Gear processing equipment remote fault diagnosis system based on wireless communication

Abstract

The invention discloses a gear processing equipment remote fault diagnosis system based on wireless communication, which relates to the field of gear processing and comprises a data acquisition module, a data transmission module, a fault diagnosis module, an early warning and feedback module, an equipment state evaluation module, a big data analysis module and a remote control module. The acquisition module acquires data at key parts by using an intelligent sensor, the transmission module adaptively retransmits the data according to the link quality by means of a wireless communication technology, the diagnosis module judges the fault type and degree in combination with transfer learning, the early warning and feedback module performs multi-form early warning, and the evaluation module evaluates the overall state of the equipment by using a formula. The analysis module carries out modeling storage and analyzes operation and fault data, and the control module operates and controls equipment through a remote server. According to the method, faults are efficiently diagnosed, problems are quickly positioned, the diagnosis time is shortened, the diagnosis precision is high, complex and early faults are recognized, equipment maintenance is assisted, hidden danger is predicted in advance, remote operation and model updating are supported, and stable operation of equipment is comprehensively guaranteed.

Description

Technical Field

[0001] The invention relates to the technical field of gear processing, and in particular to a remote fault diagnosis system for gear processing equipment based on wireless communication. Background Art

[0002] In modern manufacturing, gears are key components of mechanical transmission, and their processing accuracy and quality directly affect the performance and reliability of various mechanical equipment. The stable operation of gear processing equipment is crucial, but in the actual production process, equipment failures occur frequently. Traditional gear processing equipment mostly relies on regular inspections by on-site technicians to discover hidden faults, which is inefficient and difficult to detect subtle abnormalities inside the equipment in real time.

[0003] With the improvement of industrial automation, the structure of gear processing equipment has become more complex, and the types of faults have become increasingly diverse. When equipment fails, due to the lack of effective remote monitoring methods, technicians often need to go to the site in person to troubleshoot the fault through manual inspection and experience judgment, which not only takes a lot of time, but may also lead to inaccurate diagnosis due to differences in the professional level of technicians. Especially in large factories or remote areas, the long journey time for technicians to rush to the site greatly increases the downtime of equipment, causing serious production delays and economic losses.

[0004] Although there are some simple equipment status monitoring systems on the market, most of them are local monitoring and cannot achieve remote diagnosis. Even if some systems have remote communication functions, their fault diagnosis algorithms are relatively simple and can only detect a few common faults. It is difficult to effectively identify complex fault modes and early fault signs. Moreover, the data collection accuracy of these systems is limited and cannot fully reflect the actual operating status of the equipment. They cannot meet the needs of modern manufacturing for efficient and accurate fault diagnosis of gear processing equipment. An innovative remote fault diagnosis system is urgently needed to solve these problems. Summary of the invention

[0005] The present invention proposes a remote fault diagnosis system for gear processing equipment based on wireless communication to solve the problems mentioned in the above-mentioned prior art.

[0006] In order to achieve the above object, the present invention adopts the following technical scheme: a remote fault diagnosis system for gear processing equipment based on wireless communication, comprising: Data acquisition module: Multiple types of intelligent sensors are installed at key parts of gear processing equipment, and integrated microprocessors and machine learning algorithms are used to extract features from raw data. Equipment operation data is collected through adaptive sampling frequency. The adaptive sampling frequency f is dynamically adjusted according to the equipment operation status. The calculation formula is: , f maxis the highest frequency of the equipment operation signal, k is a coefficient greater than 1, α is the adjustment factor, S is the real-time status score of the equipment, S 0 is to set the state threshold; Data transmission module: Wi-Fi, Bluetooth, ZigBee or 5G wireless communication technology is used to send the processed data to the remote server. Multi-antenna MIMO technology is introduced to perform encryption before data transmission. An adaptive data retransmission mechanism is set up to dynamically adjust the number of retransmissions n according to the quality of the communication link. The formula is: , P r is the bit error rate of the real-time communication link; Fault diagnosis module: Use a hybrid algorithm that combines deep convolutional neural network DCNN, long short-term memory network LSTM and support vector machine SVM to analyze and process data, establish a dynamic fault feature database, use transfer learning technology to integrate and migrate fault features of different equipment, and extract fault feature entropy value E through multi-scale entropy analysis of vibration signals. The formula is: , p i It is the probability distribution at different scales, comparing the collected data with the fault characteristics in the database to determine the type and severity of equipment faults; Early warning and feedback module: Send early warning information via SMS, email, and APP push, use speech synthesis technology to generate personalized voice warning information according to the type and severity of the fault, provide corresponding solutions and maintenance suggestions, provide remote visual guidance through augmented reality AR technology, collect maintenance feedback information, optimize and update the fault diagnosis model, and use reinforcement learning algorithm to adjust model parameters.

