Intelligent supporting method and system for hydraulic support and electronic equipment

Abstract

The invention relates to the technical field of hydraulic support intelligent supporting scheme design, in particular to a hydraulic support intelligent supporting method and system and electronic equipment. The method comprises the following steps: collecting roof pressure, bracket attitude and geological structure parameters in real time through a multi-modal sensor array; performing spatio-temporal feature fusion by utilizing the edge computing nodes to generate a three-dimensional support state map; constructing a dynamic weight adjustment strategy based on a reinforcement learning algorithm, and generating multi-objective optimization support parameters; a control instruction is issued through a dual-mode communication network to realize self-adaptive adjustment of the supporting strength; and a closed-loop feedback verification mechanism is established in combination with the laser radar scanning roof sinkage. And a digital twin model and an energy consumption optimization function are integrated. The problems that a traditional method is single in data dimension, rigid in decision and insufficient in reliability are solved, dynamic optimization of support parameters, active prevention and control of risks and collaborative improvement of energy efficiency are achieved, and the support safety and the intelligent level under complex working conditions are remarkably improved.

Description

Technical Field

[0001] This invention relates to the field of intelligent support scheme design technology for hydraulic supports, specifically to an intelligent support method and system for hydraulic supports, and electronic equipment. Background Technology

[0002] Hydraulic supports, as the core support equipment in fully mechanized coal mining faces, directly impact mining efficiency and underground safety due to their level of intelligence. Traditional hydraulic support methods primarily rely on single sensor data (such as pressure or displacement) and fixed control strategies, resulting in the following technical bottlenecks: First, the data acquisition dimension is limited, failing to comprehensively perceive the synergistic effects of roof pressure distribution, support dynamic posture, and geological structural changes, leading to poor adaptability of support parameter settings to complex working conditions. Second, control strategies are mostly based on pre-set rules based on human experience, making it difficult to respond in real-time to dynamic changes such as roof pressure and coal wall spalling, easily resulting in insufficient support strength or overload. Third, the communication methods of existing systems have low reliability, easily experiencing command delays or loss in the complex electromagnetic environment underground, affecting control real-time performance. Furthermore, the lack of a closed-loop feedback verification mechanism leads to delayed evaluation of support effectiveness, making it difficult to correct support deviations in a timely manner. Although some studies have attempted to introduce intelligent algorithms, they mostly focus on single-objective optimization, failing to comprehensively address the synergistic issues of multiple objectives such as roof stability, equipment lifespan, and energy consumption. Therefore, there is an urgent need for an intelligent support solution that integrates multi-source sensing, dynamic decision-making, and reliable control to improve the adaptability, safety, and economy of the support system.

[0003] Therefore, the existing technology still needs further development. Summary of the Invention

[0004] The purpose of this invention is to overcome the above-mentioned technical deficiencies and provide a method and system for intelligent support of hydraulic supports, as well as electronic equipment, to solve the problems existing in the prior art.

[0005] To achieve the above-mentioned technical objectives, according to a first aspect of the present invention, the present invention provides an intelligent support method for hydraulic supports, comprising: S1. Real-time acquisition of working face roof pressure, support posture, and geological structure parameters through a multimodal sensor array; S2. Edge computing nodes are used to perform spatiotemporal feature fusion processing on the collected data to generate a three-dimensional support status map; S3. Construct a dynamic weight adjustment strategy based on reinforcement learning algorithm, and generate multi-objective optimized support parameters according to the state graph; S4. Control commands are sent to the hydraulic actuator via a dual-mode communication network to synchronously achieve adaptive adjustment of support strength; S5. Establish a closed-loop feedback mechanism and use lidar to scan the roof subsidence in real time to verify the support effect.

[0006] Specifically, the multimodal sensor array includes: Distributed fiber optic pressure sensor, MEMS inertial measurement unit, and ground-penetrating radar detection module.

[0007] Specifically, the spatiotemporal feature fusion processing employs an improved Kalman filter algorithm, and the spatial compensation coefficient is dynamically calculated based on the spacing between the support groups.

[0008] Specifically, the reinforcement learning algorithm adopts a deep deterministic policy gradient framework, and the reward function includes three dimensions: roof stability factor, energy efficiency coefficient, and equipment life loss rate.

[0009] Specifically, the dual-mode communication network consists of an industrial Ethernet backbone and a LoRa self-organizing network redundant channel, and the data packet retransmission mechanism adopts forward error correction coding.

[0010] Specifically, this also includes establishing a digital twin model, matching real-time support parameters with historical geological databases for similarity, and triggering expert system intervention when the matching degree is lower than a first preset threshold.

