Self-adaptive operation planning method for hydraulic support under surrounding rock condition

Abstract

The invention belongs to the field of self-adaptive control of hydraulic supports, and provides a self-adaptive operation planning method for hydraulic supports under surrounding rock conditions in order to solve the problem that the hydraulic supports are not dynamically adjusted according to the surrounding rock conditions at present. And a hydraulic-surrounding rock coupling dynamic model is built in combination with the surrounding rock model, and the dynamic behavior of the support is accurately depicted. The method comprises the following steps of: constructing a virtual scene in Unity 3D, performing visual simulation of surrounding rock by constructing a Mesh grid, realizing deformation of the Mesh grid by changing the position of a vertex in the grid, constructing a communication structure for data transmission between the Unity 3D and Simulink, and inputting a support pose of a hydraulic support in the Simulink and a stroke displacement of a hydraulic rod into the Unity 3D to drive a virtual hydraulic support in the Unity 3D to move. And the state of safely supporting the surrounding rock is achieved. Through cooperation of Mesh mesh deformation in Unity 3D and arrangement of an interaction interface, the visualization degree of the hydraulic support on surrounding rock supporting is improved.

Description

Technical Field

[0001] The present invention belongs to the field of hydraulic support adaptive control, and in particular relates to a hydraulic support adaptive operation planning method under surrounding rock conditions. Background Art

[0002] With the development of the coal industry, the requirements for coal mining efficiency and safety are increasing. Traditional hydraulic support control systems mostly rely on fixed-mode operations and lack flexibility and intelligence. Especially when faced with complex changes in surrounding rock geological conditions, traditional methods make it difficult to quickly adjust the support's operating parameters to achieve the optimal state. Moreover, in previous hydraulic support simulation systems, most of the analysis was based on the static conditions of the support and the coal seam roof, and no research was conducted from the perspective of surrounding rock changes. Existing simulation models ignore the bidirectional coupling between the nonlinear characteristics of the support's multi-body dynamics and the evolution of surrounding rock damage, resulting in prediction errors. Therefore, it is particularly important to develop a new method that can dynamically adjust the working mode of the hydraulic support according to the surrounding rock conditions. Summary of the Invention

[0003] In order to solve at least one of the above technical problems existing in the prior art, the present invention provides a method for adaptive operation planning of a hydraulic support under surrounding rock conditions.

[0004] The present invention is implemented by the following technical solution: a method for adaptive operation planning of a hydraulic support under surrounding rock conditions, comprising the following steps:

[0005] S1: Acquire surrounding rock information and construct a surrounding rock model based on the surrounding rock information;

[0006] S2: Establishing a dynamic model of a hydraulic support, the dynamic model of the hydraulic support includes a hydraulic system and a mechanical structure. The hydraulic system is constructed using the Simscape Fluids toolbox in MATLAB Simulink, and the mechanical structure is simulated using Simscape Multibody in MATLAB Simulink. A hydraulic-surrounding rock coupling model is constructed based on the dynamic model of the hydraulic support and the surrounding rock model, and preset nonlinear factors are embedded in the hydraulic-surrounding rock coupling model.

[0007] S3: establishing an operation prediction model, and performing operation simulation on the hydraulic-surrounding rock coupling model based on the surrounding rock information, and inputting the simulation results and the preset hydraulic support control instructions into the operation prediction model to obtain predicted operation data of the hydraulic support in the hydraulic-surrounding rock coupling model;

[0008] S4: Build a mine digital twin environment based on Unity3D, establish a communication connection with the hydraulic-surrounding rock coupling model through a communication interface, and then control the virtual hydraulic support in the mine digital twin environment to perform dynamic adjustment based on the predicted operation data.

[0009] Preferably, the hydraulic support dynamics model includes:

[0010] Use the Hydraulic Pump module to simulate the oil pump in the hydraulic system, use the 4-Way DirectionalValve module to simulate the control valve in the hydraulic system, use the Hydraulic Cylinder module to simulate the hydraulic cylinder in the hydraulic system, and use the Pressure Relief Valve module to simulate the behavior of the pressure relief valve in the hydraulic system;

[0011] Use Simscape Multibody to simulate the rigid body components of the mechanical structure, add rotational joint modules and translational joint modules to simulate the kinematic pairs of the mechanical structure, and then assemble the various components of the mechanical structure together.

[0012] Preferably, performing surrounding rock environment simulation based on the surrounding rock information includes:

[0013] The surrounding rock information is obtained and Simscape Multibody is used to simulate the dynamic behavior of the surrounding rock. The spring-damper system is also used to simulate the dynamic changes of the surrounding rock stiffness.