[0007] Furthermore, it also includes: Equipment status evaluation module: collect equipment operation data, use the analytic hierarchy process (AHP) combined with evidence theory to evaluate the equipment status, establish an equipment status evaluation index system, determine the weight of each index, and calculate the comprehensive score S of the equipment status. The formula is: , w i is the weight of the ith indicator, x i is the score of the ith indicator; an uncertainty reasoning mechanism is introduced to deal with the uncertainty information in the evaluation process, and the equipment status is divided into three levels: normal, warning, and fault according to the comprehensive score.

[0008] Furthermore, it also includes: Big data analysis module: Store and analyze equipment operation data and fault data, use graph neural network (GNN) for modeling, mine the correlation between components and fault propagation paths, predict equipment failures by establishing fault prediction models, and use Hadoop and Spark distributed computing technologies to process big data.

[0009] Furthermore, the sensors in the data acquisition module adopt a distributed layout, and data communication and synchronization between sensors are achieved through wireless ad hoc networking technology. Low-power design and dynamic voltage and frequency adjustment DVFS technology are used to dynamically adjust the operating voltage and frequency of the sensors according to the acquisition tasks.

[0010] Furthermore, the data transmission module adopts cognitive radio technology to perceive the spectrum resources in the communication environment in real time, automatically select the optimal communication frequency band to transmit data, adopts a multi-path routing algorithm, and automatically switches to available paths when a fault occurs. The data transmission process adopts network coding technology.

[0011] Furthermore, the fault diagnosis module uses a multi-model fusion fault diagnosis method to analyze and judge the diagnosis results, introduces fuzzy decision-making theory, and dynamically adjusts the weight of each algorithm according to the reliability and applicable scope of different algorithms. In the early stage of equipment operation, the experience-based expert system algorithm has a higher weight. In the later stage of equipment operation, the weight of the data-driven deep learning algorithm is increased.

[0012] Furthermore, the warning information of the early warning and feedback module adopts a hierarchical warning mechanism, which is divided into level one warning, level two warning and level three warning according to the severity of the fault. Level one warning is a minor fault to remind equipment management personnel to pay attention, level two warning is a moderate fault and it is recommended to arrange maintenance in time, and level three warning is a serious fault that requires immediate suspension of equipment operation for maintenance; an intelligent scheduling system is adopted to automatically arrange maintenance personnel according to their skill level, workload and geographical location factors.

[0013] Furthermore, the evaluation index system in the equipment status evaluation module is customized according to different types of gear processing equipment, and blockchain technology is introduced to encrypt, store and share evaluation indicators and evaluation results, and the evaluation index system is regularly updated and optimized.

[0014] Furthermore, the big data analysis module uses cloud computing technology for data storage and processing, and with the help of knowledge graph technology, integrates and correlates equipment failure knowledge and maintenance experience, and uses visualization technology to display the analysis results in the form of charts and reports.

[0015] Furthermore, it also includes: Remote control module: It uses virtual private network VPN technology to remotely operate and control the gear processing equipment to start, stop, and adjust parameters through a remote server. It supports remote updating of fault diagnosis models and uses incremental learning algorithms to update model parameters.

[0016] Compared with the prior art, the present invention has the following beneficial effects: In terms of fault diagnosis efficiency, the system collects equipment operation data in real time through a variety of intelligent sensors, and uses advanced wireless communication technology to quickly transmit it to the remote server. The server uses a powerful fault diagnosis algorithm to quickly analyze and process a large amount of data, quickly and accurately determine the type and severity of the fault, and greatly shortens the fault diagnosis time compared to traditional manual on-site diagnosis. The average fault diagnosis time can be shortened from several hours or even days to tens of minutes, greatly improving the maintenance efficiency of the equipment.

[0017] In terms of fault diagnosis accuracy, a multi-model fusion diagnosis method is adopted, combining algorithms such as deep convolutional neural networks, long short-term memory networks, and support vector machines, as well as dynamic fault feature databases and transfer learning technology. It can accurately identify various complex fault modes and early fault signs. The diagnostic accuracy rate far exceeds that of traditional single algorithm systems, reaching more than 95%, effectively avoiding misjudgments and missed judgments.