[0011] Specifically, the closed-loop feedback mechanism sets three levels of early warning thresholds: When the roof subsidence corresponding to the first threshold is less than or equal to the second preset threshold, the current support parameters are maintained. When the subsidence of the top plate corresponding to the secondary threshold is greater than the second preset threshold and less than or equal to the third preset threshold, parameter fine-tuning is initiated. When the subsidence of the roof exceeds the third preset threshold, the emergency support mode is triggered.

[0012] Specifically, it also includes an energy consumption optimization module, which dynamically adjusts the operating frequency of the hydraulic pump station through predictive control algorithms.

[0013] According to a second aspect of the present invention, a hydraulic support intelligent support system is provided, comprising: The acquisition module is used to collect data on the pressure on the top plate of the working face, the posture of the support, and geological structure parameters in real time through a multimodal sensor array; The control module is used to perform spatiotemporal feature fusion processing on the collected data using edge computing nodes to generate a three-dimensional support status map; it is used to construct a dynamic weight adjustment strategy based on reinforcement learning algorithms and generate multi-objective optimized support parameters according to the status map; it is used to send control commands to the hydraulic actuator through a dual-mode communication network to simultaneously realize adaptive adjustment of support strength; and it is used to establish a closed-loop feedback mechanism to verify the support effect by using lidar to scan the roof subsidence in real time.

[0014] According to a third aspect of the present invention, an electronic device is provided, comprising: a memory; and a processor, wherein the memory stores computer-readable instructions, which, when executed by the processor, implement the above-described intelligent hydraulic support method.

[0015] Beneficial effects: The intelligent hydraulic support method and system provided by this invention achieves the following core advantages through multi-dimensional technological innovation: 1. Enhanced Multi-Source Sensing and Data Fusion Capabilities: A multi-modal sensor array is constructed using distributed fiber optic pressure sensors, MEMS inertial measurement units, and ground-penetrating radar detection modules to comprehensively cover the microscopic distribution of roof pressure, the six-degree-of-freedom attitude of the support, and rock strata structure information. Combined with an improved Kalman filter algorithm, spatiotemporal feature fusion is performed to generate a high-precision three-dimensional support status map, solving the problems of local pressure blind spots and dynamic response lag caused by traditional single data sources.

[0016] 2. Dynamic Intelligent Decision Optimization: Based on the Deep Deterministic Policy Gradient (DDPG) framework, a reinforcement learning algorithm is designed. Through a multi-dimensional reward function that includes roof stability, energy efficiency and equipment life loss, the support parameters are dynamically adjusted in real time, taking into account both safe support and economical equipment operation, and overcoming the rigidity of traditional preset rule strategies.

[0017] 3. High reliability control and safety assurance: The dual-mode communication network consisting of industrial Ethernet and LoRa self-organizing network is adopted, combined with forward error correction coding mechanism to ensure reliable transmission of control commands in complex downhole environments; a closed-loop feedback is constructed by real-time scanning of roof subsidence by lidar, and a three-level early warning threshold is set to trigger a gradient response strategy, which significantly improves the proactive prevention and control capabilities against risks such as sudden roof delamination and rockburst.

[0018] 4. Enhanced System Collaboration and Scalability: The digital twin model provides expert decision support for abnormal working conditions by matching real-time data with the similarity database of historical geological data, solving the problem of lag in traditional systems that rely on human experience; the energy consumption optimization module dynamically adjusts the frequency of hydraulic pump stations based on predictive control algorithms, reducing ineffective energy consumption while ensuring support strength, and achieving green and efficient operation.

[0019] 5. Full-process autonomy and intelligence: The closed-loop design of the entire chain from data acquisition, edge computing, strategy generation to execution control eliminates manual intervention, greatly improves support response speed and consistency, and provides reliable technical support for safe and efficient mining under complex geological conditions. Attached Figure Description

[0020] Figure 1 This is a flowchart illustrating the intelligent hydraulic support method provided in a specific embodiment of the present invention. Figure 2 This is a schematic diagram of the system composition of the intelligent hydraulic support system provided in a specific embodiment of the present invention. Detailed Implementation

[0021] To enable those skilled in the art to better understand the technical solutions of the present invention, the technical solutions of the present invention will be clearly and completely described below with reference to the accompanying drawings. Based on the embodiments in this application, other similar embodiments obtained by those skilled in the art without creative effort should all fall within the scope of protection of this application. Furthermore, directional terms mentioned in the following embodiments, such as "up," "down," "left," and "right," are only for reference to the directions in the accompanying drawings; therefore, the directional terms used are for illustrative purposes and not for limiting the invention.

