CN-120667129-B - Mode conversion system and method of multimode shield tunneling machine in composite stratum

CN120667129BCN 120667129 BCN120667129 BCN 120667129BCN-120667129-B

Abstract

The invention relates to the technical field of tunneling equipment, in particular to a mode conversion system of a multimode shield tunneling machine in a composite stratum, which comprises a prediction layer, a perception layer, a data bus layer, a decision layer and an execution layer, wherein the prediction layer acquires front geological and hydrological information through TSP advanced geological prediction and an infrared water detector, the perception layer integrates multiple sensors to acquire tunneling parameters and processes the tunneling parameters through a data fusion algorithm, the data bus layer realizes data time-space alignment based on a distributed stream processing platform, the decision layer adopts a CNN+fuzzy logic+reinforcement learning mixed model to generate a conversion instruction, and the execution layer drives an execution mechanism to complete EPB/SPB/TBM mode conversion and ensures safety through feedback correction. The invention realizes the self-adaptive decision and seamless execution of mode conversion, obviously shortens the conversion time, reduces the cutter abrasion and improves the construction efficiency and the safety of the composite stratum.

Inventors

LONG WEI
XU ZIWEN
JIANG GUISHAN
LIU CHAOYIN
ZHOU KAI
LU GAOMING
JIANG YIHUI
YANG YANDONG
LU ZHIYONG
DONG CONGHUI

Assignees

中铁南方投资集团有限公司
盾构及掘进技术国家重点实验室

Dates

Publication Date: 20260512
Application Date: 20250718

Claims (7)

