Equipment fault maintenance management method and system based on augmented reality technology

Abstract

The invention discloses an equipment fault maintenance management method and system based on an augmented reality technology. The method comprises the steps of data collection, data preprocessing, equipment fault detection, maintenance scheme optimization and maintenance augmented management. The invention relates to the technical field of equipment fault maintenance, in particular to an equipment fault maintenance management method and system based on an augmented reality technology. According to the scheme, the double-line intelligent function combining equipment fault detection and maintenance scheme optimization is adopted, the depth of intelligent equipment fault maintenance management is achieved from the microcosmic angle and the macroscopic angle, the overall usability and universality of the method are improved, and meanwhile richer auxiliary performance is provided for hardware man-machine interaction; carrying out equipment fault detection by adopting a convolutional long-short-term memory network combined with an improved optimization algorithm; and optimizing a maintenance scheme by adopting a scheme optimization weighting method combining an improved optimization algorithm and reinforcement learning.

Description

Technical Field

[0001] The present invention relates to the technical field of equipment fault maintenance, and in particular to an equipment fault maintenance management method and system based on augmented reality technology. Background Art

[0002] The equipment fault inspection and maintenance management method and system based on augmented reality (AR) technology provides maintenance personnel with real-time fault diagnosis, maintenance operation guidance and resource management support by combining virtual information with real equipment. The system uses AR technology to overlay relevant data of equipment faults, maintenance steps, operation prompts and expert advice on the actual view of the equipment, helping operators to locate faults, repair and maintain them more accurately. Its role is to improve maintenance efficiency, reduce operating errors, reduce maintenance costs, and achieve long-term reliable operation and maintenance of equipment through intelligent fault prediction and optimization solutions.

[0003] However, in the existing equipment fault maintenance management methods, there is a lack of specific fault detection and macro-level maintenance overall solutions when combined with augmented reality technology. This lacks technical support for the overall equipment maintenance level in the process of AR guiding mechanical equipment maintenance, thereby reducing the technical problem of the auxiliary performance of augmented reality technology for equipment maintenance; in the existing equipment fault detection methods, there is a technical problem that in the existing equipment fault detection intelligent methods, since equipment fault detection involves both complex numerical data features and requires clear classification categories, levels and detailed identification and analysis, traditional artificial intelligence methods are difficult to meet the needs of complex output conditions; in the existing maintenance plan optimization methods, there is a problem that the existing maintenance plans often rely on manual experience to make judgments, but this method is difficult to provide more macro maintenance support when combined with augmented reality technology. Since the maintenance of equipment usually involves multiple problems of multiple devices, an optimal solution that takes into account both overall scheduling and micro-actual maintenance should be used as the main technical support for augmented reality technology maintenance guidance, and the existing intelligent methods and manual methods are difficult to achieve the technical problem of combining overall scheduling with specific maintenance. Summary of the invention