[0022] The present invention will be further described below with reference to the accompanying drawings and preferred embodiments.

[0023] Please see Figure 1 This invention provides an intelligent support method for hydraulic supports, comprising: S1. Real-time acquisition of working face roof pressure, support posture, and geological structure parameters through a multimodal sensor array.

[0024] Specifically, the multimodal sensor array includes: Distributed fiber optic pressure sensor, MEMS inertial measurement unit, and ground-penetrating radar detection module.

[0025] It should be noted that before step S1, the following steps are also included: The first preset threshold, the second preset threshold, and the third preset threshold are preset in the control module.

[0026] It is understood that the first preset threshold, the second preset threshold, and the third preset threshold can be specifically determined by the user of this invention according to actual needs, as long as they are applicable to the intelligent hydraulic support method proposed in this invention.

[0027] Preferably, the first preset threshold is 70%, the second preset threshold is 10 mm, and the third preset threshold is 25 mm. These settings were determined by the present invention's technicians through extensive testing and can effectively realize the intelligent hydraulic support method of the present invention.

[0028] It should be further explained that, regarding the configuration of the multimodal sensor array, the scheme designed in this invention includes: Distributed fiber optic pressure sensor: 32 measuring points are arranged along the top beam of the support, using wavelength-resolved fiber Bragg grating (FBG) technology. Each measuring point is spaced 5cm apart, covering a range of 1.6m (based on laboratory pressure distribution test data; a spacing greater than 5cm will result in a local pressure peak missed rate >20%). The measuring range is set to 0-50MPa (meeting the MT550 standard for coal mine hydraulic supports), with an accuracy of 0.5%FS (full-scale error) and a temperature compensation range of -20℃ to 80℃.

[0029] MEMS Inertial Measurement Unit: Employs an MPU-9250 nine-axis sensor (three-axis accelerometer + three-axis gyroscope + three-axis magnetometer), with a sampling frequency set to 200Hz (covering the support vibration fundamental frequency range of 10-80Hz). Angle measurement error compensation formula: θc=θr+0.0012·T²-0.15·T (θr is the original angle, T is the ambient temperature, unit °C), after compensation, the angle error is <0.1° (verified through high and low temperature chamber experiments).

[0030] Ground-penetrating radar (GPR) module: A Swedish MALAProEx radar with a center frequency of 400MHz is selected (the formula for balancing detection depth and resolution is: D=(v·t) / 2, where v=0.1m / ns is the propagation speed of electromagnetic waves in the rock strata; the detection depth reaches 4m at t=80ns, with a resolution of 10cm). A top-plate scan is performed every 30 seconds to generate a dielectric constant distribution map.

[0031] S2. Edge computing nodes are used to perform spatiotemporal feature fusion processing on the collected data to generate a three-dimensional support status map.

[0032] Specifically, the spatiotemporal feature fusion processing employs an improved Kalman filter algorithm, and the spatial compensation coefficient is dynamically calculated based on the spacing between the support groups.

[0033] It should be further explained that, regarding the spatiotemporal feature fusion processing, the solution designed in this invention includes: An improved Kalman filter algorithm is used, and the specific steps are as follows: Step 1: Define the state vector X = [Px, Py, Pz, α, β, γ]^T Where Px, Py, and Pz decibels are three-dimensional pressure values, α is the pitch angle, β is the roll angle, and γ is the yaw angle; Step 2: State prediction equation: X k =A·X {k-1} +B·u k +K s ·ΔS In the formula: A is a 6×6 state transition matrix (derived from the kinematic model of the support). B is the control input matrix (related to the hydraulic cylinder displacement uk). K s =d / (1.5·cosθ) is the spatial compensation coefficient (d is the distance between adjacent supports, with a reference value of 1.5m, and θ is the pitch angle of the current support). ΔS=S i -S {i-1} This is a vector representing the difference between data from adjacent sensors. Step 3: The time window is set to 150ms (based on stent action response test data), and the data within the window is processed using a moving average. P avg =(Σw i ·P i ) / Σw i Weight w i =1-0.8·(t i -t0) / Δt (Δt is the window duration) Where Pavg is the weighted average pressure, Pi is a single pressure data point, t0 is the start time of the time window, and ti is the acquisition timestamp of each pressure data point Pi, used to calculate its time offset relative to the start of the window.

[0034] S3. Construct a dynamic weight adjustment strategy based on reinforcement learning algorithm, and generate multi-objective optimized support parameters based on state graph.