1. A mode conversion system of a multimode shield tunneling machine in a composite stratum is characterized by comprising The prediction layer comprises an advanced detection module and is used for acquiring geological structure characteristics and hydrologic information in a preset range in front of the shield machine; The sensing layer comprises a real-time monitoring module and a data fusion module, wherein the real-time monitoring module is used for monitoring and collecting operation parameters and working condition data in the tunneling process of the shield machine in real time and processing the real-time monitoring data through the data fusion algorithm; The data bus layer is used for realizing alignment fusion of the advanced detection data of the prediction layer and the real-time monitoring data of the perception layer in the time dimension and the space dimension based on the distributed stream processing platform; The decision layer is used for generating a mode conversion decision instruction based on advanced detection data and real-time monitoring data aligned in time and based on a mixed model of convolutional neural network CNN geological feature extraction, fuzzy logic control and reinforcement learning; The execution layer is used for responding to the decision instruction, driving the execution mechanism of the shield machine to complete the conversion among multiple modes and dynamically adjusting the execution parameters through closed loop feedback; The real-time monitoring module comprises a cutter torque sensor, a propulsion speed sensor, a soil bin pressure sensor and a water seepage amount monitor; The data fusion algorithm comprises a wavelet transform denoising algorithm and a Kalman filtering algorithm, The wavelet transformation denoising algorithm carries out self-adaptive decomposition on the sensor signal through wavelet layered denoising, and the decomposition layer number is dynamically selected according to the signal bandwidth; the Kalman filtering algorithm fuses the data of the multiple sensors, and introduces a sensor reliability weight factor to perform state estimation; the data bus layer adopts a distributed stream processing platform to realize data integration, the distributed stream processing platform is APACHE KAFKA platform, and the data alignment is realized by the following modes: data standardization, namely adopting Protobuf protocol to process real-time monitoring data and adopting Avro + GeoJSON protocol to process advanced detection data; Time alignment, namely realizing the time unification of all devices through an NTP server, wherein the time error is less than 1ms; Space alignment, namely establishing a conversion relation between a local coordinate system of the shield tunneling machine and an engineering global coordinate system, and realizing space position matching of monitoring data and geological data; The algorithm of the hybrid model comprises: The geological feature extraction model adopts a double-branch processing structure to respectively process advanced geological detection image data and water content identification parameters and output stratum classification results, permeability coefficients and risk area information; the fuzzy logic control is used for constructing a dynamic rule base based on real-time monitoring data after denoising, wherein the real-time monitoring data comprises cutter torque, propelling speed, soil bin pressure and water seepage quantity; Reinforcement learning, namely taking the increment of earth surface subsidence, energy consumption and equipment health degree as optimization targets, adjusting a control strategy through a reward function, and outputting a mode conversion instruction and an execution parameter.
2. The system for converting modes of the multimode shield tunneling machine in the composite stratum according to claim 1, wherein the advanced detection module comprises advanced geological prediction equipment and an infrared water detector, wherein the advanced geological prediction equipment can detect at least 100m in front, and the infrared water detector can identify fault water content information in a range of at least 30m in front.
3. The system for converting modes of the multimode shield tunneling machine in the composite stratum according to claim 1, wherein the execution layer comprises an instruction analysis module, an action execution module, a safety monitoring module and a feedback correction module; Wherein, the The instruction analysis module is used for decomposing the decision instruction into a multi-system cooperative action sequence; the action execution module is used for driving the hydraulic system, the grouting system and the soil discharging system to cooperatively act and realizing the conversion among a soil pressure balance mode, a mud water balance mode and a hard rock tunneling mode; The safety monitoring module is used for checking the execution state in real time and triggering abnormal rollback or emergency shutdown; The feedback correction module dynamically adjusts the execution parameters based on sensor closed loop feedback.
4. The system for converting modes of the multimode shield tunneling machine in the composite stratum as set forth in claim 2, wherein said advanced geological prediction equipment detects the structural state and integrity of surrounding rock by adopting a seismic method, said infrared water detector identifies hidden water sources by detecting infrared radiation field distortion of the geologic body, and the two match structural anomaly regions and water-containing anomaly regions by a complementary data fusion mechanism.
5. The system for converting modes of the multimode shield tunneling machine in the composite stratum according to claim 1, wherein in a double-branch processing structure of the geological feature extraction model, one branch processes seismic wave image data, the other branch processes infrared parameters, and the action exploration range is limited within an engineering safety threshold by combining expert experience data in a reinforcement learning training process.
6. The system for switching modes of the multimode shield tunneling machine in the composite stratum according to claim 3, wherein the switching process among the soil pressure balance mode, the slurry balance mode and the hard rock tunneling mode comprises the cooperative actions of locking a screw conveyor, adjusting the pressure of a slurry pump and dynamically controlling the sealing of a shield tail, and the abnormal rollback strategy of the safety monitoring module is dynamically triggered according to the water seepage quantity and the pressure deviation of a soil bin.
7. The mode conversion method of the multimode shield machine in the composite stratum is implemented based on the mode conversion system of the multimode shield machine in the composite stratum according to any one of claims 1 to 6, and is characterized by comprising the following steps: S1, acquiring geological structure characteristics and hydrological information in a preset range in front of a shield machine through advanced detection equipment; s2, acquiring operation parameters and working condition data in the tunneling process of the shield machine through a plurality of real-time monitoring sensors, and processing the operation parameters and the working condition data by adopting a data fusion algorithm; Step S3, based on a distributed stream processing platform, aligning and fusing the geological hydrologic data acquired in the step S1 and the real-time monitoring data processed in the step S2 in a time dimension and a space dimension; S4, based on the data aligned and fused in the step S3, generating a mode conversion decision instruction and corresponding execution parameters by adopting a geological feature extraction model, a fuzzy logic control and reinforcement learning hybrid algorithm; And S5, driving an execution mechanism of the shield machine to complete the conversion among the soil pressure balance mode, the mud-water balance mode and the hard rock tunneling mode according to the mode conversion decision instruction, dynamically adjusting execution parameters through sensor closed loop feedback, and simultaneously carrying out safety monitoring on the conversion process and triggering emergency response under abnormal conditions.

