研发/创新

研发技术部分问答:钻井自动化——人员、流程和智能钻机的兴起

本文是SPE研发技术部门关于新兴能源技术的系列问答的第四篇。在这篇文章中,NOV的首席技术官兼首席营销官David Reid探讨了自动化钻井系统的演变和现状。

JPT_2025-09_RD_GI_1200780388.jpg
资料来源:Getty Images。

在本文中,SPE 研究与开发技术部门 (RDTS)的 Gaurav Agrawal 与全球油田服务提供商和技术开发商 NOV 的David Reid进行了交谈。

本系列聚焦塑造能源未来的创新理念和分析,重点关注新兴技术及其路线图、潜力和影响。我们希望通过这些对话,激发对话,加速新能源领域的发展。

David Reid 是 NOV 的首席技术官 (CTO) 兼首席营销官 (CMO),在推动公司全球创新和品牌战略方面发挥着关键作用。Reid 在能源领域拥有超过 30 年的职业生涯,在推进 NOV 的技术组合以及使公司处于行业转型前沿方面发挥了重要作用。

作为思想领袖,他领导了众多专注于集成、数字化、自动化和可持续能源解决方案的举措,确保NOV始终保持能源技术的先锋地位。他的专业知识涵盖产品开发、企业战略和企业故事讲述,使复杂的工业挑战更容易被不同的受众理解。

除了在NOV的职位之外,Reid还积极投身于人道主义事业。他积极参与非营利组织redM,该组织致力于通过提高公众意识和动员行动来打击人口贩运。他在redM的领导地位体现了他对社会影响力的热情,他利用技术和故事讲述来激发变革。Reid经常在全球会议上发表演讲,他对能源、技术和赋能人类的交汇有​​着独特的见解。他持续推动创新,同时倡导一个兼具技术先进性和社会责任感的未来。

RDTS:您在能源行业拥有超过30年的经验,拥有独特的视角,见证了各种技术路线图的演变,同时也作为技术推动者推动着变革。自动化发展对石油和天然气应用有何影响?

Reid:石油天然气行业(尤其是钻井行业)的自动化发展历程漫长而引人入胜。它始于1965年,当时Smedvig(现Seadrill)发表了一篇富有远见的SPE论文,提出通过机械化将人员从钻井平台上移除,从而提高安全性和速度。虽然最终的实现耗时更长,但这一愿景激发了数十年的创新。我们从BJ Machinery的V型钻井机的简单机械化发展到如今高度智能的系统,这些系统在安全性、效率、能力和速度方面都超越了人工操作。

如今,我们开始看到工业机器人接管所有剩余的钻井平台任务,以及能够从深入油井的实时数据中学习的系统。自动化已从理论变为现实——尤其是在深水钻井或低成本重复性页岩钻井等严苛、高成本的环境中——它正在改变我们设计、操作和维护钻井平台的方式。

RTDS:定向钻井在30年前相对较新。如今,钻井技术和钻机的功能和复杂性都得到了显著提升。您能否重点介绍一下钻机设计和技术方面有哪些重大变革?

里德:最大的转变之一始于20世纪90年代,尤其是在挪威,当时的管道处理系统从管道甲板到地面都实现了全面机械化,可以处理直径达20英寸(约51厘米)的管材。这具有开创性的意义,并成为现代机械化钻机系统的蓝图。

20世纪90年代末的深水繁荣催生了更多创新——由于海上管道运输量巨大,手动操作已不再可行。大约在同一时期,高科技控制系统应运而生,其中包括Hitec公司(现为NOV旗下公司)于1994年在斯塔万格举行的北海海上会议上首次演示的远程功能。这又是一个重要时刻,因为它预示着未来远程管理作业的时代。

1998 年,随着绞车上板盘式制动器控制的出现和普及以及交流电机在系统中的使用,钻柱控制取得了重大进步,实现了基于最佳参数调整的输入精细自动化钻井控制。

定向自动化的下一步始于对粘滑等井况的功能失调响应,然后发展为自动化泥浆脉冲管理,允许在钻机外管理反馈和输入。

如今,远程定向已成为一种可行的规范,定向自动化的最终工具正在现场开发和测试中。20多年来,有线钻杆宽带井下连接的引入不断发展,为地面、井筒和管柱测量之间的数据流提供了极致的传输。

如今,钻机上的大多数功能都已实现自动化,控压钻井、完井和电缆作业都开始展现出自动化流程和控制的迹象。钻机是一个集成平台,融合了数字控制系统、网络智能以及日益自主的机器人和井下指令,人工智能 (AI) 预测技术也正在发展成为自主控制系统和全流程自动化。在 NOV,我们的测试钻机已经运行了几年,完全没有在钻井平台上设置司钻舱,清理平台也变得可行。

RTDS:这种自动化对人员配备、绩效、安全和其他类似的 KPI 有何影响?

