A drilling and blasting method tunnel construction equipment fault diagnosis and processing method and system

Abstract

The present application relates to the field of equipment predictive maintenance, and particularly relates to a drilling and blasting tunnel construction equipment fault diagnosis and processing method and system, the method comprising the following steps: synchronously collecting equipment operation state data and tunnel environment geological data by using an equipment and geological collaborative monitoring network; identifying geological interference based on the tunnel environment geological data to extract equipment inherent state features from the equipment operation state data; tracing the equipment fault diagnosis based on the equipment inherent state features to determine the equipment fault propagation path; predicting the equipment fault based on the equipment fault propagation path to obtain equipment fault development trend information; generating an equipment maintenance strategy based on the equipment fault development trend information, and executing the equipment maintenance strategy to achieve equipment fault processing. The present application effectively realizes accurate prediction and efficient processing of equipment faults, improves the operation and maintenance efficiency of drilling and blasting tunnel construction equipment, and reduces the risk of tunnel construction interruption.

Description

TECHNICAL FIELD

[0001] The present application relates to the field of equipment predictive maintenance, and in particular to a drilling and blasting tunnel construction equipment fault diagnosis and processing method and system. BACKGROUND

[0002] In the drilling and blasting tunnel construction scene, equipment fault diagnosis and processing usually adopts an isolated equipment state monitoring method, by installing vibration sensors, temperature sensors and pressure sensors on key equipment, collecting single physical parameters of equipment operation, and performing fault alarm based on pre-set empirical threshold. Fault diagnosis relies on the experience of on-site maintenance personnel or simple signal analysis techniques, such as frequency spectrum analysis, waveform comparison, etc., to identify abnormal states of equipment. In terms of processing, the plan maintenance or after-event maintenance mode is generally adopted, that is, the equipment is maintained according to a fixed cycle, or emergency maintenance measures are taken after the equipment fails and stops. Some more advanced solutions attempt to introduce statistical analysis based on historical fault data to predict equipment life, but the analyzed equipment operation data fails to effectively remove the interference of the tunnel environment.

[0003] The above technologies have the following deficiencies. First, the equipment operation data is highly coupled with the complex and variable geological environment data, resulting in high false alarm rate and low accuracy in identifying real performance degradation of equipment and abnormality caused by external geological disturbance. Second, the fault diagnosis method is mostly superficial, lacking the ability to trace the root cause and the propagation path among multiple components and devices, so that the maintenance decision remains on the surface and it is difficult to fundamentally prevent chain failures. At the same time, the existing prediction model fails to consider the coupling effect of geological time-varying dynamics and equipment degradation mechanism, resulting in a large deviation between the predicted trend and the actual situation, and thus unable to provide reliable basis for proactive maintenance. Finally, the generation of maintenance strategies mostly relies on static rules and centralized decision-making, lacking adaptive optimization and distributed collaborative execution capabilities based on real-time state evolution, resulting in lagging maintenance response, poor resource allocation, and affecting the continuity of tunnel construction. SUMMARY

[0004] In view of the defects in the prior art, the present application provides a drilling and blasting tunnel construction equipment fault diagnosis and processing method and system.

[0005] In order to achieve the above object, in a first aspect, the present application provides a drilling and blasting tunnel construction equipment fault diagnosis and processing method, the method comprising the following steps: synchronously collecting equipment operation state data and tunnel environment geological data by using equipment and geological collaborative monitoring network; identifying geological interference based on the tunnel environment geological data to extract equipment inherent state characteristics from the equipment operation state data; determining the equipment fault propagation path by tracing the equipment fault diagnosis based on the equipment inherent state characteristics; obtaining equipment fault development trend information by predicting the equipment fault according to the equipment fault propagation path; generating equipment maintenance strategy based on the equipment fault development trend information, and executing the equipment maintenance strategy to realize equipment fault processing. The present application realizes synchronous data collection by equipment and geological collaborative monitoring, extracts equipment inherent state characteristics through geological interference identification, accurately completes fault diagnosis tracing and propagation path determination, obtains fault development trend according to prediction and generates maintenance strategy; effectively solves the problem of diagnosis deviation caused by geological environment interference, realizes accurate fault prediction and efficient processing, reduces the risk of construction interruption, and improves the operation and maintenance efficiency of drilling and blasting tunnel construction equipment.

[0006] Optionally, the synchronous collection of equipment operation state data and tunnel environment geological data by using equipment and geological collaborative monitoring network comprises: constructing the equipment and geological collaborative monitoring network by deploying equipment monitoring sensor groups on construction equipment and environment monitoring sensor groups on tunnel working face; obtaining the equipment operation state data including vibration acceleration, oil pressure, oil temperature and motor current based on the equipment monitoring sensor groups; obtaining the tunnel environment geological data including rock drilling response data and rock mass structure vibration data according to the environment monitoring sensor groups; and performing data cleaning and space-time alignment on the equipment operation state data and the tunnel environment geological data. The present application comprehensively collects equipment operation state data and tunnel environment geological data, processes through cleaning and space-time alignment, guarantees data integrity and consistency, provides high-quality data support for subsequent geological interference identification, fault diagnosis and other links, avoids the influence of data defects on analysis results, and lays a reliable data foundation for fault diagnosis and processing.

[0007] Optionally, the step of identifying geological interference in the equipment operating status data based on the tunnel environmental geological data to extract inherent state features of the equipment includes: extracting time-varying geological interference signals based on the tunnel environmental geological data; adaptively decoupling the equipment operating status data according to the time-varying geological interference signals to obtain equipment state-related signals; and obtaining equipment stability features as inherent state features of the equipment based on the equipment state-related signals in various tunnel environments, including time-domain features, frequency-domain features, and nonlinear features. This invention extracts time-varying interference signals based on tunnel geological data, adaptively decouples equipment operating data and obtains multi-dimensional stability features, removes geological environmental interference, accurately captures the inherent state of the equipment, and is adaptable to various tunnel environments, significantly improving the accuracy and applicability of equipment state feature extraction, and providing accurate basis for subsequent fault diagnosis.

[0008] Optionally, the step of adaptively decoupling the equipment operating status data to obtain equipment state-related signals based on the time-varying geological interference signals includes: analyzing the time-varying geological interference signals to obtain time-varying geological dynamic parameters, thereby establishing a dual-channel feedforward decoupling filter; and filtering the equipment operating status data using the dual-channel feedforward decoupling filter to obtain the equipment state-related signals. This invention obtains dynamic parameters by analyzing the time-varying geological interference signals, establishes a dual-channel feedforward decoupling filter to achieve data filtering, efficiently separates geological interference components from equipment operating data, obtains pure equipment state-related signals, improves the accuracy of extracting inherent equipment state features, and provides high-purity data support for subsequent fault analysis.

