Digital-twin-based full-life-cycle management platform and system for mine hydraulic support

Abstract

The application relates to the field of mine equipment management, and specifically discloses a mine hydraulic support full-life-cycle management platform and system based on digital twinning, which comprises a digital twinning body construction module, a dynamic load spectrum construction and analysis module, an adaptive control strategy generation module and a full-life-cycle optimization module; coupling mapping of geological conditions and support behavior is established by fusing geological exploration data and support working conditions; a dynamic load spectrum is constructed based on real-time load data to identify impact load characteristics; adaptive control strategies are generated according to load spectrum analysis results to realize collaborative optimization of support groups; and cumulative damage assessment and residual life prediction are performed in combination with load history to optimize maintenance decisions. The application solves problems such as disconnection between geological conditions and support control, insufficient load characteristic analysis, disconnection between life-cycle management and actual working conditions and the like, and realizes precise support, advanced control and full-life-cycle cost optimization of hydraulic supports under complex geological conditions.

Description

Technical Field

[0001] This application relates to the field of mining equipment management technology, and in particular to a mining hydraulic support full life cycle management platform and system based on digital twin. Background Technology

[0002] Hydraulic supports, as the core support equipment in fully mechanized mining faces, directly affect the safety and efficiency of coal mining due to their level of intelligence. With the advancement of smart mine construction, the application of digital twin technology in hydraulic support management is gradually deepening; however, the existing technology system still faces several technical bottlenecks that urgently need to be addressed:

[0003] The disconnect between geological conditions and support control is a significant issue: existing digital twin systems for hydraulic supports primarily focus on monitoring the condition of the equipment itself, failing to achieve deep integration of geological conditions and support behavior. When the working face advances to areas with faults, folds, or other geological structures, traditional systems cannot pre-adjust support parameters according to geological changes, leading to a mismatch between the support's working resistance and the surrounding rock stress, which can easily cause accidents such as support tilting or crushing. Although technologies such as channel wave seismic surveys and borehole drilling have been used for geological exploration, their data have not yet been effectively integrated into the hydraulic support control closed loop, leaving geological information and support management in a state of "separate operations."

[0004] Insufficient dynamic load characteristic analysis and control response capabilities: Existing systems still process hydraulic support load data at the level of basic parameter recording, lacking in-depth feature extraction of load time series. In particular, the identification and response mechanism for impact loads is imperfect. When mine pressure manifestation occurs at the working face, the system struggles to promptly identify load anomalies and initiate corresponding control strategies. Traditional load monitoring methods cannot effectively separate static loads from dynamic impact components, resulting in poor adaptability of the support under complex load conditions, and the real-time performance and accuracy of the control strategies fail to meet field requirements.

[0005] The entire lifecycle management is severely disconnected from actual operating conditions: Current maintenance decisions for hydraulic supports are mostly based on fixed cycles or simple operating times, failing to fully consider the actual load history borne by the equipment. Due to the lack of accurate load spectrum data, life prediction models generally rely on theoretical calculations and empirical formulas, resulting in significant discrepancies between predictions and actual damage accumulation. This management model, detached from actual operating conditions, leads to both under-maintenance and over-maintenance, increasing safety hazards and wasting resources.

[0006] Virtual verification does not closely match real-world operating conditions: Existing digital twin systems often use idealized load conditions in their virtual verification phase, failing to effectively reproduce the load characteristics under actual geological conditions downhole. Due to the lack of real load data, digital twin models cannot accurately simulate the mechanical response of hydraulic supports in complex geological environments, reducing the reliability of virtual commissioning and predictive maintenance.

[0007] In recent years, although some studies have attempted to incorporate geological information into hydraulic support control systems, these efforts have largely remained at the level of simple data overlay, failing to establish a complete technological chain from geological sensing to load analysis and control execution. Particularly in key areas such as dynamic load spectrum construction, impact load identification and early warning, and life prediction based on actual load history, existing technologies have yet to form systematic solutions. Therefore, there is an urgent need for a digital twin management platform for hydraulic supports that can deeply integrate geological information, achieve accurate load characteristic analysis, and support full lifecycle optimization decisions. Summary of the Invention

[0008] To address the aforementioned technical issues, this application proposes a digital twin-based full lifecycle management platform and system for mining hydraulic supports.

[0009] The technical solution adopted in this application is: a full life-cycle management platform for mining hydraulic supports based on digital twins, comprising:

[0010] The digital twin construction module is used to establish a coupled mapping model between geological conditions and support behavior based on geological structure exploration data and hydraulic support working condition data, and generate a geological-support coupled digital twin.

[0011] The dynamic load spectrum construction and analysis module is used to receive the output of the geological-support coupled digital twin and construct a dynamic load spectrum based on geological conditions and mining technology by real-time monitoring of the load changes of the hydraulic support during the advancement of the working face.

[0012] The adaptive control strategy generation module is used to dynamically generate adaptive control strategies for hydraulic support groups based on the analysis results of dynamic load spectrum and the correlation between geological change trends and support status.

[0013] The full lifecycle optimization module is used to optimize the parameters of hydraulic supports throughout their entire lifecycle, from installation and operation to maintenance, based on the execution feedback of adaptive control strategies and dynamic load spectrum data.

[0014] Furthermore, the digital twin building blocks include:

[0015] The geological structure detection unit is used to obtain geological structure data of the working face through channel wave seismic exploration and borehole inspection technology, identify geological structure anomaly areas, and form a digital model of geological structure by combining the internal images of rock strata obtained by borehole inspection technology.

[0016] The support condition monitoring unit is used to collect pressure data, displacement data, attitude data and working resistance data of the hydraulic support in real time;

[0017] The coupling mapping unit is used to establish a correlation model between geological anomaly areas and support working resistance, realize the quantitative impact analysis of geological conditions on support behavior, and fuse geological structure data with support working condition data to generate a geological-support coupled digital twin.

