A remote communication method for rock mechanics test system

Abstract

The present invention discloses a remote communication method for a rock mechanics test system, which belongs to the field of rock mechanics test equipment. The data processing flow of the present invention includes the steps of data preprocessing, feature representation, intelligent filtering, dynamic filtering processing, effect evaluation and optimization, and real-time feedback. The intelligent filtering algorithm selects an appropriate filtering strategy based on the output characteristics of the Transformer model encoder to ensure data quality. The present invention has the advantages of realizing remote communication and control, enhancing model generalization ability and robustness, and adaptive learning.

Description

Technical Field

[0001] The invention belongs to the field of rock mechanics test equipment, and in particular relates to a remote communication method for a rock mechanics test system. Background Art

[0002] Rock mechanics testing is an important part of geological engineering, mining, underground construction and geotechnical engineering. These tests usually involve measuring and analyzing the physical properties of rock samples to evaluate parameters such as strength, deformation and stability. At present, data acquisition and communication of rock mechanics test equipment usually rely on traditional methods, such as directly connecting sensors to monitoring equipment or using intermediate devices for data transmission. These methods have the following main disadvantages: (1) Independence between devices: Various rock mechanics test equipment is usually independent of each other, there is no unified terminal for operation, and they are installed in different areas. The data generated is also stored in the corresponding equipment, which makes statistical communication difficult. (2) Single computing function of the equipment: Rock mechanics test equipment usually does not have advanced computing capabilities and cannot perform complex data analysis or simulation. (3) Small data storage capacity: The data storage capacity of these devices is usually limited and cannot meet the long-term data recording needs. (4) Backward hardware equipment: Some rock mechanics test equipment may be old hardware and difficult to upgrade. (5) Slow computing time: Due to the slow computing speed, it takes a long time to process large amounts of data. (6) Data delay: Since data acquisition and transmission usually rely on limited communication channels, data delay occurs, and it may not be possible to monitor and respond to changes in real time. (7) Limited scalability: Traditional methods have scalability issues when increasing the number of sensors or modifying the experimental configuration, making it difficult to meet complex experimental requirements.

[0003] The above shortcomings will affect the accuracy and transmission rate of the data. The above problems can be solved by using remote communication technology and corresponding algorithms suitable for rock mechanics test devices. Remote communication can transfer data to more powerful computing resources for advanced data processing and analysis; it can transfer data to larger storage systems to ensure that no important test data is lost; it can improve data processing capabilities by transferring data to modern equipment while extending the service life of these old equipment; it can speed up data processing while allowing researchers to perform other tasks in parallel; it can operate multiple devices in parallel; and it can achieve faster and more convenient data transmission through the network. At present, there is no proprietary remote communication equipment and corresponding communication technology specifically for rock mechanics test devices. Summary of the invention

[0004] The present invention provides a remote communication method for a rock mechanics test system, which can achieve efficient and stable data transmission and processing. The algorithm based on the Transformer model can effectively process a variety of rock mechanics data and improve the accuracy and stability of the data through an intelligent filtering algorithm.

[0005] The technical solution adopted by the present invention is:

[0006] A remote communication method for a rock mechanics test system comprises the following steps:

[0007] Step 1: Create VLAN and port assignment

[0008] S101: Switch initialization

[0009] Before assigning VLANs and ports, first initialize the switch and gain access to the switch management interface;

[0010] S102: Create VLAN

[0011] a. Log in to the switch management interface and obtain authorized administrator credentials;

[0012] b. Configure VLANs. Create the required number of VLANs and assign a unique VLAN ID (usually an integer) to each VLAN to distinguish them;

[0013] c. Name each VLAN to identify its purpose or configuration;

[0014] S103: Ports are assigned to VLANs

[0015] Select the switch ports to be assigned to a specific VLAN, configure the PVID (Port VLAN ID) of each port to assign it to the selected VLAN, and ensure that all sensors and devices are connected to the corresponding VLAN;

[0016] Step 2: Network topology management

[0017] In rock mechanics test systems, the key step to ensure the effectiveness and reliability of data transmission and communication is to manage the network topology.

[0018] S201: Dynamic VLAN Management

[0019] In order to meet the needs and changes of different test configurations, the network topology should be dynamic and flexible.

[0020] a. VLAN creation and modification: Log in to the switch's management interface and navigate to the VLAN configuration section using the administrator credentials. Assign a unique VLAN ID to each VLAN and name it to reflect its purpose. When needed, you can dynamically create new VLANs or modify the configuration of existing VLANs.

