Search

It is inevitable to encounter through different formations in the drilling process for deep exploration, and the penetration resistance coefficient (PRC) is an uncertain parameter related to lithology. In this article, a parameter-estimation-based gain-scheduling controller is developed to eliminate undesired system performance deterioration due to the uncertain parameter. First, a drill-string axial finite element model with the uncertain PRC is established, and a control-oriented low-order model is derived via mode selection. A gain-scheduling controller is computed based on the quadratic stability condition of the closed-loop system, which can cope with the system’s parameter uncertainty using variable low-frequency gain. An adaptive observer is designed to estimate the unmeasurable scheduling variable. Field data from a geothermal drilling well is obtained to validate our model. According to this drilling well scenario, both numerical and experimental results illustrate the effectiveness of our method. The explicit relationship between the controller gain and the uncertain parameter is presented. It is found that the performance of the closed-loop system is more sensitive to the controller gain when drilling in soft formations, requiring more attention in such scenarios.

This article presents a phase-compensated equivalent-input-disturbance approach-based repetitive-control system that aims to precisely track periodic signals and effectively suppress aperiodic signals. A generalized modified repetitive control, which completely separates control and learning through two gains, is constructed to achieve independent control-learning regulation without using a 2-D model. On the other hand, the low-pass filter in the equivalent-input-disturbance estimator introduces a phase offset while ensuring causality. Such a phase delay reduces the suppression accuracy of time-varying disturbances. A phase compensator is added to accurately compensate for the phase delay through detailed analysis, thus improving the disturbance-rejection performance. Finally, theoretical analysis and experimental results demonstrate the effectiveness and superiority of the method.

Data-driven fault diagnosis methods have been widely applied at present. In actual processes, there usually exist multiple failure modes; the data frequency spectrum varies in different failure modes, which would bring challenges for feature extraction and subsequent fault diagnosis. In this article, a decision fusion scheme based on the mode decomposition and evidence theory is proposed for fault diagnosis during drilling. The raw data are decomposed into multiple series with different center frequencies, the decomposed series are reconstructed to several groups. For each group, the local diagnosis model is established, thus, several local diagnostic results are obtained. Then, all local diagnostic results are fed into the evidence theory-based decision fusion model. Meanwhile, a confidence matrices-based weight adjustment method is designed to enhance the reliability of fused results. An industrial case study based on the actual drilling data verifies that the proposed method is beneficial to improve the diagnostic effect during drilling.

Rate of penetration (ROP) optimization is critical for improving efficiency in the drilling process. Yet most optimization strategies overlook the effect of drill-string vibration, a significant ROP inhibitor. Moreover, effective methods for recognizing vibration using only surface drilling data are scarce. To address these issues, an intelligent ROP optimization strategy taking vibration mitigation into account is proposed. First, a stacked temporal convolutional network (TCN) is developed to extract the temporal features in multisensor drilling data for vibration recognition. Then, a modified ROP optimization model is developed, integrating a process evaluation index to assess vibration severity. Based on the recognition and evaluation result, an adaptive strategy is devised to constrain the optimization space of operational parameters for mitigating excessive vibration. Finally, sliding time window technique and hybrid bat algorithm (HBA) are applied to optimize the operational parameters in real-time. Experiments on industrial data from an actual drilling field demonstrate the efficiency of the proposed strategy. The vibration model achieves recognition accuracy over 90% and surpasses existing methods. Furthermore, the recommended operational parameters effectively mitigate severe vibrations induced by high weight on bit and low rotational speed, leading to a 27% improvement in ROP compared to manual operation.