[0018] The ground sound monitoring integration unit is used to analyze the spatiotemporal evolution of ground sound events by monitoring ground sound signals generated by rock mass fracturing, establish a mineral pressure manifestation prediction model based on ground sound parameters, and integrate ground sound monitoring data into a geological-support coupled digital twin.

[0019] Furthermore, geological structural anomaly zones are identified by the change in channel wave propagation velocity, and a geological structural anomaly index is obtained based on the change in channel wave propagation velocity.

[0020] The images of the rock strata obtained by the borehole inspection instrument are constructed into rock strata image feature vectors, which include fault range, coal seam thickness and lithological classification number;

[0021] A digital model of geological structure is obtained by fusing the geological structural anomaly index and the feature vector of rock strata image.

[0022] ;

[0023] in: and For weight parameters, As an index of geological structural anomaly, This is the feature vector of the rock strata image.

[0024] Furthermore, the geological-support coupled digital twin is based on the working face area divided by the geological structure digital model. The quantitative relationship between geological parameters and support working resistance is established by using multiple regression analysis, i.e., the coupled mapping model. The geological parameters include fault displacement, coal seam thickness variation rate and roof lithology index. The coupled mapping model is dynamically corrected by comparing the deviation between the predicted working resistance and the actual monitoring value in real time.

[0025] The multiple linear regression relationship between geological parameters and support resistance is expressed as follows:

[0026] ;

[0027] in: It is a fault elevation difference. The rate of change of coal seam thickness. The roof lithology index, For regression coefficients, This is the error term.

[0028] Furthermore, the dynamic load spectrum construction and analysis module includes:

[0029] The load data acquisition unit is used to collect load data of the hydraulic support under different geological conditions in real time.

[0030] The load feature extraction unit is used to extract feature parameters from the collected load data;

[0031] The dynamic load spectrum generation unit is used to construct a dynamic load spectrum database that reflects changes in geological conditions based on characteristic parameters.

[0032] The load spectrum prediction unit is used to predict the load spectrum characteristics of the future mining area based on the working face advancement plan and geological exploration data.

[0033] The load spectrum comparison unit is used to perform similarity analysis between real-time acquired load data and historical load spectra to identify abnormal load patterns.

[0034] Furthermore, the characteristic parameters include peak load, root mean square load value, number of load cycles, impact load frequency, and load asymmetry coefficient.

[0035] Furthermore, the adaptive control strategy generation module includes:

[0036] The load adaptive control unit is used to adjust the working parameters of the support according to the real-time load spectrum characteristics;

[0037] Group collaborative optimization unit is used to optimize the collaborative control strategy of hydraulic support group based on load distribution law;

[0038] The preventative parameter adjustment unit is used to adjust the support control parameters in advance based on the load spectrum prediction results.

[0039] Furthermore, the full lifecycle optimization module includes:

[0040] The cumulative damage assessment unit is used to calculate the cumulative damage degree of key components of the hydraulic support based on load spectrum data.

[0041] The remaining life prediction unit is used to predict the remaining life of a component by combining material fatigue characteristics and cumulative damage.

[0042] The maintenance strategy optimization unit is used to develop differentiated maintenance plans based on the remaining life prediction results.

[0043] Furthermore, the group collaborative optimization unit adopts a distributed collaborative control algorithm. Based on the load distribution law obtained by dynamic load spectrum analysis, it establishes a load balance optimization model for the hydraulic support group. By calculating the load concentration in different areas of the working face in real time, it dynamically adjusts the working parameters of the support group to keep the variance of the working face support strength distribution within 15%.

[0044] A digital twin-based full lifecycle management system for mining hydraulic supports includes:

[0045] The facilities and data acquisition layer includes: hydraulic support groups, coal mining machines, environmental sensors, and cloud / edge computing centers;

[0046] The data and model layer includes: a 3D model library, a historical database, a real-time database, an expert knowledge base, and an algorithm model library;

[0047] The platform layer is constructed based on the digital twin-based mining hydraulic support full life cycle management platform;

[0048] The application service layer includes: panoramic monitoring and virtual-real interaction, full lifecycle management, group collaborative intelligent control, intelligent decision-making and predictive maintenance.

[0049] The advantages of this application over the prior art are as follows:

[0050] 1. Geological perception and advanced control: By integrating multi-source geological information and support working condition data, precise support and advanced control based on geological conditions are achieved, significantly improving the safety and efficiency of mining under complex geological conditions, with a 25% improvement in support effect and an 18% increase in advance rate.

[0051] 2. Intelligent collaboration and inherent safety: Based on dynamic load spectrum analysis and a three-level response mechanism, adaptive collaborative control of hydraulic support groups is realized, which effectively suppresses the hazards of impact loads, reduces equipment failure rate by 35%, and significantly improves support stability.

[0052] 3. Accurate prediction and cost optimization: By combining the real load spectrum and the cumulative damage model, the remaining life of key components can be predicted with an accuracy of more than 90%. A three-level maintenance strategy based on health status is established, which reduces maintenance costs by 25% and extends the overhaul cycle by 20%. Attached Figure Description

[0053] The following description, in conjunction with the accompanying drawings, further illustrates this application:

[0054] Figure 1 A block diagram of a digital twin-based full lifecycle management platform for mining hydraulic supports provided in this application embodiment;

[0055] Figure 2 This is an architecture diagram of a digital twin-based full lifecycle management system for mining hydraulic supports, provided in an embodiment of this application. Detailed Implementation

[0056] like Figure 1 and Figure 2 As shown, this application provides a digital twin-based full lifecycle management platform for mining hydraulic supports, including:

[0057] The digital twin construction module is used to establish a coupled mapping model between geological conditions and support behavior based on geological structure exploration data and hydraulic support working condition data, and generate a geological-support coupled digital twin.

