提出者：

编者注：本文出现在新的 E&P 时事通讯中。请在此处订阅勘探与生产通讯 。 

有效且安全的水力压裂完井对于最大限度地提高水平井的盈利能力至关重要。为了使工程师能够继续优化油井处理设计，与生产相关的诊断数据至关重要。最近在德克萨斯州东部的一个页岩井项目展示了在生产监控过程中采用光纤传感的好处。 

近年来，页岩井随着横向长度的增加和簇间距的减小而不断发展。快速找到适合经济条件的理想横向长度和性能集群设计需要一种诊断工具，可以将这些设计变更近乎实时地反馈到生产中。但问题仍然存在，多近才算太近，将簇间距减小到一定长度是否会提高油井盈利能力并增加欧元，还是会带来潜在的挑战？ 

这些设计优化趋势推动了对诊断工具的需求，该工具可以同时测量沿整个井眼发生的所有事件，在流入过程中对真实生产流状态几乎没有干扰，同时还保持低至簇级别的高空间分辨率。光纤传感技术可以同时提供新数据，操作员可以使用这些数据将压裂增产设计提升到新的水平。

石油和天然气光纤

光纤技术是为数据电信而开发的，在 20 世纪 70 年代转变为也包括传感应用。在过去的几十年里，传感距离的增加和空间分辨率的提高增加了光纤的应用。现在可以以光速同时测量沿光纤水平距离的声场（DAS——分布式声学传感）、应变（DSS——分布式应变传感）和温度（DTS——分布式温度传感）。 

传统上，光纤的永久安装是通过将光纤电缆夹紧并粘合在套管外的适当位置，但光纤采集系统信噪比 (SNR) 的改进增加了可用的安装选项。与标准光纤相比，信噪比提高了 100 倍的工程光纤现在可以放置在有线电缆内，该电缆可以部署在套管内以获取数据，然后检索数据以用于其他油井。 

由于光纤数据是分布式测量，因此沿着光纤的每一米都可以获取声学、应变和温度数据。Silixa 的 Carina 100Xlog 是一种可检索光纤测井服务，能够比传统技术更快地实时可视化整个油井动态，从而减少延期生产。

东德克萨斯州的测试时间

为了展示该技术的能力，计划在德克萨斯州东部的海恩斯维尔页岩井上实施生产监测项目。该区域的目标地层位于大约 11,000 真实垂直深度，使其成为高温（300+ F）和高压、以干燥气体为主的储层。运营商在德克萨斯州东部的大面积区域内选择了一口井进行生产监测，因此该项目的研究结果可以应用于其每年 44 多口水平井的完井作业。

在项目开始之前，该井在压裂增产后已经生产了一个月。此前，该井已进行了 99 级压裂增产。1至74级的相同设计均包含六个集群，每英尺六次射击，平均集群间距为17英尺。75至99级的设计包括增加集群数量、减少集群间距、减少射击数量的变化每英尺和锥形性能设计。

在测井之前进行了连续油管清理操作，以增加到达趾部的机会，而不会因固体材料和沿井筒的井弯曲而增加摩擦力。选择高温牵引机将有线光纤拉至趾部。拖拉机最终能够将电缆光纤拉至总深度约 20,700 英尺，从而可以测量 99 级井眼中除前六级之外的所有阶段的产量。该井计划以可变产量生产 60 小时。选择了四个打开和关闭序列来捕获任何生产变化。 

多相分配的数据处理工作流程结合了 DAS 导数，例如声学频谱噪声记录、声速分析和热羽流分析，以约束基于熵平衡的热模型。热模型考虑了重要的井信息，包括井图、流体特性、偏差测量、地表速率和其他有助于降低模型不确定性的储层信息。

产量分析的主要结论是包含减小簇间距的阶段的产量增加。即使级长以及泵送的流体和支撑剂量相同，簇间距减小为 11 英尺的九簇设计在 100% 节流设置下的产量比簇间距为 17 英尺的六簇设计多出 140% 。减小簇间距是否会加速产量和/或增加 EUR 仍然存在不确定性，因为这些数据是在井寿命早期获得的，当时裂缝-储层流动状态可能尚未达到边界流动条件。 

积极的结果 

与六簇设计相比，在有线光缆的各种扼流设置下，九簇设计的产量显着提高了 140% 至 175%。

另一份纤维生产日志可以提供见解。沿外侧的产量以跟部为主，趾部 1 至 34 级仅占总产量的 15%，尽管这些阶段覆盖了外侧长度的 34%。从位于侧面大约一半的最极端的集水坑开始观察到水塞。从这项诊断工作中获得的积极见解将用于优化未来的设计，以根据经济环境调整竣工规模。  

Presented by:

Editor's note: This article appears in the new E&P newsletter. Subscribe to the E&P newsletter here. 