[0021] b. Port reallocation: Ports can be reallocated to different VLANs at any time by changing the port's PVID according to the test plan requirements; flexibility enables the system to adapt to changes in different test configurations without physical rewiring.

[0022] c. Delete VLANs that are no longer needed: After the experiment is over, delete VLANs that are no longer needed to free up resources and simplify the network topology; this helps maintain network clarity and efficiency.

[0023] S202: Topology Monitoring and Analysis

[0024] a. Use network management tools: Configure network management tools to monitor VLAN performance and activity; these tools can provide real-time topology maps and performance indicators to help identify potential problems.

[0025] b. Automated topology discovery: Utilize automated topology discovery tools to automatically detect the addition of new devices or the movement of devices, and perform VLAN assignment and configuration accordingly;

[0026] S203: Troubleshooting and recovery

[0027] In the event of network failures or problems, there needs to be a rapid troubleshooting and recovery mechanism.

[0028] a. Fault detection: Use network management tools to detect and report any network faults or anomalies; including communication problems between VLANs.

[0029] b. Automatic recovery: Configure an automatic recovery policy to automatically reallocate ports or VLANs when a failure is detected to minimize downtime;

[0030] Through effective network topology management, the rock mechanics test system can achieve dynamic network configuration to ensure that the network requirements of different test configurations are met, and provide a fast troubleshooting and recovery mechanism to ensure the reliability and stability of the network.

[0031] Step 3: Data encryption and transmission speed optimization

[0032] In order to protect the confidentiality of data during transmission, data encryption measures are used. Data is encrypted when it is sent and received to prevent unauthorized access.

[0033] By utilizing the SSL / TLS protocol, data transmission is encrypted using the Secure Sockets Layer (SSL) or Transport Layer Security (TLS) protocol.

[0034] In order to perform encryption and decryption efficiently, a key management strategy is adopted. The key is the key to encrypting and decrypting data. The key must be kept properly and replaced regularly to enhance security. A secure key exchange protocol is used to ensure the secure distribution and update of the key.

[0035] In order to ensure efficient data transmission and management in the rock mechanics test system, a series of data flow optimization strategies are introduced to improve data transmission speed, reliability and performance.

[0036] S301: Data compression and decompression

[0037] Data compression technology is used to reduce the size of data packets, thereby reducing the bandwidth and time required for data transmission. At the receiving end, the data is decompressed to restore the integrity of the original data while maintaining data quality.

[0038] S302: Cache and Prefetch

[0039] Use efficient caching mechanisms to cache the most frequently accessed data to reduce frequent access to remote storage. Prefetch technology can preload data that may be needed to reduce latency when accessing data.

[0040] S303: Introducing a load balancing algorithm

[0041] Load balancing algorithms are introduced to evenly distribute data traffic to multiple network paths or servers to ensure smooth network operation and optimal performance of data transmission. Load balancing ensures that each component in the system is fully utilized, thereby improving overall efficiency.

[0042] S304: Traffic Prioritization and QoS

[0043] In order to ensure the priority transmission of key data flows, traffic priority and Quality of Service (QoS) policies are configured. This ensures that key data can still maintain high performance when the network is congested.

[0044] Through the above data flow optimization strategy, the rock mechanics test system achieves high efficiency and performance improvement of data transmission.

[0045] Step 4: Use the following rock mechanics data transmission protocol to transmit data

[0046] Rock mechanics tests typically generate large amounts of data, especially in the case of high-precision measurements and high-resolution sensors. This includes data on stress, strain, pressure, displacement, and other relevant parameters. The amount of data can increase rapidly, especially when conducting long tests or high-frequency sampling, which can generate a very large amount of data, which is still large after data compression. The development of a rock mechanics data transmission protocol that adapts to data speed differences in the data acquisition unit is based on a model where a PID controller is used to dynamically adjust the transmission parameters to achieve dynamic adjustment of the transmission parameters to increase the transmission rate.

[0047] S401: Data flow rate monitoring

[0048] During implementation, the changes in data flow rate are detected in real time through the built-in monitoring module of the rock mechanics equipment to monitor the data flow rate;

[0049] S402: Dynamic parameter adjustment triggering and execution

[0050] When the data flow rate exceeds or falls below a preset threshold, the PID algorithm is used to trigger dynamic parameter adjustment of the adaptive transmission protocol so that the data flow rate is within the preset threshold range; the process of dynamic parameter adjustment is as follows: the data flow rate is divided into three levels: low, medium and high, and a transmission parameter table is created for each rate level, the transmission parameter table including data cache size, transmission frequency and compression ratio; the data flow rate is monitored in real time and matched with the preset rate level, and the transmission parameter table suitable for the current rate level is selected according to the matching result, thereby optimizing data transmission;

[0051] S403: Dynamic adjustment confirmation and application

[0052] A feedback mechanism is established to evaluate the real-time transmission performance based on events. According to the feedback results, the parameter table is adjusted in real time during the transmission process to adapt to the dynamic changes in the data flow rate.