[0058] The dynamic load spectrum construction and analysis module is used to receive the output of the geological-support coupled digital twin and construct a dynamic load spectrum based on geological conditions and mining technology by real-time monitoring of the load changes of the hydraulic support during the advancement of the working face.

[0059] The adaptive control strategy generation module is used to dynamically generate adaptive control strategies for hydraulic support groups based on the analysis results of dynamic load spectrum and the correlation between geological change trends and support status.

[0060] The full lifecycle optimization module is used to optimize the parameters of hydraulic supports throughout their entire lifecycle, from installation and operation to maintenance, based on the execution feedback of adaptive control strategies and dynamic load spectrum data.

[0061] The digital twin building blocks include:

[0062] The geological structure detection unit is used to obtain geological structure data of the working face through channel wave seismic exploration and borehole inspection technology, identify fault and collapse column geological structures, and form a digital model of geological structure.

[0063] The support condition monitoring unit is used to collect pressure data, displacement data, attitude data and working resistance data of the hydraulic support in real time;

[0064] The coupling mapping unit is used to establish a correlation model between geological anomaly areas and support working resistance, realize the quantitative impact analysis of geological conditions on support behavior, and fuse geological structure data with support working condition data to generate a geological-support coupled digital twin.

[0065] The ground sound monitoring integration unit is used to analyze the spatiotemporal evolution of ground sound events by monitoring ground sound signals generated by rock mass fracturing, establish a mineral pressure manifestation prediction model based on ground sound parameters, and integrate ground sound monitoring data into a geological-support coupled digital twin.

[0066] The workflow of each unit in the digital twin construction module is as follows:

[0067] The geological structure detection unit transmits and receives channel waves through a channel wave seismic exploration system deployed on the working face. It identifies geological structure anomaly zones based on changes in channel wave propagation speed. At the same time, it uses a borehole inspection instrument to obtain images of the rock strata. The two types of detection data are fused to form a geological structure digital model with meter-level accuracy.

[0068] The propagation speed of the channel wave ,in: The elastic modulus of the coal seam. Coal seam density;

[0069] Based on the change in the propagation speed of the channel wave Identify geological structural anomaly zones, including: Reference trough wave velocity; geological structural anomaly index ;

[0070] Images of the interior of rock strata were obtained using a borehole inspection instrument, and the feature vector of the rock strata images was analyzed. ,in: The fault range, For coal seam thickness, For lithological classification numbering;

[0071] The two types of detection data are fused to form a digital model of geological structure with meter-level accuracy: ,in: and These are the weighting parameters.

[0072] The support condition monitoring unit collects support resistance data in real time through pressure sensors installed on the hydraulic support columns and monitors the height of the top beam through displacement sensors. and the progression of the journey The spatial attitude (i.e., tilt angle) of each component of the support is measured by tilt sensors. All sensor data is transmitted to the data acquisition station via explosion-proof industrial Ethernet;

[0073] Among the support resistance ,in: For hydraulic column pressure, The area of ​​the column;

[0074] Represent all sensor data as state vectors .

[0075] The coupled mapping unit is based on the working face area divided by the geological structure digital model. The quantitative relationship between geological parameters and support working resistance is established by using multiple regression analysis, namely the coupled mapping model. The geological parameters include fault displacement, coal seam thickness variation rate and roof lithology index. The coupled mapping model is dynamically corrected by comparing the deviation between the predicted working resistance and the actual monitoring value in real time.

[0076] The multiple linear regression relationship between geological parameters and support resistance is expressed as follows:

[0077] ;

[0078] in: It is a fault elevation difference. The rate of change of coal seam thickness. The roof lithology index, For regression coefficients, This is the error term;

[0079] Deviation between predicted working resistance and actual monitored value The regression coefficients are dynamically adjusted using gradient descent based on the deviation. .

[0080] The ground sound monitoring integration unit collects rock mass fracture signals by deploying ground sound sensor arrays in the working face and two roadways, analyzes signal characteristics using wavelet transform, extracts ground sound event rate, energy rate and large event number parameters, establishes a mine pressure manifestation precursor identification model based on the spatiotemporal evolution law of the parameters, and integrates the identification results into the digital twin in real time.

[0081] The following signal features are analyzed using wavelet transform: Ground sound event feature parameters (Earth Sound Event Rate) (energy rate), (Number of major events);

[0082] The model for identifying precursors of mineral pressure manifestation, based on the spatiotemporal evolution of parameters, is as follows:

[0083] ;

[0084] in: These are the weighting coefficients. As a precursor indicator of mine pressure; determined by the threshold method. Whether a dangerous level has been reached will be determined, and the results will be integrated into the digital twin.

[0085] The dynamic load spectrum construction and analysis module includes:

[0086] The load data acquisition unit is used to collect load data such as working resistance and support strength of hydraulic supports under different geological conditions in real time.

[0087] The load feature extraction unit is used to extract feature parameters such as peak load, load cycle number, and load application time from the collected load data;

[0088] The dynamic load spectrum generation unit is used to construct a dynamic load spectrum database that reflects changes in geological conditions based on characteristic parameters.

[0089] The load spectrum prediction unit is used to predict the load spectrum characteristics of the future mining area based on the working face advancement plan and geological exploration data.

[0090] The load spectrum comparison unit is used to perform similarity analysis between real-time acquired load data and historical load spectra to identify abnormal load patterns.

