CN-121785145-B - Development machine collaborative operation real-time optimization method and system based on digital twin

CN121785145BCN 121785145 BCN121785145 BCN 121785145BCN-121785145-B

Abstract

The invention discloses a development machine collaborative operation real-time optimization method and system based on digital twinning, and relates to the technical field of digital twinning. The method comprises the steps of collecting multi-source sensing data in real time, constructing a nerve Gaussian force field digital twin body, dynamically analyzing the multi-machine operation state by combining stress prediction to generate a cooperative path instruction, adjusting hydraulic pressure by adopting self-adaptive impedance control to obtain cooperative parameters, and optimizing the heading machine cluster operation according to the cooperative parameters. The method solves the technical problems of delay path planning, unreasonable hydraulic torque distribution and unbalanced cutter abrasion caused by the lack of real-time sensing and predicting capability on physical properties of rock stratum in the existing heading machine cluster collaborative operation process, achieves the technical effects of realizing online identification of rock stratum hardness and future stress prediction based on digital twin and nerve Gaussian force fields, improving the precision of multi-machine collaborative path planning and the self-adaptive capability of impedance control, and realizing the improvement of heading efficiency, the enhancement of operation safety and the balance of cutter loss.

Inventors

LI BO
HU CHENGJUN
Pan Gege
Yuan Pengzhe
WANG SHENGZHI

Assignees

中煤(天津)地下工程智能研究院有限公司

Dates

Publication Date: 20260512
Application Date: 20260305

Claims (10)