[0018] In terms of equipment maintenance management, the equipment status assessment module uses scientific assessment methods to grade the overall status of the equipment, providing a strong basis for preventive maintenance. The big data analysis module explores the patterns of equipment failures, predicts potential failures in advance, helps companies arrange equipment maintenance plans reasonably, reduces sudden failures, and reduces equipment maintenance costs. At the same time, the remote control function facilitates managers to remotely operate equipment, and the remote update function of the fault diagnosis model enables the system to keep pace with the times, continuously improve fault diagnosis capabilities, and comprehensively guarantee the stable and efficient operation of gear processing equipment, thereby improving corporate production efficiency. BRIEF DESCRIPTION OF THE DRAWINGS

[0019] Figure 1 A schematic block diagram of a remote fault diagnosis system for gear processing equipment based on wireless communication proposed by the present invention; Figure 2 This is a schematic diagram of fault diagnosis time comparison; Figure 3 This is a schematic diagram for comparing the accuracy of fault diagnosis; Figure 4 This is a diagram comparing equipment maintenance costs. DETAILED DESCRIPTION

[0020] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0021] In the description of the present invention, it should be understood that the terms "center", "longitudinal", "lateral", "length", "width", "thickness", "up", "down", "front", "back", "left", "right", "vertical", "horizontal", "top", "bottom", "inside", "outside", "clockwise", "counterclockwise" and the like indicate orientations or positional relationships based on the orientations or positional relationships shown in the accompanying drawings, and are only for the convenience of describing the present invention and simplifying the description, rather than indicating or implying that the referred device or element must have a specific orientation, be constructed and operated in a specific orientation, and therefore should not be understood as limiting the present invention.

[0022] In addition, the terms "first" and "second" are used for descriptive purposes only and cannot be understood as indicating or implying relative importance or implicitly indicating the number of technical features indicated. Therefore, the features defined as "first" and "second" may explicitly or implicitly include one or more of the features. In the description of the present invention, the meaning of "multiple" is two or more, unless otherwise clearly and specifically defined. In addition, the terms "installed", "connected" and "connected" should be understood in a broad sense. For example, it can be a fixed connection, a detachable connection, or an integral connection; it can be a mechanical connection or an electrical connection; it can be a direct connection, or it can be indirectly connected through an intermediate medium, and it can be the internal connection of two elements. For ordinary technicians in this field, the specific meanings of the above terms in the present invention can be understood according to specific circumstances. The present invention will be further described in detail below in conjunction with the accompanying drawings.

[0023] Reference Figures 1 to 4 :Gear processing equipment remote fault diagnosis system based on wireless communication, including: Data acquisition module: Various intelligent sensors are installed in key parts of gear processing equipment, such as spindles, tools, transmission gears, etc. These sensors integrate microprocessors and machine learning algorithms, and can perform preliminary feature extraction on the collected raw data locally. The sensors collect the operation data of the equipment in real time at an adaptive sampling frequency, such as vibration frequency, temperature changes, current fluctuations, etc. The adaptive sampling frequency f is dynamically adjusted according to the operation status of the equipment, and the calculation formula is: , where f max is the highest frequency of the equipment operation signal, k is a coefficient greater than 1, α is the adjustment factor, S is the real-time status score of the equipment, S 0 Is the set state threshold. The collected raw data is initially filtered and a denoising method based on wavelet transform and empirical mode decomposition is used to remove noise interference and improve data quality.

[0024] Data transmission module: wireless communication technologies such as Wi-Fi, Bluetooth, ZigBee or 5G are used to send the data processed by the data acquisition module to the remote server. Multi-antenna MIMO technology is introduced to improve the speed and reliability of data transmission. Before data transmission, the data is encrypted and a hybrid encryption algorithm is used, combining symmetric encryption algorithms such as AES and asymmetric encryption algorithms such as RSA to ensure the security and integrity of data transmission. At the same time, an adaptive data retransmission mechanism is set to dynamically adjust the number of retransmissions n according to the quality of the communication link. The formula is: , where P r is the bit error rate of the real-time communication link.

[0025] Fault diagnosis module: After receiving the data, the remote server uses a variety of fault diagnosis algorithms, such as a hybrid algorithm combining deep convolutional neural network (DCNN), long short-term memory network (LSTM) and support vector machine (SVM) to analyze and process the data. A dynamic fault feature database is established, and the fault features of different types of gear processing equipment are integrated and transferred using transfer learning technology to improve the generalization ability of fault diagnosis. For vibration signals, the fault feature entropy value E is extracted through multi-scale entropy analysis. The formula is: , where p i It is the probability distribution at different scales. The collected data is compared with the fault characteristics in the database to determine whether the equipment has a fault and the type and severity of the fault.

