In the budgeting of all major development projects, there is always a desire to capture the financial and operational opportunities of all new and existing technologies; however, this can be difficult if not proven to be viable in advance of the development campaign commencing. This leads to great difficulty in fully capturing potential savings in future financial planning. In a cost-sensitive market, a leading East Africa exploration and production operator recognized an opportunity to trial new technologies in the exploration and appraisal campaign phase in order to reduce well cost and risk, which could be directly translated to the development campaign feasibility model. A flat-time analysis was performed on historical data to benchmark the connection performance against relevant proxies. From this analysis, it was determined that there was room for improvement from both technology and existing practices. To improve the connection time on a technology basis, offset field resistivity data were modelled to determine the feasibility of the formations drilled to efficiently propagate a bi-directional electromagnetic signal. Once the modelled feasibility was deemed acceptable to deploy from a risk perspective, a systematic field-trial plan was developed to deliver proof of concept, which was followed by the second element, running the system to reduce connection time. After two successful proof-of-concept runs, the electromagnetic system facilitated a material reduction in connection time, which could then be applied to the development project economics. Whilst primarily focusing on technology-oriented connection improvements, there was also a systematic performance improvement from the human element on the rig floor owing to the performance initiative. The secondary benefit of successfully implementing the electromagnetic telemetry system was the increased data rate and the ability to transmit annular pressure data while the pumps were off, which provides valuable data to understand wellbore hydraulic behavior during pumps-off events. With conventional mud-pulse telemetry systems, the critical path is impeded to obtain these measurements, where annular pressure data is streamed to surface after the surface event (e.g., LOT / FIT, connection ballooning check, etc.). During these field trials, the downhole equipment complexity run in conjunction with the electromagnetic telemetry transitioned from basic gamma ray and pressure measurements to a quad-combo LWD string run in conjunction with a rotary steerable system.
We tackle the challenge of Visual Question Answering in multi-image setting for the ISVQA dataset. Traditional VQA tasks have focused on a single-image setting where the target answer is generated from a single image. Image set VQA, however, comprises of a set of images and requires finding connection between images, relate the objects across images based on these connections and generate a unified answer. In this report, we work with 4 approaches in a bid to improve the performance on the task. We analyse and compare our results with three baseline models -LXMERT, HME-VideoQA and VisualBERT -and show that our approaches can provide a slight improvement over the baselines. In specific, we try to improve on the spatial awareness of the model and help the model identify color using enhanced pre-training, reduce language dependence using adversarial regularization, and improve counting using regression loss and graph based deduplication. We further delve into an in-depth analysis on the language bias in the ISVQA dataset and show how models trained on ISVQA implicitly learn to associate language more strongly with the final answer.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.
customersupport@researchsolutions.com
10624 S. Eastern Ave., Ste. A-614
Henderson, NV 89052, USA
This site is protected by reCAPTCHA and the Google Privacy Policy and Terms of Service apply.
Copyright © 2025 scite LLC. All rights reserved.
Made with 💙 for researchers
Part of the Research Solutions Family.