Vibration state forecast for electrical submersible pump unit based on dynamic gray model

This paper discussed the fundamental contents and the process developed of the GM (1,1) forecast and the dynamic gray forecast model, and predicted the vibration state of electrical submersible pump through the above-mentioned two models in the electrical submersible pump unit of oilfield. Vibration state was described by the vibration acceleration signals. The result showed that the precision of tow models was greater than 98%, prediction error was less than 5%, and the two kinds of gray forecast model were successfully applied in vibration state forecast for electrical submersible pump. The effect of dynamic gray forecast model was better than the GM (1,1) forecast model because that it added the new information, increased the white grade of gray interzone, and had the tracking feature. This gray forecast model is simple, high precision and easily calculated in computer.