Coal mine deep well elevator communication system and method based on intelligent metasurface assistance

Abstract

The invention discloses a coal mine deep well elevator communication system and method based on intelligent metasurface assistance, the system comprises Internet of Things equipment, an edge computing node, a cloud server and an intelligent metasurface, and the Internet of Things equipment comprises a tension sensor, a weighing sensor and an acceleration sensor; the tension sensor, the weighing sensor and the acceleration sensor are connected with the edge computing node through the intelligent metasurface, and the edge computing node is connected with the cloud server through a wired transmission medium. The method comprises the following steps: arranging an intelligent metasurface, and constructing an intelligent communication system; modeling is carried out on energy loss and time delay calculation problems in the unloading calculation process in the intelligent communication system; constructing a comprehensive benefit function fusing two dimensions of energy efficiency and calculation time delay; determining a maximum comprehensive benefit function as an optimization target for calculating an unloading strategy; constructing a security reinforcement learning model; and optimal calculation of unloading benefits is realized. According to the system and the method, the stability of the wireless data link in the elevator can be ensured, and the data transmission efficiency can be improved.

Description

Technical Field

[0001] The present invention belongs to the technical field of coal mine automation, intelligent communication and industrial Internet of Things, and specifically is a coal mine deep shaft hoist communication system and method based on intelligent metasurface assistance. Background Art

[0002] As a core component of the mine production system, the safety and efficiency of the operation of the deep shaft hoist system in coal mines are directly related to the stable operation of the production process and the safety of personnel. In view of the hidden dangers of overload and rope breaking caused by risk factors such as terminal load overload, loose rope jam, operating errors and maintenance negligence that may be encountered during the operation of the hoist, the current academic and industrial circles are committed to exploring innovative solutions to significantly improve the reliability and safety of the hoist system. Low-power wide area network technology, especially LoRa technology, has shown great application potential in mine remote monitoring systems due to its significant advantages of low power consumption and long-distance communication. LoRa technology achieves efficient wireless transmission of data by effectively utilizing unlicensed radio frequency bands, significantly reducing system energy consumption and communication delay. Furthermore, combined with mobile edge computing (MEC) technology, the data processing capability is pre-positioned at the edge of the device, which not only accelerates the real-time processing and analysis of data, but also significantly improves the response speed of the system, laying a solid foundation for the intelligent development of the Internet of Things (IoT) in mines. However, in the complex metal environment of coal mines, the attenuation problem of LoRa signals has become a key factor limiting its application efficiency.

[0003] To solve this technical bottleneck, intelligent metasurface (RIS) technology has been introduced into the mine communication system as a cutting-edge wireless communication solution. RIS achieves intelligent adaptive adjustment of beamforming by finely controlling the phase shift parameters of the reflection unit and integrating advanced reinforcement learning algorithms, effectively enhancing signal strength and communication stability, and providing an innovative and efficient solution for wireless communication in coal mines. However, the mine environment is complex and changeable. Factors such as electromagnetic interference, coal dust and water mist, and large mobile devices pose severe challenges to wireless communication. How to accurately obtain real-time communication models and channel state information has become a problem that needs to be solved urgently. In view of the limitations of traditional optimization methods in dealing with dynamic changes and uncertainties in mine environments, the introduction of more advanced optimization algorithms and artificial intelligence technologies has become an inevitable choice. Summary of the invention

[0004] In view of the problems existing in the above-mentioned prior art, the present invention provides a coal mine deep shaft hoist communication system and method based on intelligent metasurface assistance. The system uses LoRa wireless communication technology, combined with the characteristics of intelligent metasurface signal enhancement and adjustable self-reflection characteristics, to ensure the stability of the wireless data link inside the hoist, thereby improving the data transmission efficiency and ensuring the reliability and security of data transmission, which is conducive to the edge computing node and the cloud server to be able to perform real-time fusion processing on the data monitored by the weighing sensor, acceleration sensor and tension sensor, thereby realizing comprehensive and reliable monitoring of the hoist system status; the method has a high degree of intelligence and a simple implementation process. It aims at the signal attenuation and communication instability caused by the dense metal devices in the coal mine, and for the first time applies the intelligent metasurface technology to the special environment of the hoist tunnel. At the same time, through continuous interactive learning with the environment, the configuration strategy of the intelligent metasurface is continuously optimized using the deep reinforcement learning algorithm, which can accurately control the propagation path, amplitude and phase of the electromagnetic wave, significantly improve the strength and transmission stability of the wireless signal, effectively adapt to the dynamic changes such as the rapid movement of the hoist, and ensure the long-term stability and sustainable optimization capability of the internal communication quality of the hoist.

[0005] In order to achieve the above-mentioned object, the present invention provides a coal mine deep shaft hoist communication system assisted by an intelligent metasurface, comprising an Internet of Things device, an edge computing node, a cloud server and an intelligent metasurface; The Internet of Things device includes a tension sensor, a weighing sensor, and an acceleration sensor; the tension sensor has a wireless communication function, which is installed on the wire rope, and is used to collect the tension signal of the wire rope in real time, and transmit the tension signal to an external device through the LoRa wireless communication technology; the weighing sensor has a wireless communication function, which is installed at the connection node between the wire rope and the cage or the bucket, and is used to collect the weight signal of the cage or the bucket in real time, and transmit the weight signal to an external device through the LoRa wireless communication technology; the acceleration sensor has a wireless communication function, which is installed on the cage or the bucket, and is used to collect the acceleration signal of the cage or the bucket in real time, and transmit the acceleration signal to an external device through the LoRa wireless communication technology; The edge computing node has a wireless communication function. It is installed on the wall of the mine and is located near the wellhead. It is used to receive wireless signals through LoRa wireless communication technology, complete edge computing tasks based on the received wireless signals, and then transmit the obtained data to the cloud server through a wired transmission medium; The cloud server is set at a remote end, connected to the edge computing node, used to receive data uploaded by the edge computing node, and used to complete the core computing task; The intelligent metasurface is installed on the wall of the mine, and is used to receive and enhance the strength of wireless signals emitted by the IoT device, and transmit the received wireless signals to the edge computing node by reflection. At the same time, it is used to optimize the communication quality between the IoT device and the edge computing node by adjusting its own reflection phase shift coefficient, thereby ensuring reliable communication between the IoT device and the edge computing node.

[0006] In order to facilitate real-time monitoring of the operation of the hoist and to collect environmental information of the hoist, a monitoring camera is also included. The monitoring camera has a wireless communication function and is installed on the wall of the mine. It is used to collect image data of the monitoring area in real time and use the intelligent metasurface as a relay and enhancement medium for data transmission to ensure that the image data can be quickly and stably transmitted to the edge computing node.

[0007] Furthermore, in order to ensure the reliability of communication, the intelligent metasurface is located in the middle section of the deep shaft of the coal mine in terms of height.

[0008] As a preferred embodiment, the edge computing node consists of a wireless access point and a mobile edge computing server.

[0009] Preferably, the wired transmission medium is an optical cable or an electrical cable.