[0014] Different types of loads applied to the dynamic model of the hydraulic support are simulated based on the surrounding rock information, and multi-scenario simulation is performed through MATLAB Simulink.

[0015] Preferably, establishing a communication connection with the hydraulic-surrounding rock coupling model via a communication interface includes:

[0016] Utilizing MATLAB Simulink to configure a UDP communication module, the predicted operation data is sent to the mine digital twin environment through the communication interface;

[0017] In the mine digital twin environment, the communication interface is monitored and the predicted operation data is parsed through a C# script, and the mine digital twin environment drives the virtual hydraulic support to perform dynamic adjustments based on the parsing results.

[0018] Preferably, after step S4, the method further includes:

[0019] In the mine digital twin environment, based on the acquired surrounding rock information, mesh grids are used to simulate surrounding rock deformation, and real dynamic interaction is achieved by modifying vertex positions and adding mesh collision bodies and rigid body components.

[0020] A mathematical function is written in the C# script to control the sinking law of the surrounding rock. At the same time, the crack texture is simulated through Shader, and a particle system is added to achieve the falling effect of waste rock.

[0021] Preferably, a Mesh grid is used to simulate surrounding rock deformation, including:

[0022] In Unity3D, a dynamic terrain vertex array is generated through Perlin noise, and the vertex height is adjusted in real time to drive Mesh deformation.

[0023] Preferably, step S3 further includes:

[0024] Preprocessing the simulation results includes performing at least one of denoising, feature selection, and normalization operations on the data using a Kalman filter;

[0025] The prediction operation model is trained using a bidirectional LSTM neural network. The input data is spliced through forward and reverse window sliding, and the hydraulic-surrounding rock coupling model parameters are optimized in combination with cross validation and grid search.

[0026] Preferably, it also includes:

[0027] The preprocessed data is distributed as 75% training set, 15% validation set, and 15% test set, and the timestamps of the validation set and the test set are strictly lagging behind the training set.

[0028] Compared with the prior art, the present invention has the following beneficial effects:

[0029] 1. Compared with the existing models, this method proposes a hydraulic-surrounding rock coupling model, which breaks through the limitations of traditional static mechanical analysis and accurately describes the dynamic behavior of hydraulic supports.

[0030] 2. In the simulation of the surrounding rock environment, this method uses a non-uniform time-varying load function to simulate the roof pressure and periodic fracture process; and uses a spring-damper system to simulate the stiffness change of the surrounding rock.

[0031] 3. This method applies the bidirectional LSTM neural network system to the hydraulic support adaptation, and uses forward window sliding and reverse window sliding to incorporate the extracted data into the forward input data and reverse input data respectively, and then splices the two inputs to obtain the final input result. The impact of the data at the latter moment on the data at the previous moment can be obtained, which can make the prediction results more accurate and reliable.

[0032] 4. This method establishes a bidirectional coupling mechanism between the hydraulic support and the surrounding rock. Through an explicit-implicit hybrid solution algorithm, it realizes real-time interaction between the dynamic response of the support and the surrounding rock deformation field: the support force affects the stress distribution of the surrounding rock, and the surrounding rock deformation reacts to the stress of the support structure. The support reaction force is mapped into the surrounding rock boundary condition in real time, driving the calculation of the surrounding rock deformation in the next time step.

[0033] 5. This method uses a mesh to construct the surrounding rock in Unity3D. Mesh Collider and Rigid Body components are added to the mesh to give it dynamic characteristics, enabling collision simulation between the surrounding rock and the hydraulic support. The deformation and sinking of the surrounding rock can be achieved in Unity3D by modifying the mesh's vertex positions within the void Update function in a C# script. Mathematical functions that conform to the surrounding rock's sinking behavior are written within the void Update function to ensure the mesh sinks more realistically. Anti-gravity is also set for the top beam to ensure a more consistent support of the mesh. The mesh's vertex displacements are output in reverse to the Rigid Body to synchronize with Unity's physics engine. In the material shader, a crack texture is blended based on the vertex displacements to create surface cracks when the surrounding rock fractures. A particle system is added to create the effect of falling waste rock after the support moves forward. Finally, the dynamic mesh is marked using Mesh.MarkDynamic() to improve GPU update efficiency. BRIEF DESCRIPTION OF THE DRAWINGS

[0034] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the following briefly introduces the drawings required for use in the embodiments. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative work.