[0004] In view of the above situation, in order to overcome the defects of the prior art, the present invention provides an equipment fault inspection and maintenance management method and system based on augmented reality technology. In view of the fact that in the existing equipment fault inspection and maintenance management method, the existing technical solution lacks the combination of specific fault detection and overall maintenance plan at the macro level when combined with augmented reality technology, which lacks technical support for the overall equipment maintenance level in the process of AR guiding mechanical equipment maintenance, thereby reducing the technical problem of the auxiliary performance of augmented reality technology for equipment maintenance. This solution creatively adopts a dual-route intelligent function combining equipment fault detection and maintenance plan optimization, realizes the depth of intelligent equipment fault inspection and maintenance management from both micro and macro perspectives, improves the overall usability and versatility of the method, and also provides richer auxiliary performance for hardware human-computer interaction; In view of the fact that in the existing equipment fault detection method, the existing intelligent method of equipment fault detection involves both complex numerical data features and requires clear classification categories, levels and detailed identification and analysis, so that traditional artificial intelligence methods are difficult to meet the needs of complex output conditions, this solution creatively adopts a convolutional long short-term memory network combined with an improved optimization algorithm to perform equipment fault detection. By setting An improved sparrow search algorithm is designed to optimize the parameter combination solving ability for complex input data and detailed output data from the whole search process, and is applied to the two levels of feature optimization and model parameters, which improves the output dimension and result availability of equipment fault detection; in the existing maintenance plan optimization method, the existing maintenance plan often relies on manual experience to make judgments, but this method is difficult to provide more macro maintenance support when combined with augmented reality technology. Since the maintenance of equipment usually involves multiple problems of multiple equipment, an optimal solution that takes into account both overall scheduling and micro actual maintenance should be used as an augmented reality The main technical support for technical maintenance guidance is provided by intelligent methods and manual methods. However, it is difficult to achieve the technical problem of combining overall scheduling with specific maintenance. This solution creatively adopts a weighted method of solution optimization that combines improved optimization algorithm and reinforcement learning to optimize the maintenance solution. By combining the improved search algorithm in the fault detection stage, it provides global decision support for the maintenance solution. At the same time, it combines the solution optimization performance of reinforcement learning technology to provide suggestions for specific maintenance solutions. Through the weighted combination of the two, it effectively optimizes the scheduling capabilities of the two levels of the intelligent maintenance solution, and also provides a practical attempt for the effective combination with augmented reality technology.

[0005] The technical solution adopted by the present invention is as follows: The present invention provides an equipment fault inspection and maintenance management method based on augmented reality technology, the method comprising the following steps:

[0006] Step S1: data collection;

[0007] Step S2: data preprocessing;

[0008] Step S3: equipment failure detection;

[0009] Step S4: Optimizing maintenance plan;

[0010] Step S5: Maintenance and enhanced management.

[0011] Furthermore, in step S1, the data collection is used to collect the original data set required for equipment fault troubleshooting and maintenance management, specifically, by collecting data from industrial machinery equipment, obtaining the original data for equipment fault troubleshooting and maintenance management, and optimizing the real-time data collection through industrial Internet of Things technology;

[0012] The equipment fault inspection and maintenance management original data specifically includes mechanical vibration data, temperature data, electrical data, shaft speed data and lubrication status data.

[0013] Furthermore, in step S2, the data preprocessing is used to enhance and optimize the original data, specifically, to optimize the data content and features based on the original data of the equipment fault maintenance management, and to store the data to obtain equipment fault maintenance optimization data.

[0014] Further, in step S3, the equipment fault detection is used to intelligently analyze whether the equipment has a fault or a potential fault. Specifically, based on the equipment fault maintenance optimization data, a convolutional long short-term memory network combined with an improved optimization algorithm is used to perform equipment fault detection to obtain equipment fault detection data, which specifically includes the following steps:

[0015] Step S31: constructing an improved optimization algorithm, specifically constructing a standard sparrow search algorithm as an optimization algorithm, and improving the optimization algorithm through population initialization improvement, adaptive update rule improvement, imitation learning improvement and fitness calculation improvement, specifically including the following steps:

[0016] Step S311: Improvement of population initialization, which is used to adapt to the multi-feature and multi-parameter requirements of equipment fault detection and expand the search range of the optimization algorithm. Specifically, the population initialization stage is improved by combining Latin hypercube sampling and distribution adaptive adjustment method. Specifically, the initial population of the standard sparrow algorithm is generated by Latin hypercube sampling, and the calculation of the initial population position is improved by adding a dynamic adjustment factor. The calculation formula is:

[0017] ;

[0018] In the formula, is the population sampling sample value corresponding to the total sample size of N and the sampling interval of m, which is used to represent the specific initial value to be optimized. In step S3, it is specifically used to represent the initial value to be optimized of the original features in the equipment fault maintenance optimization data and the model parameters of the equipment fault detection model. K is the total number of samples. is a random weight with a value range of [0,1], m is the sampling interval of the input data, which is used to indicate the frequency of data collection;