[0091] The workflow of each unit in the dynamic load spectrum construction and analysis module is as follows:

[0092] The load data acquisition unit collects stress data of the support structure in real time at a sampling frequency of 100Hz through a distributed stress sensor array arranged on the top beam, column and base of the hydraulic support. At the same time, it monitors the pressure inside the column cylinder through a pressure sensor and records the cyclic movement stroke of the support through a displacement sensor, forming a raw load dataset containing timestamps.

[0093] In one specific embodiment, stress, pressure, and displacement data are acquired at a frequency of 100 Hz to form a raw dataset containing timestamps: ,in: , For stress, For pressure, For displacement, For timestamps.

[0094] The load feature extraction unit uses the db4 wavelet basis to perform 5-level wavelet decomposition on the original load data, separating the static load component of 0~6.25Hz and the dynamic impact load component of 6.25~50Hz, extracting feature parameters such as peak load, root mean square load value, load cycle number, and impact load frequency, and calculating the load asymmetry coefficient based on the influence of geological structure.

[0095] remember ,in: The static load component is 0~6.25Hz. The dynamic impact load component is 6.25~50Hz;

[0096] Extract the following feature parameters:

[0097] Peak load ;

[0098] Root mean square value of load ;

[0099] Load cycle count Number of zero-crossings ;

[0100] Impact load frequency Number of peaks exceeding threshold;

[0101] in: The number of zero-crossings represents the length of the analysis time window. The number of zero-crossings refers to the number of times the signal changes from a positive value to a negative value or vice versa. The number of peaks exceeding the threshold represents the number of peaks that exceed a certain set threshold.

[0102] Load asymmetry coefficient ,in: and These represent the loads caused by the geological structures on the left and right sides, respectively.

[0103] The dynamic load spectrum generation unit establishes an eight-level load spectrum matrix based on characteristic parameters, divides the load range into eight levels from no load to 125% of the rated load, records the number of cycles and the duration of action under each load level, and associates it with the corresponding geological condition information to form a spatiotemporally correlated dynamic load spectrum database.

[0104] The representation of the eight-level load spectrum matrix is ​​as follows:

[0105] ;

[0106] in: For the load range (from no load to 125% of rated load). The number of loops. For the duration of action, For geological conditions information, .

[0107] The load spectrum prediction unit is based on geological exploration data and working face advancement plan. It uses a weighted algorithm of geological condition influence factors to calculate the stress concentration coefficient of geological structural areas such as fault zones and fold axis. Combined with mining process parameters such as mining depth, mining height, and advancement speed, it predicts the load distribution characteristics in the next 5 cycles.

[0108] Load distribution characteristics ,in: For the current load, The weights of geological condition influencing factors, It is the stress concentration factor of geological structures (such as fault zones and fold axes). This represents the number of geological condition types.

[0109] The load spectrum comparison unit uses the dynamic time warping algorithm (DTW) to perform pattern matching between the real-time load sequence and the historical load spectrum database to identify abnormal load patterns. When the similarity between the real-time load and the typical pattern is less than 85%, an early warning mechanism is triggered.

[0110] The similarity between real-time payload and typical mode ;

[0111] in: For real-time payload sequences, For the historical load spectrum sequence, DTW(x,y) represents the similarity distance between x and y calculated by the DTW algorithm. The smaller the distance, the more similar the two are. Length(x) represents the length of x, which is usually measured by the number of sampling points or the time span.

[0112] The adaptive control strategy generation module includes:

[0113] The load adaptive control unit is used to adjust the working parameters of the support according to the real-time load spectrum characteristics;

[0114] Group collaborative optimization unit is used to optimize the collaborative control strategy of hydraulic support group based on load distribution law;

[0115] The preventative parameter adjustment unit is used to adjust the support control parameters in advance based on the load spectrum prediction results.

[0116] The workflow of each unit in the adaptive control strategy generation module is as follows:

[0117] The load adaptive control unit establishes a three-level response mechanism based on the real-time impact load characteristics in the dynamic load spectrum:

[0118] Three response level thresholds: ,in: This refers to the rated working resistance of the support.

[0119] When the peak impact load reaches 85% of the rated working resistance, that is... When this occurs, a Level 1 response is initiated, automatically adjusting the safety valve opening pressure. ,in: The initial safety valve pressure, To adjust the coefficient, This represents the peak value of the impact load.

[0120] When the peak impact load reaches 95% of the rated working resistance, that is... At that time, the secondary response is initiated, the support buffer function is activated, and the initial support force is adjusted;

[0121] When the peak impact load reaches 110% of the rated working resistance, that is... At that time, a Level 3 response is initiated, an emergency depressurization procedure is executed, and adjacent supports are coordinated to provide compensating support.

[0122] The group collaborative optimization unit adopts a distributed collaborative control algorithm. Based on the load distribution law obtained by dynamic load spectrum analysis, a load balance optimization model of hydraulic support group is established. By calculating the load concentration in different areas of the working face in real time, the working parameters of the support group are dynamically adjusted so that the variance of the working face support strength distribution is controlled within 15%.

[0123] The load concentration of each support is obtained by dividing the force on each support by the total force on the working surface:

[0124] ,in: For the first The load concentration of each support structure For the support to bear the force, For the total force on the working face, This represents the total number of stents.

[0125] Set concentration alarm threshold When the concentration of a stent exceeds When necessary, its operating parameters need to be adjusted;

[0126] Dynamically adjust the support strength of the stent: ,in: To adjust the strength of the support, The target average concentration, For adjustment coefficients;

[0127] Calculate the variance of the support strength distribution: ,make sure ,in: This represents the average support strength.

[0128] Based on the load distribution characteristics of the next 5 cycles output by the load spectrum prediction unit, the preventive parameter adjustment unit combines the fault zone and fold axis location information in the geological structure digital model with the fuzzy control algorithm to adjust the support working parameters, including initial support force, working resistance and following speed, 3 to 5 production cycles in advance, to ensure that the support completes parameter optimization configuration before entering the geological anomaly zone.