Reid:这种转变意义深远。起初,人们以为自动化只会让事情变得更快、更便宜,但它真正彻底改变的是安全性和可预测性。事实上,速度和成本之间出现了矛盾,但自动化在许多应用中的价值仍然促进了人们的采用,而加快生产速度并减少人员流失是一项非常现实的举措。

自动化管道处理和机械化钻台将人员从高风险区域转移。我们还消除了许多性能不一致性——自动化不会造成疲劳或分心。从关键绩效指标 (KPI) 的角度来看,我们看到了更长的正常运行时间、更一致的钻井参数、更少的工伤事故和更高的井质量。钻机更加智能,而且每年都在不断改进。减少人员配置和降低成本,真正见证了自动化技术重新成为行业关注的焦点,因为定制化程度的降低和合理使用工业机器人的增多,使得自动化技术重新成为行业关注的焦点。

RTDS:物料搬运长期以来一直是钻井平台上风险最高的作业之一。自动化如何提升这一领域的安全性和效率?该领域的下一步发展方向是什么?

里德:物料搬运是最先受到关注的领域之一,尤其是在 20 世纪 90 年代中期的挪威,他们实现了大型管道搬运的全面机械化。

这改变了一切。手工操作重型管材曾是钻井作业中最危险的部分之一。如今,机器人和液压臂可以精准、快速、安全地完成这项工作。我们现在正在进入预测系统,传感器和人工智能可以预测运动,自动存储和检索,甚至在不安全行为发生之前就发出警报。我们面临着一个快速发展的前沿领域。

RDTS:控制和可视化是管理日益复杂的钻机系统的关键。该领域的进步如何增强操作员的能力并提高钻机性能?

里德: 1994 年,我们在斯塔万格公开展示了远程控制系统,但直到最近我们才开始意识到它的潜力。

这一飞跃源于将井下数据与地面自动化实时连接,因此系统现在可以预测和预防故障,而不是对故障做出反应。

如今,可视化工具提供分层诊断、直观的控制界面和预测性洞察。我们已经从操作员驱动的钻机发展到系统辅助的钻机,现在我们正朝着操作员监督的系统驱动钻机迈进。

RDTS:NOV 拥有广泛而多样化的制造业务。玻璃纤维部门与其他石油和天然气垂直行业截然不同。机器人和自动化技术对你们的制造业务有何影响?

里德:我们大多数制造工序的材料和工艺都要求严格的公差。当我们生产长条玻璃纤维卷管时,自动化至关重要。

就像钻机一样,高度重复的任务也得以机械化和自动化。机器人技术使我们能够自动化重复性、高精度的任务,从而在许多制造工厂中实现了质量的稳定,并减少了浪费。

我们确实拥有一套高度定制的非金属成型系统,该系统将车间起重机改造成一台用于切割大型构件的精密切割机。该系统已用于快速建造大型、易于安装且重量轻的铁路桥梁。

在我们的制造网络中,自动化意味着更快的产能提升、更好的可重复性以及更高的工人安全性。而且,自动化不仅仅局限于机器,它还利用数据来优化从原材料到最终检验的整个制造链。我们拥有一项商业业务,帮助非NOV工厂利用数字化智能系统提升绩效。

我们经常在可以复制自动化流程并可控制成本的国家和地区开展业务,从而使全球供应网络非常可行,靠近我们所服务的市场。

RDTS:您认为软件开发是否跟上了机器人或其他机械发展的步伐?实现自动化软件开发面临哪些独特的挑战?