[0035] Specifically, the reinforcement learning algorithm adopts a deep deterministic policy gradient framework, and the reward function includes three dimensions: roof stability factor, energy efficiency coefficient, and equipment life loss rate.

[0036] It should be further explained that the dynamic weight adjustment strategy based on reinforcement learning algorithm, which generates multi-objective optimized support parameters according to the state graph, includes: 1. Design of Deep Deterministic Policy Gradient (DDPG) Model ① Network structure design: (1) Actor network (policy function μ(s|θ^μ)): Input layer: 512 nodes (corresponding to state feature dimensions); Hidden layer 1: 256 nodes, activation function LeakyReLU (α=0.01); Hidden layer 2: 128 nodes, activation function LeakyReLU; Output layer: 3 nodes (support resistance F, lifting speed v, balance valve opening k), activation function Tanh (output normalized to [-1,1]).

[0037] (2) Critic network (Q-value function Q(s,a|θ^Q)): Input layer: 512 nodes (state features) + 3 nodes (action vectors); Hidden layer 1: 512 nodes, activation function LeakyReLU; Hidden layer 2: 256 nodes, activation function LeakyReLU; Output layer: 1 node (Q-value), no activation function.

[0038] ② Reward function design: R = 0.4·Rs + 0.3·Re + 0.3·Rw in: Rs = 1 - |ΔP / Pmax|, representing the stability of the roof plate, where ΔP is the pressure fluctuation and Pmax = 50MPa; Re = 1 - E / Ebase, representing the energy efficiency coefficient, where Ebase = 5 kW·h is the baseline energy consumption. Rw = 1 - W / Wmax, representing the equipment lifespan, and Wmax = 0.8, which is the hydraulic cylinder wear failure threshold. The selection criteria for the weighting ratio of 4:3:3: Based on 200 simulation experiments, this ratio reduces the amount of top plate delamination by 18% and energy consumption by 12% (compared to the traditional 1:1:1 weighting), which greatly extends the equipment life.

[0039] 2. Model Training Process Step 1: Initialize the experience replay buffer D with a capacity of 500,000 records (covering 10 typical geological conditions); Step 2: Noise exploration was conducted using the Ornstein-Uhlenbeck process with parameters θ=0.15 and σ=0.2 (ensuring motion space coverage > 85%). Step 3: Sample 256 data points for each training batch, and set the learning rate as follows: Actor network: 3×10^-4 (to prevent oscillations caused by excessively rapid policy updates); Critic network: 1×10^-3 (accelerates Q-value convergence); Step 4: Target network soft update coefficient τ = 0.005 (experiments show that this value improves the model convergence speed by 30%) S4. Control commands are sent to the hydraulic actuator via a dual-mode communication network to synchronously achieve adaptive adjustment of support strength.

[0040] Specifically, the dual-mode communication network consists of an industrial Ethernet backbone and a LoRa self-organizing network redundant channel, and the data packet retransmission mechanism adopts forward error correction coding.

[0041] It should be further explained that the step of sending control commands to the hydraulic actuator via a dual-mode communication network to synchronously achieve adaptive adjustment of support strength includes: 1. Design a dual-mode communication protocol: ①Design the data packet structure: |Frame header (0xAA55)|Instruction code (1 byte)|Time stamp (4 bytes)|Pressure data (4 bytes)|Location data (4 bytes)|CRC16 (2 bytes)| ② Forward error correction coding: Reed-Solomon (255,223) code is used, which can correct 16-byte errors (meeting the requirement of bit error rate <10^-6 in the multipath fading environment of coal mine roadways). ③ Communication delay control: Industrial Ethernet backbone transmission delay <10ms, LoRa redundant channel delay <50ms (measured RSSI=-75dBm when LoRa transmission distance in the tunnel is 800m). 2. Design a hydraulic actuation control scheme: Design the control equation for a proportional servo valve: u(t)=Kp·e(t)+Ki·∫e(t)dt+Kd·de(t) / dt In the formula: Kp·e(t) is the proportional term, which represents the instantaneous correction amount that is linearly related to the current error. In the control of the hydraulic support, if the pressure on the top plate suddenly increases (e(t) rises sharply), the proportional term will immediately output a high control amount u(t) to drive the hydraulic cylinder to pressurize rapidly. Ki·∫e(t)dt is the integral term, which represents the cumulative compensation for historical errors and eliminates steady-state errors. If the hydraulic system pressure drops slowly due to oil leakage, the integral term will gradually increase the output until the actual pressure returns to the set value. Kd·de(t) / dt is the differential term, used to predict the trend of error change and suppress system oscillation. When the roof pressure changes rapidly (such as when the coal mining machine cuts and causes impact load), the differential term outputs the reverse correction amount in advance to prevent system oscillation.