[0019] M Nm It is the initial value of the sampling data point corresponding to the total sample size of N and the sampling interval of m, which is used to represent the specific initial value of the sparrow search algorithm population. is the inverse function of the cumulative probability distribution function, is the population expansion factor, the initial value is set to 0.1, rand(0,1) is a fine-tuned random number randomly sampled from a normal distribution;

[0020] Step S312: Adaptive update rule improvement, used to introduce global optimal information enhancement features and model parameter optimization capabilities, specifically, adaptive update improvement of forager parameter update rules and follower parameter update rules in the sparrow search algorithm, and adaptive update based on the improved forager parameter update rules and follower parameter update rules;

[0021] The improved calculation formula of the forager parameter update rule is:

[0022] ;

[0023] In the formula, is the sparrow position of the i-th forager individual at the t+1-th iteration, and the sparrow position is used to represent the search result of the sparrow algorithm. is the sparrow position of the i-th forager individual at the t-th iteration, where i is the forager individual index, which is used to represent the forager individual that performs the global optimal search in the sparrow algorithm, t is the iteration number index, which is used to control the running termination condition of the improved optimization algorithm, and T is the maximum number of iterations. is an exponential decay factor, which is used to control the update amplitude of the update rule. is the guiding factor, which is used to control the speed at which individuals search for the global optimal solution. is the optimal sparrow position at the tth iteration, used to represent the optimal solution, is the first random perturbation factor, r1 is a random update control parameter with a value range of [0,1], P is the forager environment control parameter, is the second random perturbation factor, is the average sparrow position at the tth iteration;

[0024] The improved calculation formula of the follower parameter update rule is:

[0025] ;

[0026] In the formula, is the sparrow position of the jth follower individual at the t+1th iteration, is the sparrow position of the jth follower individual at the tth iteration, where j is the follower individual index, which is used to represent the follower individual performing the local optimal search in the sparrow algorithm. is a balancing factor used to control the scope of local search;

[0027] Step S313: Imitation learning improvement is used to optimize the diversity of the search process. Specifically, by introducing the imitation learning of the best individual and the reverse incentive of the worst individual, the imitation learning improvement of the standard warning mechanism is performed. The calculation formula is:

[0028] ;

[0029] In the formula, is the sparrow position of the kth alert individual at the t+1th iteration, is the sparrow position of the kth alert individual at the tth iteration, where k is the alert individual index, which is used to represent the alert individual for optimizing the local optimal solution in the sparrow algorithm. is a random optimal individual weight in the range [0,1], is the random worst individual weight in the range [0,1], is the worst sparrow position at the tth iteration;

[0030] Step S314: Dynamically adjust the fitness calculation and improve it, specifically by introducing a gradient penalty term and dynamically adjusting the weight to improve the multi-objective fitness function. The calculation formula is:

[0031] ;

[0032] Where f(·) is the dynamically adjusted multi-objective improved fitness function, It's the sparrow position. is the sparrow position index, w is the balance factor, f org (·) is the original fitness function, g(·) is the gradient penalty term;

[0033] Step S315: improving the optimization algorithm construction, specifically, improving the standard sparrow search algorithm to adapt to the needs of equipment fault detection through the population initialization improvement, the adaptive update rule improvement, the imitation learning improvement and the dynamic adjustment fitness calculation improvement, and optimizing the parameter target by setting the operation termination condition of the improved optimization algorithm;

[0034] The parameter target optimization specifically includes feature selection optimization of equipment fault detection and model parameter optimization of equipment fault detection;

[0035] Step S32: constructing a convolutional long short-term memory network, specifically constructing a standard convolutional long short-term memory neural network including a convolutional layer, a pooling layer and a long short-term memory layer, for performing temporal and spatial feature extraction optimization for equipment fault detection, and obtaining the convolutional long short-term memory network;

[0036] Step S33: parallel optimization design, specifically, by executing the operation of step S31, using the improved optimization algorithm to respectively perform feature selection optimization in the equipment fault maintenance optimization data and model parameter optimization of the convolutional long short-term memory network, and optimizing the performance of equipment fault detection in parallel;