Reid:软件常常落后于机械创新,尤其是因为它在出现故障之前是隐形的。真正的挑战在于互操作性——将遗留系统与现代平台集成。而且,由于软件与实时操作交互,它必须坚如磐石。我们现在在设计软件时会考虑模块化,使其更容易适应和发展。但我们仍然需要整个行业建立更多协作的生态系统,才能真正释放自动化的全部潜力。

在NOV内部,我们的Max平台促进了我们庞大的收购公司生态系统中的每个系统进行通信和数据共享。我们的秘诀在于翻译多种形式的代码,而不是要求每个人都将其代码调整为一种标准。这项工作进展顺利,我们拥有一个高度互操作的数据系统和一个强大的边缘计算基础,它以类似远程云的方式运行。

RDTS:为了推进自动化并保持领先优势,变革管理和员工技能再培训面临哪些挑战?NOV 是如何应对这些挑战的?又从中获得了哪些经验教训?

里德:这不仅仅关乎技术,更关乎人。我们认识到,早期参与至关重要。

员工需要看到技术如何帮助他们,而不是取代他们。我们投资了培训项目、基于模拟器的学习和数字学院。关键的经验是什么?当你邀请员工参与到这个过程中时,他们会与你一起创新。我们的一线团队参与塑造了许多最成功的工具。

RDTS:您在工作中多久使用一次 ChatGPT 或 Perplexity 等 AI 工具?

Reid:我们在数据科学领域运用人工智能已有20多年。最新一代的生成式人工智能为每个人带来了无限机遇。我们从公司各部门抽取了一批员工作为样本,并测试了一些系统。

虽然并非所有员工都已做好迎接我们所需挑战的准备,但我们找到了一条途径,许多员工反映,在工作执行过程中大幅降低了成本并节省了开支。我们现已实施生成式人工智能培训,并将其应用于所有员工。

我的团队经常使用人工智能作为研究、构思甚至战略规划的加速器。这些工具让我们能够快速测试想法,在复杂的主题中寻找规律,并快速生成内容。它们不会取代批判性思维,但它们正在扩展我们一天中可以完成的工作。

RDTS:我们很乐意听到您对人工智能如何影响自动化路线图的看法。

里德:人工智能就像我们一直在等待的自动化神经系统。我们花了数年时间打造这些肌肉——机器人、控制系统、流体和气体处理以及联网系统。

如今,人工智能将这些系统连接起来,实时做出决策。路线图将从基于规则的自动化转向自适应系统,该系统能够随着每项作业进行学习和发展。人工智能将有助于消除停机时间,以人类无法做到的方式优化钻井、完井和加工,并使远程操作不仅成为可能,而且更受欢迎。

RDTS:总体而言,人工智能将如何影响石油和天然气领域的创新速度?

Reid:已经是了。我们能够更快地制作原型,在实际建造之前模拟整个操作过程,并在风险因素实际出现之前就发现它们。

人工智能拉平了创新曲线,缩短了从创意到落地的时间。在石油天然气这样的复杂行业,速度决定了领先与落后。

RDTS:我认为,与油气行业以外的技术开发商合作对于保持成本效益和加速商业化至关重要。我们的外部合作流程效率如何?我们可以采取哪些措施来进一步改进?

里德:我们已经取得了巨大的进步,但我们仍然可以做得更好。最好的突破往往来自我们自身的领域之外——无论是汽车领域的机器人技术、科技领域的人工智能,还是航空航天领域的材料。

我们的经验是,成功取决于信任和一致的激励机制。为了更进一步,我们需要共享的创新空间、更灵活的知识产权框架以及开放的文化——即使并非总是率先找到答案。

RDTS:您对当今进入能源行业的年轻专业人士有什么建议?

里德:要成为一座桥梁。能源行业是实体经济与数字可能性的融合。

学习设备的工作原理,同时也要理解系统思维、数据和软件。你的价值在于连接这些世界。不要只关注行业现状,而要关注行业的发展方向。这个行业正处于重塑模式,现在正是参与其中的最佳时机。

RDTS:除了担任首席技术官和首席营销官的双重身份外,您还积极投身慈善事业。您创立redM是为了致力于一项独特的人道主义事业。您能否详细介绍一下redM的使命,以及其他人如何参与支持这项事业?

