To cope with the problems of monitoring dynamic and variable quality variation in multi-type & small batch manufacturing process, diagnosing the abnormal variation and adjusting the process at the right moment, is a difficult problem that a multi-type & small batch enterprise faces in process quality control. A dynamic process quality control mode oriented to networked manufacturing, which integrated quality prevention, analysis, diagnosis and adjustment, and corresponding functional modules and framework were all put forward. This mode dealt mainly with constructing and running dynamic quality control system, integrated several enabling technologies such as the theory of similarity manufacturing, statistical process control (SPC), neural network. Furthermore, some of the key enabling technologies were studied in detail, including process quality analysis on-line based on similarity process, process quality diagnosis based on Elman and expert system of process quality adjustment. It is basis of realizing network, intelligent and automatic process quality control.