightening process data acquisition module, a processor and a fault classifier, the tightening process data acquisition module is used for acquiring bolt tightening process data in real time and transmitting the data to the processor, the processor is used for preprocessing the received bolt tightening process data, retaining tightening process data with pure rising moments, transmitting the preprocessed tightening data to the fault classifier acquired through offline training to perform fault classification and outputting fault classification results, particular processes of the fault classifier acquired through offline training include that according to historical data of tightening processes and corresponding faults of the tightening processes, the historical data of the tightening processes areclustered to form a plurality of data subsets, tags are set to form a tightening technology fault training sample set, and learning training of process curves of tightening technology is performed bythe aid of a machine learning classification algorithm based on the tightening technology fault training sample set to obtain the fault classifier.