[0009] Optionally, the step of determining the equipment fault propagation path by diagnosing and tracing the equipment fault based on the inherent state characteristics of the equipment includes: obtaining multi-scale state variables of the equipment based on the inherent state characteristics of the equipment, constructing a hierarchical network of equipment states by combining the equipment operating mechanism and correlation relationships; obtaining effective information about the equipment states based on the hierarchical network of equipment states, performing macro-causal state filtering on the effective information about the equipment states to obtain a set of macro-critical states of the equipment; constructing a cross-scale equipment fault propagation model based on the set of macro-critical states of the equipment, and determining the equipment fault propagation path by combining the equipment structure and functional coupling relationships. This invention, by constructing a hierarchical network of equipment states, filtering a set of macro-critical states, and constructing a cross-scale fault propagation model, accurately obtains the equipment fault propagation path, solving the problem of fuzzy fault tracing in traditional diagnosis, providing a clear direction for subsequent fault prediction and maintenance strategy formulation, and improving the depth and relevance of fault diagnosis.

[0010] Optionally, the step of obtaining effective equipment state information based on the equipment state hierarchy network and performing macro-causal state filtering on the effective equipment state information to obtain a set of macro-key equipment states includes: constructing a state transition probability matrix of the equipment's multi-scale state variables based on the equipment state hierarchy network to obtain the effective equipment state information; and filtering the set of macro-key equipment states through cross-scale comparative analysis based on the effective equipment state information. This invention obtains effective equipment state information through a state transition probability matrix to filter a set of macro-key states, simplifying data dimensions while retaining core state information, reducing the impact of redundant data on diagnostic efficiency, quickly identifying fault-related key states, and improving the efficiency and accuracy of fault tracing.

[0011] Optionally, the step of predicting equipment failures based on the equipment failure propagation path to obtain equipment failure development trend information includes: constructing a multi-dimensional spatial model of equipment failures based on the equipment failure propagation path; constructing an equipment failure evolution prediction model based on the equipment degradation mechanism and the multi-dimensional spatial model of equipment failures; performing forward fitting of equipment failures based on the equipment failure evolution prediction model to obtain the equipment failure evolution time-series trajectory; and performing quantitative analysis on the equipment failure evolution time-series trajectory to obtain the equipment failure development trend information. This invention constructs a multi-dimensional spatial model and an evolution prediction model of equipment failures, fits the failure evolution time-series trajectory, and quantitatively analyzes the trend, achieving accurate prediction of the development trend of equipment failures, grasping the direction of failure development in advance, breaking the passive maintenance mode, providing support for reserving maintenance time and formulating scientific response plans, and reducing failure losses.

[0012] Optionally, the step of constructing an equipment failure evolution prediction model based on the equipment degradation mechanism and the multi-dimensional space model of equipment failure includes: obtaining multi-physics field equipment degradation constraints based on the equipment degradation mechanism; and coupling and embedding the multi-physics field equipment degradation constraints into the multi-dimensional space model of equipment failure to construct the equipment failure evolution prediction model. This invention couples and embeds multi-physics field equipment degradation constraints into the multi-dimensional space model of failure, constructing a failure evolution prediction model that better reflects the actual degradation patterns of equipment, overcomes the shortcomings of traditional models that ignore physical constraints, improves the accuracy and reliability of failure development trend prediction, and provides a theoretical basis for subsequent maintenance strategy formulation.

[0013] Optionally, the step of generating an equipment maintenance strategy based on the equipment fault development trend information and executing the equipment maintenance strategy to handle equipment faults includes: establishing a fault operation and maintenance decision model based on the equipment fault development trend information and combined with digital twins; establishing a distributed autonomous decision-making mechanism based on the fault operation and maintenance decision model and combined with multi-agents; obtaining candidate maintenance strategies through the distributed autonomous decision-making mechanism; filtering the candidate maintenance strategies based on a multi-objective optimization algorithm to obtain the equipment maintenance strategy; handling equipment faults according to the equipment maintenance strategy and monitoring equipment status recovery indicators; and performing closed-loop feedback optimization on the equipment maintenance strategy based on the equipment status recovery indicators to achieve equipment fault handling. This invention combines digital twins, multi-agents, and multi-objective optimization to obtain equipment maintenance strategies, and achieves fault handling through closed-loop feedback optimization, balancing maintenance effectiveness and efficiency, generating highly adaptable personalized solutions, improving response speed through distributed decision-making, and continuously improving maintenance quality through closed-loop optimization, thus efficiently handling equipment faults.

[0014] Secondly, this invention provides a fault diagnosis and handling system for drill-and-blast tunnel construction equipment. The system executes the fault diagnosis and handling method for drill-and-blast tunnel construction equipment provided by this invention. The system includes input devices, output devices, a processor, and a memory, which are interconnected. The memory stores a computer program, which includes program instructions, and the processor is configured to call the program instructions. This invention, through the collaboration of high-performance hardware, achieves automated and precise operation from data acquisition and analysis to fault handling, avoiding human error, improving the intelligence level and efficiency of fault maintenance for drill-and-blast tunnel construction equipment, and ensuring smooth construction. Attached Figure Description

[0015] Figure 1 This is a flowchart of a fault diagnosis and handling method for tunnel construction equipment using the drill-and-blast method according to an embodiment of the present invention;

[0016] Figure 2 This is a framework diagram of a fault diagnosis and handling system for tunnel construction equipment using the drill-and-blast method, according to an embodiment of the present invention. Detailed Implementation

[0017] Specific embodiments of the present invention will now be described in detail. It should be noted that the embodiments described herein are for illustrative purposes only and are not intended to limit the invention. In the following description, numerous specific details are set forth in order to provide a thorough understanding of the invention. However, it will be apparent to those skilled in the art that these specific details are not necessary to practice the invention. In other instances, well-known circuits, software, or methods have not been specifically described to avoid obscuring the invention.

[0018] Throughout this specification, references to "an embodiment," "an embodiment," "an example," or "an example" mean that a particular feature, structure, or characteristic described in connection with that embodiment or example is included in at least one embodiment of the invention. Therefore, the phrases "in an embodiment," "in an embodiment," "an example," or "an example" appearing in various places throughout the specification do not necessarily refer to the same embodiment or example. Furthermore, specific features, structures, or characteristics can be combined in one or more embodiments or examples in any suitable combination and / or sub-combination. Moreover, those skilled in the art will understand that the illustrations provided herein are for illustrative purposes and are not necessarily drawn to scale.