里德: redM 的初衷是希望用希望和勇气去面对人类最黑暗的问题之一——人口贩卖。我们的目标是在不传递恐惧信息的情况下,提升全球意识。

我们希望人们感到力量的增强,而不是不知所措。通过教育、宣传和社区动员,我们正在帮助人们识别和应对人口贩运。任何人都可以通过捐款、志愿服务或分享信息来参与。访问joinredm.com了解更多信息,并在社交媒体上关注 redM,加入这场运动。携手共进,我们真的可以改变世界。

阅读本系列的第三篇文章,本文还与 SLB 软件技术与创新中心的高级机器学习工程师 Zikri Bayraktar 一起探讨了人工智能在上游行业的广泛应用。

原文链接/JPT
R&D/innovation

R&D Technical Section Q&A: Drilling Automation—People, Process, and the Rise of Smart Rigs

This article is the fourth in a Q&A series from the SPE Research and Development Technical Section focusing on emerging energy technologies. In this piece, David Reid, the CTO and CMO for NOV, discusses the evolution and current state of automated drilling systems.

JPT_2025-09_RD_GI_1200780388.jpg
Source: Getty Images.

In this article, Gaurav Agrawal of the SPE Research and Development Technical Section (RDTS) spoke with David Reid of NOV, a global oilfield services provider and technology developer.

This series highlights innovative ideas and analysis shaping the future of energy, with a focus on emerging technologies and their roadmaps, potential, and impact. With these conversations, we hope to inspire dialogue and accelerate progress across new energy frontiers.

David Reid is the chief technology officer (CTO) and chief marketing officer (CMO) of NOV, where he plays a key role in driving the company’s global innovation and brand strategy. With a career spanning over 3 decades in the energy sector, Reid has been instrumental in advancing NOV’s technology portfolio and positioning the company at the forefront of industry transformation.

As a thought leader, he has led numerous initiatives focused on integration, digitalization, automation, and sustainable energy solutions, ensuring NOV remains a pioneer in energy technology. His expertise spans product development, corporate strategy, and storytelling, making complex industrial challenges more accessible to diverse audiences.

Beyond his role at NOV, Reid is deeply committed to humanitarian causes. He is actively involved in redM, a nonprofit organization dedicated to combating human trafficking by raising awareness and mobilizing action. His leadership in redM reflects his passion for social impact, leveraging technology and storytelling to inspire change. A frequent speaker at global conferences, Reid brings a unique perspective on the intersection of energy, technology, and human empowerment. He continues to drive innovation while advocating for a future that is both technologically advanced and socially responsible.

RDTS: You have a unique perspective with over 30 years in the energy industry and have been a witness to various technology roadmaps unfolding while also driving change as a technology agent yourself. How has the automation evolution impacted oil and gas applications?

Reid: The evolution of automation in oil and gas—particularly in drilling—has been a long and fascinating journey. It started as far back as 1965, with a visionary SPE paper from Smedvig (now Seadrill) that proposed mechanization to remove people from the rig floor to increase safety and speed. The reality took longer to deliver, but that vision sparked decades of innovation. We went from simple mechanization with BJ Machinery’s Type V racking machines to highly intelligent systems that now outperform manual operations in safety, efficiency, capability, and speed.

Today, we’re starting to see industrial robots take over all remaining rig floor tasks and systems that can learn from real-time data that reaches into the well itself. Automation has moved from theory to necessity—especially in harsh, high-cost environments like deepwater or low-cost repetitive shale drilling—changing how we design, operate, and maintain rigs.

RTDS: Directional drilling was relatively new 30 years ago. Both drilling and the drilling rig have improved considerably in functionality and complexity. Could you highlight the step changes in the rig design and technology?

Reid: One of the biggest shifts started in the 1990s, particularly in Norway, where pipe-handling systems were fully mechanized from the pipe deck into the floor to manage tubulars up to 20-in.‑diameter. That was groundbreaking and became the blueprint for modern mechanized rig systems.

The late 1990s deepwater boom demanded even more innovation—manual handling wasn’t an option with the volumes of pipe we were moving offshore. Around the same time, high-tech control systems were introduced, including remote capabilities first demonstrated by Hitec (now part of NOV) at the Offshore Northern Seas Conference in Stavanger in 1994. That was another major moment as it hinted at a future where operations could be managed remotely.

In 1998, a big step in drillstring control emerged with fine automated drilling control of the inputs based on optimal parameter adjustment in concert with the emergence and proliferation of plate disk brake control on drawworks and the use of AC motors in the system.