[0010] In the present invention, a tension sensor is installed on the wire rope, a weighing sensor is installed at the connection node between the wire rope and the cage or bucket, and an acceleration sensor is installed on the cage or bucket, so as to facilitate the real-time collection of the tension signal of the wire rope, the weight signal of the cage or bucket, and the acceleration signal of the cage or bucket during the operation of the hoist, and then the operation data of the hoist can be collected more comprehensively. The tension sensor, the weighing sensor and the acceleration sensor all use the LoRa wireless communication technology to transmit the collected signals to the external device, and the transmission of long-distance signals can be realized under the premise of low power consumption. The edge computing node receives the wireless signal through the LoRa wireless communication technology, and the distance signal can be received under the premise of low power consumption, and then the communication distance can be extended under the premise of low power consumption through the LoRa wireless network. The intelligent super surface is arranged in the deep well of the coal mine, and the intelligent super surface is added to the LoRa wireless network system, which can significantly enhance the strength and coverage of the wireless signal, effectively improve the stability of the wireless sensor network, and then can further effectively expand the coverage of the wireless signal. Using edge computing nodes to complete edge computing tasks can effectively reduce the computing pressure of cloud servers. At the same time, it can also effectively reduce dependence on cloud servers through autonomous computing. Connecting cloud servers and edge computing nodes through wired transmission media can ensure the reliability of long-distance data transmission. Since the intelligent metasurface can optimize the propagation and direction of signals by autonomously adjusting its own reflection characteristics, it can ensure that wireless signals can bypass the original physical obstacles and signal blind spots, so that the communication link can be dynamically adjusted according to actual needs. This not only overcomes the limitations of traditional methods that are difficult to cope with dynamic changes, but also significantly improves the adaptive ability and intelligence level of the communication system, effectively ensuring the reliability and security of data transmission.

[0011] Through LoRa wireless communication technology, combined with the signal enhancement characteristics of the intelligent metasurface and its adjustable reflection characteristics, the system can ensure the stability of the wireless data link inside the hoist, thereby improving the data transmission efficiency and ensuring the reliability and security of data transmission. It is beneficial for edge computing nodes and cloud servers to integrate and process the data monitored by weighing sensors, acceleration sensors and tension sensors in real time, thereby realizing comprehensive and reliable monitoring of the hoist system status, effectively solving the challenges of severe attenuation of wireless sensor signals in deep well hoists in coal mine scenarios due to the metal environment and high requirements for data stability and latency.

[0012] The present invention also provides a coal mine deep shaft hoist communication method based on intelligent metasurface assistance, using a coal mine deep shaft hoist communication system based on intelligent metasurface assistance, including the following steps: Step 1: Arrange the smart metasurface and build a smart communication system; Deploy the smart metasurface on the mine wall; use the programmable reflection characteristics of the smart metasurface to actively adjust the direction and phase of the reflected beam, build a virtual and enhanced wireless communication link in the complex mine environment, and use the wireless communication link to establish a communication connection between the IoT device and the edge computing node; use wired transmission media to establish a communication connection between the edge computing node and the remote cloud server; Step 2: Model the energy loss and computational delay issues in the computational offloading process of the intelligent communication system; S21: According to formula (1), the channel model of the smart metasurface is established to obtain the channel gain of the smart metasurface ; (1); In the formula, represents the direct path gain; represents the gain of the nth reflection path; Smart super surface n The complex reflection coefficient of each reflection element; S22: Establish a model of the smart metasurface receiving signal according to formula (2); (2); In the formula, Indicates k The received signal of each IoT device; Indicates the transmit power; Represents edge computing nodes and k The channel model between IoT devices, is the channel estimation error; Indicates the transmission signal; represents Gaussian white noise, with variance ; S23: Establish a model of the transmission rate of the smart metasurface according to formula (3); (3); In the formula, Indicates k The transmission rate of each IoT device; Indicates bandwidth; S24: Using formula (4) to characterize the energy consumption at the IoT device ; (4); In the formula, T Indicates the duration of processing subframe; represents the energy efficiency factor of IoT devices; Indicates kCPU cycle frequency of each IoT device; S25: Use formula (5) to characterize the energy consumption at the edge computing node ; (5); In the formula, Indicates the transmit power of IoT devices; Indicate to k The proportion of transmission duration of IoT devices; represents the energy efficiency factor of the edge computing node; Represents the computing power of the mobile edge computing server; S26: Using formula (6) to characterize the energy consumption at the cloud server ; (6); In the formula, Indicates the transmit power of the edge computing node; Indicates that the edge computing node gives k The proportion of transmission duration of IoT devices; represents the energy efficiency factor of the cloud server; Indicates the CPU cycle frequency of the cloud server; S27: Define average energy consumption and average equivalent delay as indicators for measuring the performance of intelligent communication systems; use formula (7) to characterize the average energy consumption during the calculation of unloading , using formula (8) to characterize and calculate the average equivalent delay in the offloading process ; (7); (8); In the formula, ; Indicates the number of offload tasks assigned to IoT devices, Indicates the number of offload tasks assigned to edge computing nodes, Indicates the number of offload tasks assigned to the cloud server, Indicates the total number of unloaded tasks; Indicates the number of computing cycles required to process a single bit of computing task; Indicates the computing offloading mode. When it is set to 1, it triggers the task migration mechanism for different computing resource levels, activating the task offloading mode for the direct processing capability of IoT devices, the local optimization processing capability of edge computing nodes, and the powerful centralized processing capability of cloud servers. Conversely, the corresponding mode is turned off. Step 3: Based on the characteristics of coal mine hoist fault detection based on multi-source data fusion, a computational offloading strategy that takes into account both real-time performance and resource constraints is constructed; according to formula (9), a comprehensive benefit function integrating the two dimensions of energy efficiency and computational delay is constructed. ; (9); In the formula, represents the weight parameter of energy efficiency, A weight parameter representing the total computation delay; Step 4: Establish the maximization of the comprehensive benefit function as the optimization goal of the computational offloading strategy, and construct the optimization framework of the computational offloading strategy according to formulas (a), (b), (c), (d), (e) and (f); through constraint (a), ensure that the transmission power of IoT devices and edge computing nodes strikes a balance between security and energy efficiency; through constraint (b), ensure that IoT devices, edge computing nodes and cloud servers must operate at the maximum computing power limited by their respective CPU frequencies; each level measures the ability to process communication data through CPU frequency to ensure that the system is efficient and stable, and realizes efficient flow and intelligent processing of data from the source to the cloud; through constraint (c), specify the fairness of power allocation in the intelligent communication system network; through constraint (d), consider the diversity of task modes; through constraint (e), define the value range of the phase shift of the intelligent metasurface reflection coefficient matrix; through constraint (f), ensure the minimum transmission rate requirement when channel estimation errors exist; ; In the formula, Indicates k Computation offloading mode for IoT devices; Step 5: Build a secure reinforcement learning model; The dynamic computation unloading process and RIS phase shift adjustment process of the coal mine deep shaft hoist communication system assisted by intelligent metasurface are modeled as Markov decision processes, and the wireless sensor network assisted by intelligent metasurface and the intelligent metasurface are used as the environment and learning agent respectively; at the same time, a Lagrangian opportunity constrained reinforcement learning method is proposed; finally, a safety reinforcement learning model of the coal mine deep shaft hoist communication system assisted by intelligent metasurface is constructed; Step 6: Achieve optimal computation offloading benefits; S61: First, the constrained Markov decision process problem is transformed into an unconstrained dual problem using the Lagrange multiplier method according to formula (10); then, the Lagrange proportional integral is formulated using the integral separation method to prevent the policy from being overly conservative; (10); In the formula, Represents the reward of reinforcement learning, which is used to measure the performance of the algorithm; is the Lagrangian penalty factor, which is used to control the degree of relaxation of the constraints; Indicates the total transmission rate of the IoT devices used Less than rate threshold The probability of A threshold indicating the probability of outage; S62: Introduce a mining IoT computing offloading mechanism based on deep deterministic policy gradient, and use a deep deterministic policy gradient algorithm to solve the optimal task offloading strategy; A1: Initialize based on the deep deterministic policy gradient algorithm; Initialize the relevant learning parameters based on the deep deterministic policy gradient algorithm; Initialize the Actor network parameters in the deep deterministic policy gradient algorithm and Critic network parameters , by and Network parameters assigned to the Target Actor network and the Target Critic network , , complete the initialization of the Target network parameters, and randomly give the initial environmental state of the coal mine deep shaft hoist communication system assisted by the intelligent metasurface ;in, , and Respectively represent the number of tasks assigned to different devices; , and All represent the parameters of task offloading; A2: Single-step exploration and strategy sampling: Perform one-step exploration within a round, and the Actor network initializes the system state s deterministic strategy based on the current input ; Then, the mining IoT equipment is based on and the noise sampled by the OU process , according to formula (11), select and calculate the unloading strategy ; In the process of executing the computation offloading strategy, by configuring the reflection phase shift , Number of tasks , calculate the mode factor and resource allocation factors The system performance is optimized and the utility value of computing offloading is evaluated, i.e., the reward UThen, based on the real-time observation of the computational offloading process, the system state is updated to provide a new starting point for the subsequent decision-making process. In this process, the dynamic changes of the computational workload and the location information of the IoT devices follow the Markov chain model to ensure the accurate re-evaluation of the environmental channel state information. Finally, the computational offloading experience containing the above key information is Integrate into cases and store in the experience pool for subsequent learning and optimization; (11); A3: Network parameter update; determine whether the number of computational offloading experiences in the experience pool has reached Z; if not, jump directly to step A4 and continue to explore the accumulated computational offloading experiences until the computational offloading experiences have accumulated to reach Z groups; if reached, based on the experience playback technology, the mining IoT device randomly samples Z groups of computational offloading experiences from the experience pool, i.e. , and use these computational offloading experiences to iteratively update the four neural network parameters, including the Actor network parameters , Critic network parameters , Target Actor network parameters and TargetCritic network parameters ;in, Indicates the current state; Indicates the action at the current moment; Represents the reward at the current moment; Indicates the state at the next moment; A31: Use the Adam optimizer as the gradient descent algorithm and use formula (12) and formula (13) to update the critic network parameters ; (12); (13); In the formula, represents the loss function; Represents the reduction coefficient in reinforcement learning; Represents the target Critic network; Represents an Actor network; Indicates the first n A loss function; represents the Critic network; Indicates the current state; Indicates the first n Actions; A32: Update the Actor network parameters according to formula (14) using the deterministic policy gradient ascent method ; (14); In the formula, Represents the Actor network deterministic policy gradient; represents mathematical expectation; Represents the policy gradient of the Critic network; Represents the policy gradient of the Critic network parameters; represents the policy function; A33: Using soft update learning rate , update the Target Actor network parameters according to formula (15) , update the Target Critic network parameters according to formula (16) , to slow down the tracking of Actor network parameters and Critic network parameters; (15); (16); A4: Round end judgment: judge whether the number of time slots reaches the maximum number of training steps in a round. If not, set the next state Initialize the system state for the new , return to A2; if reached, the round training ends.

