Correlational Extremal System for Controlling the Beginning of Faults in Oil Field Equipment by Analyzing their Wattmeter and Dynamometer Charts

Abstract

The creation of technology for the analysis of wattmeter charts to control the main equipment of oil fields has always been of both theoretical and great practical interest. Numerous experiments have shown that the wattmeter chart of their electric motor is a highly noisy signal, and a significant part of the diagnostic information is contained in the estimates of the noise variance and the cross-correlation function between the noise and the useful signal. The lack of technologies for their analysis made it difficult and still makes it difficult to use the wattmeter chart to control the indicated equipment. The paper proposes a technology for controlling the onset of malfunctions and early diagnostics of the technical condition of sucker rod pumping units (SRPU), electric submersible pumps (ESPs), modular cluster pump stations (MCPS) of oil fields. It is shown that during operation, the wattmeter chart of the electric motors of these objects reflects information about the beginning of the latent period of change in their technical condition, which can be used as informative attributes in their diagnosis. A technology is proposed for determining the estimate of the cross-correlation function RXε(μ) between the useful signal X(iΔt) and the noise ε(iΔt) of the wattmeter chart and the technology for generating a set of reference wattmeter charts, in which the obtained informative attributes coincide with the informative attributes of the wattmeter charts of typical malfunctions. It is proposed to solve the problem of identifying faults in the equipment under consideration by the informative attributes of current and reference wattmeter charts of electric motors using correlation extremal systems. It is shown that using the proposed technology it is also possible to control the onset of similar malfunctions of SRPU by analyzing the dynamometer chart. The results of experiments showing the effectiveness and expediency of creating these control systems based on CES are presented.