[0026] Early warning and feedback module: When the fault diagnosis module determines that the equipment has a fault, it will promptly send early warning information to the equipment management personnel, including SMS, email, APP push, etc. Using speech synthesis technology, generate personalized voice warning information according to the type and severity of the fault. At the same time, according to the type and severity of the fault, provide corresponding solutions and maintenance suggestions, and use augmented reality (AR) technology to provide remote visual guidance for maintenance personnel. After the equipment is repaired, collect maintenance feedback information, optimize and update the fault diagnosis model, and use reinforcement learning algorithm to adjust the model parameters according to the maintenance effect.

[0027] The present invention also includes the following modules: Equipment status evaluation module: Based on the collected equipment operation data, the overall status of the equipment is evaluated by combining the analytic hierarchy process (AHP) and evidence theory. By establishing an equipment status evaluation index system, determining the weight of each index, and calculating the comprehensive score S of the equipment status, the formula is: , where w i is the weight of the ith indicator, x iis the score of the ith indicator. An uncertainty reasoning mechanism is introduced to process the uncertainty information in the evaluation process. The equipment status is divided into different levels such as normal, warning, and fault according to the comprehensive score, providing a basis for equipment maintenance and management.

[0028] The present invention also includes the following modules: Big data analysis module: Store and analyze a large amount of equipment operation data and fault data, use graph neural network (GNN) to model equipment operation data, and mine the correlation between equipment components and fault propagation paths. By establishing a fault prediction model, predict possible equipment failures and take preventive measures in advance. For example, through graph neural network analysis, it is found that equipment is prone to failure under certain specific operating parameter combinations, providing a reference for the optimized operation of the equipment. Use distributed computing technologies, such as Hadoop and Spark, to improve the efficiency of big data processing.

[0029] In the present invention, the data acquisition module adopts a distributed layout, and realizes data communication and synchronization between sensors through wireless ad hoc networking technology. Sensors are scientifically distributed according to the key parts and monitoring requirements of gear processing equipment, such as being deployed in positions such as gear boxes, motors, and workbenches. These sensors are interconnected through wireless ad hoc networking technology, and can form a communication network autonomously without relying on preset infrastructure. In the networking process, advanced routing protocols such as dynamic source routing protocol (DSR) are adopted, and each sensor node can automatically discover and maintain routing information to other nodes. At the same time, media access control mechanisms such as time division multiple access (TDMA) or carrier sense multiple access / collision avoidance (CSMA / CA) are used to realize data communication and synchronization between sensors, ensure the consistency of collected data in time and space, and provide an accurate data basis for subsequent fault diagnosis. The energy collection function provides a new energy source for the continuous operation of the sensor. The sensor is equipped with an efficient energy conversion device. For the collection of vibration energy, a piezoelectric transducer is often used. When the equipment generates vibration during operation, the piezoelectric material deforms to generate electrical energy. For the collection of thermal energy, thermoelectric generators are used to generate electricity through the temperature difference between the operating parts of the equipment and the surrounding environment. The collected electricity is stored in small energy storage elements, such as supercapacitors or rechargeable batteries, to power the sensor itself, reducing dependence on external power supplies, greatly extending the service life of the sensor and reducing maintenance costs. In low-power design, dynamic voltage and frequency scaling (DVFS) technology plays an important role. An intelligent monitoring module is integrated into the sensor to monitor the complexity of the acquisition task in real time. When collecting simple data, such as the normal operating temperature of the equipment, the monitoring module will reduce the operating voltage and frequency of the sensor to reduce energy consumption. When complex data needs to be collected, such as vibration spectrum data when the gear is running at high speed, the operating voltage and frequency will be increased accordingly to ensure the accuracy and completeness of data collection. Through this dynamic adjustment method, while meeting the data collection needs, the power consumption of the sensor is minimized, so that the entire data acquisition module can operate stably for a long time.

