Recently, the team of Zhang Kai has made new progress in the field of smart wellbore diagnosis and optimization., and the related research has been published in the SPE journal. The paper is titled " Imbalanced Working States Recognition of Sucker Rod Well Dynamometer Cards Based on Data Generation and Diversity Augmentation ".
Innovation:The existing research of dynamometer cards diagnosis mostly focuses on improving and optimizing classification models, with little attention paid to improving the quality of the data itself. The actual dynamometer cards often have the problem of significant differences in the amount of data under different working conditions. When using this type of data for diagnostic model training, the model will focus on learning the categories with the larger amount of samples and seriously ignore those uncommon but need to be recognized categories. This study addresses the aforementioned challenges by using conditional generative adversarial neural networks for data generation of subcategory working conditions. Considering the insufficient diversity of generated samples, a mini batch strategy is adopted to interact sample information during the training process of the discriminator, further strengthening the adversarial learning process between the generator and discriminator, and achieving diversity enhancement of generated samples. The results section of the article provides a detailed analysis of the performance of the generative model, sample diversity, and classification model. The calculation results show that this method can significantly and stably improve the recognition accuracy of small categories working conditions.
Abstract: During sucker rod pump production, there is a commonly seen problem of class imbalance, which refers to the differences in the amount of data accumulated under different working conditions. This problem in rod pump diagnosis can lead to unsatisfactory classification results of surface dynamometer cards under working conditions with fewer samples. Therefore, this study adopts the conditional generative adversarial nets (CGANs) improved by mini-batch method to address the problem of class imbalance. CGAN is an efficient method of multiclass data generation, which learns the properties of dynamometer cards by training the generator and discriminator networks. CGAN is modified by mini-batch strategy to avoid mode collapse and enable the interaction among input samples of discriminator, so that the generated dynamometer cards can be much more diverse. Results show that the shapes of generated dynamometer cards are basically consistent with those of real samples, and the structural similarity (SSIM) among the generated samples decreases, indicating that the generated dynamometer cards have more types of shape. Meanwhile, as the generated dynamometer cards become more diverse, their differences with real samples in data distribution are reduced, according to the calculation of sliced Wasserstein (SW) distance. Based on real and generated dynamometer cards, we developed the classifiers for working condition diagnosis of rod pump through convolutional neural network (CNN). The classification results of the validation set indicate that without the mini-batch method, the recall of generated categories for pump hitting down and leakage has increased by 12 and 5.3%, respectively; in contrast, with the mini-batch method, the recall has increased more obviously by 7, 24, and 2%, respectively, for gas lock, pump hitting down, and leakage. Our research results have demonstrated that the proposed method can effectively solve the problem of insufficient data accumulation in the oil field.
The SPE Journal covers new theories and concepts in various aspects of oil and gas exploration and production engineering, including drilling and completion, geomechanics, production and facilities, oilfield chemistry, CO2 storage and injection, reservoir assessment and engineering, numerical simulation, data analysis, economics and externalities, including health, safety, environment, and sustainability. The journal has an impact factor of 3.602 in 2023, with JCR divided into Q1 zone and Chinese Academy of Sciences engineering technology category 3 zone.
Paper link:
https://doi.org/10.2118/214661-PA
Citation:
Yin, C., Zhang, K., Zhang, L. et al. 2023. Imbalanced Working States Recognition of Sucker Rod Well Dynamometer Cards Based on Data Generation and Diversity Augmentation. SPE J. SPE-214661-PA. https://doi.org/10.2118/214661-PA.
