Search

Copyright of Science Technology & Engineering is the property of Chinese Society of Technology Economics and its content may not be copied or emailed to multiple sites or posted to a listserv without the copyright holder's express written permission. However, users may print, download, or email articles for individual use. This abstract may be abridged. No warranty is given about the accuracy of the copy. Users should refer to the original published version of the material for the full abstract. (Copyright applies to all Abstracts.)

Traditional machine learning methods for mechanical rate of penetration (ROP) prediction are affected by complex feature extraction and imitations of human understanding, which make the prediction accuracy difficult to meet the on-site demand. Based on this, a new ROP prediction method combining feature extraction and regression prediction was proposed. Firstly, data in drilling engineering were pre-processed by box-plot method and one-hot encoding to eliminate abnormal data and to make discrete characteristic continuous. Secondly, convolutional neural network (CNN) was applied to extracting data features, and channel attention mechanism squeeze-and-excitation network (SENet) was introduced to construct a SE-CNN model, which could adjust the importance of CNN feature channel. Finally, SE-CNN model was compared with CNN model. The results show that the goodness of fit of SE-CNN model is increased by 2. 1%, the mean absolute error and the root mean squared error are decreased by 1. 1% and 1. 5% . SE-CNN model has a better prediction accuracy, and can forecast the ROP of drilling field and provides a scientific reference for increasing the ROP during drilling. [ABSTRACT FROM AUTHOR]

Copyright of Science Technology & Engineering is the property of Chinese Society of Technology Economics and its content may not be copied or emailed to multiple sites or posted to a listserv without the copyright holder's express written permission. However, users may print, download, or email articles for individual use. This abstract may be abridged. No warranty is given about the accuracy of the copy. Users should refer to the original published version of the material for the full abstract. (Copyright applies to all Abstracts.)

Aiming at the problems of low efficiency, low precision and resource loss in the fault diagnosis of ground-driven screw pump, a diagnosis method of back propagation(BP) neural network optimized by introducing the idea of power spectrum refinement was proposed to improve wavelet packet transform and cuckoo search(CS). Firstly, the active power of the screw pump was decomposed and reconstructed by the improved wavelet transform to obtain the feature vector. Secondly, it was normalized with the parameters such as instantaneous flow and inlet back pressure as the output layer information of the BP neural network. Thirdly, the weights and thresholds of BP neural network were obtained by CS, and the CS-BP fault diagnosis model was established. Finally, it was applied to the diagnosis of different fault types of screw pump, and the diagnosis effect was analyzed and compared with the current mainstream diagnosis methods. The results show that the average accuracy of fault diagnosis for different types of screw pump is 95. 6%, and the feasibility and superiority of the proposed method are proved by comparative analysis. [ABSTRACT FROM AUTHOR]

At present, most of the oilfield centrifugal pump injection stations use traditional manual inspection and other methods for anomaly detection, which wastes a lot of resources and has low detection accuracy. To address this problem, a genetic algorithm optimized one-class support vector machine (GA-OC-SVM) based anomaly detection method was proposed for water injection stations. Firstly, the data of water injection station were standardized, normalized and feature extracted. Secondly, genetic algorithm was used to find the best individual value of the population as a parameter of single class support vector machine, and the detection model was established. Finally, GA-OC-SVM algorithm was compared with isolated forest algorithm, local outlier factor algorithm and other mainstream methods for anomaly detection of test data sets, and the accuracy of the algorithm was analyzed. The receiver operating characteristics (ROC) curve was used for model evaluation. The results show that the proposed GA-OC-SVM algorithm is superior, with a detection accuracy of 99%, and can save a lot of human and material resources at the same time. [ABSTRACT FROM AUTHOR]