[0019] Please see Figure 1 One embodiment of the present invention provides a method for fault diagnosis and handling of drilling and blasting tunnel construction equipment, the method comprising the following steps:

[0020] S1. Utilize the equipment and geological collaborative monitoring network to synchronously collect equipment operation status data and tunnel environmental geological data.

[0021] In this embodiment, the equipment and geological collaborative monitoring network consists of equipment monitoring sensor groups deployed on construction equipment such as rock drilling rigs and wet spraying rigs, and environmental monitoring sensor groups deployed at the tunnel face and surrounding key areas. The equipment monitoring sensor groups are deeply integrated into the equipment's hydraulic system, transmission system, and motors, and include, but are not limited to, high-precision vibration acceleration sensors, embedded oil pressure and oil temperature sensors, and non-invasive current transformers; the environmental monitoring sensor groups include, but are not limited to, drilling response monitors and vibration sensors. All of the above sensors transmit raw data to a cloud data platform in real time via a wireless network, laying the data foundation for subsequent collaborative analysis.

[0022] Equipment operating status data, including but not limited to vibration acceleration, oil pressure, oil temperature, and motor current, is acquired using a sensor array. This data is then analyzed and extracted in depth on a data platform to accurately determine the equipment's condition. Vibration acceleration signals are decomposed into energy distributions at different frequency bands using a Fast Fourier Transform (FFT). Abnormal energy concentrations in the high-frequency band often indicate early wear of critical components such as bearings and gears, while low-frequency vibrations synchronized with the drilling rhythm reflect the equipment's operational stability. Hydraulic system oil pressure and temperature data are used for preliminary early warning; for example, if the drilling rig's hydraulic pressure remains consistently high while the drilling speed decreases, it initially indicates an encounter with hard rock formations or a risk of stuck drill bit. Motor current data is compared with the equipment's power output to detect abnormal current harmonics or overload signals.

[0023] Tunnel environmental geological data is obtained through an environmental monitoring sensor array, including but not limited to rock drilling response data and rock mass vibration data. Rock drilling response data (drilling speed, thrust, torque) is the best data for identifying rock mass strength. Through data fusion algorithms, these parameters are cross-validated with the corresponding equipment operating status data to dynamically plot the drillability index of the rock in front of the tunnel face. Based on the rock mass vibration data, the integrity and joint development of the rock mass are inferred by analyzing the blasting vibration wave velocity, amplitude, and frequency, in order to assess the stability of the surrounding rock.

[0024] In one optional embodiment, a geological inversion model is constructed, which integrates historical geological survey data and tunnel environmental geological data, continuously iterates and optimizes, and outputs key geological inversion parameters such as rock mass quality grade and geostress distribution, providing a scientific basis for the dynamic adjustment of construction plans.

[0025] In this embodiment, equipment operation status data and tunnel environmental geological data undergo data cleaning and spatiotemporal alignment. Data cleaning ensures data usability by performing preliminary filtering, outlier removal, handling data loss, and correcting sensor drift. Spatiotemporal alignment enables collaboration between equipment and geological data. All data is marked with high-precision timestamps, and each set of equipment status data is bound to its precise coordinates (mileage, axis offset, elevation) within the tunnel using a positioning system deployed inside the tunnel. Similarly, each set of geological monitoring data is associated with the spatial coordinates of its collection point. A unified spatiotemporal coordinate system is established within the data platform to accurately match and fuse equipment operation data with geological data under the same spatiotemporal coordinates.

[0026] S2. Based on the tunnel environmental geological data, perform geological interference identification on the equipment operating status data to extract the inherent state characteristics of the equipment.

[0027] In this embodiment, geological disturbance time-varying signals are obtained by fusing and feature mining tunnel environmental geological data. Key analytical data include: rock drilling response data, i.e., the time series of ternary parameters of drilling speed, thrust, and torque recorded in real time by the drilling rig during drilling, which directly reflects the spatial variation of rock strength at the tunnel face; rock mass vibration data, which, after processing, yields time-series signals reflecting the integrity of the rock mass, including vibration velocity, dominant frequency, and attenuation coefficient; and dynamic geological inversion parameters, calculated in real time by the geological inversion model based on the first two types of data and historical exploration data. These heterogeneous data streams are unified into a spatiotemporal framework based on tunnel mileage and high-precision timestamps. A dimensionless time-varying signal is generated using a Kalman filter as the geological disturbance time-varying signal. Its peaks correspond to deteriorating geological conditions (such as encountering fault fracture zones, hard rock masses, or stress concentration areas), while troughs correspond to favorable geological conditions. The geological disturbance time-varying signal reflects the dynamic characteristics of the additional load and excitation caused by the geological environment on construction equipment, providing a basis for subsequent equipment state decoupling.

[0028] In this embodiment, in-depth analysis of the time-varying geological disturbance signal is performed to obtain geological time-varying dynamic parameters. The geological disturbance signal is considered as an input excitation to the equipment-geological coupled system, and a transfer function from the geological disturbance signal to the typical equipment response is established using the recursive least squares method. Based on the transfer function, key parameters including gain, damping ratio, and dominant frequency are obtained as dynamically changing geological time-varying dynamic parameters, quantitatively describing the intensity and pattern of the influence of current geological conditions on the equipment's dynamic behavior.

[0029] The above transfer function satisfies the following relationship:

[0030]

[0031] in, For transfer functions, Let Z be the variable for transformation. This represents a typical device response. This is a time-varying signal caused by geological interference. and These are the transfer function coefficients.

[0032] It should be noted that the transfer function coefficients are dynamically adjusted using an adaptive algorithm.

[0033] Furthermore, a dual-channel feedforward decoupling filter is dynamically established using geological time-varying dynamic parameters. This includes: a geological interference prediction channel, which uses current and historical geological interference time-varying signals, combined with the aforementioned transfer function, to obtain the components of geological interference generated in equipment vibration, oil pressure, and other signals; and an equipment status compensation channel, which receives raw equipment operating status data and removes the geological interference components obtained from the geological interference prediction channel. The filter coefficients are dynamically fine-tuned using an adaptive algorithm to ensure that they can track the drift of interference characteristics caused by changes in geological conditions, thereby guaranteeing accurate removal of geological influences.