1. The real-time optimization method for the collaborative operation of the heading machine based on digital twinning is characterized by comprising the following steps: acquiring multisource sensing data of the heading machine cluster in real time through a multisource sensor, and constructing a digital twin body based on the Gaussian neural field; Dynamically analyzing a multi-machine operation state in the digital twin body, extracting a propulsion speed characteristic set and a cutter load characteristic set of the heading machine cluster, obtaining future force field distribution by combining stress prediction, executing space-time conflict interference risk analysis, adjusting a path, and generating a collaborative path planning instruction; adjusting the pressure of each heading machine hydraulic system in the heading machine cluster by utilizing a self-adaptive impedance control algorithm to obtain an impedance cooperative control parameter, wherein the impedance cooperative control parameter is dynamically adjusted according to an online identification result of the rock stratum hardness output by the explicit force field representation in the digital twin body; and carrying out collaborative operation optimization on the heading machine cluster according to the impedance collaborative control parameters and the collaborative path planning instruction.
2. A digital twinning-based heading machine collaborative work real-time optimization method according to claim 1 wherein the digital twinning body contains an explicit object-centric force field representation describing the physical interaction mechanism between the formation and the tool.
3. The method for optimizing the collaborative operation of a heading machine in real time based on digital twinning according to claim 1, wherein the steps of acquiring multisource sensing data of a heading machine cluster in real time by a multisource sensor, constructing a digital twinning body based on a neurogaussian force field include: Capturing a geometric texture image of a rock wall and a pose image of a heading machine cluster by utilizing a multi-view RGB-D camera in a multi-source sensor to obtain a rock wall geometric texture image set and a heading machine pose image set; Sensing a cutter head bearing by using a fiber bragg grating strain sensor in the multi-source sensor to obtain cutter load data; sensing the root of the cutting pick by using a three-way acceleration sensor in the multi-source sensor to obtain rock stratum vibration data; Integrating the rock wall geometric texture image set, the heading machine pose image set, the cutter load data and the rock stratum vibration data to obtain the multi-source sensing data; and jointly encoding the multisource sensing data, determining an explicit force field representation of the heading machine cluster by combining a nerve Gaussian force field and a framework constructed based on a nerve operator, and constructing the digital twin body.
4. A digital twinning-based real-time optimization method for collaborative operation of a heading machine according to claim 3, wherein dynamically analyzing a plurality of machine operation states in the digital twinning body, extracting a propulsion speed feature set and a cutter load feature set of the heading machine cluster, obtaining future force field distribution in combination with stress prediction, performing space-time conflict interference risk analysis, adjusting a path, and generating a collaborative path planning instruction, comprising: traversing the heading machine cluster to perform flow calculation in the digital twin body to obtain a propulsion speed feature set; Analyzing the time-frequency domain characteristics of the cutter load in the digital twin body to obtain a cutter load characteristic set; Carrying out stress prediction based on the explicit force field representation to obtain future force field distribution; acquiring real-time operation path planning of the heading machine cluster, and performing space-time conflict interference risk analysis according to the propulsion speed feature set, the cutter load feature set and future force field distribution to acquire heading machine cluster interference risk analysis results; and executing real-time collaborative path planning adjustment by combining the heading machine cluster interference risk analysis result to obtain a collaborative path planning instruction.
5. The method for optimizing the collaborative operation of a heading machine in real time based on digital twinning as claimed in claim 4, wherein the step of predicting the stress based on the explicit force field representation to obtain future force field distribution comprises the steps of: Acquiring a heading machine pose image set in the multi-source sensing data, and extracting a heading machine pose state characteristic set; And carrying out force field distribution prediction based on the explicit force field representation and the heading machine pose state feature set to obtain the future force field distribution.
6. The method for optimizing the collaborative operation of a heading machine in real time based on digital twinning according to claim 4, wherein the step of obtaining a real-time operation path plan of the heading machine cluster, performing space-time conflict interference risk analysis according to the propulsion speed feature set, the cutter load feature set and future force field distribution, and obtaining a heading machine cluster interference risk analysis result comprises the following steps: performing motion trail projection in the digital twin body according to the real-time operation path planning to construct a space-time occupied body; and carrying out space-time conflict interference risk analysis by using the propulsion speed feature set, the cutter load feature set and the future force field distribution as constraints and combining the space-time occupied body by using a continuous collision detection algorithm to obtain a heading machine cluster interference risk analysis result.
7. The digital twinning-based real-time optimization method for development machine collaborative operation according to claim 1, wherein adjusting the pressure of each development machine hydraulic system in the development machine cluster by using an adaptive impedance control algorithm to obtain an impedance collaborative control parameter comprises: acquiring an online identification result of the rock stratum hardness according to the explicit force field representation and the rock stratum vibration data in the multi-source sensing data; determining a desired stiffness and a desired damping based on the formation hardness on-line recognition result; And aiming at improving the total tunneling efficiency and the cutter load balance, carrying out multi-objective optimization solution on the expected rigidity and expected damping and the online identification result of the rock stratum hardness to obtain the impedance cooperative control parameter.
8. The digital twinning-based real-time optimization method for the development machine collaborative operation according to claim 7, wherein the method for performing multi-objective optimization solution on the expected rigidity and expected damping and the formation hardness online identification result with the aim of improving the total development efficiency and the cutter load balance to obtain the impedance collaborative control parameter comprises the following steps: carrying out lithology recognition based on the online rock stratum hardness recognition result to obtain lithology mechanical characteristics, wherein the lithology mechanical characteristics comprise rock uniaxial compressive strength, elastic modulus and internal friction angle; determining initial impedance cooperative control parameters according to the lithologic characteristics, the expected rigidity and the expected damping; and optimizing and adjusting the initial impedance cooperative control parameters to obtain the impedance cooperative control parameters by taking the improvement of the total tunneling efficiency and the cutter load balance as targets.
9. The digital twinning-based real-time optimization method for the collaborative operation of the heading machines according to claim 8, wherein the impedance collaborative control parameters comprise pressure set values and proportional valve opening of hydraulic systems of the heading machines.
10. A digital twinning-based real-time optimization system for collaborative operation of a heading machine, which is used for implementing the digital twinning-based real-time optimization method for collaborative operation of a heading machine according to any one of claims 1-9, the system comprising: The twin body construction module is used for acquiring multisource sensing data of the heading machine cluster in real time through a multisource sensor and constructing a digital twin body based on the Gaussian neural field; The path planning module dynamically analyzes the multi-machine operation state in the digital twin body, extracts a propulsion speed characteristic set and a cutter load characteristic set of the heading machine cluster, obtains future force field distribution by combining stress prediction, executes space-time conflict interference risk analysis, adjusts the path and generates a collaborative path planning instruction; the impedance control module is used for adjusting the pressure of each heading machine hydraulic system in the heading machine cluster by utilizing a self-adaptive impedance control algorithm to obtain an impedance cooperative control parameter, wherein the impedance cooperative control parameter is dynamically adjusted according to an online identification result of the rock stratum hardness, which is output by the explicit force field representation in the digital twin body; And the operation optimization module is used for carrying out collaborative operation optimization on the heading machine cluster according to the impedance collaborative control parameter and the collaborative path planning instruction.