[0013] As a preferred embodiment, in step A1 of step 6, the relevant learning parameters include the learning rate , Discount Factor , memory pool size, OU noise , soft update learning rate , , batch size Z, number of rounds, and exploration step size within a round.

[0014] In the present invention, the intelligent reflective surface is first arranged in the deep well of the coal mine, and the wireless communication link constructed by it is used to establish the communication connection between the Internet of Things device and the edge computing node. Based on the advantages of the signal enhancement and adjustable reflection characteristics of the intelligent metasurface, combined with the low-power long-distance transmission characteristics of the LoRa wireless communication technology, the coverage of the internal wireless communication network of the hoist can be further expanded. At the same time, the reliability and stability of the wireless communication link can be further improved, thereby ensuring that the monitoring data can be transmitted to the edge computing node in a timely and reliable manner. For the fault detection of coal mine hoists based on multi-source data fusion, for the problem of high data stability and delay requirements, while designing the communication unloading benefit function of the coal mine deep well hoist, the present invention also considers the problem of improving the minimum energy efficiency limit and reducing the calculation delay, and constructs a model of energy loss and calculation delay problems in the calculation unloading process of the mine Internet of Things device in the network system assisted by the intelligent metasurface, and then the optimization goal of maximizing the calculation unloading benefit function can be achieved by jointly designing the reflection phase shift and calculation unloading rate of the intelligent metasurface. Since it is difficult to accurately obtain real-time wireless communication models and channel state information underground in coal mines, and it will be subject to the non-convex constraints of the intelligent supersurface reflection phase shift coefficient and the minimum energy loss, in order to effectively solve the problem of difficulty in obtaining real-time information underground and non-convex constraints, the dynamic calculation unloading process of intelligent supersurface assisted coal mine deep well hoist communication and the intelligent supersurface phase shift adjustment process are modeled into Markov decision processes. At the same time, the present invention also introduces a Lagrangian opportunity constraint reinforcement learning scheme to ensure the real-time data processing and the timeliness of emergency response. As an emerging technology in the field of artificial intelligence, deep reinforcement learning has demonstrated excellent capabilities in handling optimization problems in complex dynamic environments. By constructing a model based on a finite state Markov decision process, the deep reinforcement learning algorithm can adaptively select the optimal calculation unloading strategy through continuous interactive learning with the environment without fully understanding the system model and specific channel state information, thereby ensuring efficient real-time data processing and rapid emergency response capabilities of mine Internet of Things equipment. In order to ensure that the requirements of the computing delay threshold are met and to ensure real-time data processing and accident emergency response, a Lagrangian opportunity-constrained reinforcement learning method is proposed, and finally a safe reinforcement learning model for coal mine deep shaft hoist communication assisted by intelligent metasurface is constructed. Then, the deep deterministic policy gradient (DDPG) algorithm is used to optimize the safety reinforcement model to achieve the optimal computing offloading benefit in a complex environment, ensuring the efficiency and safety of coal mine deep shaft hoist communication. By cleverly integrating the safety learning mechanism into the DRL framework, the safety of the exploration process can be ensured. The mechanism evaluates the risk level of the state-action pair by setting safety indicators and automatically avoids high-risk behaviors, effectively preventing the problem of excessive computing delay caused by random exploration and ensuring the stable operation of the mine Internet of Things system.Driven by the reinforcement learning algorithm, the elevator communication system can continuously monitor and analyze changes in the communication environment, autonomously adjust the phase shift parameters of the reflection unit, and respond to communication needs under different environmental conditions with the most optimized beam shape.

[0015] This method has a high degree of intelligence and a simple implementation process. It applies intelligent metasurface technology to the special environment of the hoist tunnel for the first time to address the problems of signal attenuation and communication instability caused by dense metal devices in coal mines. At the same time, through continuous interactive learning with the environment, the configuration strategy of the intelligent metasurface is continuously optimized using a deep reinforcement learning algorithm. It can accurately control the propagation path, amplitude and phase of electromagnetic waves, significantly improve the strength and transmission stability of wireless signals, effectively adapt to dynamic changes such as rapid movement of the hoist, and ensure the long-term stability and sustainable optimization capability of the internal communication quality of the hoist. BRIEF DESCRIPTION OF THE DRAWINGS

[0016] Figure 1 It is a structural schematic diagram of the hoist communication system part in the present invention; Figure 2 It is a principle block diagram of the circuit part in the present invention; Figure 3 It is a flow chart of the hoist communication method part in the present invention; Figure 4 It is a flowchart of the improved DDPG algorithm in the present invention.