[0035] Figure 1 This is a flow chart of a method for adaptive operation planning of a hydraulic support under surrounding rock conditions provided by the present invention;

[0036] Figure 2 It is a schematic diagram of a hydraulic-surrounding rock coupling model constructed by Simulink in a method for adaptive operation planning of a hydraulic support under surrounding rock conditions provided by the present invention;

[0037] Figure 3 This is a schematic diagram of the operation of the Simulink end in a method for adaptive operation planning of a hydraulic support under surrounding rock conditions provided by the present invention;

[0038] Figure 4This is a schematic diagram of the Unity end construction in the method for adaptive operation planning of a hydraulic support under surrounding rock conditions provided by the present invention;

[0039] Figure 5 It is a virtual surrounding rock schematic diagram constructed by the Unity end in a method for adaptive operation planning of a hydraulic support under surrounding rock conditions provided by the present invention;

[0040] Figure 6 Schematic diagram of adaptive closed-loop control of an adaptive operation planning method of a hydraulic support under surrounding rock conditions provided by the present invention;

[0041] Figure 7 It is a structural schematic diagram of a method for adaptive operation planning of a hydraulic support under surrounding rock conditions provided by the present invention. DETAILED DESCRIPTION

[0042] The technical solutions in the embodiments of the present invention are clearly and completely described in conjunction with the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of them. All other implementations derived by ordinary technicians in this field based on the embodiments of the present invention without making any creative efforts are within the scope of protection of the present invention.

[0043] It should be noted that the structures, proportions, sizes, etc. illustrated in the drawings of this specification are only used to match the contents disclosed in the specification for people familiar with this technology to understand and read, and are not used to limit the conditions under which the present invention can be implemented. Therefore, they have no substantive technical significance. Any structural modification, change in proportional relationship or adjustment of size should fall within the scope of the technical content disclosed in the present invention without affecting the efficacy and purpose that can be achieved by the present invention. It should be noted that in this specification, relational terms such as first and second are only used to distinguish one entity from several other entities, and do not necessarily require or imply any actual relationship or order between these entities.

[0044] In related technical fields, digital twin technology is often used to build a three-dimensional model of a hydraulic support in Unity3D. A hydraulic support posture prediction model that integrates neural networks and genetic algorithms is used to achieve adaptive adjustment of the hydraulic support posture. However, the above method has the following drawbacks: 1) Unity3D is very complex in achieving dynamic simulation and precise perception, and is not as accurate and convenient as Simulink. 2) The surrounding rock system is not taken into consideration. As coal resource development continues to deepen, the complex phenomena of movement, fracture, and delamination of the surrounding rock in the fully-mechanized mining face during mining have placed higher demands on the support performance of the hydraulic support.

[0045] The present invention mainly solves the problem of how to construct a more realistic surrounding rock environment so that the hydraulic support can accurately perceive the surrounding environment and make autonomous decisions to achieve the optimal support posture.

[0046] like Figure 1 As shown, an embodiment of the present invention provides a flow chart of a method for adaptive operation planning of a hydraulic support under surrounding rock conditions, including the following steps:

[0047] S1: Obtain surrounding rock information and construct a surrounding rock model based on the surrounding rock information.

[0048] In this embodiment, a variety of sensors (such as pressure sensors, displacement sensors, and strain sensors) are used to monitor the pressure distribution, deformation, and other information of the surrounding rock in real time. A high-precision surrounding rock model is created using three-dimensional laser scanning technology.

[0049] In this embodiment, the pressure sensor uses a high-precision fiber optic pressure sensor, which is resistant to electromagnetic interference, corrosion-resistant, and adaptable to the humid environment of the underground mine. The displacement sensor uses a laser displacement sensor, which has high measurement accuracy and is suitable for long-distance and non-contact measurement. The strain sensor incorporates resistance strain gauges embedded in key structural components of the support (top beam and columns), forming a Wheatstone bridge circuit. The data acquisition and transmission system utilizes a distributed acquisition module to improve system reliability and security. The backbone network utilizes an industrial-grade gigabit fiber optic ring network.

[0050] In this embodiment, the collected rock pressure, stress, and displacement data are processed and feature extracted, including noise filtering and outlier monitoring. Linear interpolation is used when the data that should have been collected is missing due to various reasons.

[0051] S2: Establish a dynamic model of a hydraulic support, which includes a hydraulic system and a mechanical structure. The hydraulic system is constructed using the Simscape Fluids toolbox in MATLAB Simulink, and the mechanical structure is simulated using Simscape Multibody in MATLAB Simulink; and a hydraulic-surrounding rock coupling model is constructed based on the hydraulic support dynamic model and the surrounding rock model, and preset nonlinear factors are embedded in the hydraulic-surrounding rock coupling model.

