Collecting seismic data requires an energy source to generate waves and sensors to receive those waves. The appropriate energy source and receiver depend on the location and the application. This article describes different types of equipment used for seismic data acquisition. A variety of seismic sources exist that can apply vertical impulse forces to the surface of the ground. These devices are viable energy sources for onshore seismic work.
Imrie, Andrew (Halliburton Energy Services) | Negenman, Brendon (Halliburton Energy Services) | Lee, Chung Yee (Halliburton Energy Services) | Iyer, Mahadevan S. (Halliburton Energy Services) | Parashar, Sarvagya (Halliburton Energy Services) | Shata, Mohamed Raouf (Halliburton Energy Services) | Helton, Sean (ConocoPhillips)
The identification of low-rate leaks along with low annular-pressure buildup rates in any type of completion presents challenges in the well-integrity domain. This paper emphasizes the importance of understanding the well-diagnostic problem to determine feasibility, isolate interest zones, enhance stimulation strategies, and ultimately optimize the acquisition of high-resolution acoustical data from the wellbore with a latest-generation advanced leak-detection tool.
This case study discusses the methodology that underlies the successful determination of the depths and the radial locations in the outer casing strings of multiple leaks in an offshore well. In the study presented, emphasis had been placed on the job planning to provide adequate or substantial leak stimulation for the accurate determination of the leak points in terms of radial distance away from the tool axis within the wellbore. Rather than a shut-in and flowing or venting acquisition, it was proposed that the optimal method for the successful determination of an outer casing string leak involved invoking a range of flow rates and, therefore, acoustic levels, across an extended period. The study also demonstrates the advantages of integrating acoustic-based tools with conventional production logging tools.
Two outer string casing leaks with annulus to formation communication areas were identified from high-resolution leak-detection logging coupled with conventional pressure and temperature measurements. The interpretation process included the computation of a 2D radial map of the flow activity across each zone of interest. This process resulted in less ambiguity and clearer results obtained in real time during the acquisition. The location of each leak point was triangulated using an error-minimization algorithm from the received acoustic waveforms at the tool receiver array. Further, the optimized stimulation strategy enabled leak-stimulation responses to be tracked in the computed power spectral density (PSD) at each leak. This process enabled the operator to promptly move on with the well abandonment strategy without waiting for further data analysis.
Attention to detail from the outset and a complete understanding of the well and its annular pressure and fluid behavior enabled an optimized and focused electric line diagnostic strategy to be used. The use of high-resolution acoustic data from an advanced leak-detection tool with an array of hydrophones ensured that the multiple leak locations were identified and characterized.
Acoustic hydrophones have been used for detection and localization of leak flows in wellbores. In addition to the location of detected leak flows, fluid phases of the flows are also valuable information to extract from the measured acoustic signals.
This work introduces a support-vector-machine (SVM) approach to classify four different flow phase scenarios: liquid-to-liquid (L2L), gas-to-gas (G2G), gas-to-liquid (G2L), and liquid-to-gas (L2G). The proposed algorithm consists of three steps: spectrum estimation, principal component analysis (PCA), and SVM classification. First, the frequency spectrum is computed from the time-domain acoustic signals. Second, PCA is used to extract features from the spectrum profile, during which principal components of the spectrum that contain most of the information are extracted. Third, the extracted features from Step 2 are used as inputs to the SVM classifier, which labels the input feature values with one of the previously mentioned scenarios.
Experimental data were used to train and test the SVM classifier. For each of the four flow phase scenarios, multiple leak types were simulated to account for variations of actual leak flows downhole. For each leak type, various leak flow rates were produced using different orifice sizes and differential pressures across the orifices. An acoustic leak-detection tool was used to record the acoustic signal of the leak sources. Field data were used to validate the SVM classifier, achieving an accuracy of 91%. In addition, it was determined that the combination of spectrum estimation, PCA, and SVM outperformed the algorithm that used only spectrum estimation and SVM.