[0017] In the figure: 1. Edge computing node, 2. Cloud server, 3. Smart metasurface, 4. Tension sensor, 5. Weighing sensor, 6. Acceleration sensor, 7. Wire rope, 8. Cage or bucket. DETAILED DESCRIPTION

[0018] The present invention will be further described below in conjunction with the accompanying drawings.

[0019] As a key equipment in mine production, deep shaft hoists in coal mines are responsible for the vertical transportation of materials and personnel. Their operating safety and efficiency are directly related to the production efficiency and personnel safety of the entire mine. Therefore, it is necessary to monitor their operating status and environmental parameters in real time. For this purpose, sensor nodes based on LoRa technology are used to collect data in real time and transmit them through a low-power wide area network. At the same time, edge computing nodes 1 are deployed nearby to perform preliminary processing on sensor data, reduce data transmission volume and reduce latency. However, the unique environmental characteristics of deep shafts in coal mines, such as deep depth, narrow tunnels, rough rock walls, and high concentrations of coal dust and ore dust, greatly hinder the transmission of wireless communication signals, forming serious signal blind spots. To this end, the Reconfigurable Intelligence Surface (RIS) technology is introduced in the coal mine deep shaft hoist scenario. The intelligent metasurface 3 is deployed at the appropriate position of the mine shaft wall or at the appropriate height around the hoist, and its programmable reflection characteristics are used to actively adjust the direction and phase of the reflected beam, thereby dynamically adjusting the electromagnetic characteristics to optimize the signal propagation path, and constructing a virtual and enhanced wireless communication link in the complex mine environment to ensure the stability and coverage of the signal in the complex underground environment. Finally, all processed data are gathered to the cloud server 2 for advanced analysis and report generation, providing real-time and comprehensive decision support for coal mine managers and realizing remote monitoring and control of deep shaft hoists.

[0020] Specifically, Figure 1 and Figure 2 As shown, the present invention provides a coal mine deep shaft hoist communication system based on intelligent metasurface assistance, including an Internet of Things device, an edge computing node 1, a cloud server 2 and an intelligent metasurface 3; The Internet of Things device includes a tension sensor 4, a weighing sensor 5, and an acceleration sensor 6; the tension sensor 4 has a wireless communication function, which is installed on the wire rope 7, and is used to collect the tension signal of the wire rope 7 in real time, and transmit the tension signal to an external device through the LoRa wireless communication technology; the weighing sensor 5 has a wireless communication function, which is installed at the connection node between the wire rope 7 and the cage or the bucket 8, and is used to collect the weight signal of the cage or the bucket 8 in real time, and transmit the weight signal to an external device through the LoRa wireless communication technology; the acceleration sensor 6 has a wireless communication function, which is installed on the cage or the bucket 8, and is used to collect the acceleration signal of the cage or the bucket 8 in real time, and transmit the acceleration signal to an external device through the LoRa wireless communication technology; The edge computing node 1 has a wireless communication function, which is installed on the mine wall and located near the wellhead, and is used to receive wireless signals through LoRa wireless communication technology, complete edge computing tasks according to the received wireless signals, and then transmit the obtained data to the cloud server 2 through a wired transmission medium; The cloud server 2 is set at a remote end, connected to the edge computing node 1, and is used to receive data uploaded by the edge computing node 1 and to complete the core computing task; The smart metasurface 3 is installed on the wall of the mine, and is used to receive and enhance the strength of the wireless signal emitted by the Internet of Things device, and transmit the received wireless signal to the edge computing node 1 by reflection. At the same time, it is used to optimize the communication quality between the Internet of Things device and the edge computing node 1 by adjusting its own reflection phase shift coefficient, thereby ensuring reliable communication between the Internet of Things device and the edge computing node 1.

[0021] In order to facilitate real-time monitoring of the operation of the elevator, a monitoring camera is also included. The monitoring camera has a wireless communication function and is installed on the wall of the mine to collect image data of the monitored area in real time. The smart metasurface 3 is used as a relay and enhancement medium for data transmission to ensure that the image data can be quickly and stably transmitted to the edge computing node 1.

[0022] In order to ensure the reliability of communication, the intelligent super surface 3 is located in the middle section of the deep shaft of the coal mine in the height direction.

[0023] As a preferred embodiment, the edge computing node 1 is composed of a wireless access point and a mobile edge computing (MEC) server. The wireless access point (AP) is used to receive wireless signals and transmit them to the mobile edge computing server, and the mobile edge computing server is used to process computing tasks from Internet of Things (IoT) devices.

[0024] Preferably, the wired transmission medium is an optical cable or an electrical cable.

[0025] Placing the intelligent metasurface (RIS) in the deep well of the coal mine and adding the intelligent metasurface to the LoRa wireless network system can significantly enhance the signal coverage and provide a stable wireless sensor network signal, which can effectively solve the problem that the wireless sensor signal inside the hoist is easily affected by the dense metal devices, resulting in serious signal attenuation, and thus effectively expand the coverage of the wireless signal. Specifically, through the two reflection paths of RIS, combined with Lora wireless communication technology, it is ensured that the IoT devices (such as various sensors, monitoring cameras, control terminals, etc.) on the deep well hoist of the coal mine can more effectively transmit the measurement signals, images and data to the edge computing node or server or control center. In this process, the intelligent metasurface adjusts its own reflection characteristics by timely obtaining channel state information, thereby optimizing the propagation and direction of the signal, thereby enhancing the signal coverage and the stability of the data link. This virtual link not only bypasses the original physical obstacles and signal blind spots, but also significantly improves the reliability and coverage of signal transmission. At the same time, the flexible configuration capability of RIS enables the communication link to be dynamically optimized according to actual needs, further improving the efficiency and security of data transmission.

[0026] In the present invention, a tension sensor is installed on the wire rope, a weighing sensor is installed at the connection node between the wire rope and the cage or bucket, and an acceleration sensor is installed on the cage or bucket, so as to facilitate the real-time collection of the tension signal of the wire rope, the weight signal of the cage or bucket, and the acceleration signal of the cage or bucket during the operation of the hoist, and then the operation data of the hoist can be collected more comprehensively. The tension sensor, the weighing sensor and the acceleration sensor all use the LoRa wireless communication technology to transmit the collected signals to the external device, and the transmission of long-distance signals can be realized under the premise of low power consumption. The edge computing node receives the wireless signal through the LoRa wireless communication technology, and the distance signal can be received under the premise of low power consumption, and then the communication distance can be extended under the premise of low power consumption through the LoRa wireless network. The intelligent super surface is arranged in the deep well of the coal mine, and the intelligent super surface is added to the LoRa wireless network system, which can significantly enhance the strength and coverage of the wireless signal, effectively improve the stability of the wireless sensor network, and then can further effectively expand the coverage of the wireless signal. Using edge computing nodes to complete edge computing tasks can effectively reduce the computing pressure of cloud servers. At the same time, it can also effectively reduce dependence on cloud servers through autonomous computing. Connecting cloud servers and edge computing nodes through wired transmission media can ensure the reliability of long-distance data transmission. Since the intelligent metasurface can optimize the propagation and direction of signals by autonomously adjusting its own reflection characteristics, it can ensure that wireless signals can bypass the original physical obstacles and signal blind spots, so that the communication link can be dynamically adjusted according to actual needs. This not only overcomes the limitations of traditional methods that are difficult to cope with dynamic changes, but also significantly improves the adaptive ability and intelligence level of the communication system, effectively ensuring the reliability and security of data transmission.