[0053] Assume the current data flow rate measured is V C , the preset target rate is V T , then the rate error is:

[0054] e=V T -V C ;PID controller calculation:

[0055] Where Kp, Ki, Kd are the proportional, integral and differential coefficients respectively; e(t) is the change of error over time; u(t) is the controller output, indicating the parameter value that needs to be adjusted;

[0056] Adjustment of transmission frequency: new frequency = original frequency + u(t);

[0057] Adjustment of data compression ratio: new compression ratio = original compression ratio + u(t);

[0058] Cache size adjustment: new cache size = original cache size + u(t);

[0059] This adaptive parameter adjustment system can dynamically select and adjust transmission parameters according to the real-time monitored data flow rate, thereby improving data transmission efficiency and ensuring efficient data transmission under different rate conditions.

[0060] S404: Effect monitoring and data transmission efficiency evaluation

[0061] Compare the transmission rates before and after the adjustment, evaluate whether data loss or damage occurs during the data transmission process, whether the data transmission delay is reduced, and check the stability and reliability of the transmission system to ensure the improvement or optimization of the data transmission rate;

[0062] S405: Cycle Optimization and Continuous Adjustment

[0063] Based on monitoring data and effect evaluation, the system will continuously monitor changes in data flow rate and optimize transmission parameters to continuously improve data transmission efficiency.

[0064] Furthermore, in the data acquisition unit, a compression and filtering algorithm for various rock mechanics test data is constructed, including the following steps:

[0065] First, the rock mechanics test data (including time series data of stress, strain, pressure, displacement and other indicators) are normalized to ensure that the data are within a similar scale range;

[0066] Secondly, according to the characteristics of rock mechanics test data, a RS-LSTM (Rock Statistics-Long Short-Term Memory) model suitable for time series is constructed to adapt to the complex characteristics of the data;

[0067] The rock mechanics test data is serialized into time step data for training the RS-LSTM model. The data set is divided into a training set and a validation set, and the RS-LSTM model is used for training. The trained RS-LSTM model obtains the hidden state or output of the middle layer as a feature representation. These features are used for further analysis and application.

[0068] Finally, according to the set feature threshold, the data is intelligently filtered to remove noise interference and retain higher quality data.

[0069] Furthermore, the compression and filtering algorithm for rock mechanics test data includes the following steps:

[0070] S01, data preprocessing: preprocess the original rock mechanics test data, including outlier processing, missing value processing, and data normalization to ensure data quality and consistency;

[0071] S02, applying the Transformer model for data representation and feature extraction: using the encoder of the Transformer model, the preprocessed rock mechanics test data is converted into high-quality feature representation; at the same time, the Transformer model captures the time series characteristics and complex correlations in the rock mechanics test data through the self-attention mechanism and multi-head attention mechanism;

[0072] S03, intelligent filtering: Based on the output characteristics of the Transformer model encoder, intelligently select an appropriate filtering strategy, and select the filtering strategy based on the characteristics of the real-time data and the preset noise level to ensure that the filtering process effectively removes noise and retains key signals; Apply the selected intelligent filtering strategy to dynamically filter the feature representation output by the Transformer model encoder, including smoothing of time domain data and spectral analysis of frequency domain data, to optimize the quality and accuracy of the data;

[0073] S04, effect evaluation and optimization: compare the data quality and feature retention before and after filtering, evaluate the filtering effect, and optimize the algorithm according to the evaluation results to further improve the efficiency and accuracy of data processing;

[0074] S05, real-time application and feedback: Through the feedback information of real-time data stream, the parameters and intelligent filtering strategies of the Transformer model are dynamically adjusted to ensure that the algorithm remains efficient and stable in a changing environment.

[0075] As a deep learning model, the Transformer model has made remarkable achievements in the fields of natural language processing and time series modeling. Its core mechanisms include self-attention mechanism and position encoding, which can process input sequences in parallel and effectively capture long-distance dependencies while maintaining relevance. In the rock mechanics test system, the application of the Transformer model can improve the accuracy and efficiency of data processing, especially in the feature extraction and filtering processing of complex time series data, and has significant advantages.

[0076] Furthermore, the various rock mechanics test data include compressive strength, tensile strength, shear strength, rock density, deformation characteristics, rock structure and fracture characteristics, and acoustic wave velocity.

