# US 2023/0144184 A1 · Advanced geological prediction method and system based on perception while drilling

> Agents: this data is available as JSON from the patentgrep API and MCP server — see https://patentgrep.com/llms.txt

- Number: US 2023/0144184 A1
- Document ID: US20230144184A1
- Status: Published application
- Kind: A1
- Application number: 17/881,158
- Filed: 2022-08-04
- Published: 2023-05-11
- Earliest claimed: 2021-11-11
- Assignees: Shandong University
- Applicants: Shandong University
- Inventors: Zhenhao XU; Peng LIN; Shucai LI; Huihui XIE; Heng SHI; Zhaoyang WANG; Fumin LIU; Ruiqi SHAO
- Web page: https://patentgrep.com/patent/US20230144184A1

## CPC

- E21B 49/003: {by analysing drilling variables or conditions (E21B49/005 takes precedence; systems specially adapted for monitoring a plurality of drilling variables or conditions E21B44/00)}
- E21B 49/005: {Testing the nature of borehole walls or the formation by using drilling mud or cutting data}
- E21D 9/003: {Arrangement of measuring or indicating devices for use during driving of tunnels, e.g. for guiding machines}
- G06F 30/27: using machine learning, e.g. artificial intelligence, neural networks, support vector machines [SVM] or training a model
- E21B 2200/20: Computer models or simulations, e.g. for reservoirs under production, drill bits
- E21B 2200/22: Fuzzy logic, artificial intelligence, neural networks or the like
- E21B: EARTH OR ROCK DRILLING; OBTAINING OIL, GAS, WATER, SOLUBLE OR MELTABLE MATERIALS OR A SLURRY OF MINERALS FROM WELLS
- E21D: SHAFTS; TUNNELS; GALLERIES; LARGE UNDERGROUND CHAMBERS (soil-conditioning or soil-stabilising materials C09K17/00; drilling or cutting machines for mining or quarrying E21C; safety devices, transport, rescue, ventilation or drainage E21F)
- G06F: ELECTRIC DIGITAL DATA PROCESSING (computer systems based on specific computational models G06N)

## Abstract

An advanced geological prediction method and system based on perception while drilling, and relates to advanced geological prediction. The solution includes: acquiring drilling parameters during drilling; obtaining physical and mechanical parameters of tunnel surrounding rocks by inversion based on drilling parameters; acquiring rock slag or powder based on flushing fluid collected during drilling; acquiring geochemical characteristic parameters of rock slag or powder; and obtaining at least one adverse geology recognition result and surrounding rock classification result using a pre-trained deep learning model, and realizing advanced geological prediction. Combined with advanced geological drilling, the solution reflects geological characteristics from changes of physical and mechanical properties of tunnel surrounding rocks and changes of geochemical characteristic parameters. Advanced prediction of geology ahead of a tunnel face is realized by collection and analysis of drilling parameters and flushing fluid during advanced drilling and the fusion of big data and a deep learning algorithm.

## Claims

### Claim 1

An advanced geological prediction method based on perception while drilling, comprising:
acquiring drilling parameters during drilling;
obtaining physical and mechanical parameters of tunnel surrounding rocks by inversion based on the drilling parameters;
acquiring rock slag or rock powder based on a flushing fluid collected during drilling;
acquiring geochemical characteristic parameters of the rock slag or the rock powder; and
analyzing, according to the acquired physical and mechanical parameters of tunnel surrounding rocks and geochemical characteristic parameters, engineering geological conditions ahead of a tunnel face by using a pre-trained deep learning model respectively, obtaining an adverse geology recognition result and a surrounding rock classification result respectively, and realizing advanced geological prediction by using at least one of the two results.

### Claim 2 (depends on 1)

The advanced geological prediction method based on perception while drilling according to claim 1, wherein the geochemical characteristic parameters comprise types and content of elements in rock mass, types and content of minerals, and types and content of anions and cations in an aqueous solution.

### Claim 3 (depends on 1)

The advanced geological prediction method based on perception while drilling according to claim 1, wherein the physical and mechanical parameters of tunnel surrounding rocks comprise compressive strength, cohesion, internal friction angle, abradability, and integrity of rock mass.

### Claim 4 (depends on 1)

The advanced geological prediction method based on perception while drilling according to claim 1, wherein training of the deep learning model specifically comprises:
constructing a training set for adverse geology recognition based on an existing data set, and training the deep learning model by using the training set to obtain a trained adverse geology recognition model; and constructing a training set for surrounding rock classification based on the existing data set, and training the deep learning model by using the training set to obtain a trained surrounding rock classification model.

### Claim 5 (depends on 4)

The advanced geological prediction method based on perception while drilling according to claim 4, wherein a process of mining the existing data set comprises: collecting physical and mechanical parameters of compressive strength, cohesion, internal friction angle, abradability, and integrity of rock mass in various adverse geologies and influence areas thereof on a tunneling route, as well as types and content of elements, types and content of minerals, and types and content of anions and cations in an aqueous solution, and mining, based on a data mining mode, physical and mechanical parameters capable of reflecting geology precursor characteristic information and geochemical characteristic gradual evolution information in the rock mass on the tunneling route.

### Claim 6 (depends on 1)

The advanced geological prediction method based on perception while drilling according to claim 1, wherein the corresponding deep learning model is continuously updated and optimized according to the physical and mechanical parameters of tunnel surrounding rocks, the geochemical characteristic parameters and the adverse geology recognition result as a drilling process progresses; and the corresponding deep learning model is continuously updated and optimized according to the physical and mechanical parameters of tunnel surrounding rocks, the geochemical characteristic parameters and the surrounding rock classification result.

### Claim 7 (depends on 1)

The advanced geological prediction method based on perception while drilling according to claim 1, wherein the deep learning model performs multi-level characteristic extraction on input data by using fully connected layers and residual fully connected layers while introducing an attention mechanism.

### Claim 8 (depends on 1)

The advanced geological prediction method based on perception while drilling according to claim 1, wherein a process of fusing input data of the deep learning model specifically comprises: performing characteristic extraction on the input data respectively based on the fully connected layers, and concatenating extracted characteristics.

### Claim 9 (depends on 1)

The advanced geological prediction method based on perception while drilling according to claim 1, wherein the obtaining physical and mechanical parameters of tunnel surrounding rocks by inversion based on the drilling parameters specifically comprises: constructing a mapping relation between the drilling parameters and the physical and mechanical parameters of tunnel surrounding rocks based on historical data; and determining the physical and mechanical parameters of tunnel surrounding rocks based on the mapping relation and the acquired drilling parameters.

### Claim 10

An advanced geological prediction system based on perception while drilling, comprising:
a drilling parameter acquisition unit, configured to acquire drilling parameters during drilling;
a mechanical property analysis unit, configured to obtain physical and mechanical parameters of tunnel surrounding rocks by inversion based on the drilling parameters;
a rock slag collection unit, configured to acquire rock slag or rock powder based on a flushing fluid collected during drilling;
a geochemical characteristic analysis unit, configured to acquire geochemical characteristic parameters of the rock slag or the rock powder; and
an advanced geological prediction unit, configured to analyze, according to the acquired physical and mechanical parameters of tunnel surrounding rocks and geochemical characteristic parameters, engineering geological conditions ahead of a tunnel face by using a pre-trained deep learning model respectively, obtain an adverse geology recognition result and a surrounding rock classification result respectively, and realize advanced geological prediction by using at least one of the two results.