[0027] Through LoRa wireless communication technology, combined with the signal enhancement characteristics of the intelligent metasurface and its adjustable reflection characteristics, the system can ensure the stability of the wireless data link inside the hoist, thereby improving the data transmission efficiency and ensuring the reliability and security of data transmission. It is beneficial for edge computing nodes and cloud servers to integrate and process the data monitored by weighing sensors, acceleration sensors and tension sensors in real time, thereby realizing comprehensive and reliable monitoring of the hoist system status, effectively solving the challenges of severe attenuation of wireless sensor signals in deep well hoists in coal mine scenarios due to the metal environment and high requirements for data stability and latency.

[0028] like Figure 3 and Figure 4 As shown, the present invention also provides a coal mine deep shaft hoist communication method based on intelligent metasurface assistance, using a coal mine deep shaft hoist communication system based on intelligent metasurface assistance, including the following steps: Step 1: Arrange the intelligent metasurface 3 and build an intelligent communication system; The smart metasurface 3 is deployed on the wall of the mine. The direction and phase of the reflected beam are actively adjusted by utilizing the programmable reflection characteristics of the smart metasurface 3 to construct a virtual and enhanced wireless communication link in the complex mine environment, and the wireless communication link is used to establish a communication connection between the IoT device and the edge computing node 1. The wired transmission medium is used to establish a communication connection between the edge computing node 1 and the remote cloud server 2. Step 2: Model the energy loss and computational delay issues in the computational offloading process of the intelligent communication system; S21: In the LoRa communication system in coal mines assisted by the intelligent metasurface 3, the construction of the channel model is crucial. N The smart metasurface is composed of independently controllable reflective elements that can flexibly adjust the phase of the reflected signal to optimize the performance of the communication link. In the communication link between the Internet of Things (IoT) device and the wireless access point (AP), the signal will experience two reflection paths through the smart metasurface 3 and a possible direct path. Channel gain h Taking into account the direct path gain , reflection path gain And the RIS reflection coefficient matrix The influence of is used to comprehensively characterize the quality and characteristics of the communication link. Specifically, the channel model of the smart metasurface 3 is established according to formula (1), and the channel gain of the smart metasurface 3 is obtained. ; (1); In the formula, represents the direct path gain; represents the gain of the nth reflection path; Smart Super Surface 3 n The complex reflection coefficient of each reflection element (including amplitude and phase information) is calculated, and the amplitude is set to 1, and only the phase is adjusted; S22: The Internet of Things (IoT) device is responsible for transmitting computing task data to the wireless access point (AP) via a wireless channel. Given the unique physical conditions of the mine, direct communication links often face severe interference and coverage blind spots, which greatly challenges the reliability and efficiency of signal transmission. In order to solve this problem, the deployment of intelligent metasurface (RIS) 3 is particularly important. As a key means of signal enhancement, it has immeasurable value in expanding signal coverage and improving transmission quality. For the process of receiving signals, a model of intelligent metasurface 3 receiving signals can be established according to formula (2); (2); In the formula, Indicatesk The received signal of each IoT device; Indicates the transmit power; Indicates edge computing node 1 and k The channel model between IoT devices, is the channel estimation error; Indicates the transmission signal; represents Gaussian white noise, with variance ; S23: In the overall architecture of the coal mine deep shaft communication system, the transmission rate is an important criterion for measuring the transmission capacity of the system. The transmission rate model of the intelligent metasurface 3 can be established according to formula (3); (3); In the formula, Indicates k The transmission rate of each IoT device; Indicates bandwidth; S24: The energy consumption of sensor nodes is a key indicator to measure their sustainable operation capability. It not only affects the cost of a single task execution, but also directly affects the long-term stable operation and energy efficiency optimization of the entire system. Formula (4) can be used to characterize the energy consumption of IoT devices. ; (4); In the formula, T Indicates the duration of processing subframe; It represents the energy efficiency factor of IoT devices; Indicates k CPU cycle frequency of each IoT device; S25: Use formula (5) to characterize the energy consumption at edge computing node 1 ; (5); In the formula, Indicates the transmit power of IoT devices; Indicate to k The ratio of transmission duration of each user; represents the energy efficiency factor of edge computing node 1; Represents the computing power of the mobile edge computing server; S26: Assume that the delay of transmitting the result information after the edge computing task is negligible, because the calculated information is usually small. For cloud computing, all the calculated data are first transmitted to the MEC server and then forwarded to the remote cloud server 2 via optical fiber transmission. The average data computing capacity and energy consumption of cloud server 2 can be characterized by formula (6); (6); In the formula, Indicates the transmit power of edge computing node 1; Indicates that edge computing node 1 gives k The ratio of transmission duration of each user; represents the energy efficiency factor of cloud server 2; Indicates the CPU cycle frequency of cloud server 2; S27: In order to comprehensively evaluate the performance of the entire offloading strategy, the average energy consumption and average equivalent delay are defined as indicators to measure the performance of the intelligent communication system; the average energy consumption in the offloading process is characterized and calculated using formula (7): , using formula (8) to characterize and calculate the average equivalent delay in the offloading process ; (7); (8); In the formula, ; Indicates the number of offload tasks assigned to IoT devices, represents the number of offload tasks assigned to the edge computing node (1), represents the number of offload tasks assigned to cloud server (2), Indicates the total number of unloaded tasks; Indicates the number of computing cycles required to process a single bit of computing task; Indicates the computing offloading mode. When it is set to 1, the task migration mechanism for different computing resource levels is triggered, which activates the task offloading mode for the direct processing capability of IoT devices, the local optimization processing capability of edge computing node 1, and the powerful centralized processing capability of cloud server 2. Conversely, the corresponding mode is turned off. Step 3: For coal mine hoist fault detection based on multi-source data fusion, in view of the characteristics of data stability and high latency requirements, while designing the unloading benefit function of coal mine deep shaft hoist communication, the problem of improving the minimum energy efficiency limit and reducing the computing latency is fully considered. The energy loss and computing latency problems of the computing unloading process of the mine Internet of Things equipment in the network system assisted by the intelligent metasurface (RIS) 3 are modeled, and the optimization goal of maximizing the computing unloading benefit function is achieved by jointly designing the intelligent metasurface (RIS) 3 reflection phase shift and computing unloading rate; Specifically, a computing offloading strategy that takes into account both real-time performance and resource constraints is constructed. This strategy needs to ensure the immediate processing of computing tasks to maintain the safety and efficiency of mining operations. To this end, a comprehensive benefit function is constructed, which integrates the two dimensions of energy efficiency and computing latency. It aims to comprehensively evaluate and optimize offloading decisions. The comprehensive benefit function integrating the two dimensions of energy efficiency and computing latency can be constructed according to formula (9): ; (9); In the formula, represents the weight parameter of energy efficiency, A weight parameter representing the total computation delay, used to balance the effects of the utility function.

[0029] Step 4: Establish the maximization of the comprehensive benefit function as the optimization goal of the computing offloading strategy, that is, maximize the function through strategy adjustment, so as to take into account both the real-time requirements of computing tasks and the efficiency of resource utilization; Specifically, an optimization framework for the computational offloading strategy is constructed according to formulas (a), (b), (c), (d), (e) and (f). In the process of constructing the optimization framework for the computational offloading strategy, multidimensional constraints are carefully set to reflect the complexity of system operation and actual needs. Among them, constraint (a) is used to ensure that the transmission power of the IoT device and the edge computing node 1 strikes a balance between security and energy efficiency; constraint (b) is used to ensure that the IoT device, edge computing node 1 and cloud server 2 must operate at the maximum computing power limited by their respective CPU frequencies; each level measures the ability to process communication data through the CPU frequency to ensure the efficiency and stability of the system and achieve efficient flow and intelligent processing of data from the source to the cloud; constraint (c) is used to specify the fairness of power allocation in the intelligent communication system network; constraint (d) is used to consider the diversity of task modes; constraint (e) is used to define the value range of the phase shift of the reflection coefficient matrix of the intelligent metasurface 3; constraint (f) is used to ensure the minimum transmission rate requirement when channel estimation errors exist; this series of constraints together constitute a high-dimensional, continuous and non-convex optimization problem, and its solution requires the use of advanced optimization theories and algorithms to achieve the optimization of system performance.