[0077] Beneficial effects of the present invention:

[0078] (1) Improved real-time performance and response speed: Traditional rock mechanics test systems usually rely on delayed data acquisition and transmission methods, which may result in the real-time performance of test data being insufficient to meet the needs of monitoring and control. The switch introduced in the present invention can realize real-time data transmission, ensuring that the test data is almost immediately available for monitoring and decision-making. This improvement in real-time performance is crucial for responding to changes or abnormal conditions in rock samples, and helps to improve the accuracy and operability of the test. (2) Through the connection of the optical fiber switch, real-time data transmission and remote wake-up control between the test equipment and the terminal server are realized, improving the remote operation capability and convenience of the system. (3) The Transformer model is introduced for data processing and feature extraction. By using the self-attention mechanism and multi-head attention mechanism, it can effectively capture complex time series features and improve the efficiency and accuracy of data processing. (4) Edge computing nodes are deployed in the system for localized data processing and analysis, which reduces data transmission delays and optimizes the response speed and reliability of the system. (5) Expanded application scope: Traditional methods may have limitations when facing complex and special test configurations. The technical solution of the present invention expands the application scope of rock mechanics test systems by providing more efficient, flexible and real-time data acquisition and communication. It can adapt to various test requirements, including different types of rock samples and test conditions, thereby improving the versatility and adaptability of the system. The system adopts a modular design, which is easy to maintain and expand, and can be flexibly configured and adjusted according to different experimental requirements. (6) Reduced energy consumption: Energy consumption and equipment loss are reduced, and the terminal server can be set to sleep mode and does not need to run continuously. However, even in sleep mode, through WOL technology (remote wake-up LAN), the test equipment can still remotely wake up the terminal server for data transmission and further processing. This method effectively reduces energy waste because the terminal server is only woken up when needed. BRIEF DESCRIPTION OF THE DRAWINGS

[0079] Figure 1 It is a schematic diagram of the system structure. DETAILED DESCRIPTION

[0080] like Figure 1 As shown, in this embodiment, the components of the present invention are configured as follows:

[0081] 1. Test equipment:

[0082] Quantity: At least 5 units (can be increased as needed).

[0083] Data port: Each device has one data port, which is an LC fiber optic interface.

[0084] Function: Collect rock mechanics data with a data sampling rate of 1000 times / second.

[0085] 2. Fiber Optic Switch:

[0086] Model: XYZ series fiber optic switch.

[0087] Data communication port: at least 10 LC fiber optic interfaces.

[0088] Wake-up control port: 2 RJ45 interfaces, used to connect the wake-up module and the terminal server.

[0089] Bandwidth: Each port supports 10Gbps.

[0090] 3. Terminal Server:

[0091] Quantity: 2 units (can be increased as needed).

[0092] Data port: Each server has two LC fiber ports, one for data communication and one for backup.

[0093] Wake-up port: 1 RJ45 port.

[0094] Processing capacity: Each server has a 16-core CPU and 64GB of memory, supporting real-time data processing and storage.

[0095] 4. Wake-up module:

[0096] Interface: 2 RJ45 interfaces, one connected to the fiber optic switch and one connected to the terminal server.

[0097] Function: Send a wake-up signal to activate the terminal server.

[0098] 5. Edge Computing Nodes

[0099] Model: GigabitEthernet port embedded computing board

[0100] It is suitable for small edge computing scenarios, especially when space is limited or power consumption requirements are low, and can effectively support edge computing nodes deployed near test equipment.

[0101] Edge computing nodes 1 and 2 are located between the test equipment and the terminal server, respectively, and are responsible for localized data processing and analysis. The specific equipment of edge computing nodes includes high-performance computing servers, data storage devices, and intelligent analysis modules.

[0102] In this embodiment, the test equipment collects rock mechanics data and transmits it to the optical fiber switch through the optical fiber cable, and the optical fiber switch transmits the data to the terminal server. The terminal server is configured to communicate with the test equipment remotely and realize real-time data transmission and reception. The wake-up module wakes up the terminal server when needed to process the data.

[0103] The test equipment collects rock mechanics data in real time, and the data is transmitted to the fiber optic switch via a fiber optic cable. The acquisition frequency is 1000 times / second, and the size of each data transmission packet is 1KB. The fiber optic switch transmits the received test equipment data to the terminal server via a fiber optic cable. The amount of data that can be processed per second is 10Gbps, ensuring the stability of data transmission when multiple devices work simultaneously. The data is pre-processed by the edge computing node device to reduce the transmission size of the data. The processed data is stored in the server database according to demand, and automatic backup is supported. The wake-up module sends a wake-up signal to the terminal server through the wake-up control port of the fiber optic switch. The wake-up signal transmission time is 0.1 seconds to ensure a quick response.