Description

Mode conversion system and method of multimode shield tunneling machine in composite stratum Technical Field The invention relates to the technical field of tunneling equipment, in particular to a mode conversion system and method of a multi-mode shield tunneling machine in a composite stratum. Background Along with the extension of urban underground space development to complicated geological condition areas, the multimode shield tunneling machine becomes core equipment because of being capable of adapting to composite strata such as alternating soft and hard rocks, karst cave, fracture zones and the like. The multi-mode shield machine is mainly divided into two main types of a dual-mode shield machine and a three-mode shield machine, wherein the dual-mode shield machine is divided into an earth pressure balance EPB/hard rock tunneling TBM dual-mode shield, an earth pressure balance EPB/slurry balance SPB dual-mode shield and a slurry balance SPB/hard rock tunneling TBM dual-mode shield. The soil pressure balance EPB/SPB dual-mode shield combines two pressure control modes of soil pressure balance and slurry balance, mode conversion is realized through a conversion slag discharging system, low-permeability stratum such as clay, silt and the like are subjected to pressure balance through a soil bin, and the excavation surface is stabilized by slurry circulation through high-permeability stratum such as a sand layer and a gravel layer or high-water pressure stratum. The EPB/TBM dual-mode shield has soil pressure balance and hard rock tunneling functions, mode conversion is realized by changing a cutter head or adjusting a thrust system, a soft rock, weathered rock and soft soil composite stratum adopts a soil pressure mode, and a full-section hard rock stratum adopts a TBM mode. The SPB/TBM dual-mode shield integrates mud water circulation and hard rock breaking functions, and is suitable for high-water pressure hard rock and permeability soft soil alternate stratum. The three-mode shield tunneling machine further integrates three modes on the basis of double modes, realizes the full-adaptability tunneling of extremely complex stratum, and is compatible with mud circulation, spiral deslagging and hob rock breaking systems in a limited space. The prior art has the following defects: 1. The traditional multimode shield mode conversion relies on manual experience judgment or off-line simulation of preset geological parameters, cannot adapt to dynamic changes of complex stratum, has response lag and misjudgment risks, and is statistically displayed with a manual misjudgment rate reaching 18%, 2. The traditional online multimode shield shortens conversion time, but the slag discharging system is fixed in layout, and lacks geological suitability, and meanwhile abnormal abrasion of a cutter is extremely easy to occur, 3. The mode conversion decision is not closed-loop, namely conversion safety is ensured through grouting reinforcement, but is not linked with real-time stratum parameters, lag risks exist, and 4. The mechanical system in the mode conversion process has poor cooperativity, and potential safety hazards such as hydraulic impact, sealing failure and the like exist. Therefore, development of a full-link adaptive system integrating multisource sensing, dynamic decision and mechanical linkage is needed to solve the real-time and safety problems of mode conversion in the composite stratum. Disclosure of Invention In order to solve the technical problems, the invention adopts the following technical scheme. A mode conversion system of a multimode shield tunneling machine in a composite stratum is designed, which comprises The prediction layer comprises an advanced detection module and is used for acquiring geological structure characteristics and hydrologic information in a preset range in front of the shield machine; The sensing layer comprises a real-time monitoring module and a data fusion module, wherein the real-time monitoring module is used for monitoring and collecting operation parameters and working condition data in the tunneling process of the shield machine in real time and processing the real-time monitoring data through the data fusion algorithm; The data bus layer is used for realizing alignment fusion of the advanced detection data of the prediction layer and the real-time monitoring data of the perception layer in the time dimension and the space dimension based on the distributed stream processing platform; The decision layer is used for generating a mode conversion decision instruction based on advanced detection data and real-time monitoring data aligned in time and based on a mixed model of convolutional neural network CNN geological feature extraction, fuzzy logic control and reinforcement learning; And the execution layer is used for responding to the decision instruction, driving the execution mechanism of the shield machine to complete the conversion among multiple modes and dynamica