[0129] In a specific embodiment, the preventive parameter tuning unit is implemented as follows:

[0130] Using fuzzy membership degree The degree of load anomaly when the quantification support approaches a geologically abnormal area, including: and These represent the fuzzy membership degrees of the stent near faults and folds, respectively. and The corresponding influence weights;

[0131] according to The initial support force, working resistance, and following speed of the support are adjusted through fuzzy control rules:

[0132] Initial support force adjustment formula: ;

[0133] Working resistance adjustment formula: ;

[0134] Follow-up speed adjustment formula: ;

[0135] in: These are adjustment coefficients used to control the adjustment range of each parameter; These are the unadjusted initial support force, working resistance, and following speed, respectively. These are the adjusted initial support force, working resistance, and following speed, respectively.

[0136] The full lifecycle optimization module includes:

[0137] The cumulative damage assessment unit is used to calculate the cumulative damage degree of key components of the hydraulic support based on load spectrum data.

[0138] The remaining life prediction unit is used to predict the remaining life of a component by combining material fatigue characteristics and cumulative damage.

[0139] The maintenance strategy optimization unit is used to develop differentiated maintenance plans based on the remaining life prediction results.

[0140] The workflow of each unit in the full lifecycle optimization module is as follows:

[0141] The cumulative damage assessment unit is based on the eight-level load spectrum matrix in the dynamic load spectrum database. It uses Miner's linear cumulative damage theory to calculate the cumulative damage degree of key components of the hydraulic support and displays the damage degree distribution cloud map of each component in real time.

[0142] The cumulative damage is:

[0143] ,in: To accumulate damage, For load level The actual number of loops, For load level The material fatigue life is measured; the damage distribution cloud map is displayed in real time to intuitively present the damage status of each component.

[0144] The remaining life prediction unit combines material fatigue characteristic curves and real-time cumulative damage to establish a remaining life prediction model based on actual load history. Through finite element analysis, the stress concentration factors of key components such as the top beam and columns under different load conditions are obtained, and the crack propagation rate formula is used... Calculate the fatigue crack propagation rate;

[0145] The remaining lifespan ;

[0146] , The stress concentration factor is... For stress, The length of the crack. and is the crack propagation parameter of the material, and N is the number of fatigue load cycles.

[0147] Based on the remaining life prediction results and the cumulative damage distribution, the maintenance strategy optimization unit establishes a three-level maintenance response mechanism: when the component damage reaches 60%, an early warning is initiated and planned maintenance is arranged; when the component damage reaches 80%, an emergency maintenance plan is initiated and spare parts are prepared; when the component damage reaches 95%, a forced replacement procedure is executed, and maintenance resource allocation is optimized according to the component damage distribution characteristics.

[0148] In one specific embodiment, the full lifecycle management platform of this application also includes a virtual testing function based on dynamic load spectrum. By reproducing typical load conditions in a digital twin environment, the structural strength and control system reliability of the hydraulic support under different geological conditions are verified. The specific steps are as follows:

[0149] Based on historical data from the dynamic load spectrum database, we selected four typical geological conditions for ultimate load conditions: fault zone mining, passage through fold axis, initial pressure on the working face, and periodic pressure. We then extracted the peak load, load duration, and load cycle characteristic parameters for each condition.

[0150] Import the 3D model of the hydraulic support into the ANSYS Twin Builder environment, set the material constitutive relation and boundary condition constraints, establish a finite element analysis model containing key components such as the top beam, column and base, and map the geological condition parameters in the geological-support coupled digital twin to the finite element model;

[0151] The selected typical load condition data are input into the finite element analysis model. The dynamic response of the hydraulic support under various geological conditions is simulated through transient dynamic analysis to obtain the stress distribution cloud map, displacement deformation and fatigue hot spot area of ​​key components.

[0152] The maximum equivalent stress region is identified based on the stress distribution cloud map, and the structural safety margin is assessed by comparing the material yield strength. When the maximum equivalent stress exceeds 85% of the material yield strength, it is marked as a high-risk region and structural optimization suggestions are generated.

[0153] The support response data during the load condition reproduction process is input into the adaptive control strategy generation module to verify the response speed and stability of the control strategy under different geological conditions, and to record the response time of the control system from load anomaly identification to parameter adjustment completion.

[0154] The results of structural strength verification and control system reliability verification are fed back to the life cycle optimization module to optimize the cumulative damage assessment model and maintenance strategy, forming a closed-loop feedback mechanism from virtual verification to actual optimization.

[0155] This application also proposes a digital twin-based full lifecycle management system for mining hydraulic supports, including:

[0156] The facilities and data acquisition layer includes: hydraulic support groups, coal mining machines, environmental sensors, and cloud / edge computing centers;

[0157] This layer serves as the interface between the system and the physical working plane, undertaking the dual functions of data source acquisition and control command execution. Specifically:

[0158] Hydraulic support group: As the core execution and sensing object, it receives control commands to complete support actions such as raising, lowering, and pushing the column. On the other hand, it outputs working condition data such as pressure, displacement, attitude and working resistance in real time through built-in sensors.

[0159] Coal mining machine: Provides coal mining process parameters and location information. Its advance speed, drum height and coal cutting trajectory directly affect the load distribution and action sequence of the hydraulic support, and are an important data source for the boundary conditions of mining process in the construction of dynamic load spectrum;

[0160] Environmental sensors: Independent of the support body, they are used to collect environmental parameters of the working face, including but not limited to gas concentration, temperature, humidity, dust concentration and roof delamination data, to provide environmental boundary constraints for the digital twin;

[0161] Cloud / edge computing centers: Deployed underground or on the ground, they undertake the local preprocessing, caching, and lightweight computing tasks of raw data, reducing data transmission latency and providing computing power support for complex algorithm models.