[0030] ; In the formula, Indicates k Computation offloading mode for IoT devices; Step 5: Build a secure reinforcement learning model; Since it is difficult to accurately obtain real-time wireless communication models and channel state information underground in coal mines, and it is subject to the non-convex constraints of the RIS reflection phase shift coefficient and the minimum energy loss, the dynamic computation unloading process and the RIS phase shift adjustment process of the coal mine deep shaft hoist communication system assisted by the intelligent metasurface 3 are modeled as Markov decision processes, and the wireless sensor network assisted by the intelligent metasurface 3 and the intelligent metasurface 3 are used as the environment and the learning agent respectively; at the same time, in order to ensure that the computational delay threshold requirements are met and to ensure real-time data processing and accident emergency response, a Lagrangian opportunity-constrained reinforcement learning method is proposed; finally, a safety reinforcement learning model of the coal mine deep shaft hoist communication system assisted by the intelligent metasurface 3 is constructed; Step 6: Achieve optimal computation offloading benefits; S61: In order to avoid the difficulty of solving the problem caused by non-convex constraints, the Lagrange multiplier method is first used to transform the constrained Markov decision process problem into an unconstrained dual problem according to formula (10); then the Lagrange proportional integral is formulated through the integral separation method to prevent the policy from being overly conservative; (10); In the formula, Represents the reward of reinforcement learning, which is used to measure the performance of the algorithm; is the Lagrangian penalty factor, which is used to control the degree of relaxation of the constraints; Indicates the total transmission rate of the IoT devices used Less than rate threshold The probability of A threshold indicating the probability of outage; S62: In response to the complex high-dimensional continuous strategy optimization problem in the LoRa-MEC system assisted by the intelligent metasurface (RIS) 3 in coal mines, this application introduces a mining IoT computing offloading mechanism based on deep deterministic policy gradient (DDPG). This mechanism relies on the secure reinforcement learning model constructed in step five, and uses the deep deterministic policy gradient (DDPG) algorithm to solve the optimal task offloading strategy, aiming to achieve the unconstrained optimization goal. The specific implementation path covers the precise definition of state space, action space and reward function, approximation of strategy and value function through deep neural network, and the use of technical means such as experience replay and dual network structure to ensure the stability and convergence of the learning process, and finally reach an efficient computing task offloading decision; In order to dynamically explore the computational offloading strategy in the complex and changeable hoist communication scenario, it is necessary to design a reasonable RIS reflection phase shift and computational offloading rate to achieve the goal of optimal computational offloading benefit, which specifically includes the following steps: A1: Initialize based on the deep deterministic policy gradient algorithm; initialize the relevant learning parameters based on the deep deterministic policy gradient algorithm, including the learning rate , Discount Factor , memory pool size, OU noise , soft update learning rate , , batch size Z, number of rounds, and exploration step size within a round; initialize the Actor network parameters based on the deep deterministic policy gradient algorithm and Critic network parameters , by and Network parameters assigned to the TargetActor network and the Target Critic network , , complete the initialization of the Target network parameters, and randomly give the initial environmental state of the coal mine deep shaft hoist communication system assisted by intelligent metasurface 3 ;in, , and They represent the number of tasks assigned to different devices (IoT devices, edge computing nodes, and cloud servers); , and All represent the parameters of task offloading; A2: Single-step exploration and strategy sampling: Perform one-step exploration within a round, and the Actor network initializes the system state s deterministic strategy based on the current input ; Then, the mining IoT equipment is based on and the noise sampled by the OU process , according to formula (11), select and calculate the unloading strategy ; In the process of implementing the computational offloading strategy, the reflection phase shift is carefully configured. , Number of tasks , calculate the mode factor and resource allocation factors The system performance is optimized and the utility value of offloading is evaluated (which quantifies the effectiveness of the offloading strategy), that is, the reward U Then, based on the real-time observation of the computational offloading process, the system state is updated to provide a new starting point for the subsequent decision-making process. In this process, the dynamic changes of the computational workload and the location information of the IoT devices follow the Markov chain model to ensure the accurate re-evaluation of the environmental channel state information. Finally, the computational offloading experience containing the above key information is Integrate into cases and store in the experience pool for subsequent learning and optimization; (11); A3: Network parameter update; determine whether the number of computational offloading experiences in the experience pool has reached Z; if not, jump directly to step A4 and continue to explore the accumulated computational offloading experiences until the computational offloading experiences have accumulated to reach Z groups; if reached, based on the experience playback technology, the mining IoT device randomly samples Z groups of computational offloading experiences from the experience pool, i.e. , and use these computational offloading experiences to iteratively update the four neural network parameters, including the Actor network parameters , Critic network parameters , Target Actor network parameters and TargetCritic network parameters ;in, Indicates the current state; Indicates the action at the current moment; Represents the reward at the current moment; Indicates the state at the next moment; A31: Adopt Adam optimizer as gradient descent algorithm and use formula (12) and formula (13) to update the critic network parameters ; (12); (13); In the formula, represents the loss function; Represents the reduction coefficient in reinforcement learning; Represents the target Critic network; Represents an Actor network; Indicates the first n A loss function; represents the Critic network; Indicates the current state; Indicates the first n Actions; A32: Update the Actor network parameters according to formula (14) using the deterministic policy gradient ascent method ; (14); In the formula, Represents the Actor network deterministic policy gradient; represents mathematical expectation; Represents the policy gradient of the Critic network; Represents the policy gradient of the Critic network parameters; represents the policy function; A33: Using soft update learning rate , update the Target Actor network parameters according to formula (15) , update the Target Critic network parameters according to formula (16) , to slow down the tracking of Actor network parameters and Critic network parameters; (15); (16); A4: Round end judgment: judge whether the number of time slots reaches the maximum number of training steps in a round. If not, set the next state Initialize the system state for the new , return to A2; if reached, the round training ends.