[0037] Step S34: Equipment fault detection model training, specifically, training the equipment fault detection model through the improved optimization algorithm, the convolutional long short-term memory network and the parallel optimization design to obtain the equipment fault detection model Model SC ;

[0038] Step S35: Device fault detection, specifically using the device fault detection model Model SC , performing equipment fault detection according to the equipment fault repair optimization data to obtain equipment fault detection data;

[0039] The equipment fault detection data specifically includes numerical detection data and categorical detection data;

[0040] The numerical detection data specifically includes vibration fault detection data, temperature fault detection data, current fault detection data, voltage fault detection data, shaft speed fault detection data and lubrication state fault detection data;

[0041] The categorized detection data specifically includes normal equipment operation category and equipment operation failure category;

[0042] The equipment operation fault category specifically includes a fault detail category and a fault level category;

[0043] The detailed fault categories include mechanical faults, electrical faults and thermal faults;

[0044] The fault level categories specifically include minor faults, moderate faults and severe faults.

[0045] Further, in step S4, the maintenance scheme optimization is used to automatically generate the optimal maintenance scheme according to the result of the fault detection, specifically, based on the equipment fault detection data and the equipment fault maintenance optimization data, a scheme optimization weighting method combining an improved optimization algorithm and reinforcement learning is used to optimize the maintenance scheme to obtain maintenance scheme optimization reference data, specifically including the following steps:

[0046] Step S41: constructing an improved optimization algorithm, specifically executing the operation of step S31 to construct an improved optimization algorithm for optimizing the maintenance plan;

[0047] Step S42: reinforcement learning optimization of the maintenance plan, specifically, by constructing a maintenance plan parameter environment and a maintenance plan intelligent agent, and adopting a reinforcement learning method combined with an improved optimization algorithm to perform reinforcement learning optimization of the maintenance plan, and obtain reinforcement learning optimized maintenance plan reference data;

[0048] Step S43: optimizing the maintenance scheme optimization algorithm, specifically, searching and solving the maintenance scheme using the improved optimization algorithm according to the equipment fault detection data and the equipment fault maintenance optimization data, and obtaining the maintenance scheme reference data of the optimization algorithm;

[0049] Step S44: weighting the output of the solution, specifically, performing weighted average summation of the optimized maintenance time, optimized maintenance cost and optimized resource allocation solution in the reinforcement learning optimized maintenance solution reference data and the optimized algorithm maintenance solution reference data to obtain weighted output maintenance solution data;

[0050] Step S45: optimizing the maintenance plan, specifically optimizing the maintenance plan by constructing the improved optimization algorithm, optimizing the maintenance plan through reinforcement learning, optimizing the maintenance plan optimization algorithm, and weighting the plan output, to obtain maintenance plan optimization reference data;

[0051] The maintenance plan optimization reference data specifically includes optimizing maintenance time, optimizing maintenance costs, optimizing resource allocation plans, and optimizing fault recovery levels.

[0052] Further, in step S5, the maintenance enhancement management is used to enhance the equipment maintenance management by combining the equipment fault detection results and the maintenance optimization plan through augmented reality technology and intelligent feedback system, specifically, by using augmented reality equipment, providing equipment maintenance reference for equipment maintenance personnel, and by combining the equipment fault detection data and the maintenance plan optimization reference data, intelligent maintenance management is performed to obtain equipment maintenance management enhancement data, and equipment maintenance enhancement management is performed based on the equipment maintenance management enhancement data;

[0053] The equipment maintenance enhanced management specifically includes equipment maintenance operation guidance, equipment fault location marking and equipment maintenance step demonstration;

[0054] The use of augmented reality equipment provides equipment maintenance personnel with equipment maintenance references, specifically by superimposing equipment fault information and optimal maintenance solutions based on the equipment fault detection data and the maintenance solution optimization reference data in the control interface of the augmented reality equipment, and providing equipment maintenance references by introducing three-dimensional animation demonstration tools and maintenance tool prompts, and recording, analyzing and maintaining maintenance data through an intelligent data management system.