[0042] Kp=2.5, Ki=0.8, Kd=0.3 (tuned through step response experiment, overshoot <5%). e(t) = Fset - Factual is the pressure deviation, and the control period is 1ms; The hydraulic cylinder displacement closed-loop control accuracy is ±0.5mm (meets the requirements of MT550-1996 standard).

[0043] S5. Establish a closed-loop feedback mechanism and use lidar to scan the roof subsidence in real time to verify the support effect.

[0044] Specifically, this also includes establishing a digital twin model, matching real-time support parameters with historical geological databases for similarity, and triggering expert system intervention when the matching degree is lower than a first preset threshold.

[0045] Specifically, the closed-loop feedback mechanism sets three levels of early warning thresholds: When the roof subsidence corresponding to the first threshold is less than or equal to the second preset threshold, the current support parameters are maintained. When the subsidence of the top plate corresponding to the secondary threshold is greater than the second preset threshold and less than or equal to the third preset threshold, parameter fine-tuning is initiated. When the subsidence of the roof exceeds the third preset threshold, the emergency support mode is triggered.

[0046] It should be further explained that the establishment of the closed-loop feedback mechanism, which uses lidar to scan the roof subsidence in real time to verify the support effect, includes: 1. Design a digital twin matching algorithm Dynamic Time Warping (DTW) is used to calculate the similarity between real-time data and the historical database: In the formula: D is the minimum normalized distance between the two sets of data sequences; This refers to the i-th data point in a reference sequence (such as a standard action template). Let w(i) be the w-th data point corresponding to the sequence to be matched (such as real-time sensor data) under path w; w is the alignment path in Dynamic Time Warping (DTW), which indicates how to match point i in sequence x with point w(i) in sequence y, and N is the total number of matching point pairs in path w (not the length of the original sequence). The similarity threshold is preferably set at 70% (D≤0.3). This preferred value is based on the analysis of 100 cases of roof fall accidents. This threshold can identify 85% of abnormal working conditions (ROC curve AUC=0.92).

[0047] 2. As shown in Table 1, a three-level early warning mechanism is designed. Table 1. Three-level early warning mechanism It should be noted here that in the above formula: v min Minimum boost rate is the lowest pressure rise rate required for the hydraulic system to reach the target support force, used to ensure that the support can provide sufficient support strength within a specified time when the roof sinks rapidly; F max The ultimate support force that a hydraulic support can provide; t req This represents the maximum allowable response latency of the system. F curr This refers to the actual support force of the stent as monitored in real time. A represents the effective pressure-bearing area of ​​the hydraulic cylinder piston.

[0048] Specifically, it also includes an energy consumption optimization module, which dynamically adjusts the operating frequency of the hydraulic pump station through predictive control algorithms.

[0049] It should be further explained that, regarding the energy consumption optimization module, the solution designed in this invention includes: Design a Model Predictive Control (MPC) algorithm: ① Objective function: Constraints: -25Hz≤fk≤50Hz (pump station frequency); ΔP_k≤5MPa (pressure fluctuation threshold) In the formula: The objective function is... This is a frequency tracking term used to penalize the actual frequency. With set frequency To minimize deviations and ensure that control variables accurately track target values; For a specific moment; This is a power fluctuation term used to suppress power variations. This reduces system fluctuations and energy loss. α=0.7, β=0.3 (optimal weights determined through Pareto front analysis); Predict the time domain N=10 steps (corresponding to an actual duration of 30 seconds); The solver uses the IPOPT algorithm, with a computation time of <50ms / cycle.

[0050] The following specific examples further illustrate this point: 1. Set parameters: Hz (corresponding to a support force of 30MPa), ; 2. Control Process: The pressure on the top plate suddenly increased. Adjust to 50Hz.

[0051] Optimization results: The frequency tracking term dominates, and the frequency rises rapidly to 48Hz (deviation |48-50|=2Hz); Power change kW, squared term contribution .

[0052] Controller action: Adjusts the proportional valve opening to balance frequency tracking and power fluctuations, so that... Minimum.

[0053] Understandably, this formula is implemented in intelligent hydraulic support through multi-objective optimization: Precise control: The frequency tracking feature ensures that the support force responds quickly to changes in roof pressure.

[0054] Energy consumption optimization: Power fluctuation terms suppress ineffective energy loss and extend equipment life.

[0055] Dynamic balance: The weighting coefficients α and β can be flexibly adjusted according to the working conditions to adapt to complex downhole environments.