[0052] In this embodiment, the hydraulic support is decomposed into multiple rigid-body subsystems, including columns, top beams, bases, four-bar linkages, and side guards. Pre-set nonlinear factors are then embedded into the hydraulic-rock coupling model. These factors include hydraulic system nonlinearity and contact nonlinearity. Hydraulic system nonlinearity at least includes: a hydraulic circuit transfer function that considers the compressibility of the oil in the hydraulic cylinder, pipeline pressure loss, and valve port flow-pressure characteristics; contact nonlinearity includes a friction contact force model between the top beam and the roof. Furthermore, during the construction of the hydraulic-rock coupling model, a non-uniform time-varying load function is used to simulate the roof pressure and cyclic fracture process.

[0053] In this embodiment, the Simulink tool module in MATLAB is used to assemble and adjust the hydraulic support. The Hydraulic Pump module is used to simulate the oil pump in the hydraulic system and set the flow and pressure parameters. The 4-Way Directional Valve module is used to simulate the control valve in the hydraulic system and define the relationship between the valve core displacement and the flow rate. The Hydraulic Cylinder module is used to simulate the hydraulic cylinder in the hydraulic system. Parameters such as cylinder diameter, rod diameter, and stroke are input, pipelines are connected, and oil properties (viscosity and density) are considered. The Pressure Relief Valve module is used to simulate the behavior of the pressure safety valve in the hydraulic system and set the safety pressure threshold.

[0054] Use Simscape Multibody to simulate the rigid body components of the mechanical structure, define the rigid body components of the bracket (top beam, column, base, etc.), add revolute joint (Revolute Joint) and translation joint (Prismatic Joint) modules to simulate the kinematic pairs of the mechanical structure, and then assemble the various components of the mechanical structure together.

[0055] In this embodiment, an External Force module can be added to the mechanical mechanism to simulate roof pressure or coal rock load, and a Spatial Contact Force module is used to define the contact force between the hydraulic support and the roof / bottom plate.

[0056] In this embodiment, Figure 2 As shown in the figure, the specific steps for constructing the dynamic model of the hydraulic support are as follows:

[0057] Export the hydraulic support 3D model created in SolidWorks as ".STL" files according to different components, including the base, front connecting rod, rear connecting rod and other components;

[0058] Add the exported 3D model file to the "File Solid" module of the simulation model and define the object's mass, density and other parameters;

[0059] Add a "Rigid Transform" coordinate system transformation relationship module between the components of the hydraulic support to connect the joints and define the coordinate transformation parameters. When changing the coordinate position, you can directly click the "Rigid Transform" module, select Cartesian and enter the target position.

[0060] Analyze the motion relationships between the components of the hydraulic support and add constraint modules between them. Add joint components such as "Prismatic Joint" and "Revolute Joint" to connect the components of the hydraulic support. Build a hydraulic system for the hydraulic components, add a Fixed-Displacement Pump module for fluid supply, and then set the pump's operating efficiency and volumetric efficiency. Add a 4-Way 3-Position Directional Valve module to simulate the control valve and define the relationship between valve core displacement and flow rate. Add a Hydraulic Cylinder module and enter parameters such as cylinder diameter, rod diameter, and stroke.

[0061] After assembling the hydraulic support, embed the nonlinear factors:

[0062] Valve port flow nonlinearity: Use the Hydraulic Resistive Tube module to set the pipe inner diameter and inner surface roughness to simulate the nonlinear flow characteristics of the valve port;

[0063] Define the flow coefficient curve: simulate flow-pressure data through the Lookup Table; set the deadband and hysteresis parameters to simulate the opening and closing characteristics of the actual valve.

[0064] Oil compressibility and leakage: Use the Hydraulic Chamber module and set parameter options; define the nonlinear leakage model of the piston seal by setting the Leakage Area parameter in the three-position four-way valve.

[0065] Top beam-top plate contact nonlinearity:

[0066] Normal contact force: Use the Spatial Contact Force module and set the parameters after activating Provided by Input.

[0067] Tangential friction: Enable the Friction option and set the static friction coefficient and dynamic friction coefficient;

[0068] Use Ideal Force Source to simulate the initial contact preload. The force source signal can be controlled by the Step module to determine when to cancel the preload.

