Rough set based intelligence diagnostic system for valves in reciprocating pumps

The paper presents a novel approach to fault diagnosis of valves in three-cylinder reciprocating pumps. Since the vibration signals collected from pumps apparently show the existence of nonstationary signals and the interference of neighboring valves, the wavelet packet transform is introduced as a preprocessing means of extracting time-frequency information from vibration signals to obtain the fault characteristics of the valves. Furthermore, to reduce the dimensions of the character vectors and extract diagnosis rules of the faulty valves, a rough set based intelligence diagnostic system is constructed and used in the valve faults diagnosis. It is proved that fault types and positions can be identified and diagnosed by the above method