[0104] A remote communication method for a rock mechanics test system comprises the following steps:

[0105] Step 1: Create VLAN and port assignment

[0106] S101: Switch initialization

[0107] Before assigning VLANs and ports, first initialize the switch and gain access to the switch management interface;

[0108] S102: Create VLAN

[0109] a. Log in to the switch management interface and obtain authorized administrator credentials;

[0110] b. Configure VLANs. Create the required number of VLANs and assign a unique VLAN ID (usually an integer) to each VLAN to distinguish them;

[0111] c. Name each VLAN to identify its purpose or configuration;

[0112] S103: Ports are assigned to VLANs

[0113] Select the switch ports to be assigned to a specific VLAN, configure the PVID (PortVLANID) of each port to assign it to the selected VLAN, and ensure that all sensors and devices are connected to the corresponding VLAN;

[0114] Step 2: Network topology management

[0115] In rock mechanics test systems, the key step to ensure the effectiveness and reliability of data transmission and communication is to manage the network topology.

[0116] S201: Dynamic VLAN Management

[0117] In order to meet the needs and changes of different test configurations, the network topology should be dynamic and flexible.

[0118] a. VLAN creation and modification: Log in to the switch's management interface and navigate to the VLAN configuration section using the administrator credentials. Assign a unique VLAN ID to each VLAN and name it to reflect its purpose. When needed, you can dynamically create new VLANs or modify the configuration of existing VLANs.

[0119] b. Port reallocation: Ports can be reallocated to different VLANs at any time by changing the port's PVID according to the test plan requirements; flexibility enables the system to adapt to changes in different test configurations without physical rewiring.

[0120] c. Delete VLANs that are no longer needed: After the experiment is over, delete VLANs that are no longer needed to free up resources and simplify the network topology; this helps maintain network clarity and efficiency.

[0121] S202: Topology Monitoring and Analysis

[0122] a. Use network management tools: Configure network management tools to monitor VLAN performance and activity; these tools can provide real-time topology maps and performance indicators to help identify potential problems.

[0123] b. Automated topology discovery: Utilize automated topology discovery tools to automatically detect the addition of new devices or the movement of devices, and perform VLAN assignment and configuration accordingly;

[0124] S203: Troubleshooting and recovery

[0125] In the event of network failures or problems, there needs to be a rapid troubleshooting and recovery mechanism.

[0126] a. Fault detection: Use network management tools to detect and report any network faults or anomalies; including communication problems between VLANs.

[0127] b. Automatic recovery: Configure an automatic recovery policy to automatically reallocate ports or VLANs when a failure is detected to minimize downtime;

[0128] Through effective network topology management, the rock mechanics test system can achieve dynamic network configuration to ensure that the network requirements of different test configurations are met, and provide a fast troubleshooting and recovery mechanism to ensure the reliability and stability of the network.

[0129] Step 3: Data encryption and transmission speed optimization

[0130] In order to protect the confidentiality of data during transmission, data encryption measures are used. Data is encrypted when it is sent and received to prevent unauthorized access.

[0131] By utilizing the SSL / TLS protocol, data transmission is encrypted using the Secure Sockets Layer (SSL) or Transport Layer Security (TLS) protocol.

[0132] In order to perform encryption and decryption efficiently, a key management strategy is adopted. The key is the key to encrypting and decrypting data. The key must be kept properly and replaced regularly to enhance security. A secure key exchange protocol is used to ensure the secure distribution and update of the key.

[0133] In order to ensure efficient data transmission and management in the rock mechanics test system, a series of data flow optimization strategies are introduced to improve data transmission speed, reliability and performance.

[0134] S301: Data compression and decompression

[0135] Data compression technology is used to reduce the size of data packets, thereby reducing the bandwidth and time required for data transmission. At the receiving end, the data is decompressed to restore the integrity of the original data while maintaining data quality.

[0136] S302: Cache and Prefetch

[0137] Use efficient caching mechanisms to cache the most frequently accessed data to reduce frequent access to remote storage. Prefetch technology can preload data that may be needed to reduce latency when accessing data.

[0138] S303: Introducing a load balancing algorithm

[0139] Load balancing algorithms are introduced to evenly distribute data traffic to multiple network paths or servers to ensure smooth network operation and optimal performance of data transmission. Load balancing ensures that each component in the system is fully utilized, thereby improving overall efficiency.