[0030] In the present invention, the data transmission module is equipped with a highly sensitive spectrum sensing device to monitor the use of spectrum resources in the communication environment in real time. It can not only detect the occupied frequency bands, but also analyze the signal strength, interference degree and other detailed information of each frequency band. Based on these sensing data, the system uses intelligent algorithms to evaluate and screen the available spectrum. For example, by calculating the signal-to-noise ratio, bit error rate and other indicators of each frequency band, the optimal communication frequency band is comprehensively judged. When the device needs to transmit data, it automatically switches to the frequency band for transmission, effectively avoiding the congestion and interference problems of the frequency band, and greatly improving the speed and quality of data transmission. The multipath routing algorithm provides a strong guarantee for the stability of data transmission. The module will pre-detect and establish multiple data transmission paths, which cover different network nodes and communication links. During the data transmission process, the system continuously monitors the status of each path, including the bandwidth, delay, packet loss rate and other parameters of the link. Once a communication path failure is detected, such as link interruption or network congestion, the multipath routing algorithm will be started immediately. It will quickly select the best alternative path from the pre-established available paths according to the real-time network conditions and data transmission requirements, seamlessly switch the data transmission channel, ensure uninterrupted data transmission, and ensure that the remote fault diagnosis system can obtain the operating data of the equipment in real time. Network coding technology plays a key role in improving data transmission performance. At the sending end, the data is divided into multiple small blocks, encoded and processed, and redundant information is added. These encoded data blocks can be transmitted through different paths. At the receiving end, even if some data blocks are lost during transmission, the original data can be restored through the decoding algorithm using other received data blocks and redundant information. This method not only improves the data transmission throughput and reduces the time wasted due to retransmission of lost data, but also enhances the reliability of data transmission and effectively copes with various interference and data loss problems in complex wireless communication environments.

[0031] In the present invention, the fault diagnosis module integrates a variety of classic and efficient fault diagnosis algorithms on the fault diagnosis method of multi-model fusion, such as spectrum analysis based on vibration signal analysis, thermal imaging diagnosis based on temperature monitoring, motor fault diagnosis based on current data analysis, etc. When the equipment is running, various sensors will collect multi-dimensional data such as vibration, temperature, and current in real time, and input them into the corresponding algorithm model for preliminary analysis. Each algorithm model will output the judgment result of the equipment fault, such as whether there is a fault, the type of fault, and the possible fault location. The introduced fuzzy decision theory plays a key role in this process. The system will evaluate the reliability and scope of application of different algorithms based on the performance data of historical fault diagnosis. For example, the spectrum analysis method has a high accuracy rate in diagnosing gear tooth surface wear faults, but the diagnosis effect of motor internal winding faults is poor. Based on these evaluations, the system can dynamically assign weights to each algorithm. In the early stage of equipment operation, due to the relatively small amount of historical data, the experience-based expert system algorithm can make a more accurate judgment on the fault by relying on the rich knowledge and practical experience accumulated by industry experts, so it is given a higher weight. At this point, the expert system algorithm will comprehensively consider factors such as the equipment's design parameters, common failure modes, and operating specifications, and output fault diagnosis results. As the equipment's operating time increases, the system accumulates a large amount of operating data. Data-driven deep learning algorithms, such as convolutional neural networks (CNN) and long short-term memory networks (LSTM), can mine complex features and patterns from massive amounts of data to diagnose faults more accurately. At this point, the system will automatically and gradually increase the weight of the deep learning algorithm, while adjusting the weights of other algorithms accordingly, thereby combining the advantages of multiple algorithms to draw more accurate and reliable fault diagnosis conclusions, providing strong guarantees for the stable operation of the equipment.

[0032] In the present invention, the early warning and feedback module uses advanced data analysis and fault assessment algorithms. When the equipment operation data is collected, the system will analyze various parameters in real time and compare them with the pre-set fault threshold. For minor faults of the first-level early warning, such as the motor temperature of the gear processing equipment is slightly increased but still at the edge of the normal working range, the system will determine whether it has a trend of further deterioration based on historical data and machine learning models. Once it is determined to be a minor fault, it will remind the equipment management personnel to pay attention in a gentle prompt tone and interface flashing, and provide a detailed equipment operation parameter change trend chart to facilitate the management personnel to grasp the equipment status. The second-level early warning is for moderate faults, such as a certain degree of wear on the gear, which affects the processing accuracy. At this time, the system will not only issue a stronger alarm, but also highlight the fault information with a red mark on the equipment management interface. At the same time, the system will automatically generate a fault analysis report, including the possible causes of the fault, the scope of impact, and the recommended maintenance measures, and it is recommended that the management personnel arrange maintenance in time to prevent the fault from expanding. The third-level early warning corresponds to serious faults, such as the risk of fracture of key components of the equipment. The system will immediately trigger an emergency alarm and automatically send a stop operation instruction to the equipment to avoid causing more serious equipment damage and safety accidents. Alarm information will be sent to relevant persons in charge through various channels such as text messages and emails to ensure that they are informed at the first time. In terms of the intelligent scheduling system, it has built a huge maintenance personnel information database, which records in detail each maintenance personnel's skill level (such as the type of equipment they are good at repairing, the technical expertise they have mastered, etc.), workload (the current amount of tasks on hand, the expected completion time, etc.) and geographical location. When a maintenance task is generated, the system will use an optimization algorithm to comprehensively consider these factors and quickly select the most suitable maintenance personnel to achieve efficient allocation of maintenance resources, minimize equipment maintenance time, and ensure production continuity.

