Method for inverting ground power diagram from electric power diagram of oil pumping unit

Abstract

The invention discloses a method for inverting a ground power diagram from an electric power diagram of an oil pumping unit. The method comprises the following steps of: constructing a BP neural network model, and inputting training samples, wherein input parameters in the training samples comprise periodic curve data of the electric power diagram, parameters of various types of motors, and parameters of various types of oil pumping units, output parameters in the training samples are ground power diagram loads, classifying the oil pumping units and the motors according to the sensitive parameters of the oil pumping units and the motors, and forming the training samples by using the input parameters in different composite modes according to classification results; training the training samples in the BP neural network model, so that the ground power diagram can be inverted from the electric power diagram of the oil pumping unit; and inverting the collected periodic curve data of the electric power diagram by using the trained BP neural network model to obtain the ground power diagram. According to the method, different types of motor parameters and oil pumping unit parameters are added on the basis of original electric parameter data, the multi-parameter learning function is realized, and the accuracy of the inverted ground power diagram is relatively high.

Description

technical field [0001] The invention relates to a method for diagnosing working conditions of pumping wells. More specifically, the present invention relates to a method for inverting the surface power map from the electric power map of the pumping unit. Background technique [0002] In recent years, the international oil price has dropped sharply. At the same time, most of the newly discovered blocks in oilfields are low-permeability, heavy oil or high-condensation oil, which is difficult to mine and high in development costs. In the context of the overall decline in oilfield profits, oilfield companies generally implement the "retire from five to one" personnel management system, and there is an urgent need for low-cost IoT technology to replace traditional labor-intensive work and reduce labor intensity in oil production. Single well IoT equipment mainly includes surface power map instrument, electric parameter module, etc. The emergency analysis work mainly depends on ...

AI Technical Summary

Problems solved by technology

However, this method ignores some actual physical models in the inversion process, and the accuracy and scope of application are greatly limited.

If a unified training model and training results are used to invert the ground power map, although the scope of application of the model is enhanced, the pertinence of the model is also reduced, thereby reducing the accuracy of the inversion results

Images