[0140] S304: Traffic Prioritization and QoS

[0141] In order to ensure the priority transmission of key data flows, traffic priority and Quality of Service (QoS) policies are configured. This ensures that key data can still maintain high performance when the network is congested.

[0142] Through the above data flow optimization strategy, the rock mechanics test system achieves high efficiency and performance improvement of data transmission.

[0143] Step 4: Use the following rock mechanics data transmission protocol to transmit data

[0144] Rock mechanics tests typically generate large amounts of data, especially in the case of high-precision measurements and high-resolution sensors. This includes data on stress, strain, pressure, displacement, and other relevant parameters. The amount of data can increase rapidly, especially when conducting long tests or high-frequency sampling, which can generate a very large amount of data, which is still large after data compression. The development of a rock mechanics data transmission protocol that adapts to data speed differences in the data acquisition unit is based on a model where a PID controller is used to dynamically adjust the transmission parameters to achieve dynamic adjustment of the transmission parameters to increase the transmission rate.

[0145] S401: Data flow rate monitoring

[0146] During implementation, the changes in data flow rate are detected in real time through the built-in monitoring module of the rock mechanics equipment to monitor the data flow rate;

[0147] S402: Dynamic parameter adjustment triggering and execution

[0148] When the data flow rate exceeds or falls below a preset threshold, the PID algorithm is used to trigger dynamic parameter adjustment of the adaptive transmission protocol so that the data flow rate is within the preset threshold range; the process of dynamic parameter adjustment is as follows: the data flow rate is divided into three levels: low, medium and high, and a transmission parameter table is created for each rate level, the transmission parameter table including data cache size, transmission frequency and compression ratio; the data flow rate is monitored in real time and matched with the preset rate level, and the transmission parameter table suitable for the current rate level is selected according to the matching result, thereby optimizing data transmission;

[0149] S403: Dynamic adjustment confirmation and application

[0150] A feedback mechanism is established to evaluate the real-time transmission performance based on events. According to the feedback results, the parameter table is adjusted in real time during the transmission process to adapt to the dynamic changes in the data flow rate.

[0151] Assume the current data flow rate measured is V C , the preset target rate is V T , then the rate error is:

[0152] e=V T -V C ;PID controller calculation:

[0153] Where Kp, Ki, Kd are the proportional, integral and differential coefficients respectively; e(t) is the change of error over time; u(t) is the controller output, indicating the parameter value that needs to be adjusted;

[0154] Adjustment of transmission frequency: new frequency = original frequency + u(t);

[0155] Adjustment of data compression ratio: new compression ratio = original compression ratio + u(t);

[0156] Cache size adjustment: new cache size = original cache size + u(t);

[0157] This adaptive parameter adjustment system can dynamically select and adjust transmission parameters according to the real-time monitored data flow rate, thereby improving data transmission efficiency and ensuring efficient data transmission under different rate conditions.

[0158] S404: Effect monitoring and data transmission efficiency evaluation

[0159] Compare the transmission rates before and after the adjustment, evaluate whether data loss or damage occurs during the data transmission process, whether the data transmission delay is reduced, and check the stability and reliability of the transmission system to ensure the improvement or optimization of the data transmission rate;

[0160] S405: Cycle Optimization and Continuous Adjustment

[0161] Based on monitoring data and effect evaluation, the system will continuously monitor changes in data flow rate and optimize transmission parameters to continuously improve data transmission efficiency.

[0162] Among them, in the data acquisition unit, a compression and filtering algorithm for various rock mechanics test data is constructed, including the following steps:

[0163] First, the rock mechanics test data (including compressive strength, tensile strength, shear strength, rock density, deformation characteristics, rock structure and fracture characteristics, acoustic wave velocity, that is, time series data of stress, strain, pressure, displacement and other indicators) are normalized to ensure that the data are within a similar scale range;

[0164] Secondly, according to the characteristics of rock mechanics test data, a RS-LSTM (Rock Statistics-Long Short-Term Memory) model suitable for time series is constructed to adapt to the complex characteristics of the data;

[0165] The rock mechanics test data is serialized into time step data for training the RS-LSTM model. The data set is divided into a training set and a validation set, and the RS-LSTM model is used for training. The trained RS-LSTM model obtains the hidden state or output of the middle layer as a feature representation. These features are used for further analysis and application.

[0166] Finally, according to the set feature threshold, the data is intelligently filtered to remove noise interference and retain higher quality data.