[0055] The present invention provides an equipment fault repair management system based on augmented reality technology, comprising a data management module, an equipment fault repair module, an augmented reality guidance module and a hardware management module;

[0056] The data management module is used for data collection and data preprocessing, and obtains equipment fault repair optimization data through data collection and data preprocessing, and sends the equipment fault repair optimization data to the equipment fault repair module;

[0057] The equipment fault repair module is used for equipment fault detection and repair scheme optimization. Through equipment fault detection and repair scheme optimization, equipment fault detection data and repair scheme optimization reference data are obtained, and the equipment fault detection data and repair scheme optimization reference data are sent to the augmented reality guidance module;

[0058] The augmented reality guidance module is used for maintenance enhancement management, through which equipment maintenance management enhancement data is obtained, and the equipment maintenance management enhancement data is sent to the hardware management module;

[0059] The hardware management module is used for hardware management. Through hardware management, augmented reality fault marking and maintenance operation guidance are performed in the maintenance scene, and fault maintenance confirmation is confirmed through human-computer interaction to obtain fault maintenance management data records. Based on the fault maintenance management data records, management and replacement of augmented reality hardware equipment and maintenance equipment are performed.

[0060] The beneficial effects achieved by the present invention using the above scheme are as follows:

[0061] (1) In view of the fact that the existing equipment fault maintenance management methods lack the combination of specific fault detection and overall maintenance solutions at the macro level when combined with augmented reality technology, the technical support for the overall equipment maintenance level is missing in the process of AR guiding mechanical equipment maintenance, thereby reducing the technical problem of the auxiliary performance of augmented reality technology for equipment maintenance. This solution creatively adopts a dual-route intelligent function that combines equipment fault detection and maintenance solution optimization, and realizes the depth of intelligent equipment fault maintenance management from both micro and macro perspectives, thereby improving the overall usability and versatility of the method, and also providing richer auxiliary performance for hardware human-computer interaction;

[0062] (2) In view of the technical problem that in the existing equipment fault detection intelligent methods, traditional artificial intelligence methods are difficult to meet the needs of complex output conditions because equipment fault detection involves complex numerical data features and requires clear classification categories, levels and detailed identification and analysis, this solution creatively adopts a convolutional long short-term memory network combined with an improved optimization algorithm to perform equipment fault detection. By designing an improved sparrow search algorithm, the ability to solve the parameter combination of complex input data and detailed output data is optimized from the perspective of the entire search process, and it is applied to the two levels of feature optimization and model parameters, thereby improving the output dimension of equipment fault detection and the availability of results;

[0063] (3) In the existing maintenance plan optimization methods, the existing maintenance plans often rely on manual experience to make judgments. However, this method is difficult to provide more macro-maintenance support when combined with augmented reality technology. Since equipment maintenance usually involves multiple problems of multiple equipment, an optimal plan that takes into account both overall scheduling and micro-actual maintenance should be used as the main technical support for augmented reality technology maintenance guidance. However, both existing intelligent methods and manual methods are difficult to achieve the technical problem of combining overall scheduling with specific maintenance. This scheme creatively adopts a scheme optimization weighted method that combines improved optimization algorithms and reinforcement learning to optimize maintenance plans. By combining the improved search algorithm in the fault detection stage, it provides global decision support for maintenance plans. At the same time, combined with the scheme optimization performance of reinforcement learning technology, it provides suggestions for specific maintenance plans. Through the weighted combination of the two, the scheduling capabilities of the two levels of intelligent maintenance plans are effectively optimized, and a practical attempt is also provided for the effective combination with augmented reality technology. BRIEF DESCRIPTION OF THE DRAWINGS

[0064] Figure 1 A flowchart of an equipment fault inspection and maintenance management method based on augmented reality technology provided by the present invention;

[0065] Figure 2A schematic diagram of an equipment fault inspection and maintenance management system based on augmented reality technology provided by the present invention;

[0066] Figure 3 This is a schematic diagram of the process of data preprocessing in step S2;

[0067] Figure 4 This is a schematic diagram of the process of device fault detection in step S3;

[0068] Figure 5 A schematic flow chart of constructing an improved optimization algorithm for step S31;

[0069] Figure 6 Schematic diagram of the process flow for optimizing the maintenance plan for step S4.