[0034] In this embodiment, during the signal separation execution phase, a dual-channel feedforward decoupling filter is used to filter the equipment operating status data to obtain equipment status-related signals. Taking the vibration acceleration signal of a rock drilling rig as an example, the original vibration signal is a complex sequence that mixes the equipment's own mechanical operation, geological interactions, and background noise. When the filter is working, it acquires the original vibration signal in real time, while simultaneously inputting the time-varying geological interference signal and its derived dynamic parameters. Inside the filter, the geological interference prediction channel generates a simulated interference signal corresponding to the current geological conditions. Then, in the equipment status compensation channel, the simulated interference signal is subtracted from the original mixed signal, and the output signal is the equipment status-related signal. This retains components directly related to the health status of the equipment's mechanical body, such as high-frequency resonance caused by bearing defects, and significantly reduces low-frequency load fluctuations and random impact responses caused by changes in rock hardness. As a reliable data source for fault list generation and model analysis, the equipment status-related signal greatly improves diagnostic accuracy and avoids misjudging severe geological conditions as internal equipment faults.

[0035] The above-mentioned device status-related signals satisfy the following relationship:

[0036]

[0037] in, For equipment status related signals, The original mixed signal, To simulate interference signals.

[0038] In various tunnel environments, equipment stability characteristics are obtained from the decoupled equipment state-related signals as inherent equipment state features. A feature extraction engine is embedded within the digital twin platform to perform multi-dimensional, holographic feature mining on the pure equipment state-related signals. First, time-domain feature extraction includes calculating the signal's effective value and peak factor, providing an intuitive trend in operational health. Second, frequency-domain features are extracted by converting vibration, current, and other signals to the frequency domain using Fast Fourier Transform (FFT), analyzing the fault characteristic frequencies of specific bearings and gears, their harmonic amplitude changes, and the emergence and evolution of sidebands. Frequency-domain information can accurately locate specific faulty components (such as inner ring faults in spindle bearings) during equipment fault diagnosis. Finally, nonlinear features, such as the multi-scale entropy of the signal, are extracted. This is highly sensitive to early, weak nonlinear dynamic anomalies and can reveal potential faults such as early fatigue crack propagation and deterioration of lubrication conditions in components.

[0039] The above time-domain features satisfy the following relationship:

[0040]

[0041]

[0042] in, The effective value of the signal. The number of sampling points. For sampling point index, These are discrete sampled values ​​of the signal. This is the peak factor.

[0043] The above frequency domain features satisfy the following relationship:

[0044]

[0045] in, It is a frequency domain signal. For Fourier transform operators, These are signals related to the device status.

[0046] The above nonlinear characteristics satisfy the following relationship:

[0047]

[0048] in, For multi-scale entropy, For the first Sample entropy at each time scale Index for the time scale.

[0049] In an optional embodiment, a stability feature vector is constructed based on time-domain features, frequency-domain features, and nonlinear features to characterize the inherent mechanical state of the equipment. This stability feature vector is stored in a historical database for predicting equipment status and reliability in construction simulations and task scheduling. It also serves as a core basis for determining whether the equipment can adapt to complex geological environments, thus achieving a shift from passive maintenance to proactive prediction.

[0050] S3. Based on the inherent state characteristics of the equipment, perform equipment fault diagnosis and source tracing to determine the equipment fault propagation path.

[0051] In this embodiment, based on the inherent state characteristics of the equipment, multi-scale state variables of the equipment are obtained through multi-scale analysis. At the micro scale, the variables include features reflecting the behavior of specific components, such as the amplitude of a specific frequency of the bearing, the energy of the gear meshing sideband, and the approximate entropy of oil pressure pulsation. At the meso scale, equipment subsystem-level variables are formed through feature fusion, such as the hydraulic system stability index (synthesized from the characteristics of the main pump pressure, oil temperature, and the operating current of multiple valve cores) and the transmission system health (fused from the vibration characteristics of each level of transmission shaft). At the macro scale, equipment whole-machine variables are defined, such as the equipment comprehensive efficiency coefficient and the operational stability index.

[0052] Furthermore, a hierarchical network of equipment states is constructed by combining the equipment's operating mechanisms and interrelationships. This network is a directed weighted graph, where nodes are state variables at various scales, and edges represent causal or strong correlations between variables. The weights of the relationships are learned from historical data and verified and corrected based on the equipment's physical model (such as hydraulic circuit diagrams and transmission chain dynamic equations). For example, at the microscale, the node "bearing inner ring failure frequency amplitude" will point to the mesoscale node "total vibration of the spindle box," ultimately affecting the macroscale "overall machine stability index"; while the node "main pump pressure fluctuation" will simultaneously affect multiple mesoscale nodes such as "hydraulic system stability" and "actuator response delay." The hierarchical network of equipment states connects isolated feature points to form a topology that comprehensively reflects the causal interactions within the equipment's internal states.

[0053] The above device state hierarchy network satisfies the following relationship:

[0054]

[0055] in, This represents a device state hierarchy network. For a set of nodes, Let it be the set of edges.

[0056] In this embodiment, based on a device state hierarchical network, the importance of each state variable in the network is quantified by calculating the effective information content of the state. The core of this approach is constructing a state transition probability matrix. The time-series data of each state variable is discretized and divided into multiple state levels such as "normal," "warning," and "abnormal." For each state variable in the network, its conditional probability of transitioning to each state level at the next time step is calculated using its parent node's state as a condition, thus forming the state transition probability matrix for that variable. Next, combining the state transition probability matrix, the effective information content is calculated using the effective information measurement method in causal emergence theory as the effective information content of the device state. The determinism and degeneracy of the state transition probability matrix are evaluated by intervening in the input distribution and calculating its KL divergence with the true output distribution.

[0057] First, the state variables are discretized into state levels, satisfying the following relationship:

[0058]

[0059] in, Represents the set of state levels. For the first Each status level This is an index for the status level.

[0060] Secondly, obtain the state. Transition to state The probabilities are used to construct the state transition probability matrix, satisfying the following relationship:

[0061]

[0062] in, For state Transition to state The probability, This indicates the calculation of conditional probability. Indicates at time Status is , Indicates at time Status is , This refers to the state of the parent node.

[0063] Finally, the above effective information content satisfies the following relationship:

[0064]

[0065] in, For effective information content, Represents the mathematical expectation. To intervene in the distribution, Let KL divergence be the KL divergence. This represents the conditional probability after intervention. For state variables, This indicates intervention in the state variables. This represents the marginal probability.

[0066] It should be noted that the higher the amount of effective information about a device's state in a state variable, the stronger its causal effect in the network. That is, the state variable can influence the future evolution of the device with greater certainty, and the richer the effective information it carries, thus giving all nodes in the network a quantitative causal importance index.