Ziolkowski et al.,1982, Parkes et al., 1984) is a well-established procedure and is now recognized as important for accurate broadband designature of seismic data. In the case of a single NFH per source element, it is necessary to define a model for the propagation of sound from each source element (a translating and pulsating bubble of air) to each receiver, comprising phase shift and amplitude scaling terms. The precise layout of sources and receivers must be known together with the reflectivity of the sea-surface to account for direct and reflected paths (ghosts). With n hydrophone measurements and n notional sources to solve for, the problem is well posed. Parkes and Hatton (1986) proposed that this scheme could be extended, whereby 2n hydrophones are used and a series of n virtual sources (in mirror positions above the sea-surface) are additionally solved for. This then removes both the need to model the ghost path, and to parameterize the sea-surface reflectivity. Hampson (2017) and Kryvohuz and Campman (2017) recently revived interest in these ideas, separating the wave-field to better-understand the physics occurring in the vicinity of the array and to improve characterization of the source ghost in the far-field signature. Here we describe an acquisition test carried out early in 2018, where near-field hydrophone data and the associated (far-field) streamer data were recorded with the aim of evaluating the benefit of additional measurements and of the virtual notional concept in particular.
Summary The extension of full waveform inversion (FWI) into elastic models can potentially address the limitations of acoustic FWI due to amplitude versus offset effects. We show the steps used to pre-process the data and discuss a strategy for creating initial elastic models for inversion. Finally, we apply our inversion workflow and show results for frequencies up to 10 Hz. Results show inverted elastic models for Vp, Vs and density, the potential of extracting a Vp-Vs ratio from such inversions and elastic reverse time migration images for each parameter. Introduction Full Waveform Inversion (FWI) is one of the most prolific research topics in seismic imaging. Although the fundamentals of FWI were proposed over 30 years ago (Tarantola, 1987), computational demands and unsuitable acquisition configurations limited its generalized adoption until recently.
We present the theory and initial results for deblending multicomponent simultaneous source data using a pattern-based approach based on multidimensional prediction-error filters (PEFs). In using this pattern-based approach, we provide a method for PEF estimation that makes use of the directional information recorded on the geophone components in order to improve the source separation on the hydrophone. We provide synthetic numerical examples and an example from a FreeCableTM data set that demonstrate that using PEFs estimated on all data components results in better separation than using only the hydrophone component.
Presentation Date: Wednesday, October 17, 2018
Start Time: 1:50:00 PM
Location: 210C (Anaheim Convention Center)
Presentation Type: Oral
Efficiently and accurately estimating fluid-flow movement information from time-lapse data is a prime deliverable of any 4D acquisition and analysis. The key to success in this depends on a few factors including optimum 4D seismic acquisition, the seismic frequency bandwidth at the reservoir level and being able to deliver the 4D analysis or results in a very rapid and efficient manner. Maximum value of 4D is derived not from the data quality alone but also from the efficiency of delivering a 4D image and analysis. The value of the 4D decreases significantly with time, as results and analysis need to be delivered promptly to make an impact on the in-fill well program as well as on the reservoir development.
From the 4D seismic image analysis (based on a calibrated broadband PSDM seismic processing), a dynamic warping algorithm was implemented for estimation of time-shift and delta velocity on this non-conventional typical “broadband�? 4D seismic i.e. new multi-component over a conventional streamer legacy survey. The 4D analysis results were then compared with the 4D rock physics analysis at the available wells over the survey and related to the production-injection mechanism. This paper will review the project results and their impact in term of reservoir management understanding.
Presentation Date: Monday, October 15, 2018
Start Time: 1:50:00 PM
Location: 204C (Anaheim Convention Center)
Presentation Type: Oral
This paper presents a new technology and methods that can detect leak locations in a well and illustrate the flow profile of the leak. A substantial amount of time and effort can be expended in repairing leaks in wells, and these methods can reduce that time. The paper shows results and compares them to those of other techniques for a well that had been shut in as a result of a small leak. Noise tools have been used to detect the sound of leak flow to provide an estimated description on the basis of the magnitude of the noise and the frequency properties. Typically, these tools consisted of one hydrophone or receiver that was limited to frequency and information recorded.
Many innovative advances in the seismic method have been introduced over recent years. In this discussion, I will focus on the topic of sampling. A key example is the azimuthal sampling in full-azimuth 3D surveys--surveys that are needed, for instance, to characterize fractures. Full-azimuth geometries typically call for an expensive explosion in the amount of data needing to be acquired. A way to reduce the acquisition cost is to rely instead on interpolation, but aliasing issues limit the spatial frequencies that can be recovered.