[0033] In the present invention, the characteristics of different types of gear processing equipment are fully considered in the customization of the evaluation index system of the equipment status evaluation module. For the gear hobbing machine, the focus is on indicators such as the degree of tool wear, the rotation accuracy of the worktable, and the axial and radial feed stability. Tool wear can be measured by monitoring the changes in cutting force and the geometric size changes of the tool cutting edge; the rotation accuracy of the worktable uses a high-precision angle sensor to collect data in real time and analyze the angle deviation and jitter during the rotation process. For the gear grinding machine, the evaluation indicators will focus on the wear state of the grinding wheel, the uniformity of the grinding pressure, the surface roughness of the workpiece, etc. The grinding wheel wear analyzes the abrasive wear of the grinding wheel surface through image recognition technology; the uniformity of the grinding pressure uses a pressure sensor array to monitor the pressure values ​​at different positions. Through this customized setting for different equipment, the key features of the equipment operation status can be accurately captured to ensure that the evaluation results are accurate and effective. In terms of data storage and sharing, blockchain technology is introduced. Each evaluation indicator and evaluation result will be converted into an encrypted data block, and the data will be encrypted using the asymmetric encryption algorithm of the blockchain, such as the RSA algorithm. The data blocks are linked into a chain in chronological order, and each data block contains the hash value of the previous data block, forming an unalterable chain structure. When different departments or personnel need to share evaluation data, permission management is performed through the smart contract mechanism of the blockchain. Only authorized users can access specific data, and any operation on the data will be recorded on the blockchain to achieve data traceability. For example, after obtaining the equipment status evaluation results, the equipment maintenance personnel perform maintenance operations, and their operation records will be added to the blockchain to facilitate subsequent inquiries and audits. In order to keep the evaluation index system up to date, regular updates and optimizations are carried out. First, collect the actual operation data of the equipment after the application of new technologies and changes in the operating environment, and analyze whether the existing indicator system can fully reflect the equipment status. For example, when a new intelligent control system is introduced into the equipment, the original evaluation indicators may not cover the operating parameters and performance of the new system, and relevant indicators need to be added. Secondly, use cluster analysis and association rule mining algorithms in machine learning to conduct in-depth analysis of historical evaluation data and equipment failure data to discover potential key indicators or new associations between indicators. Finally, industry experts and equipment technicians are organized to review the updated and optimized indicator system to ensure its scientificity and practicality so as to better adapt to the technical upgrades of equipment and changes in the operating environment.

[0034] In the present invention, the big data analysis module uses cloud computing technology to store massive amounts of gear processing equipment operating data in multiple cloud nodes through a distributed storage architecture. These nodes are intelligently classified and stored according to the type of data (such as real-time operating parameters, historical fault records, etc.) and access frequency to achieve fast data reading and writing. When processing data, the powerful parallel computing capabilities of cloud computing are used to split complex data analysis tasks into multiple subtasks and assign them to different computing resources for simultaneous processing. For example, for the spectrum analysis of equipment vibration data, multiple computing cores can be called simultaneously to perform operations such as fast Fourier transform, which greatly improves the efficiency of data processing compared to traditional single-machine processing. In addition, the cloud computing platform has a data redundancy backup mechanism, which saves the same data copies in multiple physical locations. When a node fails, it can automatically switch to other copies to ensure data reliability and avoid data loss affecting fault diagnosis. In terms of knowledge integration, knowledge graph technology is used. First, the equipment's fault knowledge and maintenance experience are comprehensively sorted out and digitally entered. The fault knowledge covers fault type, fault cause, fault manifestation, etc., and the maintenance experience includes maintenance steps, replacement parts records, etc. Then, the correlation between these information is mined through natural language processing and machine learning algorithms. For example, it is found that there is a high correlation between a certain abnormal vibration frequency and gear wear failure, and this correlation is presented in the form of a graph. With the accumulation of new fault cases and maintenance experience, the knowledge graph will be automatically updated and expanded to provide richer and more comprehensive knowledge support for fault diagnosis and prediction, so that the diagnosis process is no longer limited to a single fault feature, but can comprehensively consider multiple related factors. In the display of analysis results, visualization technology is used. The system will select the appropriate chart type according to the content of data analysis and the needs of equipment managers. For the trend analysis of equipment operating parameters, a line chart is used to clearly show the changes of parameters over time; for the distribution statistics of different fault types, a bar chart or pie chart is used to intuitively present the proportion of various types of faults. At the same time, the report not only contains data summaries, but also provides detailed interpretations and early warning prompts for key indicators. For example, when the temperature of a key component of the equipment exceeds the normal threshold, the report will highlight the anomaly and provide possible causes and recommended treatment measures, so that equipment managers can quickly understand the equipment status and make scientific decisions.