Description

Development machine collaborative operation real-time optimization method and system based on digital twin Technical Field The invention relates to the technical field of digital twinning, in particular to a real-time optimization method and a real-time optimization system for collaborative operation of a heading machine based on digital twinning. Background Along with the continuous expansion of underground mining, tunnel engineering and large-scale underground space construction scale, the development of the heading machine is gradually carried out towards the direction of clustering and synergism so as to improve the construction efficiency and the resource utilization rate. Under a complex geological environment, the collaborative operation of a plurality of heading machines becomes an important means for improving the engineering progress. However, the rock stratum structure has obvious heterogeneity and uncertainty, the rock stratum hardness, joint fracture development degree and stress distribution difference of different areas are obvious, so that the cutter load fluctuation, uneven abrasion and equipment interference risk increase in the tunneling process are caused, and even equipment faults and construction safety problems are caused when the cutter load fluctuation is serious. The existing heading machine control system is based on a single machine closed-loop control mode, is mainly adjusted by depending on preset parameters or simple feedback signals, lacks real-time accurate perception capability for physical properties of rock stratum, and is difficult to realize prospective prediction for complex working conditions. Meanwhile, under a multi-machine collaborative operation scene, the traditional path planning method is generally based on a static environment model, and the dynamic state and future stress change of equipment cannot be comprehensively considered, so that space-time conflict and interference risks are easy to generate, and the collaborative efficiency is lower. Disclosure of Invention The application provides a real-time optimization method and a real-time optimization system for collaborative operation of a heading machine based on digital twinning, which solve the technical problems of delay path planning, unreasonable hydraulic torque distribution and unbalanced cutter abrasion caused by lack of real-time sensing and prediction capability of physical properties of rock stratum in the existing collaborative operation process of a heading machine cluster. According to a first aspect of the application, a digital twinning-based real-time optimization method for collaborative operation of a heading machine is provided, and the method comprises the following steps: The method comprises the steps of acquiring multisource sensing data of a development machine cluster in real time through multisource sensors, constructing a digital twin body based on a nerve Gaussian force field, dynamically analyzing a multisystem operation state in the digital twin body, extracting a propulsion speed characteristic set and a cutter load characteristic set of the development machine cluster, obtaining future force field distribution by combining stress prediction, executing space-time conflict interference risk analysis, adjusting paths to generate a collaborative path planning instruction, adjusting the pressure of each development machine hydraulic system in the development machine cluster by utilizing a self-adaptive impedance control algorithm to obtain an impedance collaborative control parameter, dynamically adjusting the impedance collaborative control parameter according to an online identification result of rock stratum hardness output by an explicit force field representation in the digital twin body, and carrying out collaborative operation optimization on the development machine cluster according to the impedance collaborative control parameter and the collaborative path planning instruction. In a second aspect of the present application, there is provided a digital twinning-based real-time optimization system for collaborative work of a heading machine, the system comprising: The system comprises a twin body construction module, a path planning module, an impedance control module and an operation optimization module, wherein the twin body construction module acquires multisource sensing data of a heading machine cluster in real time through multisource sensors, builds a digital twin body based on a nerve Gaussian force field, the path planning module dynamically analyzes a multisource operation state in the digital twin body, extracts a propulsion speed feature set and a cutter load feature set of the heading machine cluster, obtains future force field distribution through stress prediction, executes space-time conflict interference risk analysis, adjusts paths to generate a collaborative path planning instruction, the impedance control module adjusts the pressure of each heading machi