Effective and safe hydraulic fracture completions are crucial to maximizing the profitability of horizontal wells. To enable engineers to continue to optimize well treatment designs, diagnostic data tied to production is crucial. A recent project at a shale well in East Texas showed the benefits of employing fiber-optic sensing during production monitoring. 

Shale wells have evolved in recent years with increasing lateral lengths and decreasing cluster spacing. Quickly finding the ideal lateral length and perf cluster design for the economic conditions requires a diagnostic tool that can tie those design changes back to production in near-real time.  The questions remain, though, how close is too close and does decreasing cluster spacing to a certain length improve well profitability and add to EUR, or does it create potential challenges? 

These design optimization trends have driven a demand for a diagnostic tool that can simultaneously measure all events happening along the entire wellbore, with little interference to the true production flow regime during in-flow, while also maintaining high spatial resolution down to the cluster level. Fiber-optic sensing technology can do both providing new data that are being used by operators to take their fracture stimulation designs to the next level.

Fibre optics for oil and gas

Developed for data telecommunications, fiber-optics technology transitioned to also include sensing applications in the 1970s. Over the past decades increased sensing distances and higher spatial resolution have increased fiber optic’s applications. Simultaneous measurements of acoustic fields (DAS – distributed acoustic sensing), strain (DSS - distributed strain sensing) and temperature (DTS - distributed temperature sensing) along the horizontal distance of the fiber can now be made instantaneously at the speed of light. 

Fiber has traditionally been permanently installed by clamping and cementing in place a fiber cable outside the casing, but improvements in the fiber acquisition system signal-to-noise ratio (SNR) increased the installation options available. An engineered fiber with 100x improvement in SNR over standard fibers can now be placed inside a wireline cable that can be deployed inside the casing to acquire data and then subsequently be retrieved to be used on additional wells. 

As fiber data are a distributed measurement, at every meter along the optical fiber acoustic, strain and temperature data can be acquired. Silixa's Carina 100Xlog is a retrievable fiber-optic well logging service that visualizes entire well dynamics in real time much more rapidly than conventional technologies, reducing deferred production.

Testing times in East Texas

To demonstrate the capability of the technology, a production monitoring project was planned at a well in East Texas targeting the Haynesville Shale. The target formation in this area is located at approximately 11,000 true vertical depth, making it a high-temperature (300-plus F) and high-pressure dominantly dry gas producing reservoir. A well was chosen for production monitoring within the operator’s large acreage position in East Texas so the findings from the project could be applied across their 44-plus horizontal well completions per year.

Before the project began, the well had been producing for a month after having been fracture stimulated. Previously, the well had been fracture stimulated with 99 stages. The identical design of stages 1 to 74 all contained six clusters, at six shots per foot, with an average cluster spacing of 17 ft. The design of stages 75 to 99 included variations of increased clusters counts, decreased cluster spacing, reduced number of shots per foot and tapered perf designs.

A coiled tubing clean-out operation was performed before logging the well to increase the chances of reaching the toe without experiencing increased friction caused by solid materials and well tortuosity along the wellbore. A high-temperature tractor was chosen to pull the wireline fiber to the toe. The tractor was ultimately able to pull the wireline fiber to a total depth of approximately 20,700 ft, allowing production to be measured from all but the first six stages of the 99-stage wellbore. The plan was to produce the well for 60 hours at variable rates. Four opening and closing sequences were chosen to capture any production changes. 

The data processing workflow for multiphase allocation combines DAS derivates, such as acoustics spectral noise logging, sound speed analysis and thermal plumes analysis, to constrain an enthalpy balance based thermal model. The thermal model takes into consideration essential well information including a well diagram, fluid properties, deviation survey, surface rates and other reservoir information that can help reduce the uncertainty of the model.

The main takeaway from the production analysis is the increase in production for the stages that contained the reduced cluster spacing. The nine-cluster design with reduced cluster spacing of 11 ft produced 140% more at a 100% choke setting than the six-cluster design with 17-ft cluster spacing, even though the stage length and the amount of fluid and proppant pumped were equivalent. Uncertainty remains whether the downspaced cluster spacing is accelerating production and/or adding to EUR as these data were acquired early in the life of the well when the fracture-reservoir flow regime may not have reached boundary flow conditions. 

Positive results 

An extremely positive 140% to  175% production uplift was observed from the nine-cluster design compared to the six-cluster designs at various choke settings by the wireline fiber-optic cable.

Another fiber production log can offer insight. Production along the lateral is heel-dominated with toe-stages 1 to 34 contributing only 15% to total production, even though those stages cover 34% of the lateral length. Water slugs are observed starting at the most extreme sump located approximately halfway down the lateral. The positive insight gained from this diagnostic job will be used to optimize future designs to right-size the completion to the economic environment.