Pumping unit abnormal working condition monitoring method

Abstract

The invention relates to a pumping unit abnormal working condition monitoring method. The pumping unit abnormal working condition monitoring method includes the steps that indicator diagram data of apumping unit under normal operating conditions are collected as training data, after the training data are normalized, a KPCA method is used for extracting the non-linear feature from the pumping unittraining data as the core component, a first derivative of a probability density function of the core component is considered and combined with KLD to obtain the trained KLD component, and based on the trained KLD component, statistics are calculated and corresponding control limits are determined; and indicator diagram data of the pumping unit under failure conditions are collected as testing data, KPCA is used for extracting the corresponding core component, the online KLD component of the testing data is further calculated, based on the online KLD component, the statistics are calculated,and the control limits are used for monitoring. According to the pumping unit abnormal working condition monitoring method, the KLD is introduced into a traditional KPCA method, and probability information contained in pumping unit process data is fully used to improve the monitoring ability to the pumping unit abnormal working conditions.

Description

technical field [0001] The invention belongs to the technical field of working condition monitoring of industrial systems, and in particular relates to a method for monitoring abnormal working conditions of an oil pumping unit. Background technique [0002] The pumping unit is the main device for oil extraction and plays an important role in the oil extraction industry. Since most pumping units operate in harsh environments, there are usually different failures that will lead to a deterioration in the safety of the pumping unit and a reduction in production profits. Therefore, it is necessary to monitor the pumping unit in real time to ensure the safe and stable operation of the oil extraction process. In the field of pumping unit fault diagnosis, the dynamometer diagram is the most commonly used tool. Based on the dynamometer diagram, some methods for pumping unit fault diagnosis have been developed and successfully applied, but these methods are essentially supervised lea...

AI Technical Summary

Problems solved by technology

[0004] The present invention provides a method for monitoring abnormal working conditions of pumping units, aiming at the problems of ignoring the probability information in the process information in the traditional KPCA method, resulting in low fault detection performance.

Images