Search

The dynamometer card (DC) combined with computer-aided intelligent diagnosis technology (pattern recognition) is an essential and useful means to monitor and diagnose working conditions of oil wells. However, the difficulty to collect the faulty DC album and the property differences between wells results in the decrease of diagnosis accuracy. In this paper, a novel approach regarding generating DCs is proposed based on the analysis of the mechanism of a sucker rod pump at normal and several faulty scenarios. According to the simulating DCs and Back Propagation Neural Network (BPNN), a self-evolution diagnosis method is presented to obtain higher diagnosis accuracy. The results of field application show that the average accuracy of the self evolution diagnosis method is improved from 84.6% to 93.1% compared with the original diagnosis method. And the new method is more effective for oil wells with high pumping speed or containing flexible sucker rods.

A carbon/glass hybrid fiber continuous sucker rod (CR) with lower specific gravity, better corrosion resistance and higher strength was developed to replace the steel rod in the application of underground oil extraction. To meet the actual needs of oil fields, the computer-aided design techniques based on mathematical optimization model are used to help engineers in optimizing the CR pumping system. Because the performance of CR pumping system, including the energy consumption and the service life of sucker rod string (SRS) is the complex function related to the system parameters, and the constraints are nonlinear, the enumeration method is used to solve the optimization problem. Moreover, the existing methods for predicting the SRS behavior, including wave equation and finite element analysis (FEA), require iterative calculation, which may result in a low prediction efficiency. Hence, the huge optimization space, i.e. the set of all possible CR pumping systems, will lead to a very long optimization process. Currently, the typical solution for improving the optimization efficiency is to greatly reduce the optimization space based on the experience of steel sucker rod application in the oilfield. However, due to the difference of properties between CR and steel sucker rod, the CR pumping system designed by the current method cannot achieve the best performance. In order to address this tough issue, an equivalent vibration model (EVM) of SRS is established to quickly predict the plunger displacement and the polished rod load. The new model is verified by comparing it with experimental results. In addition, to formulate the design criteria as the theoretical basis for reducing the optimization space of CR pumping system, the influence mechanism of system parameters on the plunger motion is analyzed by decomposing the motion of the plunger into forced vibration and intermittent free vibration. The results show that due to the amplification effect of the forced vibration and the plunger being not in the balanced position when the intermittent free vibration is terminated, the plunger of the CR pumping system has obvious overstroke. Besides, the long stroke of polished rod, medium pumping speed and concentrated weighting of CR is favorable for the CR pumping system to obtain high plunger overstroke. According to the overstroke analysis results and previous experience, three design criteria are formulated. Then, based on the formulated design criteria and EVM, an efficient optimization design method of CR pumping system is proposed. At last, the proposed method is applied to production oil wells in oilfield. The obtained results demonstrate the effectiveness and efficiency of proposed method in reducing the energy consumption and improving the SRS service life of CR pumping system.