[0167] The compression and filtering algorithm of rock mechanics test data includes the following steps:

[0168] S01, data preprocessing: preprocess the original rock mechanics test data, including outlier processing (such as removing data beyond a reasonable range), missing value processing (such as using interpolation to fill missing data), and data normalization processing (such as scaling the data to the [0,1] interval) to ensure data quality and consistency;

[0169] S02, applying the Transformer model for data representation and feature extraction: using the encoder of the Transformer model, the preprocessed rock mechanics test data is converted into high-quality feature representation; at the same time, the Transformer model captures the time series characteristics and complex correlations in the rock mechanics test data through the self-attention mechanism and multi-head attention mechanism;

[0170] S03, intelligent filtering: Based on the output characteristics of the Transformer model encoder, intelligently select an appropriate filtering strategy, and select the filtering strategy based on the characteristics of the real-time data and the preset noise level to ensure that the filtering process effectively removes noise and retains key signals; Apply the selected intelligent filtering strategy to dynamically filter the feature representation output by the Transformer model encoder, including smoothing of time domain data and spectral analysis of frequency domain data, to optimize the quality and accuracy of the data;

[0171] S04, effect evaluation and optimization: compare the data quality and feature retention before and after filtering, evaluate the filtering effect, and optimize the algorithm according to the evaluation results to further improve the efficiency and accuracy of data processing;

[0172] S05, real-time application and feedback: Through the feedback information of real-time data stream, the parameters and intelligent filtering strategies of the Transformer model are dynamically adjusted to ensure that the algorithm remains efficient and stable in a changing environment.

[0173] As a deep learning model, the Transformer model has made remarkable achievements in the fields of natural language processing and time series modeling. Its core mechanisms include self-attention mechanism and position encoding, which can process input sequences in parallel and effectively capture long-distance dependencies while maintaining relevance. In the rock mechanics test system, the application of the Transformer model can improve the accuracy and efficiency of data processing, especially in the feature extraction and filtering processing of complex time series data, and has significant advantages.

[0174] The present invention improves the model module mainly in the following aspects:

[0175] (1) The Transformer model is introduced as a core component to replace the traditional model module. Through the self-attention mechanism and position encoding, it can handle long-distance dependencies and complex time series data features.

[0176] (2) A multi-task learning framework is used to simultaneously process multiple key tasks in rock mechanics test data, such as compressive strength prediction and rock density estimation. By sharing model parameters and feature representations, the generalization ability of the model and data utilization are optimized.

[0177] (3) Introducing data augmentation technology to expand the diversity and coverage of the data set and improve the robustness and generalization ability of the model by synthesizing data samples or transforming existing data.

[0178] (4) Design and implement adaptive learning algorithms to dynamically adjust model parameters and learning strategies based on real-time data streams and environmental changes, optimize the performance of the model under different experimental conditions, and ensure the stability and continuous learning ability of the model.