[0067] The theory of causal emergence states that in complex multi-scale systems, macroscopic scales often exhibit more powerful and simpler causal laws than microscopic scales.

[0068] In this embodiment, the macroscopic key state set of the equipment is selected through cross-scale comparative analysis based on the effective information content of the equipment state. First, the average effective information content of all state variables within each micro, meso, and macro scale is calculated. According to the theory of causal emergence, the effective information content of macro-scale variables is significantly higher than that of micro-scale variables, indicating that the overall causal laws of the equipment can be more clearly grasped from a macro perspective. Next, within each scale, variables are ranked according to their effective information content, and state variables with effective information content higher than the average level of that scale are selected as candidate key variables for that scale. Then, a cross-scale causal coverage analysis is performed. For example, if the causal effect of a candidate key variable at the macro scale (such as the hydraulic system stability index) can cover and explain the evolutionary behavior of multiple candidate key variables at the meso and micro scales (such as main pump pressure fluctuation and valve core response entropy), then this macro variable is included in the final macroscopic key state set of the equipment. Conversely, if a micro variable has a high effective information content, but its causal influence cannot be summarized by any macro variable, it may represent local noise or redundant information and will be discarded in the selection process. By combining top-down and bottom-up screening methods, a concise set of key macroscopic states of the equipment is obtained, which includes core state variables that have high causal power at the macroscopic scale and can predict the overall behavior evolution of the equipment to the greatest extent.

[0069] In this embodiment, a cross-scale equipment fault propagation model is constructed based on the selected set of macroscopic key states of the equipment to ultimately determine the equipment fault propagation path. The cross-scale equipment fault propagation model is a dynamic, weighted directed graph model, where the nodes are the set of macroscopic key states (such as the vibration intensity of the whole machine, the stability of the hydraulic system, and the drive efficiency). The directed edges between nodes and their weights are dynamically updated by analyzing the time-delay mutual information and causal flow between key state variables.

[0070] Furthermore, to accurately depict the transmission chain of faults, the cross-scale equipment fault propagation model deeply integrates the coupling relationship between equipment structure and function. For example, structurally, the model identifies the physical connection points between the hydraulic system and the transmission system (such as the interface between the hydraulic motor and the gearbox) based on the equipment's three-dimensional digital twin model; functionally, it clarifies the signal flow between "drive commands" and "actuator actions" based on the control logic. When a key state variable (such as hydraulic system stability) is identified as the origin of an anomaly, the model simulates the most probable equipment fault propagation path based on its causal strength and time series relationship with downstream variables (such as drive efficiency and overall machine vibration). For example, a typical fault path might be determined as: "Main pump wear (microscopic) → decreased hydraulic system stability (macroscopic key state) → drive torque fluctuation → increased transmission chain impact load → bearing overheating → overall machine vibration intensity exceeding limits." The finally determined multiple equipment fault propagation paths are visualized and rendered in the three-dimensional twin scene of equipment fault diagnosis and stored in the fault list as part of the equipment fault diagnosis conclusion, achieving a fundamental leap from condition monitoring to causal tracing.

[0071] S4. Based on the equipment fault propagation path, perform equipment fault prediction to obtain equipment fault development trend information.

[0072] In this embodiment, a multi-dimensional space model of equipment faults is constructed based on a determined equipment fault propagation path, providing a quantifiable mathematical framework for fault evolution. This model is a high-dimensional state space, where each coordinate axis represents a key state variable identified along the fault propagation path. These variables originate from the macroscopic key state set of the equipment; they also include microscopic features that play a crucial role in the propagation path (such as the specific frequency amplitude of a bearing). Taking a typical fault propagation path as an example, such as "pump wear → pressure fluctuation → bearing impact → overall machine vibration," this path is represented in the multi-dimensional space as an abstract trajectory composed of corresponding state variables. A subspace model is independently constructed for each important fault propagation path. Within the model, a relatively healthy piece of equipment will cluster in a compact region near the spatial origin, while the occurrence and development of faults manifest as the movement and diffusion of state points along a specific path (i.e., the fault propagation direction) in the high-dimensional space.

[0073] The above-mentioned multidimensional space model of equipment failure satisfies the following relationship:

[0074]

[0075] in, For state vectors, For time, Let be the state evolution function. For external input.

[0076] In this embodiment, physical rules are injected based on the equipment degradation mechanism, i.e., multi-physics field equipment degradation constraints are obtained. These constraints are mathematical expressions derived from the equipment's design principles, materials science, and physicochemical laws, used to limit the arbitrariness of state variables evolving in the multi-dimensional space of the fault. Multi-physics field equipment degradation constraints are constructed by integrating technical manuals provided by the equipment manufacturer, failure analysis reports from historical maintenance records, and physical simulation models. For example, in the mechanical field, the constraints manifest as the relationship between bearing wear depth and accumulated vibration energy, or the functional relationship between gear fatigue crack propagation rate and stress intensity; in the fluid dynamics and thermodynamics field, the constraints are reflected in the equation showing that hydraulic oil viscosity decreases with increasing contaminants, leading to equipment efficiency degradation, or the positive feedback relationship between brake friction pair temperature and wear rate; in the electromagnetic field, the constraints manifest as the correlation model between motor winding insulation aging and partial discharge activity intensity. These degradation equations from different physical fields together constitute a set of constraint equations as multi-physics field equipment degradation constraints, ensuring that the fault evolution simulation follows basic physical laws and providing physical boundaries for the prediction model.

[0077] The aforementioned multiphysics device degradation constraints include mechanical wear constraints, thermodynamic constraints, and electromagnetic field constraints, including:

[0078] Mechanical wear constraints satisfy the following relationship:

[0079]

[0080] in, This refers to mechanical wear. For time, The wear coefficient is... For the action force, This refers to relative velocity.

[0081] Thermodynamic constraints satisfy the following relationship:

[0082]

[0083] in, For temperature, For time, For heat generation rate, For heat capacity, For heat dissipation coefficient, The ambient temperature.

[0084] Electromagnetic field constraints satisfy the following relationship:

[0085]

[0086] in, To indicate the degree of insulation aging, For time, This is the material aging rate constant. The intensity of partial discharge activity. The discharge intensity affects the index. It is a natural constant. This is the temperature influence coefficient. This refers to the winding temperature.