The next step in directional automation began with dysfunctional responses to well conditions like stick/slip and then evolved to automated mud pulse management, allowing the feedback and input to be managed off the rig.

Today, remote directional is a feasible norm, and the final tools for directional automation are being developed and tested in the field. The introduction of broadband downhole connectivity with wired drillpipe has been growing over 20 years, providing a nirvana of data flow between the surface, complete wellbore, and string measurements.

Now you see most functions on a rig being automated with managed pressure drilling, completions, and wireline all starting to show signs of automation process and controls. Rigs are integrated platforms of digital control systems, networked intelligence, and increasingly autonomous robots and downhole commands, with artificial intelligence (AI) prediction being advanced to autonomous systems of control and full process automation. At NOV, we have now operated our test rig for a few years without a driller’s cabin on the floor at all, and clearing the floor is becoming feasible.

RTDS: How has this automation impacted the staffing, performance, safety, and other similar KPIs?

Reid: The shift has been profound. Initially, people thought automation would just make things faster and cheaper, but what it truly revolutionized was safety and predictability. In reality, speed and cost took the wrong turn, but the value in many applications still facilitated adoption, and the speed and removal of people is a very current activity.

Automated pipe handling and mechanized drill floors removed people from high-risk zones. We also eliminated a lot of performance inconsistency—automation doesn’t fatigue or get distracted. From a key performance indicator standpoint, we’ve seen stronger uptime, more consistent drilling parameters, fewer injuries, and higher well quality. The rigs are smarter, and they’re getting smarter every year. Removing personnel and reducing the price point is really seeing a return to the forefront with the implementation of less customization and more use of justifiable industrial robotics.

RTDS: Material handling has long been one of the highest-risk activities on a rig. How has automation in this sub-area improved safety and efficiency, and what’s next for this space?

Reid: Material handling was one of the first areas to get attention, especially in Norway in the mid‑1990s, where they fully mechanized large pipe handling.

That changed everything. Manual work with heavy tubulars was one of the riskiest parts of rig operations. Today, robots and hydraulic arms do that job with precision, speed, and safety. We’re now moving into predictive systems where sensors and AI can anticipate movement, automate storage and retrieval, and even flag unsafe behavior before it happens. We have a rapidly evolving frontier on our hands.

RDTS: Controls and visualization are key to managing increasingly complex rig systems. How have advances in this area empowered operators and improved rig performance?

Reid: In 1994, we saw remote control systems demonstrated publicly in Stavanger, but it’s only recently that we’ve begun realizing that potential.

The leap has come from connecting downhole data with surface automation in real time, so instead of reacting to dysfunction, systems now can predict and prevent it.

Visualization tools today offer layered diagnostics, intuitive control interfaces, and predictive insights. We’ve gone from operator-driven rigs to system-assisted, and now we’re heading toward system-driven rigs with operator oversight.

RDTS: NOV has a broad and diverse manufacturing footprint. The fiberglass division is distinctive from other oil and gas verticals. How have robotics and automation influenced your manufacturing?

Reid: The materials and processes require tight tolerances in most of our manufacturing processes. When we produce long strings of reeled fiberglass pipe, automation becomes essential.

Like drilling rigs, it’s the highly repetitive tasks that get mechanized and then automated. Robotics has allowed us to automate repetitive, high-precision tasks, resulting in consistent quality and reduced waste in many of our manufacturing plants.

We do have a highly customized nonmetallic shape-forming system that involved converting a workshop crane into a sophisticated cutting machine for large shapes. It has been used to create fast construction bridges over railway lines in large, easy-to-install, lightweight structures.

Across our manufacturing network, automation has meant faster ramp-up times, better repeatability, and improved worker safety. And it’s not just machines—we’re using data to optimize the entire manufacturing chain, from raw materials to final inspection. We have a commercial business that helps non-NOV plants improve their performance with digitized smart systems.

We have often operated in countries and states where the use of automated processes can be replicated and costs managed to make the global supply network very feasible, close to the markets we serve.

RDTS: In your opinion, has software development kept pace with robotics or other machinery development? What are the unique challenges in enabling software development for automation?