[0056] It is understood that the intelligent hydraulic support method and system provided by this invention achieves the following core advantages through multi-dimensional technological innovation: 1. Enhanced Multi-Source Sensing and Data Fusion Capabilities: A multi-modal sensor array is constructed using distributed fiber optic pressure sensors, MEMS inertial measurement units, and ground-penetrating radar detection modules to comprehensively cover the microscopic distribution of roof pressure, the six-degree-of-freedom attitude of the support, and rock strata structure information. Combined with an improved Kalman filter algorithm, spatiotemporal feature fusion is performed to generate a high-precision three-dimensional support status map, solving the problems of local pressure blind spots and dynamic response lag caused by traditional single data sources.

[0057] 2. Dynamic Intelligent Decision Optimization: Based on the Deep Deterministic Policy Gradient (DDPG) framework, a reinforcement learning algorithm is designed. Through a multi-dimensional reward function that includes roof stability, energy efficiency and equipment life loss, the support parameters are dynamically adjusted in real time, taking into account both safe support and economical equipment operation, and overcoming the rigidity of traditional preset rule strategies.

[0058] 3. High reliability control and safety assurance: The dual-mode communication network consisting of industrial Ethernet and LoRa self-organizing network is adopted, combined with forward error correction coding mechanism to ensure reliable transmission of control commands in complex downhole environments; a closed-loop feedback is constructed by real-time scanning of roof subsidence by lidar, and a three-level early warning threshold is set to trigger a gradient response strategy, which significantly improves the proactive prevention and control capabilities against risks such as sudden roof delamination and rockburst.

[0059] 4. Enhanced System Collaboration and Scalability: The digital twin model provides expert decision support for abnormal working conditions by matching real-time data with the similarity database of historical geological data, solving the problem of lag in traditional systems that rely on human experience; the energy consumption optimization module dynamically adjusts the frequency of hydraulic pump stations based on predictive control algorithms, reducing ineffective energy consumption while ensuring support strength, and achieving green and efficient operation.

[0060] 5. Full-process autonomy and intelligence: The closed-loop design of the entire chain from data acquisition, edge computing, strategy generation to execution control eliminates manual intervention, greatly improves support response speed and consistency, and provides reliable technical support for safe and efficient mining under complex geological conditions.

[0061] Please see Figure 2 The present invention provides another embodiment, which provides an intelligent hydraulic support system, the intelligent hydraulic support system comprising: The acquisition module 100 is used to collect data on the pressure on the top plate of the working face, the posture of the support, and geological structure parameters in real time through a multimodal sensor array; The control module 200 is used to perform spatiotemporal feature fusion processing on the collected data using edge computing nodes to generate a three-dimensional support status map; it is used to construct a dynamic weight adjustment strategy based on reinforcement learning algorithm and generate multi-objective optimized support parameters according to the status map; it is used to send control commands to the hydraulic actuator through a dual-mode communication network to simultaneously realize adaptive adjustment of support strength; and it is used to establish a closed-loop feedback mechanism to verify the support effect by using lidar to scan the roof subsidence in real time.

[0062] It should be noted that the intelligent hydraulic support method and system provided by this invention achieves the following core advantages through multi-dimensional technological innovation: 1. Enhanced Multi-Source Sensing and Data Fusion Capabilities: A multi-modal sensor array is constructed using distributed fiber optic pressure sensors, MEMS inertial measurement units, and ground-penetrating radar detection modules to comprehensively cover the microscopic distribution of roof pressure, the six-degree-of-freedom attitude of the support, and rock strata structure information. Combined with an improved Kalman filter algorithm, spatiotemporal feature fusion is performed to generate a high-precision three-dimensional support status map, solving the problems of local pressure blind spots and dynamic response lag caused by traditional single data sources.

[0063] 2. Dynamic Intelligent Decision Optimization: Based on the Deep Deterministic Policy Gradient (DDPG) framework, a reinforcement learning algorithm is designed. Through a multi-dimensional reward function that includes roof stability, energy efficiency and equipment life loss, the support parameters are dynamically adjusted in real time, taking into account both safe support and economical equipment operation, and overcoming the rigidity of traditional preset rule strategies.

[0064] 3. High reliability control and safety assurance: The dual-mode communication network consisting of industrial Ethernet and LoRa self-organizing network is adopted, combined with forward error correction coding mechanism to ensure reliable transmission of control commands in complex downhole environments; a closed-loop feedback is constructed by real-time scanning of roof subsidence by lidar, and a three-level early warning threshold is set to trigger a gradient response strategy, which significantly improves the proactive prevention and control capabilities against risks such as sudden roof delamination and rockburst.