[0069] S3: Establish an operation prediction model, and perform operation simulation on the hydraulic-surrounding rock coupling model based on the surrounding rock information, and input the simulation results and preset hydraulic support control instructions into the operation prediction model to obtain predicted operation data of the hydraulic support in the hydraulic-surrounding rock coupling model.

[0070] In this embodiment, a mapping relationship of dynamic responses between the operation prediction model and the hydraulic-coupling model is established.

[0071] Optionally, the surrounding rock information is obtained and Simscape Multibody is used to simulate the dynamic behavior of the surrounding rock, and a spring-damper system is used to simulate the dynamic change of the surrounding rock stiffness; different types of loads applied to the dynamic model of the hydraulic support are simulated based on the surrounding rock information, and multi-scenario simulation is performed through MATLAB Simulink.

[0072] In this example, the dynamic behavior of the surrounding rock primarily includes stress, strain, and displacement. Because the surrounding rock is not an absolutely rigid body in reality and exhibits a certain degree of elasticity, a spring-damper system is used for simulation. Variations in the surrounding rock's stiffness are achieved by adjusting the spring-damper coefficients.

[0073] In this embodiment, different types of loads are applied to the hydraulic support dynamic model by collecting different surrounding rock information (such as normal mining, basic roof pressure, roof instability, cantilever beam rotation, etc.). Multi-scenario simulation is performed through MATLAB Simulink to collect the operating data of the hydraulic support dynamic model under various load and surrounding rock movement conditions, including column stroke, pressure distribution, etc.

[0074] In this embodiment, Figure 2 As shown in the figure, the surrounding rock and its non-rigid properties are set in Simscape Multibody:

[0075] A non-uniform time-varying load function was used to simulate the roof pressure and periodic fracture process. The Sine Wave module was used to generate the fundamental frequency as the fracture period, and the Pulse Generator module was used to simulate the instantaneous pressure.

[0076] Then use the spring-damper system to simulate the stiffness change of the surrounding rock. Build the basic components, select the spring element, and then set the initial parameters; select the damping element and add the Translational Damper module; select the mass block connection and use the Mass module to simulate the surrounding rock action point.

[0077] Dynamic stiffness control methods include:

[0078] Time-driven stiffness changes, using Clock+MATLAB Function to generate time-varying curves; displacement feedback stiffness adjustment, modifying parameters in real time through feedback signals.

[0079] Finally, perform dynamic parameter injection:

[0080] Online parameter modification: Right-click the spring module and check Specify stiffness via input port to add a new input port to connect the control signal.

[0081] Random Perturbation Overlay: Add Band-Limited White Noise module to simulate geological uncertainties.

[0082] Optionally, it also includes: preprocessing the simulation results, including using Kalman filtering to perform at least one of denoising, feature selection and normalization operations on the data; using a bidirectional LSTM neural network to train the prediction operation model, sliding and splicing input data through forward and reverse windows, and optimizing the hydraulic-surrounding rock coupling model parameters in combination with cross-validation and grid search.

[0083] In this example, the non-rigid properties of the surrounding rock are addressed using a spring-damper system in Simulink, with stiffness changes achieved by adjusting its coefficients. A bidirectional coupling mechanism is built into the operational prediction model, and an explicit-implicit hybrid solution algorithm is used to achieve real-time interaction between the dynamic response of the hydraulic support and the surrounding rock deformation field: the hydraulic support's support force influences the surrounding rock stress distribution, and the surrounding rock deformation reacts to the hydraulic support's structural stress.

[0084] In this embodiment, Figure 3 As shown in the figure, the multidimensional data obtained through simulation is first preprocessed, including data denoising, feature selection, and normalization using Kalman filtering. A bidirectional LSTM neural network is then used to train the processed data. The operational prediction model parameters are optimized through cross-validation and grid validation. The trained model is then applied to a real-time monitoring system, dynamically adjusting the hydraulic support parameter settings based on the current surrounding rock conditions and the mapping relationship between the operational prediction model and the hydraulic-coupling model.

[0085] Optionally, the preprocessed data is distributed as 75% training set, 15% validation set, and 15% test set, and the timestamps of the validation set and the test set are strictly lagging behind the training set to avoid future information leakage.

[0086] In this embodiment, a bidirectional coupling mechanism is constructed: model segmentation. An explicit subsystem (surrounding rock deformation field) and an implicit subsystem (support dynamics) are used. Data exchange channels are: support force → explicit surrounding rock boundary conditions; surrounding rock displacement field → implicit support constraint force. Support reaction forces are mapped to surrounding rock boundary conditions in real time, driving the calculation of surrounding rock deformation in the next time step.