Claims

1. A remote communication method for a rock mechanics test system, characterized in that: The following steps are involved: Step 1: Create VLAN and port assignment S101: Switch initialization Before assigning VLANs and ports, first initialize the switch and gain access to the switch management interface; S102: Create VLAN a. Log in to the switch management interface and obtain authorized administrator credentials; b. Configure VLANs: Create the required number of VLANs and assign a unique VLAN ID to each VLAN to distinguish them; c. Name each VLAN to identify its purpose or configuration; S103: Ports are assigned to VLANs Select the switch ports to be assigned to a specific VLAN, configure the PVID of each port to assign it to the selected VLAN, and ensure that all sensors and devices are connected to the corresponding VLAN; Step 2: Network topology management S201: Dynamic VLAN Management a. VLAN creation and modification: Log in to the switch's management interface and navigate to the VLAN configuration section using the administrator credentials. Assign each VLAN a unique VLAN ID and name it to reflect its purpose. b. Port reallocation: According to the test plan requirements, the port can be reallocated to a different VLAN at any time by changing the port's PVID; c. Delete VLANs that are no longer needed: After the experiment is over, delete VLANs that are no longer needed to free up resources and simplify the network topology; S202: Topology Monitoring and Analysis a. Use network management tools: Configure network management tools to monitor VLAN performance and activity; b. Automated topology discovery: Utilize automated topology discovery tools to automatically detect the addition of new devices or the movement of devices, and perform VLAN assignment and configuration accordingly; S203: Troubleshooting and recovery a. Fault detection: Use network management tools to detect and report any network faults or anomalies; b. Automatic recovery: Configure an automatic recovery policy to automatically reallocate ports or VLANs when a failure is detected to minimize downtime; Step 3: Data encryption and transmission speed optimization Encrypted data transmission using Secure Sockets Layer or Transport Layer Security protocols via SSL / TLS protocols; S302: Cache and Prefetch Use caching mechanisms to cache the most frequently accessed data to reduce frequent access to remote storage; S303: Introducing a load balancing algorithm Introducing load balancing algorithms to evenly distribute data traffic to multiple network paths or servers, ensuring smooth network operation and optimal performance of data transmission; S304: Traffic Prioritization and QoS Configure traffic prioritization and quality of service to ensure that critical data flows are delivered first; Step 4: Transmit data using the rock mechanics data transmission protocol S401: Data flow rate monitoring During implementation, the changes in data flow rate are detected in real time through the built-in monitoring module of the rock mechanics equipment to monitor the data flow rate; S402: Dynamic parameter adjustment triggering and execution When the data flow rate exceeds or falls below a preset threshold, the PID algorithm is used to trigger dynamic parameter adjustment of the adaptive transmission protocol so that the data flow rate is within the preset threshold range; the process of dynamic parameter adjustment is as follows: the data flow rate is divided into three levels: low, medium and high, and a transmission parameter table is created for each rate level, the transmission parameter table including data cache size, transmission frequency and compression ratio; the data flow rate is monitored in real time and matched with the preset rate level, and the transmission parameter table suitable for the current rate level is selected according to the matching result, thereby optimizing data transmission; S403: Dynamic adjustment confirmation and application Assume the current data flow rate measured is V C , the preset target rate is V T , then the rate error is: e=V T -V C ;PID controller calculation: Where Kp, Ki, Kd are the proportional, integral and differential coefficients respectively; e(t) is the change of error over time; u(t) is the controller output, indicating the parameter value that needs to be adjusted; Adjustment of transmission frequency: new frequency = original frequency + u(t); Adjustment of data compression ratio: new compression ratio = original compression ratio + u(t); Cache size adjustment: new cache size = original cache size + u(t); S404: Effect monitoring and data transmission efficiency evaluation Compare the transmission rates before and after the adjustment, evaluate whether data loss or damage occurs during the data transmission process, whether the data transmission delay is reduced, and check the stability and reliability of the transmission system to ensure the improvement or optimization of the data transmission rate; S405: Cycle Optimization and Continuous Adjustment Based on monitoring data and effect evaluation, the system will continuously monitor changes in data flow rate and optimize transmission parameters to continuously improve data transmission efficiency.

2. A remote communication method for a rock mechanics test system according to claim 1, characterized in that: In the data acquisition unit, a compression and filtering algorithm for various rock mechanics test data is constructed, including the following steps: First, the rock mechanics test data are normalized to ensure that the data are within a similar scale range; Secondly, according to the characteristics of rock mechanics test data, an RS-LSTM model suitable for time series is constructed to adapt to the complex characteristics of the data; Finally, according to the set feature threshold, the data is intelligently filtered to remove noise interference and retain higher quality data.

3. A remote communication method for a rock mechanics test system according to claim 2, characterized in that: The compression and filtering algorithm of rock mechanics test data includes the following steps: S01, data preprocessing: preprocess the original rock mechanics test data, including outlier processing, missing value processing, and data normalization to ensure data quality and consistency; S02, applying the Transformer model for data representation and feature extraction: using the encoder of the Transformer model, the preprocessed rock mechanics test data is converted into high-quality feature representation; at the same time, the Transformer model captures the time series characteristics and complex correlations in the rock mechanics test data through the self-attention mechanism and multi-head attention mechanism; S03, intelligent filtering: Based on the output characteristics of the Transformer model encoder, intelligently select an appropriate filtering strategy, and select the filtering strategy based on the characteristics of the real-time data and the preset noise level to ensure that the filtering process effectively removes noise and retains key signals; Apply the selected intelligent filtering strategy to dynamically filter the feature representation output by the Transformer model encoder, including smoothing of time domain data and spectral analysis of frequency domain data, to optimize the quality and accuracy of the data; S04, effect evaluation and optimization: compare the data quality and feature retention before and after filtering, evaluate the filtering effect, and optimize the algorithm according to the evaluation results to further improve the efficiency and accuracy of data processing; S05, real-time application and feedback: Through the feedback information of real-time data stream, the parameters and intelligent filtering strategies of the Transformer model are dynamically adjusted to ensure that the algorithm remains efficient and stable in a changing environment.

4. A remote communication method for a rock mechanics test system according to claim 3, characterized in that: The various rock mechanics test data include compressive strength, tensile strength, shear strength, rock density, deformation characteristics, rock structure and fracture characteristics, and acoustic wave velocity.

Images