[0065] 4. Enhanced System Collaboration and Scalability: The digital twin model provides expert decision support for abnormal working conditions by matching real-time data with the similarity database of historical geological data, solving the problem of lag in traditional systems that rely on human experience; the energy consumption optimization module dynamically adjusts the frequency of hydraulic pump stations based on predictive control algorithms, reducing ineffective energy consumption while ensuring support strength, and achieving green and efficient operation.

[0066] 5. Full-process autonomy and intelligence: The closed-loop design of the entire chain from data acquisition, edge computing, strategy generation to execution control eliminates manual intervention, greatly improves support response speed and consistency, and provides reliable technical support for safe and efficient mining under complex geological conditions.

[0067] In a preferred embodiment, this application also provides an electronic device, the electronic device comprising: The computer device includes a memory and a processor, wherein the memory stores computer-readable instructions that, when executed by the processor, implement the intelligent support method for hydraulic supports. The computer device can be broadly categorized as a server, terminal, or any other electronic device with the necessary computing and / or processing capabilities. In one embodiment, the computer device may include a processor, memory, network interface, communication interface, etc., connected via a system bus. The processor of the computer device can be used to provide the necessary computing, processing, and / or control capabilities. The memory of the computer device may include a non-volatile storage medium and internal memory. The non-volatile storage medium may store an operating system, computer programs, etc. The internal memory can provide an environment for the operation of the operating system and computer programs in the non-volatile storage medium. The network interface and communication interface of the computer device can be used to connect and communicate with external devices via a network. When the computer program is executed by the processor, it performs the steps of the method of the present invention.

[0068] This invention can be implemented as a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, causes the steps of the methods of embodiments of the invention to be performed. In one embodiment, the computer program is distributed across multiple network-coupled computer devices or processors, such that the computer program is stored, accessed, and executed in a distributed manner by one or more computer devices or processors. A single method step / operation, or two or more method steps / operations, may be executed by a single computer device or processor or by two or more computer devices or processors. One or more method steps / operations may be executed by one or more computer devices or processors, and one or more other method steps / operations may be executed by one or more other computer devices or processors. One or more computer devices or processors may execute a single method step / operation, or execute two or more method steps / operations.

[0069] It will be understood by those skilled in the art that the method steps of the present invention can be performed by a computer program instructing related hardware, such as a computer device or processor, which may be stored in a non-transitory computer-readable storage medium. When the computer program is executed, the steps of the present invention are performed. Depending on the context, any references herein to memory, storage, database, or other media may include non-volatile and / or volatile memory. Examples of non-volatile memory include read-only memory (ROM), programmable ROM (PROM), electrically programmable ROM (EPROM), electrically erasable programmable ROM (EEPROM), flash memory, magnetic tape, floppy disk, magneto-optical data storage device, optical data storage device, hard disk, solid-state drive, etc. Examples of volatile memory include random access memory (RAM), external cache memory, etc.

[0070] It is understood that the intelligent hydraulic support method and system provided by this invention achieves the following core advantages through multi-dimensional technological innovation: 1. Enhanced Multi-Source Sensing and Data Fusion Capabilities: A multi-modal sensor array is constructed using distributed fiber optic pressure sensors, MEMS inertial measurement units, and ground-penetrating radar detection modules to comprehensively cover the microscopic distribution of roof pressure, the six-degree-of-freedom attitude of the support, and rock strata structure information. Combined with an improved Kalman filter algorithm, spatiotemporal feature fusion is performed to generate a high-precision three-dimensional support status map, solving the problems of local pressure blind spots and dynamic response lag caused by traditional single data sources.

[0071] 2. Dynamic Intelligent Decision Optimization: Based on the Deep Deterministic Policy Gradient (DDPG) framework, a reinforcement learning algorithm is designed. Through a multi-dimensional reward function that includes roof stability, energy efficiency and equipment life loss, the support parameters are dynamically adjusted in real time, taking into account both safe support and economical equipment operation, and overcoming the rigidity of traditional preset rule strategies.

[0072] 3. High reliability control and safety assurance: The dual-mode communication network consisting of industrial Ethernet and LoRa self-organizing network is adopted, combined with forward error correction coding mechanism to ensure reliable transmission of control commands in complex downhole environments; a closed-loop feedback is constructed by real-time scanning of roof subsidence by lidar, and a three-level early warning threshold is set to trigger a gradient response strategy, which significantly improves the proactive prevention and control capabilities against risks such as sudden roof delamination and rockburst.