[0070] The accompanying drawings are used to provide further understanding of the present invention and constitute a part of the specification. They are used to explain the present invention together with the embodiments of the present invention and do not constitute a limitation of the present invention. DETAILED DESCRIPTION

[0071] The technical solutions in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, rather than all the embodiments; based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative work are within the scope of protection of the present invention.

[0072] In the description of the present invention, it should be understood that terms such as “upper”, “lower”, “front”, “back”, “left”, “right”, “top”, “bottom”, “inside” and “outside” indicating directions or positional relationships are based on the directions 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 device or element referred to must have a specific direction, be constructed and operated in a specific direction, and therefore should not be understood as limiting the present invention.

[0073] Example 1, see Figure 1 The present invention provides an equipment fault inspection and maintenance management method based on augmented reality technology, the method comprising the following steps:

[0074] Step S1: data collection;

[0075] Step S2: data preprocessing;

[0076] Step S3: equipment failure detection;

[0077] Step S4: Optimizing maintenance plan;

[0078] Step S5: Maintenance and enhanced management.

[0079] By performing the above operations, in the existing equipment fault maintenance management methods, there is a problem that the existing technical solutions lack the combination of specific fault detection and overall maintenance solutions at the macro level when combined with augmented reality technology. This lacks technical support for the overall equipment maintenance level in the process of AR-guided mechanical equipment maintenance, thereby reducing the auxiliary performance of augmented reality technology for equipment maintenance. This solution creatively adopts a dual-route intelligent function that combines equipment fault detection and maintenance solution optimization, and realizes the depth of intelligent equipment fault maintenance management from both micro and macro perspectives, thereby improving the overall usability and versatility of the method, and also provides richer auxiliary performance for hardware human-computer interaction.

[0080] Example 2, see Figure 1 and Figure 2 In step S1, the data collection is used to collect the original data set required for equipment fault troubleshooting and maintenance management, specifically, through industrial machinery equipment data collection, the original data of equipment fault troubleshooting and maintenance management is obtained and the real-time data collection is optimized through the industrial Internet of Things technology;

[0081] The equipment fault inspection and maintenance management raw data specifically includes mechanical vibration data, temperature data, electrical data, shaft speed data and lubrication status data;

[0082] The mechanical vibration data is specifically collected by a vibration sensor to characterize the bearing vibration during the operation of the industrial machinery and equipment, and the specific collection frequency is set to 1000 Hz;

[0083] The temperature data is specifically collected through a temperature sensor to characterize the temperature changes of the spindle, click and coolant parts of the industrial machinery and equipment. The specific collection frequency is once every 1 second, once every 5 seconds and once every 10 seconds.

[0084] The electrical data is specifically collected by current sensors and voltage sensors to characterize the electrical changes of the current fluctuations and voltage fluctuations of the motors of industrial machinery and equipment. The specific collection frequency is set to 100 Hz.

[0085] The shaft speed data is specifically collected through an encoder and a rotation meter to characterize the spindle speed fluctuation of the industrial machinery equipment, and the specific collection frequency is set to once per second;

[0086] The lubrication status data is specifically collected through a pH sensor and a viscosity sensor to characterize the pH value and viscosity of the lubricating oil during the working process of the industrial machinery and equipment. The specific collection frequency is set to once every 10 minutes;

[0087] The industrial machinery and equipment specifically refers to CNC machine tool equipment;

[0088] The industrial machinery equipment data collection is specifically carried out through sensor arrangement to collect real-time data of CNC machine tool equipment.