[0087] After preprocessing the collected signals, they are time-segmented to prevent future information leakage. Stratified sampling of operating conditions ensures uniform distribution across all operating conditions. The LSTM neural network architecture and time window construction are then performed. LSTM can handle long-term data and mitigate the problems of vanishing and exploding gradients. A reverse LSTM neural network is constructed on top of the existing LSTM neural network, forming a bidirectional LSTM. Forward and reverse window sliding are used to incorporate the extracted data into the forward and reverse inputs, respectively. The two inputs are then concatenated to produce the final input. This captures the impact of the data at the later moment on the data at the previous moment, ensuring more accurate and reliable predictions. After constructing the input structure, deviations are normalized. After the prediction model is run, the output of the new predictions is also denormalized and incorporated into the loss function for comparison with the original data. Cross-validation is used to assess model stability, and grid search is performed to find the optimal hyperparameter combination.

[0088] S4: Build a mine digital twin environment based on Unity3D, establish a communication connection with the hydraulic-surrounding rock coupling model through a communication interface, and then control the virtual hydraulic support in the mine digital twin environment to perform dynamic adjustment based on the predicted operation data.

[0089] Optionally, a communication connection is established with the hydraulic-surrounding rock coupling model through a communication interface, including: using MATLAB Simulink to configure a UDP communication module, and sending the predicted operation data to the mine digital twin environment through the communication interface; in the mine digital twin environment, the communication interface is monitored and the predicted operation data is parsed through a C# script, and the mine digital twin environment drives the virtual hydraulic support to perform dynamic adjustments based on the parsing results.

[0090] In this embodiment, Figure 4As shown, based on the mapping relationship between the operational prediction model and the hydraulic-rock coupling model, data interaction between Unity3D and Simulink is established. A highly simulated mine digital twin environment is created in Unity3D. The hydraulic-rock coupling model is imported and precisely assembled and physically constrained. Using Unity3D's UI display panel and control panel, users can intuitively monitor various support parameters and input different operating conditions. Through a TCP communication interface, the mine digital twin environment constructed in Unity3D seamlessly integrates with the hydraulic-rock coupling model. Consequently, when users specify pre-set hydraulic support control instructions, such as load conditions and roof displacement parameters, or send control commands in the UI display panel and control panel, these pre-set hydraulic support control instructions are transmitted in real time to the hydraulic-rock coupling model. Based on these inputs, the hydraulic-rock coupling model calculates the corresponding travel and posture changes and returns the results to Unity3D, driving the virtual hydraulic support to dynamically adjust according to the predicted results.

[0091] In this example, the hydraulic-rock coupling model was imported into 3D Max for preprocessing. The model was then reduced in size and its dimensions adjusted. The corresponding model file was then imported into the Unity3D environment in ".fbx" format. Physical constraints were added to each structural model to achieve virtual assembly of the hydraulic support within the mine's digital twin environment.

[0092] Write control scripts on the digital twin end to control the movements of the virtual hydraulic support, so that the virtual hydraulic support can achieve the basic movements required for virtual simulation;

[0093] Design an interactive interface and use UGUI to design the hydraulic support operation panel. Users can input the working parameters of the support through the graphical interface. Simulate the manual control simulation of the actual hydraulic support controller. You can also input different working conditions through the interface to communicate with the Python model.

[0094] Add a UDP Send / Receive module in Simulink, configure the receiving and sending ends, set the IP address and signal port, select Block Parameters, and configure the parameters. Connect the data to be sent (such as hydraulic cylinder displacement and pressure) to the input port of the UDP Send module, and then connect the Byte Pack module. For the data to be received, add a UDPRecieve module, then connect the Byte Unpack module and the Terminator module to convert it into a recognizable input signal. Create a new C# script in Unity3D, write the listening port and received data, etc., and bind specific function buttons to the communication script. After the communication settings are made, the stroke change data output by the Simulink model will be fed back to Unity3D in real time, driving the virtual hydraulic support to complete the corresponding actions. When different working conditions are selected in the interactive interface, the Simulink model can also be updated accordingly. Through the visual interface, users can observe the dynamic response of the support.

[0095] Optionally, it also includes: in the digital twin environment of the mine, based on the acquired surrounding rock information, using Mesh grid to simulate the deformation of the surrounding rock, and realizing real dynamic interaction by modifying the vertex position and adding mesh collision bodies and rigid body components; writing mathematical functions in C# scripts to control the sinking law of the surrounding rock, and simulating crack textures through Shader, and adding particle systems to realize the falling effect of waste rock.