[0035] The present invention also includes the following modules: Remote control module: Equipment managers can use remote servers to perform a variety of remote operations and precise control of gear processing equipment through stable and reliable wireless communication links. Specifically, the startup operation is not a simple signal transmission, but a multiple handshake verification between the server and the built-in intelligent controller of the equipment. After confirming the operation authority and equipment status, a command containing specific startup parameters is sent to ensure that the equipment can start according to the preset optimal mode. The stop operation adopts a progressive power-off and braking strategy, first cutting off the power input of the equipment and activating the braking device at the same time to avoid equipment damage and processing errors caused by inertia. For parameter adjustment, the system supports numerical settings with precision to multiple decimal places, covering key processing parameters such as speed, feed rate, and cutting depth. Managers can make real-time adjustments according to different processing requirements and material properties. In order to ensure data security and operation safety during remote control, the system adopts virtual private network (VPN) technology. At the data transmission level, advanced encryption algorithms, such as AES-256-bit encryption, are used to perform high-intensity encryption processing on remote control instructions and equipment feedback data to prevent data from being stolen or tampered with during transmission. At the same time, the VPN network has a strict identity authentication mechanism. In addition to the traditional username and password authentication, it also introduces multi-factor authentication, such as dynamic verification codes, biometrics (fingerprint or facial recognition), etc., to ensure that only authorized managers can access the remote control network. In terms of the remote update function of the fault diagnosis model, the system has designed an efficient and intelligent update mechanism. When the latest fault diagnosis algorithm and model are released, the remote server will automatically detect the model version on the device and compare it with the latest version. If there is a version difference, the server will intelligently select the appropriate time window for update push based on the current workload and network conditions of the device. During the update process, an incremental learning algorithm is used, which can accurately identify the parts of the new algorithm and model that are different from the original model, and only transmit data and update parameters for these different parts, greatly reducing the amount of data transmission and update time. At the same time, the system will monitor and evaluate the model performance in real time before and after the update. Once it is found that the update may have a negative impact on the performance of the original model, it will immediately suspend the update and automatically roll back to the previous stable version to ensure that the device can maintain reliable fault diagnosis capabilities under any circumstances.

[0036] The above are only preferred specific implementation modes of the present invention, but the protection scope of the present invention is not limited thereto. Any technician familiar with the technical field can make equivalent replacements or changes according to the technical solutions and inventive concepts of the present invention within the technical scope disclosed by the present invention, which should be covered by the protection scope of the present invention.

Claims

1. A remote fault diagnosis system for gear processing equipment based on wireless communication, characterized in that: Includes the following modules: Data acquisition module: Multiple types of intelligent sensors are installed at key parts of gear processing equipment, and integrated microprocessors and machine learning algorithms are used to extract features from raw data. Equipment operation data is collected through adaptive sampling frequency. The adaptive sampling frequency f is dynamically adjusted according to the equipment operation status. The calculation formula is: , f max is the highest frequency of the equipment operation signal, k is a coefficient greater than 1, α is the adjustment factor, S is the real-time status score of the equipment, and S0 is the set status threshold; Data transmission module: Wi-Fi, Bluetooth, ZigBee or 5G wireless communication technology is used to send the processed data to the remote server. Multi-antenna MIMO technology is introduced to perform encryption before data transmission. An adaptive data retransmission mechanism is set up to dynamically adjust the number of retransmissions n according to the quality of the communication link. The formula is: , P r is the bit error rate of the real-time communication link; Fault diagnosis module: Use a hybrid algorithm that combines deep convolutional neural network DCNN, long short-term memory network LSTM and support vector machine SVM to analyze and process data, establish a dynamic fault feature database, use transfer learning technology to integrate and migrate fault features of different equipment, and extract fault feature entropy value E through multi-scale entropy analysis of vibration signals. The formula is: , p i It is the probability distribution at different scales, comparing the collected data with the fault characteristics in the database to determine the type and severity of equipment faults; Early warning and feedback module: Send early warning information via SMS, email, and APP push, use speech synthesis technology to generate personalized voice warning information according to the type and severity of the fault, provide corresponding solutions and maintenance suggestions, provide remote visual guidance through augmented reality AR technology, collect maintenance feedback information, optimize and update the fault diagnosis model, and use reinforcement learning algorithm to adjust model parameters.

