生产分析被视为比较的黄金标准。虽然生产是收入的来源，但许多其他诊断方法最近在经济上变得可行，可以帮助解释刺激效果的差异。诊断必须具有成本效益，才能实现运营商在从油井完井中创造更多价值的过程中寻求的投资回报（ROI）。

由于设备成本低、操作影响最小以及分析所创建数据的强大能力，压力监测已成为最具成本效益的诊断应用。为了更多地了解这项技术，Reveal Energy Services 结合了基于压力的裂缝图 (PBFM) 和增产后泄漏分析，以确定全油田压力监测应用的油藏增产效果。此外，该方法评估每个阶段的增产，而不是在井级进行粗略聚合，例如生产分析。

这两种诊断方法在特拉华盆地的多个垫上实施。这些独立的诊断应用程序监测油藏的不同区域，并结合起来，提供每个监测处理阶段的全现场评估。本文回顾了两种方法与处理设计变量的组合，以了解导致最佳岩石增产的完井趋势。

压力监测研究

两种独立的压力监测技术实现了全场刺激评估。这项研究以合作的方式利用来自多个运营商的数据集。运营商同意对数据进行汇总和匿名处理，以便进行适当的分布以确保得出有效的结论。

该研究基于特拉华盆地的 Wolfcamp A 地层。所考虑的阶段级数据通常比井级数据包含更多的数据点。几个环境因素必须相似才能进行比较。使用来自相似地理和地质角度的数据有助于使这些环境因素尽可能接近。

研究进行如下：

• 首先，对每种压力监测方法进行独立分析。
• 其次，每种方法的相对分数为低（1）、中（2）和高（3）。
• 第三，通过将两种方法的分数相乘来计算代理刺激岩石体积（SRV）。
• 最后，计算平均完井参数（每英尺阶段长度）并进行比较，以确定相关的独特趋势。

方法 1 应用了 PBFM，这是一种 Reveal Energy Services 应用程序，其中数字孪生模型呈现给定阶段主要裂缝的水力范围，以确定裂缝高度、半长和方位角（图 1a）。从附近子井或填充井的隔离阶段监测压力。监测阶段和治疗阶段的位置被输入到数字孪生模型中，迭代求解器会更新几何结构，以便建模的压力响应与观察到的压力响应相匹配。这是一个远场监测应用，因为应力通过水库延伸到监测仪。

方法2采用增产后泄漏分析，利用处理井上的压力表来监测泵停止长达1小时后的压力下降情况。压力下降速率与增产期间实现的储层接触量成正比，因此可用于阶段与阶段之间的比较（图 1b）。该监测是近场测量，因为压力下降是从处理阶段直接测量的。远场和近场监测的结合为每个完成阶段提供了全场评估。全场方法改进了给定阶段的 SRV 估计。

面积来自主要裂缝的 PBFM，长度是阶段长度，效率因子来自泄漏分析。井的 SRV 是每个阶段的单独 SRV 贡献的总和。通过将刺激后泄漏分析的分数和 PBFM 相对大小与面积相乘来使用 SRV 的代理，并且每个阶段都单独评估。

观察和见解

这项研究得出了三个主要见解。对于给定的裂缝区域，SRV 分数较高的阶段具有较低的支撑剂负载（磅/英尺）。例如，在 PBFM 计算中裂缝面积评级为中 (2) 的所有阶段中，最高 SRV 组的支撑剂负载最低，最低 SRV 组的支撑剂负载最高（图 2） 。

Production analysis has been regarded as the gold standard of comparison. While production is where revenue is accrued, many other diagnostics have recently become economically viable to help explain differences in stimulation effectiveness. Diagnostics must be cost-effective to render the return on invest­ ment (ROI) that operators seek in the drive to create more value from oil well completions.

Pressure monitoring has emerged as the most cost-effective diagnostic application because of the low cost of devices, minimal operational impacts and vast ability to analyze the data created. For the purposes of understanding more about this technol­ogy, Reveal Energy Services combined pressure-based fracture maps (PBFMs) and post-stimulation leak-off analysis to identify reservoir stimulation effective­ness for a full-field pressure monitoring application. Additionally, this method evaluated stimulation at each stage rather than having coarse aggregations at the well level, such as production analysis.

The two diagnostic methods were implemented on multiple pads across the Delaware Basin. These independent diagnostic applications monitored different areas of the reservoir and, when combined, provide a full-field evaluation of each monitored treatment stage. This article reviews the combina­tion of the two methods with the treatment design variables to understand the completion trends that have resulted in optimal rock stimulation.

Pressure monitoring study

The two independent pressure monitoring tech­niques enabled the full-field stimulation evaluation. This study leverages datasets from multiple opera­tors in a cooperative way. The operators agreed to aggregate and anonymize the data so there is an appropriate distribution to ensure valid conclusions.

The study was based in the Wolfcamp A Forma­tion in the Delaware Basin. Stage-level data that typically include many more datapoints than well level data were considered. Several environmental factors must be similar for comparisons. Using data from similar geographical and geological perspec­tives helped to keep these environmental factors as close as possible.

The study proceeded as follows:

• First, an independent analysis was conducted for each pressure monitoring method.
• Second, a relative score was given for each method as low (1), mid (2) and high (3).
• Third, a proxy stimulated rock volume (SRV) was calculated by multiplying the scores from both methods.
• Finally, the average completion parameters were calculated (per foot of stage length) and com­pared to identify relevant distinctive trends.

Method 1 applied PBFMs, a Reveal Energy Ser­vices application in which a digital twin model ren­ders the hydraulic extent for a dominant fracture in a given stage to determine fracture height, half-length and azimuth (Figure 1a). The pressure is monitored from an isolated stage in a nearby child, or infill, well. The location of the monitoring stage and treatment stages are inputs into the digital twin model, and the iterative solver updates the geometries so the modeled pressure response matches the observed pressure response. This is a far-field monitoring application because the stress extends through the reservoir to the monitor.

Method 2 applied post-stimulation leak-off analysis, which utilizes the pressure gauge on the treatment well to monitor the pressure decline after the pumps have been stopped for a period up to 1 hour. The rate of pressure decline is proportional to the amount of reservoir contact achieved during stimulation and, therefore, can be used in a stage to-stage comparison (Figure 1b). This monitoring is a near-field measurement because the pressure decline is directly measured from the treatment stage. The combination of the far-field and near-field monitoring provides a full-field evaluation for each completion stage. The full-field method improves the estimation of the SRV for a given stage.

Area is from the PBFM of the dominant fracture, the Length is the stage length and the efficiency factor is derived from the leak-off analysis. The SRV for a well is the summation of each stage’s individ­ual SRV contribution. A proxy to the SRV is used by multiplying the score from the post-stimulation leak-off analysis and the PBFM relative size with Area, and each stage is evaluated individually.

Observations and insights

This study arrived at three main insights. The stages with better SRV scores had lower proppant loading (pounds per foot) for a given fracture area. For example, of all the stages with a mid (2) rating for fracture area in the PBFM calculation, the high­est SRV group had the lowest proppant loading, and the lowest SRV group had the highest prop­pant loading (Figure 2).