[0089] Example 3, see Figure 1 , Figure 2 and Figure 3 This embodiment is based on the above embodiment. In step S2, the data preprocessing is used to enhance and optimize the original data. Specifically, according to the original data of the equipment fault maintenance management, the data content and characteristics are optimized, and the data is stored to obtain the equipment fault maintenance optimization data. Specifically, the following steps are included:

[0090] Step S21: data content cleaning, specifically, performing low-pass filter denoising on the mechanical vibration data in the raw data of the equipment fault inspection and maintenance management, performing smoothing processing on the temperature data, performing missing value processing on the electrical data, shaft speed data and lubrication status data, and filling the missing values ​​by linear interpolation to obtain content cleaning optimization data;

[0091] Step S22: data classification and standardization, specifically, performing a Z-score standardization operation on all data in the content cleaning and optimization data, and performing a maximum and minimum normalization operation on the temperature data and the electrical data to obtain classification and standardized data;

[0092] Step S23: time series data processing, specifically optimizing and aligning the time series data in the equipment fault inspection and maintenance management raw data to obtain timestamp aligned data;

[0093] Step S24: initial optimization of feature data, specifically, initial feature extraction is performed through Fourier transform and principal component analysis methods to obtain initial feature data for equipment fault maintenance;

[0094] Step S25: data storage organization, specifically, preprocessing the data through data content cleaning, data classification standardization, time series data processing and initial optimization of feature data to obtain equipment fault maintenance optimization data, and storing and organizing index management of the equipment fault maintenance optimization data.

[0095] Example 4, see Figure 1 , Figure 2 , Figure 4 and Figure 5 This embodiment is based on the above embodiment. In step S3, the equipment fault detection is used to intelligently analyze whether the equipment has a fault or a potential fault. Specifically, based on the equipment fault maintenance optimization data, a convolutional long short-term memory network combined with an improved optimization algorithm is used to perform equipment fault detection to obtain equipment fault detection data. Specifically, the following steps are included:

[0096] Step S31: constructing an improved optimization algorithm, specifically constructing a standard sparrow search algorithm as an optimization algorithm, and improving the optimization algorithm through population initialization improvement, adaptive update rule improvement, imitation learning improvement and fitness calculation improvement, specifically including the following steps:

[0097] Step S311: Improvement of population initialization, which is used to adapt to the multi-feature and multi-parameter requirements of equipment fault detection and expand the search range of the optimization algorithm. Specifically, the population initialization stage is improved by combining Latin hypercube sampling and distribution adaptive adjustment method. Specifically, the initial population of the standard sparrow algorithm is generated by Latin hypercube sampling, and the calculation of the initial population position is improved by adding a dynamic adjustment factor. The calculation formula is:

[0098] ;

[0099] In the formula, is the population sampling sample value corresponding to the total sample size of N and the sampling interval of m, which is used to represent the specific initial value to be optimized. In step S3, it is specifically used to represent the initial value to be optimized of the original features in the equipment fault maintenance optimization data and the model parameters of the equipment fault detection model. K is the total number of samples. is a random weight with a value range of [0,1], m is the sampling interval of the input data, which is used to indicate the frequency of data collection;

[0100] M Nm It is the initial value of the sampling data point corresponding to the total sample size of N and the sampling interval of m, which is used to represent the specific initial value of the sparrow search algorithm population. is the inverse function of the cumulative probability distribution function, is the population expansion factor, the initial value is set to 0.1, rand(0,1) is a fine-tuned random number randomly sampled from a normal distribution;

[0101] Step S312: Adaptive update rule improvement, used to introduce global optimal information enhancement features and model parameter optimization capabilities, specifically, adaptive update improvement of forager parameter update rules and follower parameter update rules in the sparrow search algorithm, and adaptive update based on the improved forager parameter update rules and follower parameter update rules;

[0102] The improved calculation formula of the forager parameter update rule is:

[0103] ;