[0096] In this embodiment, Figure 5 As shown in the figure, in Unity3D, based on the surrounding rock environment in Simulink, a mesh is used to simulate the surrounding rock. Mesh Collider and Rigid Body components are added to give it dynamic characteristics, allowing collision between the surrounding rock and the hydraulic support. At the same time, anti-gravity is set for the top beam to ensure that the top beam supports the mesh more closely.

[0097] In this example, the deformation and sinking effect of the surrounding rock is achieved by changing the positions of mesh vertices within the void Update function of a C# script. A mathematical function that conforms to the laws of surrounding rock sinking is written within the void Update function to make the mesh sinking more realistic. A shader simulates surface cracks when the surrounding rock breaks. A particle system is used to create the falling waste rock effect.

[0098] In this embodiment, a procedural Mesh is created in a C# script, vertex coordinates are defined, triangle patches are set, UV mapping is added, normals are calculated, and a Mesh grid is formed. When it is necessary to generate a dynamic terrain that matches the sinking of the surrounding rock, a vertex array is first generated, and for complex surrounding rocks, Perlin noise is used to generate the terrain. Then, triangle indexing is performed to generate the mesh, and finally, the dynamic mesh is realized by modifying the vertex height and recalculating the normal in real time. The real-time height of the surrounding rock vertices is data-driven by the surrounding rock model in Simulink, and the Mesh vertex displacement is output in reverse to the Rigidbody to achieve synchronization with Unity's physics engine. In the material Shader, the crack texture is mixed according to the vertex displacement to add surface cracks when the surrounding rock breaks. A particle system is added to realize the effect of falling waste rock after the support moves forward. Finally, Mesh.MarkDynamic() is used to mark the dynamic Mesh to improve the GPU update efficiency.

[0099] The present invention provides a method for adaptive operation planning of hydraulic supports under surrounding rock conditions, such as Figure 6 、 Figure 7This method combines the powerful mathematical modeling capabilities of MATLAB Simulink with the excellent 3D graphics processing and interactive design features of Unity3D, achieving an integrated process from model building, simulation testing, to final solution output. The hydraulic support components are imported into Simulink and assembled through coordinate transformation and the addition of kinematic joints. A hydraulic drive system is then added, or force is directly applied, to drive the hydraulic device's motion. After obtaining parameters such as surrounding rock pressure and displacement from sensors in a real coal mine, a coal mine simulation environment consistent with the actual surrounding rock conditions is added to Simulink, enabling the hydraulic support to safely support the working face. To achieve a more realistic model, nonlinear factors are embedded in the hydraulic support, using a spring-damper system to simulate changes in surrounding rock stiffness. By applying different surrounding rock conditions (such as normal mining, basic roof pressure, roof instability, and cantilever beam rotation), the hydraulic support dynamically adjusts to surrounding rock changes, improving its support efficiency. A bidirectional coupling mechanism is established to enable real-time interaction between the support's dynamic response and the surrounding rock deformation field, resulting in more accurate simulation results. The data was divided into training, validation, and test sets—for example, 70% of the data was used for training, 15% for validation, and 15% for testing. The predictive model was then run to learn from different surrounding rock conditions, ensuring that the hydraulic support always maintained a safe support state. Cross-validation was used to assess the model's stability, and grid search was used to find the optimal hyperparameter combination for model optimization and evaluation. A virtual scene was constructed in Unity3D, and a mesh was constructed to visualize the surrounding rock. The mesh was deformed by changing the positions of its vertices. Mathematical functions that conform to the laws of surrounding rock subsidence were added to the code that modified the vertex positions to achieve a visual effect of surrounding rock subsidence. A communication structure was established between Unity3D and Simulink for data transmission. The support pose and hydraulic rod travel displacement of the hydraulic support in Simulink were input into Unity3D to drive the movement of the virtual hydraulic support in Unity3D, ensuring that it maintained a safe support state for the surrounding rock. By combining mesh deformation with interactive interface settings in Unity3D, the visualization of the hydraulic support's support of the surrounding rock was enhanced.

[0100] The foregoing description is merely a preferred embodiment of the present invention, but the scope of protection of the present invention is not limited thereto. Any modifications or substitutions that can be readily conceived by a person skilled in the art within the technical scope disclosed herein should be included within the scope of protection of the present invention. Therefore, the scope of protection of the present invention should be based on the scope of protection of the claims.

