- Home
- Conferences
- Conference Proceedings
- Conferences
Seventh EAGE Borehole Geophysics Workshop
- Conference date: September 18 - 20, 2023
- Location: Milan, Italy
- Published: 18 September 2023
1 - 20 of 35 results
-
-
Dual-Signal Processing of DAS-VSP Data of the Kizildere Geothermal Field
SummaryWe present the approach and initial QC results of the dual-signal processing of VSP data acquired using semipermanent DAS technology during a baseline survey of a CO2 injection- monitoring project in the Kizildere (Turkey) geothermal-production reservoir. The data were recorded in the framework of the SUCCEED project in two wells using a high-sensitivity engineered fibre with the cable suspended in the vertical cased wells. The source was a new electric seismic vibrator operated at the surface with a 3D configuration, supported with measurements on two bi-axial geophone lines.
Good-quality VSP results were obtained during the initial QC performed by in-field and remote control and from the prompt data processing after the survey acquisition. The VSP-data processing takes advantage of the dual-field separation method effectively applied with the DAS well data densely sampled every 1 m. This approach enabled us to quickly separate up- going and down-going VSP wavefields. This technique does not require first-arrival picking, which is advantageous for processing extensive 3D-VSP datasets.
The results from sample VSP revealed the reflection information contained in the data, relevant for target characterization. This analysis demonstrates the potential of the dataset for carbonate- reservoir monitoring purposes, to be compared in the future with time-lapse measurements.
-
-
-
LWD Image Logs Interpretation (LILI) : An Automated Approach Using Deep Learning
By A. MolossiSummaryImage logs can provide critical information to reduce drilling risk, as they allow for early detection of structures such as faults or fractures. The timeliness of such information is undermined by the time required to manually interpret the data and the subjectivity of interpretations. The proposed methodology is a supervised Deep Learning - based method built on U-Net architecture for segmentaton of image logs acquired while drilling, i.e., automated detection of geological edges in borehole images. The proposed network has been trained on synthetic data and tested on field data. Different learning strategies, namely Curriculum and standard learning (CL,SL), were compared to observe the impact of segmentation process on final results: CL shows greater potential in early fracture detection due to its superior performance in chaotic and heterogeneous intervals
-