Process information
| Key Data Set Information | |
| Location | EEDS-NMG-CN |
| Geographical representativeness description | Limestone of Taiyuan Formation in Central and Eastern Ordos Basin |
| Reference year | 1987 |
| Name |
|
| Use advice for data set | Users of this data set should explicitly account for the technological specificity of the micro-fault fine characterization seismic processing in their assessment models. Data application should consider the temporal exploration phases as influential factors affecting the reliability of the results. Take note of the non-linear mapping relationship established through deep learning for the identification of micro-faults and fractures, and the use of guided filtering combined with AI for seismic data interpretation. It is crucial to adapt and calibrate the LCA models to the complex 'sandwich' reservoir forming pattern indicated by this data set for the Taiyuan Formation limestone. |
| Technical purpose of product or process | The seismic processing technology used in the prediction of earthquakes, and exploration and characterization of limestone reservoirs of the Taiyuan Formation is primarily intended for the identification and mapping of micro-faults within geological structures. This advanced technology is pivotal in enhancing the clarity, continuity, and confidence of fault predictions which has applications in the field of natural gas exploration, specifically in detecting and analyzing potential gas reservoirs with small concealment and micro-faults in the Ordos Basin. |
| Classification |
Class name
:
Hierarchy level
|
| General comment on data set | The technical flow of micro-fault fine characterization seismic processing consists of three parts: ① the original seismic data are processed by structural guidance filtering under the constraint of horizon, and the interference of background noise is suppressed to highlight fault breakpoints; ② Carry out neural network deep learning to obtain deep learning coherence body (fault) and curvature body (fracture), and establish nonlinear mapping relationship from seismic data to micro-fault; Select the dominant information for weighted fusion to realize the fine identification of faults with small concealment and micro-faults. |
| Copyright | No |
| Owner of data set | |
| Quantitative reference | |
| Reference flow(s) | |
| Time representativeness | |
| Data set valid until | 2022 |
| Time representativeness description | Exploration began in the 1980s |
| Technological representativeness | |
| Technology description including background system | Seismic processing adopts the combination of "guided filtering + artificial intelligence + key seismic information fusion", and the fault prediction results have better continuity, clear boundary and higher confidence. |
| Flow diagram(s) or picture(s) | |
Modelling and validation
Administrative information
Inputs and Outputs