[0031] In the present invention, the intelligent reflective surface is first arranged in the deep well of the coal mine, and the wireless communication link constructed by it is used to establish the communication connection between the Internet of Things device and the edge computing node. Based on the advantages of the signal enhancement and adjustable reflection characteristics of the intelligent metasurface, combined with the low-power long-distance transmission characteristics of the LoRa wireless communication technology, the coverage of the internal wireless communication network of the hoist can be further expanded. At the same time, the reliability and stability of the wireless communication link can be further improved, thereby ensuring that the monitoring data can be transmitted to the edge computing node in a timely and reliable manner. For the fault detection of coal mine hoists based on multi-source data fusion, for the problem of high data stability and delay requirements, while designing the communication unloading benefit function of the coal mine deep well hoist, the present invention also considers the problem of improving the minimum energy efficiency limit and reducing the calculation delay, and constructs a model of energy loss and calculation delay problems in the calculation unloading process of the mine Internet of Things device in the network system assisted by the intelligent metasurface, and then the optimization goal of maximizing the calculation unloading benefit function can be achieved by jointly designing the reflection phase shift and calculation unloading rate of the intelligent metasurface. Since it is difficult to accurately obtain real-time wireless communication models and channel state information underground in coal mines, and it will be subject to the non-convex constraints of the intelligent supersurface reflection phase shift coefficient and the minimum energy loss, in order to effectively solve the problem of difficulty in obtaining real-time information underground and non-convex constraints, the dynamic calculation unloading process of intelligent supersurface assisted coal mine deep well hoist communication and the intelligent supersurface phase shift adjustment process are modeled into Markov decision processes. At the same time, the present invention also introduces a Lagrangian opportunity constraint reinforcement learning scheme to ensure the real-time data processing and the timeliness of emergency response. As an emerging technology in the field of artificial intelligence, deep reinforcement learning has demonstrated excellent capabilities in handling optimization problems in complex dynamic environments. By constructing a model based on a finite state Markov decision process, the deep reinforcement learning algorithm can adaptively select the optimal calculation unloading strategy through continuous interactive learning with the environment without fully understanding the system model and specific channel state information, thereby ensuring efficient real-time data processing and rapid emergency response capabilities of mine Internet of Things equipment. In order to ensure that the requirements of the computing delay threshold are met and to ensure real-time data processing and accident emergency response, a Lagrangian opportunity-constrained reinforcement learning method is proposed, and finally a safe reinforcement learning model for coal mine deep shaft hoist communication assisted by intelligent metasurface is constructed. Then, the deep deterministic policy gradient (DDPG) algorithm is used to optimize the safety reinforcement model to achieve the optimal computing offloading benefit in a complex environment, ensuring the efficiency and safety of coal mine deep shaft hoist communication. By cleverly integrating the safety learning mechanism into the DRL framework, the safety of the exploration process can be ensured. The mechanism evaluates the risk level of the state-action pair by setting safety indicators and automatically avoids high-risk behaviors, effectively preventing the problem of excessive computing delay caused by random exploration and ensuring the stable operation of the mine Internet of Things system.Driven by the reinforcement learning algorithm, the elevator communication system can continuously monitor and analyze changes in the communication environment, autonomously adjust the phase shift parameters of the reflection unit, and respond to communication needs under different environmental conditions with the most optimized beam shape.

[0032] This method has a high degree of intelligence and a simple implementation process. It applies intelligent metasurface technology to the special environment of the hoist tunnel for the first time to address the problems of signal attenuation and communication instability caused by dense metal devices in coal mines. At the same time, through continuous interactive learning with the environment, the configuration strategy of the intelligent metasurface is continuously optimized using a deep reinforcement learning algorithm. It can accurately control the propagation path, amplitude and phase of electromagnetic waves, significantly improve the strength and transmission stability of wireless signals, effectively adapt to dynamic changes such as rapid movement of the hoist, and ensure the long-term stability and sustainable optimization capability of the internal communication quality of the hoist.

Claims

1. A coal mine deep shaft hoist communication system based on intelligent metasurface assistance, including an Internet of Things device, characterized in that: It also includes an edge computing node (1), a cloud server (2), and a smart metasurface (3); The Internet of Things device comprises a tension sensor (4), a weighing sensor (5), and an acceleration sensor (6); the tension sensor (4) has a wireless communication function, is mounted on the steel wire rope (7), and is used to collect the tension signal of the steel wire rope (7) in real time, and transmit the tension signal to an external device through the LoRa wireless communication technology; the weighing sensor (5) has a wireless communication function, is mounted at a connection point between the steel wire rope (7) and the cage or bucket (8), and is used to collect the weight signal of the cage or bucket (8) in real time, and transmit the weight signal to an external device through the LoRa wireless communication technology; the acceleration sensor (6) has a wireless communication function, is mounted on the cage or bucket (8), and is used to collect the acceleration signal of the cage or bucket (8) in real time, and transmit the acceleration signal to an external device through the LoRa wireless communication technology; The edge computing node (1) has a wireless communication function and is installed on the wall of the mine and located near the mine mouth. It is used to receive wireless signals through LoRa wireless communication technology, complete edge computing tasks based on the received wireless signals, and then transmit the obtained data to the cloud server (2) through a wired transmission medium; The cloud server (2) is arranged at a remote end and is connected to the edge computing node (1), and is used to receive data uploaded by the edge computing node (1) and to complete core computing tasks; The intelligent metasurface (3) is installed on the wall of a mine shaft, and is used to receive and enhance the strength of wireless signals emitted by an IoT device, and transmit the received wireless signals to an edge computing node (1) by means of reflection. At the same time, it is used to optimize the communication quality between the IoT device and the edge computing node (1) by adjusting its own reflection phase shift coefficient, thereby ensuring reliable communication between the IoT device and the edge computing node (1).

2. According to claim 1, a coal mine deep shaft hoist communication system based on intelligent metasurface assistance is characterized in that: It also includes a monitoring camera, which has a wireless communication function and is installed on the wall of the mine shaft to collect image data of the monitoring area in real time, and uses the intelligent metasurface (3) as a relay and enhancement medium for data transmission to ensure that the image data can be quickly and stably transmitted to the edge computing node (1).

3. A coal mine deep shaft hoist communication system based on intelligent metasurface assistance according to claim 1 or 2, characterized in that: The intelligent super surface (3) is located in the middle section of the coal mine deep shaft in the height direction.

4. According to claim 3, a coal mine deep shaft hoist communication system based on intelligent metasurface assistance is characterized in that: The edge computing node (1) is composed of a wireless access point and a mobile edge computing server.

5. The coal mine deep shaft hoist communication system based on intelligent metasurface assistance according to claim 4 is characterized in that: The wired transmission medium is an optical cable or an electrical cable.