[0087] Furthermore, the aforementioned multiphysics-based equipment degradation constraints are coupled and embedded into the multidimensional space model of equipment failure to establish an equipment failure evolution prediction model. This is achieved by introducing a physics-based penalty term into the state space. The state evolution of the multidimensional space model is viewed as a controlled differential dynamic system. In this case, state transition relationships learned solely from historical data may produce physically infeasible predictions. Embedding constraints adds constraints to this dynamic system to make the constraints conform to physical laws, thereby establishing an equipment failure evolution prediction model. This is a hybrid model that integrates data-driven and physical mechanisms. It possesses the flexibility to learn failure evolution patterns from historical big data while being strictly constrained by the multiphysics degradation mechanism, thus obtaining future failure evolution scenarios that conform to both statistical laws and physical laws.

[0088] The above equipment failure evolution prediction model satisfies the following relationship:

[0089]

[0090]

[0091] in, For state vectors, For time, Let be the state evolution function. For external input, For physical constraint evolution function, For random disturbance terms, Indicates mechanical wear constraint. Indicates thermodynamic constraints. Indicates electromagnetic field constraint. This indicates transpose.

[0092] In this embodiment, a forward fitting of the constructed equipment failure evolution prediction model is performed to obtain the equipment failure evolution time-series trajectory. Using the equipment's current real-time state point in the multi-dimensional space as the initial condition and the estimated equipment load from planned future construction tasks as the external excitation input, the equipment failure evolution prediction model iteratively solves the problem in the state space at discrete time steps (e.g., 1 hour, 1 day, 1 week). At each time step, the model first calculates the initial change in state based on the current state and load using a data-driven component; subsequently, the physical constraint component verifies and corrects this change to ensure its rationality, thereby obtaining the equipment failure evolution time-series trajectory.

[0093] The forward fitting of the above equipment faults satisfies the following relationship:

[0094]

[0095] in, For the first The state vector at each time step For the index of time steps, For the first The state vector at each time step For time step, For the first The state evolution function at each time step For the first Physical constraint evolution function at each time step, For the first External input at each time step, For the first The physical time of a time step For the first Discrete random perturbation term at each time step.

[0096] For example, when predicting the vibration level of the main bearing of a rock drilling rig, the model comprehensively considers the current vibration value, the future drilling workload (rock hardness estimation), and the physical constraint of irreversible bearing wear, predicting the trajectory of the vibration level increasing over time. Through iterative calculations and the use of the Monte Carlo method to account for uncertainties, one or more state-space trajectories (corresponding to different confidence intervals) starting from the current state point and extending along the future time axis are generated as the time-series trajectory of equipment failure evolution. This quantitatively depicts the evolution process of key state variables (such as vibration, temperature, and efficiency) over a period of time, clearly showing the complete timeline of the failure's development from the current state through the propagation path.

[0097] In this embodiment, a deep quantitative analysis is performed on the fitted equipment fault evolution time-series trajectory to extract equipment fault development trend information. First, the remaining useful life of the equipment is calculated by identifying the time points when the equipment fault evolution time-series trajectory reaches the fault threshold, and its probability distribution is given. Second, the fault evolution rate, i.e., the rate of change of key state variables on the trajectory, is analyzed to determine whether the fault is in a slow incubation period or about to enter an accelerated failure period. Finally, the scope of fault impact is assessed to determine which functional indicators of the equipment are mainly affected by the fault (such as decreased accuracy, power loss, etc.). The above quantitative information, including the expected value and confidence interval of the remaining useful life of the equipment, the key inflection points of the evolution rate, and the main affected functional indicators, is integrated into equipment fault development trend information. This information is then pushed to the equipment fault list, triggering different levels of warnings (pop-up windows, highlighting, model coloring).

[0098] S5. Generate an equipment maintenance strategy based on the equipment fault development trend information, and execute the equipment maintenance strategy to handle equipment faults.

[0099] In this embodiment, a fault operation and maintenance decision-making model is established based on equipment fault development trend information and combined with digital twin technology. This model is based on the digital twin model of the equipment for simulation decision-making. When the equipment fault development trend information shows that the main bearing of a rock drilling rig has only 50 hours of remaining life and is in an accelerated deterioration period, the decision-making model uses the current fault development trend information as the initial condition to simulate the execution of various potential operation and maintenance plans, such as: "immediately shut down the machine and dispatch a maintenance team to the site to replace the bearing", "adjust subsequent construction tasks and reduce its load by 30% to extend its life to the planned maintenance time window", and "activate the same model of spare equipment to replace its task". The simulation comprehensively considers the overall construction plan, spare parts inventory and maintenance resource distribution, and the status of skilled worker teams. By calculating the impact of each plan on the construction progress, the costs incurred (including downtime costs, spare parts costs, and labor costs), and the expected effect of equipment status recovery, the model pre-evaluates the comprehensive utility of each potential maintenance strategy, forming a decision space containing multi-dimensional evaluation indicators, providing a quantitative basis for subsequent automatic decision-making.

[0100] Furthermore, based on the aforementioned fault operation and maintenance decision-making model, a distributed autonomous decision-making mechanism is established using multiple agents to achieve efficient and collaborative operation and maintenance response. Multiple agents are created, including: a task scheduling agent (responsible for ensuring overall construction progress), an equipment health management agent (focusing on equipment reliability), a resource management agent (managing spare parts, tools, and maintenance personnel), and a field safety agent (ensuring that operation and maintenance strategies comply with safety regulations). These agents engage in distributed game theory through methods such as utility function coordination, ultimately forming one or more candidate maintenance strategies that satisfy the core interests of all parties under current constraints. This improves the agility and robustness in responding to sudden faults in complex and dynamic unmanned tunnel construction environments.

[0101] When a fault warning is triggered, multiple agents autonomously negotiate and collaborate within the shared decision space provided by the fault operation and maintenance decision-making model, based on preset rules and objectives. For example, the equipment health management agent, based on the remaining useful life of the equipment, will prioritize and strongly recommend "immediate repair"; the task scheduling agent will assess the irreplaceability of the equipment on the current critical path and propose an alternative solution of "delaying repair until the current drilling process is completed"; the resource management agent will simultaneously check the bearing spare parts inventory and the location of the maintenance team to provide resource accessibility confirmation for feasible maintenance strategies; and the field safety agent will assess the safety of the maintenance strategy execution.