[0162] The data and model layer includes: a 3D model library, a historical database, a real-time database, an expert knowledge base, and an algorithm model library. This layer serves as the system's data asset and knowledge hub, providing the platform layer with structured data, geometric models, and algorithmic capabilities. Specifically:

[0163] 3D Model Library: Stores high-precision 3D geometric models and physical property parameters of key components and working surface environment of hydraulic supports, providing digital ontology for geometric mapping and finite element simulation of digital twins;

[0164] Historical database: Archives geological exploration data, load spectrum records, fault cases and maintenance logs accumulated during long-term operation, used for model training, trend analysis and knowledge mining;

[0165] Real-time database: Stores the raw data stream collected by the sensors on the current working face with high temporal resolution, supports millisecond-level or second-level data query, and serves as the data foundation for real-time construction of dynamic load spectrum;

[0166] Expert Knowledge Base: It gathers the experience of experts in the field of mining engineering, including geological anomaly criteria, rockburst early warning rules, support control strategy rule base and maintenance decision threshold, etc., to assist intelligent decision-making and anomaly identification;

[0167] Algorithm Model Library: Centrally manages various mathematical and artificial intelligence algorithm instances, including multivariate regression models, wavelet transform algorithms, dynamic time warping algorithms, Miner cumulative damage models, fuzzy control algorithms, and finite element analysis models, which can be called by various modules of the platform layer as needed.

[0168] The platform layer (digital twin) includes: geometric model, physical model, behavioral model, rule model, and a digital twin of the hydraulic support. This layer is the core technology of the system, built upon a digital twin-based full lifecycle management platform for mining hydraulic supports, achieving a closed-loop process from geological sensing to lifespan prediction. Specifically:

[0169] Digital twin construction module: Based on geological structure exploration data and hydraulic support working condition data, a coupled mapping model between geological conditions and support behavior is established to generate a geological-support coupled digital twin;

[0170] Dynamic load spectrum construction and analysis module: Receives the output of the geological-support coupled digital twin, and constructs a dynamic load spectrum based on geological conditions and mining technology by real-time monitoring of load changes of the hydraulic support during the advancement of the working face;

[0171] Adaptive control strategy generation module: Based on the analysis results of dynamic load spectrum, combined with the correlation between geological change trend and support status, dynamically generate adaptive control strategies for hydraulic support groups;

[0172] Full lifecycle optimization module: Based on the execution feedback of the adaptive control strategy and dynamic load spectrum data, it realizes the full lifecycle parameter optimization of hydraulic supports from installation, operation to maintenance.

[0173] The application service layer includes: panoramic monitoring and virtual-real interaction, full lifecycle management, group collaborative intelligent control, intelligent decision-making, and predictive maintenance. This layer faces end users and external systems, encapsulating the platform layer's technical capabilities into directly callable business services. Specifically:

[0174] Panoramic monitoring and virtual-real interaction: Real-time mapping of equipment status and geological environment on the working face through a 3D visualization interface, supporting users to remotely monitor and intervene in a virtual-real fusion manner;

[0175] Full lifecycle management: Provides full lifecycle file management and process tracking services from hydraulic support installation and commissioning, operation monitoring to scrapping and replacement;

[0176] Group collaborative intelligent control: The adaptive control strategy and collaborative optimization parameters generated at the platform layer are sent to the execution terminal to realize centralized management and distributed collaboration of the support group;

[0177] Intelligent decision-making and predictive maintenance: Based on the life prediction and damage assessment results output by the platform layer, maintenance plans, spare parts requirement lists and production optimization suggestions are automatically generated to support management decisions.

[0178] The facility and data acquisition layer collects real-time environmental data from the working face through environmental sensors and acquires working condition data through sensors built into the hydraulic support group and the coal mining machine. After preprocessing by the cloud / edge computing center, the data is uploaded to the data and model layer for real-time updates and historical archiving. The data and model layer serves as the data support and model driver for the platform layer, providing the platform layer with geometric models, historical data, real-time data, expert knowledge, and algorithm examples. Based on the above resources, the platform layer constructs a geological-support coupled digital twin, performs dynamic load spectrum analysis, adaptive control strategy generation, and full life cycle simulation optimization to ensure real-time mapping between the digital twin and the actual actions of the hydraulic support. The platform layer encapsulates its core capabilities into service interfaces. The application service layer calls the above service interfaces to provide panoramic monitoring, full life cycle management, group collaborative control, and intelligent decision-making services. It also sends business-level control decision requests to the platform layer, which parses and generates equipment-level control commands. These commands are then sent to the hydraulic support group and the coal mining machine through the facility and data acquisition layer to execute the corresponding actions, forming a complete closed loop from perception to decision-making to execution.

[0179] The present application will be further described below with reference to a specific embodiment.

[0180] Example 1

[0181] 1. Implementation environment:

[0182] The application was verified in a fully mechanized mining face of a mining group. The working face has a strike length of 1850m, an dip length of 250m, an average coal seam thickness of 3.5m, and contains two fault structures (fault displacement of 1.8~2.5m) and a fold structure zone.

[0183] 2. Implementation method:

[0184] (1) Deploy a geological coupling detection system: arrange a channel wave seismic transmitter / receiver point every 30m along the working face, construct a geological borehole every 50m and install a borehole inspection instrument to form a digital model of the geological structure.

[0185] (2) Install support monitoring sensors: Install pressure sensors, displacement sensors and tilt sensors on 125 hydraulic supports, and add distributed stress sensor arrays on 30 key supports.