Claims

1. A method for adaptive operation planning of hydraulic supports under surrounding rock conditions, characterized in that: The steps include: S1: Acquire surrounding rock information and construct a surrounding rock model based on the surrounding rock information; S2: Establishing a dynamic model of a hydraulic support, the dynamic model of the hydraulic support includes a hydraulic system and a mechanical structure. The hydraulic system is constructed using the Simscape Fluids toolbox in MATLAB Simulink, and the mechanical structure is simulated using Simscape Multibody in MATLAB Simulink. A hydraulic-surrounding rock coupling model is constructed based on the dynamic model of the hydraulic support and the surrounding rock model, and preset nonlinear factors are embedded in the hydraulic-surrounding rock coupling model. S3: establishing an operation prediction model, and performing operation simulation on the hydraulic-surrounding rock coupling model based on the surrounding rock information, and inputting the simulation results and the preset hydraulic support control instructions into the operation prediction model to obtain predicted operation data of the hydraulic support in the hydraulic-surrounding rock coupling model; S4: Build a mine digital twin environment based on Unity3D, establish a communication connection with the hydraulic-surrounding rock coupling model through a communication interface, and then control the virtual hydraulic support in the mine digital twin environment to perform dynamic adjustment based on the predicted operation data.

2. The method for adaptive operation planning of a hydraulic support under surrounding rock conditions according to claim 1, characterized in that: The hydraulic support dynamic model includes: Use the Hydraulic Pump module to simulate the oil pump in the hydraulic system, use the 4-Way Directional Valve module to simulate the control valve in the hydraulic system, use the Hydraulic Cylinder module to simulate the hydraulic cylinder in the hydraulic system, and use the Pressure Relief Valve module to simulate the behavior of the pressure relief valve in the hydraulic system; Use Simscape Multibody to simulate the rigid body components of the mechanical structure, add rotational joint modules and translational joint modules to simulate the kinematic pairs of the mechanical structure, and then assemble the various components of the mechanical structure together.

3. The method for adaptive operation planning of a hydraulic support under surrounding rock conditions according to claim 1, characterized in that: Performing surrounding rock environment simulation based on the surrounding rock information includes: The surrounding rock information is obtained and Simscape Multibody is used to simulate the dynamic behavior of the surrounding rock. The spring-damper system is also used to simulate the dynamic changes of the surrounding rock stiffness. Different types of loads applied to the dynamic model of the hydraulic support are simulated based on the surrounding rock information, and multi-scenario simulation is performed through MATLAB Simulink.

4. The method for adaptive operation planning of a hydraulic support under surrounding rock conditions according to claim 1, characterized in that: And establish a communication connection with the hydraulic-surrounding rock coupling model through a communication interface, including: Utilizing MATLAB Simulink to configure a UDP communication module, the predicted operation data is sent to the mine digital twin environment through the communication interface; In the mine digital twin environment, the communication interface is monitored and the predicted operation data is parsed through a C# script, and the mine digital twin environment drives the virtual hydraulic support to perform dynamic adjustments based on the parsing results.

5. The method for adaptive operation planning of a hydraulic support under surrounding rock conditions according to claim 4, characterized in that: After step S4, the method further includes: In the mine digital twin environment, based on the acquired surrounding rock information, mesh grids are used to simulate surrounding rock deformation, and real dynamic interaction is achieved by modifying vertex positions and adding mesh collision bodies and rigid body components. A mathematical function is written in the C# script to control the sinking law of the surrounding rock. At the same time, the crack texture is simulated through Shader, and a particle system is added to achieve the falling effect of waste rock.

6. The method for adaptive operation planning of a hydraulic support under surrounding rock conditions according to claim 5, characterized in that: Use Mesh to simulate surrounding rock deformation, including: In Unity3D, a dynamic terrain vertex array is generated through Perlin noise, and the vertex height is adjusted in real time to drive Mesh deformation.

7. The method for adaptive operation planning of a hydraulic support under surrounding rock conditions according to claim 1, characterized in that: Step S3 further includes: Preprocessing the simulation results includes performing at least one of denoising, feature selection, and normalization operations on the data using a Kalman filter; The prediction operation model is trained using a bidirectional LSTM neural network. The input data is spliced through forward and reverse window sliding, and the hydraulic-surrounding rock coupling model parameters are optimized in combination with cross validation and grid search.

8. The method for adaptive operation planning of a hydraulic support under surrounding rock conditions according to claim 7, characterized in that: Also includes: The preprocessed data is distributed as 75% training set, 15% validation set, and 15% test set, and the timestamps of the validation set and the test set are strictly lagging behind the training set.

Images