[0073] 4. Enhanced System Collaboration and Scalability: The digital twin model provides expert decision support for abnormal working conditions by matching real-time data with the similarity database of historical geological data, solving the problem of lag in traditional systems that rely on human experience; the energy consumption optimization module dynamically adjusts the frequency of hydraulic pump stations based on predictive control algorithms, reducing ineffective energy consumption while ensuring support strength, and achieving green and efficient operation.

[0074] 5. Full-process autonomy and intelligence: The closed-loop design of the entire chain from data acquisition, edge computing, strategy generation to execution control eliminates manual intervention, greatly improves support response speed and consistency, and provides reliable technical support for safe and efficient mining under complex geological conditions.

[0075] The technical features described above can be combined arbitrarily. Although not all possible combinations of these technical features are described, any combination of these technical features should be considered to be covered by this specification, provided that such combination does not contain contradictions.

[0076] The specific embodiments of the present invention described above do not constitute a limitation on the scope of protection of the present invention. Any other corresponding changes and modifications made in accordance with the technical concept of the present invention should be included within the scope of protection of the claims of the present invention.

Claims

1. A method for intelligent support of hydraulic supports, characterized in that, The method includes: S1. Real-time acquisition of working face roof pressure, support posture, and geological structure parameters through a multimodal sensor array; S2. Edge computing nodes are used to perform spatiotemporal feature fusion processing on the collected data to generate a three-dimensional support status map; S3. Construct a dynamic weight adjustment strategy based on reinforcement learning algorithm, and generate multi-objective optimized support parameters according to the state graph; S4. Control commands are sent to the hydraulic actuator via a dual-mode communication network to synchronously achieve adaptive adjustment of support strength; S5. Establish a closed-loop feedback mechanism and use lidar to scan the roof subsidence in real time to verify the support effect.

2. The intelligent support method for hydraulic supports according to claim 1, characterized in that, The multimodal sensor array includes: Distributed fiber optic pressure sensor, MEMS inertial measurement unit, and ground-penetrating radar detection module.

3. The intelligent support method for hydraulic supports according to claim 2, characterized in that, The spatiotemporal feature fusion processing employs an improved Kalman filter algorithm, and the spatial compensation coefficient is dynamically calculated based on the spacing between the support groups.

4. The intelligent support method for hydraulic supports according to claim 1, characterized in that, The reinforcement learning algorithm adopts a deep deterministic policy gradient framework, and the reward function includes three dimensions: roof stability factor, energy efficiency coefficient, and equipment life loss rate.

5. The intelligent support method for hydraulic supports according to claim 1, characterized in that, The dual-mode communication network consists of an industrial Ethernet backbone and a LoRa self-organizing network redundant channel, and the data packet retransmission mechanism adopts forward error correction coding.

6. The intelligent support method for hydraulic supports according to claim 1, characterized in that, It also includes establishing a digital twin model, matching real-time support parameters with historical geological databases for similarity, and triggering expert system intervention when the matching degree is lower than a first preset threshold.

7. The intelligent support method for hydraulic supports according to claim 1, characterized in that, The closed-loop feedback mechanism is configured with three levels of early warning thresholds: When the roof subsidence corresponding to the first threshold is less than or equal to the second preset threshold, the current support parameters are maintained. When the subsidence of the top plate corresponding to the secondary threshold is greater than the second preset threshold and less than or equal to the third preset threshold, parameter fine-tuning is initiated. When the subsidence of the roof exceeds the third preset threshold, the emergency support mode is triggered.

8. The intelligent support method for hydraulic supports according to claim 1, characterized in that, It also includes an energy consumption optimization module, which dynamically adjusts the operating frequency of the hydraulic pump station through predictive control algorithms.

9. A hydraulic support intelligent support system, characterized in that, include: The acquisition module is used to collect data on the pressure on the top plate of the working face, the posture of the support, and geological structure parameters in real time through a multimodal sensor array; The control module is used to perform spatiotemporal feature fusion processing on the collected data using edge computing nodes to generate a three-dimensional support status map; it is used to construct a dynamic weight adjustment strategy based on reinforcement learning algorithms and generate multi-objective optimized support parameters according to the status map; it is used to send control commands to the hydraulic actuator through a dual-mode communication network to simultaneously realize adaptive adjustment of support strength; and it is used to establish a closed-loop feedback mechanism to verify the support effect by using lidar to scan the roof subsidence in real time.

10. An electronic device, characterized in that, include: Memory; The system includes a processor, wherein the memory stores computer-readable instructions that, when executed by the processor, implement the intelligent support method for hydraulic supports according to any one of claims 1 to 8.

Images