[0102] In this embodiment, a series of candidate maintenance strategies are obtained through a distributed autonomous decision-making mechanism. Candidate maintenance strategies are a set of preliminary fault solutions proposed by various agents after game theory and negotiation. Candidate strategies are structured data objects that define the set of maintenance actions, their timing, required resources, and expected execution effects. For example, for a rock drilling rig bearing failure, possible candidate strategies include: Strategy A (Immediate Repair): Defines the "bearing replacement" procedure for "rock drilling rig-01," requiring "maintenance team-B," "spare bearings," and an estimated time of "4 hours," and is associated with the task adjustment sub-strategy of "activating backup rig-03" to replace it. Strategy B (Delayed Repair): Defines reducing the equipment load to 70% for "6 hours" and performing the repair during the "night shift." This strategy does not require activating backup equipment but increases the risk of failure. Strategy C (Remote Intervention): Defines the temporary measure of first performing "lubrication flushing and parameter adjustment," postponing the repair to "24 hours later," and is associated with the monitoring sub-strategy of "increasing the equipment status monitoring frequency to once per minute." The above candidate strategies cover different response levels, from emergency response to risk mitigation, providing a wide range of options for the final decision.

[0103] Furthermore, candidate maintenance strategies are screened using multi-objective optimization algorithms to obtain the optimal equipment maintenance strategy. The optimization objectives of multi-objective optimization algorithms (such as multi-objective particle swarm optimization) include: minimizing the impact on construction schedule, minimizing overall operation and maintenance costs, maximizing equipment reliability recovery, and minimizing operational safety risks. These optimization objectives are often conflicting (for example, immediate repair is most beneficial to reliability but has the greatest impact on schedule). The multi-objective optimization algorithm searches in a multi-dimensional space comprised of these objectives to find a Pareto optimal solution set. Then, based on preset weight preferences, it selects the best solution from the Pareto optimal solution set as the equipment maintenance strategy and decomposes it into specific work orders, instructions, and resource allocation plans.

[0104] In this embodiment, based on the finalized equipment maintenance strategy, equipment fault handling is executed through equipment control and task scheduling. Maintenance work orders are automatically sent to the smart terminals of relevant maintenance personnel, spare parts issuance instructions are sent to the warehouse management system, and task adjustment instructions update the global construction plan. Throughout the maintenance process, equipment status recovery indicators are continuously monitored. This includes not only the static "equipment status normal" confirmation after maintenance but also a dynamic recovery verification period, during which vibration, temperature, pressure, and other inherent equipment status characteristics are continuously collected after the equipment is put back into operation and compared with the equipment's health benchmark.

[0105] It should be noted that equipment status recovery indicators include, but are not limited to: whether the amplitude of the fault characteristic frequency has dropped below the threshold, whether the equipment operating efficiency has recovered to the expected level, and whether the equipment status-related signals after geological interference identification processing are stable.

[0106] Furthermore, based on the monitored equipment status recovery indicators, a closed-loop feedback optimization of the equipment maintenance strategy is implemented to achieve continuous self-improvement of the process. First, all fault prediction information, equipment maintenance strategies, actual resource and time consumption, and the actual recovery status of the equipment after repair are stored in a historical database. Then, machine learning algorithms are used to analyze the deviation between the actual effect of the strategy and the predicted effect of the decision model. For example, if the "bearing replacement" strategy shows a better recovery of equipment vibration levels than the model predicts, the parameters related to "bearing replacement effectiveness" in the model are adjusted. Conversely, if the recovery effect is less than expected, the model will self-correct, potentially favoring a major overhaul strategy of "complete machine replacement" in similar future situations. This closed-loop learning process makes the fault operation and maintenance decision model more accurate and efficient over time. Ultimately, intelligent adaptive equipment fault handling is achieved in drill-and-blast tunnel construction.

[0107] Please see Figure 2In an optional embodiment, the present invention provides a fault diagnosis and handling system for drill-and-blast tunnel construction equipment. The system includes an input device, an output device, a processor, and a memory, all interconnected. The memory stores a computer program comprising program instructions, and the processor is configured to invoke the program instructions to execute specific steps as described in the relevant embodiments of the drill-and-blast tunnel construction equipment fault diagnosis and handling method provided by the present invention. The fault diagnosis and handling system for drill-and-blast tunnel construction equipment provided by the present invention has a complete structure, is objective and stable, and enhances the overall applicability and practical application capability of the present invention.

[0108] In summary, the present invention provides a method and system for fault diagnosis and handling of drilling and blasting tunnel construction equipment. Based on a collaborative monitoring network between the equipment and the geology, it synchronously collects equipment operating status and tunnel environmental geological data. By identifying and decoupling geological interference, it extracts the inherent state characteristics of the equipment, thereby tracing the fault source to determine the fault propagation path. Based on the fault propagation path, it constructs a model to predict the development trend of equipment faults, and finally combines digital twin technology and multi-agent decision-making mechanisms to generate and execute equipment maintenance strategies. This forms a closed-loop intelligent operation and maintenance system from state perception, diagnosis and prediction to decision execution, improving the accuracy and foresight of equipment fault handling. The method of this invention is easy to understand, computationally simple, requires less workload, and is convenient for engineering applications, providing a theoretical foundation and technical support for the further development of predictive maintenance.

[0109] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, and not to limit them. Although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some or all of the technical features therein. Such modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the scope of the technical solutions of the embodiments of the present invention, and they should all be covered within the scope of the claims and specification of the present invention.

