Stick-slip vibration has been a major problem in drilling operations in B basin, Chad, where rocks in basement formation are hard granite. Premature failure of drill bits and string has profoundly undermined the safe and economical drilling operations. Hence, it is urgent to reveal the mechanism of stick-slip vibration and find a way to eliminate or relieve stick-slip problem. Through the numerical and experimental study on the effects of the cutting parameters on cutting performance, root cause on depth of cut (DOC) or cutting depth is revealed as the mechanism of stick-slip vibration, optimized cutting depths for different rock layers are recommended as well. With it, a novel design of auto-response DOC controller is proposed and those structural parameters are modified and via numerical simulation research. After machining and testing, the structure of the auto-response DOC controller is preliminary determined and it has acquired the satisfactory motion performance and met the testing standards as it can slowly retract into the body by resisting the load and quickly extend outward the body. The auto-response DOC controller and the anti-stick-slip PDC bit have profound application potential in developing the ultra-deep and unconventional reservoirs, especially in this era shocked by low oil price and the pandemic.

No abstract

No abstract

Traditional sequence tagging methods for named entity recognition (NER) face challenges when handling nested entities, where an entity is nested in another. Most previous methods for nested NER ignore the effect of entity boundary information or type information. Considering that entity boundary information and type information can be utilized to improve the performance of boundary detection, we propose a nested NER model with a multi-agent communication module. The type tagger and boundary tagger in the multi-agent communication module iteratively utilize the information from each other, which improves the boundary detection and the final performance of nested NER. Empirical experiments conducted on two nested NER datasets show the effectiveness of our model.