6. A coal mine deep shaft hoist communication method based on intelligent metasurface assistance, using a coal mine deep shaft hoist communication system based on intelligent metasurface assistance as claimed in any one of claims 1 to 5, characterized in that: The following steps are involved: Step 1: Arrange the smart metasurface (3) and build a smart communication system; The intelligent metasurface (3) is deployed on the wall of a mine shaft; the direction and phase of the reflected beam are actively adjusted by utilizing the programmable reflection characteristics of the intelligent metasurface (3), thereby constructing a virtual and enhanced wireless communication link in a complex mine environment, and utilizing the wireless communication link to establish a communication connection between an IoT device and an edge computing node (1); and utilizing a wired transmission medium to establish a communication connection between the edge computing node (1) and a remote cloud server (2); Step 2: Model the energy loss and computational delay issues in the computational offloading process of the intelligent communication system; S21: According to formula (1), the channel model of the smart metasurface (3) is established to obtain the channel gain of the smart metasurface (3) ; （1）； In the formula, represents the direct path gain; Indicates n The gain of the reflection path; Represents the smart metasurface (3) n The complex reflection coefficient of each reflection element; S22: Establish a model of the smart metasurface (3) receiving signals according to formula (2); （2）； In the formula, Indicates k The received signal of each IoT device; Indicates the transmit power; Represents the edge computing node (1) and k Channel model between IoT devices; is the channel estimation error; Indicates the transmission signal; represents Gaussian white noise, with variance ; S23: Establish a model of the transmission rate of the smart metasurface (3) according to formula (3); （3）； In the formula, Indicates k The transmission rate of each IoT device; Indicates bandwidth; S24: Using formula (4) to characterize the energy consumption at the IoT device ; （4）； In the formula, T Indicates the duration of processing subframe; represents the energy efficiency factor of IoT devices; Indicates k CPU cycle frequency of each IoT device; S25: Use formula (5) to characterize the energy consumption at the edge computing node (1) ; （5）； In the formula, Indicates the transmit power of IoT devices; Indicate to k The proportion of transmission duration of IoT devices; represents the energy efficiency factor of the edge computing node (1); Indicates the CPU cycle frequency of the mobile edge computing server; S26: Using formula (6) to characterize the energy consumption at the cloud server (2) ; （6）； In the formula, represents the transmit power of the edge computing node (1); Indicates that the edge computing node (1) gives k The proportion of transmission duration of IoT devices; represents the energy efficiency factor of the cloud server (2); represents the CPU cycle frequency of the cloud server (2); S27: Define average energy consumption and average equivalent delay as indicators for measuring the performance of intelligent communication systems; use formula (7) to characterize the average energy consumption during the calculation of unloading , using formula (8) to characterize and calculate the average equivalent delay in the offloading process ; （7）； （8）； In the formula, ; Indicates the number of offload tasks assigned to IoT devices, represents the number of offload tasks assigned to the edge computing node (1), represents the number of offload tasks assigned to cloud server (2), Indicates the total number of unloaded tasks; Indicates the number of computing cycles required to process a single bit of computing task; Indicates the computing offloading mode. When it is set to 1, it triggers the task migration mechanism for different computing resource levels, activating the task offloading mode for the direct processing capability of IoT devices, the local optimization processing capability of edge computing nodes (1), and the powerful centralized processing capability of cloud servers (2). Conversely, the corresponding mode is turned off. Step 3: Based on the characteristics of coal mine hoist fault detection based on multi-source data fusion, a computational offloading strategy that takes into account both real-time performance and resource constraints is constructed; according to formula (9), a comprehensive benefit function integrating the two dimensions of energy efficiency and computational delay is constructed. ; （9）； In the formula, represents the weight parameter of energy efficiency, A weight parameter representing the total computation delay; Step 4: Establish the maximization of the comprehensive benefit function as the optimization goal of the computational offloading strategy, and construct the optimization framework of the computational offloading strategy according to formulas (a), (b), (c), (d), (e) and (f); through constraint (a), ensure that the transmission power of the IoT device and the edge computing node (1) strikes a balance between security and energy efficiency; through constraint (b), ensure that the IoT device, edge computing node (1) and cloud server (2) must operate at the maximum computing power limited by their respective CPU frequencies; each layer measures the ability to process communication data through the CPU frequency to ensure the efficiency and stability of the system and realize the efficient flow and intelligent processing of data from the source to the cloud; through constraint (c), specify the fairness of power allocation in the intelligent communication system network; through constraint (d), consider the diversity of task modes; through constraint (e), define the value range of the phase shift of the reflection coefficient matrix of the intelligent metasurface (3); through constraint (f), ensure the minimum transmission rate requirement when channel estimation errors exist; ； In the formula, Indicates k Computation offloading mode for IoT devices; Step 5: Build a secure reinforcement learning model; The dynamic computation unloading process and RIS phase shift adjustment process of the coal mine deep shaft hoist communication system assisted by the intelligent metasurface (3) are modeled as Markov decision processes, and the wireless sensor network assisted by the intelligent metasurface (3) and the intelligent metasurface (3) are used as the environment and learning agent respectively; at the same time, a Lagrangian opportunity constrained reinforcement learning method is proposed; finally, a safety reinforcement learning model of the coal mine deep shaft hoist communication system assisted by the intelligent metasurface (3) is constructed; Step 6: Achieve optimal computation offloading benefits; S61: First, the constrained Markov decision process problem is transformed into an unconstrained dual problem using the Lagrange multiplier method according to formula (10); then, the Lagrange proportional integral is formulated using the integral separation method to prevent the policy from being overly conservative; （10）； In the formula, Represents the reward of reinforcement learning, which is used to measure the performance of the algorithm; is the Lagrangian penalty factor, which is used to control the degree of relaxation of the constraints; Indicates the total transmission rate of the IoT devices used Less than rate threshold probability; A threshold indicating the probability of outage; S62: Introduce a mining IoT computing offloading mechanism based on deep deterministic policy gradient, and use a deep deterministic policy gradient algorithm to solve the optimal task offloading strategy; A1: Initialize based on the deep deterministic policy gradient algorithm; Initialize the relevant learning parameters based on the deep deterministic policy gradient algorithm; Initialize the Actor network parameters in the deep deterministic policy gradient algorithm and Critic network parameters , by and Network parameters assigned to the Target Actor network and the Target Critic network , , complete the initialization of the Target network parameters, and randomly give the initial environmental state of the coal mine deep shaft hoist communication system assisted by the intelligent metasurface (3) ;in, , and Respectively represent the number of tasks assigned to different devices; , and All represent the parameters of task offloading; A2: Single-step exploration and strategy sampling: Perform one-step exploration within a round, and the Actor network initializes the system state s deterministic strategy based on the current input ; Then, the mining IoT equipment is based on and the noise sampled by the OU process , according to formula (11), select and calculate the unloading strategy ; In the process of executing the computation offloading strategy, by configuring the reflection phase shift , Number of tasks , calculate the mode factor and resource allocation factors The system performance is optimized and the utility value of computing offloading is evaluated, i.e., the reward U Then, based on the real-time observation of the computational offloading process, the system state is updated to provide a new starting point for the subsequent decision-making process. In this process, the dynamic changes of the computational workload and the location information of the IoT devices follow the Markov chain model to ensure the accurate re-evaluation of the environmental channel state information. Finally, the computational offloading experience containing the above key information is Integrate into cases and store in the experience pool for subsequent learning and optimization; （11）； A3: Network parameter update; determine whether the number of computational offloading experiences in the experience pool has reached Z; if not, jump directly to step A4 and continue to explore the accumulated computational offloading experiences until the computational offloading experiences have accumulated to reach Z groups; if reached, based on the experience playback technology, the mining IoT device randomly samples Z groups of computational offloading experiences from the experience pool, i.e. , and use these computational offloading experiences to iteratively update the four neural network parameters, including the Actor network parameters , Critic network parameters , Target Actor network parameters and TargetCritic network parameters ;in, Indicates the current state; Indicates the action at the current moment; Represents the reward at the current moment; Indicates the state at the next moment; A31: Adopt Adam optimizer as gradient descent algorithm and use formula (12) and formula (13) to update the critic network parameters ; （12）； （13）； In the formula, represents the loss function; Represents the reduction coefficient in reinforcement learning; Represents the target Critic network; Represents an Actor network; Indicates the first n A loss function; represents the Critic network; Indicates the current state; Indicates the first n Actions; A32: Update the Actor network parameters according to formula (14) using the deterministic policy gradient ascent method ; （14）； In the formula, Represents the Actor network deterministic policy gradient; represents mathematical expectation; Represents the policy gradient of the Critic network; Represents the policy gradient of the Critic network parameters; represents the policy function; A33: Using soft update learning rate , update the Target Actor network parameters according to formula (15) , update the Target Critic network parameters according to formula (16) , to slow down the tracking of Actor network parameters and Critic network parameters; （15）； （16）； A4: Round end judgment: judge whether the number of time slots reaches the maximum number of training steps in a round. If not, set the next state Initialize the system state for the new , return to A2; if reached, the round training ends.

7. The method for coal mine deep shaft hoist communication based on intelligent metasurface assistance according to claim 6 is characterized in that: In step 6, A1, the relevant learning parameters include the learning rate , Discount Factor , memory pool size, OU noise , soft update learning rate , , batch size Z, number of rounds, and exploration step size within a round.

Images