[0186] (3) Construct a digital twin platform: Deploy a full life cycle management platform, including a digital twin construction module for geological coupling, a dynamic load spectrum construction and analysis module, an adaptive control strategy generation module, and a full life cycle optimization module.

[0187] (4) Operational verification: The platform operated continuously for 6 months, covering the entire process of advancing 850m of the working face, including special geological conditions such as passing through two fault zones and a fold structure area.

[0188] 3. Verification method:

[0189] (1) Verification of the accuracy of geological prediction: Compare the consistency between the location of the geological anomaly area predicted by the platform and the actual location revealed.

[0190] (2) Load control effect verification: Statistical analysis of the failure rate and support quality qualification rate of the support when passing through geological structural areas.

[0191] (3) Verification of the accuracy of life prediction: Compare the degree of agreement between the remaining life of the component predicted by the platform and the actual test results.

[0192] (4) Economic benefit analysis: Statistics on the direct economic benefits generated by reducing downtime and extending equipment life.

[0193] 4. Comparison of verification results:

[0194] The comparison of the effects of the platform using this application with existing technologies is shown in Table 1 below:

[0195] Table 1. Comparison of Results

[0196]

[0197] 5. Specific implementation results:

[0198] (1) Geological prediction and control effect: The platform accurately predicted the fault location 12 days in advance with a deviation of only 3.2m. During the passage through the fault, the system automatically adjusted the working resistance of the support from 28MPa to 35MPa, effectively controlling the subsidence of the roof and reducing the support tilt failure rate from the expected 25% to the actual 6%.

[0199] (2) Load analysis and safety control: In the folded structure area, the system detected an increase in the frequency of impact loads and automatically activated the secondary response mechanism. By adjusting the initial support force and activating the buffer function, it successfully avoided three potential support crushing accidents.

[0200] (3) Life prediction and maintenance optimization: The system predicts that the remaining life of the top beam of the No. 45 to No. 48 supports is 7 months. After actual disassembly and inspection, the fatigue damage degree is consistent with the prediction result by 93%. Based on this, the preventive replacement plan avoids unplanned downtime losses.

[0201] This embodiment fully demonstrates the significant effects of the platform of this application in improving the geological adaptability of hydraulic supports, enhancing safety assurance, and optimizing the whole life cycle management, providing effective technical support for the intelligent construction of coal mines.

[0202] The embodiments described above in this application achieve the following technical effects:

[0203] 1. It has achieved precise support and advanced control based on geological perception, which has significantly improved the safety and efficiency of mining under complex geological conditions.

[0204] The core innovation of this application lies in breaking through the technical bottleneck of traditional hydraulic support management being disconnected from geological conditions. Through a geologically coupled digital twin construction module, it is the first to deeply integrate multi-source geological information, such as channel seismic exploration, borehole inspection, and ground acoustic monitoring, with support operating data, establishing a quantitative correlation model between geological conditions and support behavior. This innovation enables the platform to accurately identify geological anomalies such as faults and folds, and based on dynamic load spectrum prediction technology, predict load change trends 3-5 production cycles in advance, thereby achieving pre-optimization of support parameters. Practical application shows that this technology improves the support effect of hydraulic supports in geologically structural areas by approximately 25%, increases the monthly progress of the working face by more than 18%, and successfully provides early warning of multiple working face pressure surges, preventing major safety accidents.

[0205] 2. A closed-loop technology system from load sensing to adaptive control was created, realizing intelligent collaboration and inherent safety of hydraulic support groups.

[0206] This application introduces a dynamic load spectrum construction and analysis module, enabling unprecedented in-depth analysis of the load characteristics of hydraulic supports. This module employs advanced signal processing techniques such as wavelet transform to achieve precise separation of static loads and dynamic impact loads, and establishes an eight-level load spectrum matrix. Based on this, the adaptive control strategy generation module can initiate a three-level response mechanism according to real-time load characteristics. When abnormal impact loads are detected, it can automatically adjust the support's operating parameters within milliseconds, effectively avoiding overload damage. Simultaneously, through a distributed collaborative control algorithm, load balancing control of the hydraulic support group at the working face is achieved, keeping the variance of the support strength distribution within 15%, fundamentally improving the overall stability and reliability of the support system, and reducing the failure rate of on-site application display equipment by 35%.

[0207] 3. A precise lifetime prediction and closed-loop optimization mechanism based on actual load history was constructed, achieving the optimization of the entire life cycle cost.

[0208] Another major innovation of this application lies in the deep integration of dynamic load spectrum analysis and full lifecycle management. Through a cumulative damage assessment unit, the system can accurately calculate the cumulative damage of key components based on real eight-level load spectrum data using Miner's theory, and combined with finite element analysis, achieve a remaining life prediction accuracy of over 90%. This mechanism completely changes the traditional maintenance model that relies on empirical formulas or fixed cycles, establishing a three-level maintenance response strategy based on the actual health status of components. This not only avoids the waste caused by "over-maintenance" but also prevents unexpected downtime caused by "under-maintenance." Application data shows that maintenance costs are reduced by more than 25%, and the overhaul cycle of hydraulic supports is extended by 20%. Furthermore, by reproducing typical load conditions in a digital twin environment for virtual verification, a complete closed loop from design optimization and operation control to maintenance decision-making is formed, continuously improving the system's reliability and economy.

[0209] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of this application, and are not intended to limit them. Although this application 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 this application.