Reid: Software has often lagged behind mechanical innovation, particularly because it’s invisible until it fails. The real challenge is interoperability—integrating legacy systems with modern platforms. And because software interacts with live operations, it has to be rock solid. We’re now designing software with modularity in mind, making it easier to adapt and evolve. But we still need more collaborative ecosystems across the industry to truly unlock automation’s full potential.

Inside NOV, our Max platform facilitated every system within our vast ecosystem of acquired companies to communicate and share data. The secret for us was to translate many forms of code rather than ask everyone to adapt their code to one standard. It has gone very well, and we have a highly interoperable system of data and a powerful base of edge computing that operates in a remote cloud-like fashion.

RDTS: To progress automation and maintain a leading edge, what have been the change management and employee reskilling challenges? How has NOV addressed them, and what have the learnings been?

Reid: It’s never just about the tech—it’s about the people. We’ve learned that early engagement is key.

Employees need to see how the technology helps them, not replaces them. We’ve invested in training programs, simulator-based learning, and digital academies. The key lesson? When you invite people into the process, they innovate with you. Our field teams have helped shape many of our most successful tools.

RDTS: How often do you use AI tools like ChatGPT or Perplexity in your work?

Reid: We have used AI in our data science for over 20 years now. The latest generation of generative AI has everyone buzzing with opportunities. We took a sample of employees from across the board and tested a few systems.

Not all were ready for what we needed, but we have found a path where many are reporting large cost reductions and savings in the execution of our work. We have now implemented generative AI training and use it for all employees.

My team uses AI regularly as accelerators for research, ideation, and even strategic planning. These tools let us test ideas quickly, find patterns in complex topics, and generate content at speed. They’re not replacing critical thinking, but they are expanding what’s possible to do in a day.

RDTS: We’d love to hear your opinion on how AI will impact automation roadmaps.

Reid: AI is the nervous system we’ve been waiting for in automation. We’ve spent years building the muscles—robots, control systems, fluids and gas processing, and networked systems.

Now AI connects those systems to make decisions in real time. The roadmap shifts from rule-based automation to adaptive systems that learn and evolve with each job. AI will help eliminate downtime, optimize drilling, completing, and processing in ways humans simply can’t, and make remote operations not just possible, but preferable.

RDTS: How might AI impact our innovation velocity, in general, in oil and gas?

Reid: It already is. We are able to prototype faster, simulate entire operations before we build them, and spot risk factors before they appear in the field.

AI flattens the innovation curve—it reduces the time between idea and implementation. In a complex sector like oil and gas, that speed makes the difference between leading and lagging.

RDTS: I presume that collaboration with technology developers from outside the oil and gas industry is necessary to maintain both cost efficiency and accelerate commercialization. How effective are our external collaboration processes? What can we do to further enhance them?

Reid: We’ve made great progress, but we can still get better. The best breakthroughs often come from outside our bubble—whether it’s robotics from automotive, AI from tech, or materials from aerospace.

What we’ve learned is that success depends on trust and aligned incentives. To go further, we need shared innovation spaces, more agile IP frameworks, and cultural openness—being okay with not always having the answer first.

RDTS: What advice would you give to young professionals entering the energy sector today?

Reid: Be a bridge. The energy sector is a fusion of physical grit and digital possibility.

Learn how the equipment works, but also understand systems thinking, data, and software. Your value will be in connecting those worlds. And don’t just look at where the industry is—look at where it’s going. This is a sector in reinvention mode, and there’s never been a better time to be part of that.

RDTS: Beyond your dual jobs as CTO and CMO, you’re deeply involved in philanthropy. You founded redM to address a unique humane cause. Can you share more about redM’s mission and how others can get involved in supporting this cause?

Reid: The drive for redM is about facing one of humanity’s darkest problems—human trafficking—with hope and courage. Our goal is to build global awareness without fear-based messaging.

We want people to feel empowered, not overwhelmed. Through education, advocacy, and community mobilization, we’re equipping people to recognize and respond to trafficking. Anyone can get involved by donating, volunteering, or simply sharing the message. Learn more at joinredm.com and follow redM Join the Movement on social media. Together, we really can change the world.

Read the third article in this series, which also looks at the expanding use of artificial intelligence in the upstream sector with Zikri Bayraktar, a senior machine learning engineer with SLB’s Software Technology and Innovation Center.