Claims

1. A method for fault diagnosis and handling of drill-and-blast tunnel construction equipment, characterized in that, Includes the following steps: The equipment operation status data and tunnel environmental geological data are collected synchronously using an equipment and geological collaborative monitoring network. Based on the tunnel environmental geological data, geological interference identification is performed on the equipment operating status data to extract the inherent state characteristics of the equipment. Based on the inherent state characteristics of the equipment, fault diagnosis and source tracing are performed to determine the fault propagation path. Based on the equipment fault propagation path, equipment fault prediction is performed to obtain equipment fault development trend information; Based on the equipment failure development trend information, an equipment maintenance strategy is generated, and the equipment maintenance strategy is executed to handle equipment failures. The step of identifying geological interference in the equipment operating status data based on the tunnel environmental geological data to extract the inherent state characteristics of the equipment includes: Extract time-varying signals of geological interference based on the tunnel environmental geological data; Based on the time-varying geological interference signal, the equipment operating status data is adaptively decoupled to obtain equipment status related signals; In various tunnel environments, the equipment stability characteristics are obtained based on the equipment state-related signals as the inherent state characteristics of the equipment, including time-domain characteristics, frequency-domain characteristics and nonlinear characteristics; The extraction of time-varying geological interference signals based on the tunnel environmental geological data includes: Obtain rock drilling response data, rock mass vibration data, and dynamic geological inversion parameters as a heterogeneous data stream; The heterogeneous data stream is passed through a Kalman filter to generate the time-varying signal of the geological interference. The step of adaptively decoupling the equipment operating status data based on the time-varying geological interference signal to obtain equipment status-related signals includes: The geological time-varying dynamic parameters are obtained by analyzing the geological interference time-varying signal in order to establish a dual-channel feedforward decoupling filter; The device status-related signal is obtained by filtering the device operating status data using the dual-channel feedforward decoupling filter. The step of analyzing the time-varying geological disturbance signal to obtain the time-varying geological dynamic parameters includes: A transfer function from the time-varying geological disturbance signal to the response of a typical device is established using the recursive least squares method. Based on the transfer function, key parameters are obtained, including gain, damping ratio and dominant frequency, and these key parameters are used as the geological time-varying dynamic parameters. The transfer function satisfies the following relationship: in, For transfer functions, Let Z be the variable for transformation. This represents a typical device response. This is a time-varying signal caused by geological interference. and These are the transfer function coefficients; The establishment of the dual-channel feedforward decoupling filter includes: The geological disturbance prediction channel uses current and historical time-varying geological disturbance signals, combined with the transfer function, to obtain the geological disturbance components. The equipment status compensation channel receives the equipment operating status data and removes the geological interference component obtained from the geological interference prediction channel from the equipment operating status data; The step of filtering the device operating status data using the dual-channel feedforward decoupling filter to obtain the device status-related signal includes: The device status-related signals satisfy the following relationship: in, For equipment status related signals, The original mixed signal, To simulate interference signals; In various tunnel environments, the device stability characteristics obtained based on the device state-related signals are used as the inherent state characteristics of the device, including time-domain characteristics, frequency-domain characteristics, and nonlinear characteristics, including: The time-domain features satisfy the following relationship: in, The effective value of the signal. The number of sampling points. For sampling point index, These are discrete sampled values ​​of the signal. Peak factor; The frequency domain features satisfy the following relationship: in, It is a frequency domain signal. For Fourier transform operators, For equipment status related signals; The nonlinear characteristic satisfies the following relationship: in, For multi-scale entropy, For the first Sample entropy at each time scale Index for the time scale.

2. The method for fault diagnosis and handling of tunnel construction equipment using the drill-and-blast method according to claim 1, characterized in that, The method of using equipment and geological collaborative monitoring network to synchronously collect equipment operation status data and tunnel environmental geological data includes: The equipment and geological collaborative monitoring network is formed by the equipment monitoring sensor group deployed on the construction equipment and the environmental monitoring sensor group at the tunnel working face. The equipment operating status data, including vibration acceleration, oil pressure, oil temperature and motor current, are obtained based on the equipment monitoring sensor group. The tunnel environmental geological data, including rock drilling response data and rock mass structure vibration data, are obtained based on the environmental monitoring sensor group. The equipment operating status data and the tunnel environmental geological data are cleaned and spatiotemporally aligned.

3. The method for fault diagnosis and handling of tunnel construction equipment using the drill-and-blast method according to claim 1, characterized in that, The step of diagnosing and tracing equipment faults based on the inherent state characteristics of the equipment to determine the fault propagation path includes: Based on the inherent state characteristics of the equipment, multi-scale state variables of the equipment are obtained, and a hierarchical network of equipment state is constructed by combining the equipment operation mechanism and correlation relationship. Based on the device state hierarchy network, obtain the effective information of device state, and perform macro-causal state filtering on the effective information of device state to obtain the set of macro-key states of the device. A cross-scale equipment fault propagation model is constructed based on the set of macroscopic key states of the equipment, and the fault propagation path of the equipment is determined by combining the equipment structure and functional coupling relationship; The device state hierarchy network is a directed weighted graph, where nodes are the device's multi-scale state variables, and edges represent the causal or strong correlation relationships between these multi-scale state variables, satisfying the following relationships: in, This represents a device state hierarchy network. For a set of nodes, Let it be the set of edges; The process involves acquiring effective device state information based on the device state hierarchy network, and then performing macroscopic causal state filtering on this effective device state information to obtain a set of macroscopic key device states, including: Based on the device state hierarchical network, a state transition probability matrix of the device multi-scale state variables is constructed to obtain the effective information of the device state. Based on the amount of valid information about the equipment status, a set of macroscopic key statuses of the equipment is obtained through cross-scale comparative analysis.

4. The method for fault diagnosis and handling of tunnel construction equipment using the drill-and-blast method according to claim 1, characterized in that, The step of predicting equipment failures based on the equipment failure propagation path to obtain equipment failure development trend information includes: Construct a multi-dimensional space model of equipment faults based on the described equipment fault propagation path; Based on the equipment degradation mechanism, an equipment failure evolution prediction model is constructed by combining the aforementioned equipment failure multidimensional space model; Based on the equipment fault evolution prediction model, the equipment fault evolution time trajectory is obtained by forward fitting of the equipment fault. Quantitative analysis is performed on the time-series trajectory of the equipment failure evolution to obtain information on the development trend of the equipment failure.

5. The method for fault diagnosis and handling of tunnel construction equipment using the drill-and-blast method according to claim 4, characterized in that, The method for constructing a device failure evolution prediction model based on the device degradation mechanism and the multi-dimensional space model of device failure includes: Based on the aforementioned equipment degradation mechanism, multiphysics-based equipment degradation constraints are obtained. The multiphysics device degradation constraints are coupled and embedded into the device fault multidimensional space model to construct the device fault evolution prediction model.

6. The method for fault diagnosis and handling of tunnel construction equipment using the drill-and-blast method according to claim 1, characterized in that, The process of generating an equipment maintenance strategy based on the equipment fault development trend information and executing the equipment maintenance strategy to handle equipment faults includes: Based on the equipment failure development trend information, a failure operation and maintenance decision model is established by combining digital twins; Based on the aforementioned fault operation and maintenance decision-making model, a distributed autonomous decision-making mechanism is established by combining multiple agents; Candidate maintenance strategies are obtained through the distributed autonomous decision-making mechanism, and the equipment maintenance strategy is obtained by screening the candidate maintenance strategies based on a multi-objective optimization algorithm. Based on the equipment maintenance strategy, equipment faults are handled and equipment status recovery indicators are monitored. The equipment maintenance strategy is then optimized through closed-loop feedback based on the equipment status recovery indicators to achieve equipment fault handling.

7. A fault diagnosis and handling system for drill-and-blast tunnel construction equipment, characterized in that, The system includes an input device, an output device, a processor, and a memory, which are interconnected. The memory stores a computer program, which includes program instructions. The processor is configured to invoke the program instructions to execute the fault diagnosis and handling method for tunnel construction equipment using the drill-and-blast method as described in any one of claims 1-6.

Images