Claims

1. A mining hydraulic support full lifecycle management platform based on digital twin, characterized in that: include: The digital twin construction module is used to establish a coupled mapping model between geological conditions and support behavior based on geological structure exploration data and hydraulic support working condition data, and generate a geological-support coupled digital twin. The dynamic load spectrum construction and analysis module is used to receive the output of the geological-support coupled digital twin and construct a dynamic load spectrum based on geological conditions and mining technology by real-time monitoring of the load changes of the hydraulic support during the advancement of the working face. The adaptive control strategy generation module is used to dynamically generate adaptive control strategies for hydraulic support groups based on the analysis results of dynamic load spectrum and the correlation between geological change trends and support status. The full lifecycle optimization module is used to optimize the parameters of hydraulic supports throughout their entire lifecycle, from installation and operation to maintenance, based on the execution feedback of adaptive control strategies and dynamic load spectrum data.

2. The mining hydraulic support full life cycle management platform based on digital twin as described in claim 1, characterized in that: The digital twin building blocks include: The geological structure detection unit is used to obtain geological structure data of the working face through channel wave seismic exploration and borehole inspection technology, identify geological structure anomaly areas, and form a digital model of geological structure by combining the internal images of rock strata obtained by borehole inspection technology. The support condition monitoring unit is used to collect pressure data, displacement data, attitude data and working resistance data of the hydraulic support in real time; The coupling mapping unit is used to establish a correlation model between geological anomaly areas and support working resistance, realize the quantitative impact analysis of geological conditions on support behavior, and fuse geological structure data with support working condition data to generate a geological-support coupled digital twin. The ground sound monitoring integration unit is used to analyze the spatiotemporal evolution of ground sound events by monitoring ground sound signals generated by rock mass fracturing, establish a mineral pressure manifestation prediction model based on ground sound parameters, and integrate ground sound monitoring data into a geological-support coupled digital twin.

3. The mining hydraulic support full life cycle management platform based on digital twin as described in claim 2, characterized in that: Geological structural anomaly zones are identified by the change in channel wave propagation velocity, and the degree of geological structural anomaly is obtained based on the change in channel wave propagation velocity. The images of the rock strata obtained by the borehole inspection instrument are constructed into rock strata image feature vectors, which include fault range, coal seam thickness and lithological classification number; A digital model of geological structure is obtained by fusing the geological structural anomaly index and the feature vector of rock strata image. ； in: and For weight parameters, As an index of geological structural anomaly, This is the feature vector of the rock strata image.

4. The mining hydraulic support full life cycle management platform based on digital twin as described in claim 2, characterized in that: The geological-support coupled digital twin is based on the working face area divided by the geological structure digital model. The quantitative relationship between geological parameters and support working resistance is established by multiple regression analysis, namely the coupled mapping model. The geological parameters include fault displacement, coal seam thickness variation rate and roof lithology index. The coupled mapping model is dynamically corrected by comparing the deviation between the predicted working resistance and the actual monitoring value in real time. The multiple linear regression relationship between geological parameters and support resistance is expressed as follows: ； in: It is a fault elevation difference. The rate of change of coal seam thickness. The roof lithology index, For regression coefficients, This is the error term.

5. The mining hydraulic support full life cycle management platform based on digital twin as described in claim 1, characterized in that: The dynamic load spectrum construction and analysis module includes: The load data acquisition unit is used to collect load data of the hydraulic support under different geological conditions in real time. The load feature extraction unit is used to extract feature parameters from the collected load data; The dynamic load spectrum generation unit is used to construct a dynamic load spectrum database that reflects changes in geological conditions based on characteristic parameters. The load spectrum prediction unit is used to predict the load spectrum characteristics of the future mining area based on the working face advancement plan and geological exploration data. The load spectrum comparison unit is used to perform similarity analysis between real-time acquired load data and historical load spectra to identify abnormal load patterns.

6. The mining hydraulic support full life cycle management platform based on digital twin as described in claim 5, characterized in that: Characteristic parameters include peak load, root mean square load value, number of load cycles, impact load frequency, and load asymmetry coefficient.

7. The mining hydraulic support full life cycle management platform based on digital twin as described in claim 1, characterized in that: The adaptive control strategy generation module includes: The load adaptive control unit is used to adjust the working parameters of the support according to the real-time load spectrum characteristics; Group collaborative optimization unit is used to optimize the collaborative control strategy of hydraulic support group based on load distribution law; The preventative parameter adjustment unit is used to adjust the support control parameters in advance based on the load spectrum prediction results.

8. The mining hydraulic support full life cycle management platform based on digital twin as described in claim 1, characterized in that: The full lifecycle optimization module includes: The cumulative damage assessment unit is used to calculate the cumulative damage degree of key components of the hydraulic support based on load spectrum data. The remaining life prediction unit is used to predict the remaining life of a component by combining material fatigue characteristics and cumulative damage. The maintenance strategy optimization unit is used to develop differentiated maintenance plans based on the remaining life prediction results.

9. A mining hydraulic support full lifecycle management platform based on digital twin as described in claim 7, characterized in that: The group collaborative optimization unit adopts a distributed collaborative control algorithm. Based on the load distribution law obtained by dynamic load spectrum analysis, a load balance optimization model of hydraulic support group is established. By calculating the load concentration in different areas of the working face in real time, the working parameters of the support group are dynamically adjusted to keep the variance of the working face support strength distribution within 15%.

10. A full life-cycle management system for mining hydraulic supports based on digital twins, characterized in that: include: The facilities and data acquisition layer includes: hydraulic support groups, coal mining machines, environmental sensors, and cloud / edge computing centers; The data and model layer includes: a 3D model library, a historical database, a real-time database, an expert knowledge base, and an algorithm model library; The platform layer constructed according to any one of claims 1-9 for the full life cycle management platform of mining hydraulic supports based on digital twins; The application service layer includes: Panoramic monitoring and virtual-real interaction, full lifecycle management, group collaborative intelligent control, intelligent decision-making and predictive maintenance.

Images