2. The remote fault diagnosis system for gear processing equipment based on wireless communication according to claim 1 is characterized in that: Also includes: Equipment status evaluation module: collect equipment operation data, use the analytic hierarchy process (AHP) combined with evidence theory to evaluate the equipment status, establish an equipment status evaluation index system, determine the weight of each index, and calculate the comprehensive score S of the equipment status. The formula is: , w i is the weight of the ith indicator, x i is the score of the ith indicator; an uncertainty reasoning mechanism is introduced to deal with the uncertainty information in the evaluation process, and the equipment status is divided into three levels: normal, warning, and fault according to the comprehensive score.

3. The remote fault diagnosis system for gear processing equipment based on wireless communication according to claim 1, characterized in that: Also includes: Big data analysis module: Store and analyze equipment operation data and fault data, use graph neural network (GNN) for modeling, mine the correlation between components and fault propagation paths, predict equipment failures by establishing fault prediction models, and use Hadoop and Spark distributed computing technologies to process big data.

4. The remote fault diagnosis system for gear processing equipment based on wireless communication according to claim 1, characterized in that: The sensors in the data acquisition module adopt a distributed layout, and data communication and synchronization between sensors are achieved through wireless ad hoc networking technology. Low-power design and dynamic voltage and frequency adjustment DVFS technology are adopted to dynamically adjust the operating voltage and frequency of the sensors according to the acquisition tasks.

5. The remote fault diagnosis system for gear processing equipment based on wireless communication according to claim 1, characterized in that: The data transmission module adopts cognitive radio technology to perceive the spectrum resources in the communication environment in real time, automatically select the optimal communication frequency band to transmit data, adopt a multi-path routing algorithm, and automatically switch available paths when a fault occurs. The data transmission process adopts network coding technology.

6. The remote fault diagnosis system for gear processing equipment based on wireless communication according to claim 1, characterized in that: The fault diagnosis module uses a multi-model fusion fault diagnosis method to analyze and judge the diagnosis results, introduces fuzzy decision-making theory, and dynamically adjusts the weight of each algorithm according to the reliability and applicable scope of different algorithms. In the early stage of equipment operation, the experience-based expert system algorithm has a higher weight. In the later stage of equipment operation, the weight of the data-driven deep learning algorithm is increased.

7. The remote fault diagnosis system for gear processing equipment based on wireless communication according to claim 1, characterized in that: The warning information of the early warning and feedback module adopts a graded warning mechanism, which is divided into level one warning, level two warning and level three warning according to the severity of the fault. Level one warning is a minor fault to remind equipment management personnel to pay attention, level two warning is a moderate fault and it is recommended to arrange maintenance in time, and level three warning is a serious fault that requires immediate suspension of equipment operation for maintenance. An intelligent scheduling system is adopted to automatically arrange maintenance personnel according to their skill level, workload and geographical location.

8. The remote fault diagnosis system for gear processing equipment based on wireless communication according to claim 2, characterized in that: The evaluation index system in the equipment status evaluation module is customized according to different types of gear processing equipment. Blockchain technology is introduced to encrypt, store and share evaluation indicators and evaluation results, and the evaluation index system is regularly updated and optimized.

9. The remote fault diagnosis system for gear processing equipment based on wireless communication according to claim 3, characterized in that: The big data analysis module uses cloud computing technology for data storage and processing, and uses knowledge graph technology to integrate and correlate equipment failure knowledge and maintenance experience, and uses visualization technology to display analysis results in the form of charts and reports.

10. The remote fault diagnosis system for gear processing equipment based on wireless communication according to claim 1, characterized in that: Also includes: Remote control module: It uses virtual private network VPN technology to remotely operate and control the gear processing equipment to start, stop, and adjust parameters through a remote server. It supports remote updating of fault diagnosis models and uses incremental learning algorithms to update model parameters.

Images