Ant Colony Clustering Procedure Used in Vibration Fault Diagnosis of Beam Pumping Unit

The study of vibration fault diagnosis of beam pumping unit is an important task to reasonable production, scientific management and improvement of the oil recovery. The vibration fault diagnosis is a good method with direct result. In view of measurement data during operation of a beam pumping unit, to extract characteristic attribute value of vibration frequency under different conditions. Firstly, clustering procedure and classification based on the distance of some common faults similarity relation are used to establish beam pumping unit failure classification system to avoid the subjective factors. Then ant colony algorithm is applied in the fault type system of beam pumping unit and fault diagnosis model is established by training fault data sample. Experimental results show that the combination of two mathematical methods can diagnose the fault of beam pumping unit with accurate results. It is a good diagnosis to diagnose beam pumping unit fault.