[0104] In the formula, is the sparrow position of the i-th forager individual at the t+1-th iteration, and the sparrow position is used to represent the search result of the sparrow algorithm. is the sparrow position of the i-th forager individual at the t-th iteration, where i is the forager individual index, which is used to represent the forager individual that performs the global optimal search in the sparrow algorithm, t is the iteration number index, which is used to control the running termination condition of the improved optimization algorithm, and T is the maximum number of iterations. is an exponential decay factor, which is used to control the update amplitude of the update rule. is the guiding factor, which is used to control the speed at which individuals search for the global optimal solution. is the optimal sparrow position at the tth iteration, used to represent the optimal solution, is the first random perturbation factor, r1 is a random update control parameter with a value range of [0,1], P is the forager environment control parameter, is the second random perturbation factor, is the average sparrow position at the tth iteration;

[0105] The improved calculation formula of the follower parameter update rule is:

[0106] ;

[0107] In the formula, is the sparrow position of the jth follower individual at the t+1th iteration, is the sparrow position of the jth follower individual at the tth iteration, where j is the follower individual index, which is used to represent the follower individual performing the local optimal search in the sparrow algorithm. is a balancing factor used to control the scope of local search;

[0108] Step S313: Imitation learning improvement is used to optimize the diversity of the search process. Specifically, by introducing the imitation learning of the best individual and the reverse incentive of the worst individual, the imitation learning improvement of the standard warning mechanism is performed. The calculation formula is:

[0109] ;

[0110] In the formula, is the sparrow position of the kth alert individual at the t+1th iteration, is the sparrow position of the kth alert individual at the tth iteration, where k is the alert individual index, which is used to represent the alert individual for optimizing the local optimal solution in the sparrow algorithm. is a random optimal individual weight in the range [0,1], is the random worst individual weight in the range [0,1], is the worst sparrow position at the tth iteration;

[0111] Step S314: Dynamically adjust the fitness calculation and improve it, specifically by introducing a gradient penalty term and dynamically adjusting the weight to improve the multi-objective fitness function. The calculation formula is:

[0112] ;

[0113] Where f(·) is the dynamically adjusted multi-objective improved fitness function, It's the sparrow position. is the sparrow position index, w is the balance factor, f org (·) is the original fitness function, g(·) is the gradient penalty term;

[0114] Step S315: improving the optimization algorithm construction, specifically, improving the standard sparrow search algorithm to adapt to the needs of equipment fault detection through the population initialization improvement, the adaptive update rule improvement, the imitation learning improvement and the dynamic adjustment fitness calculation improvement, and optimizing the parameter target by setting the operation termination condition of the improved optimization algorithm;

[0115] The parameter target optimization specifically includes feature selection optimization of equipment fault detection and model parameter optimization of equipment fault detection;

[0116] Step S32: constructing a convolutional long short-term memory network, specifically constructing a standard convolutional long short-term memory neural network including a convolutional layer, a pooling layer and a long short-term memory layer, for performing temporal and spatial feature extraction optimization for equipment fault detection, and obtaining the convolutional long short-term memory network;

[0117] Step S33: parallel optimization design, specifically, by executing the operation of step S31, using the improved optimization algorithm to respectively perform feature selection optimization in the equipment fault maintenance optimization data and model parameter optimization of the convolutional long short-term memory network, and optimizing the performance of equipment fault detection in parallel;

[0118] Step S34: Equipment fault detection model training, specifically, training the equipment fault detection model through the improved optimization algorithm, the convolutional long short-term memory network and the parallel optimization design to obtain the equipment fault detection model Model SC ;

[0119] Step S35: Device fault detection, specifically using the device fault detection model Model SC , performing equipment fault detection according to the equipment fault repair optimization data to obtain equipment fault detection data;

[0120] The equipment fault detection data specifically includes numerical detection data and categorical detection data;

[0121] The numerical detection data specifically includes vibration fault detection data, temperature fault detection data, current fault detection data, voltage fault detection data, shaft speed fault detection data and lubrication state fault detection data;

[0122] The categorized detection data specifically includes normal equipment operation category and equipment operation failure category;

[0123] The equipment operation fault category specifically includes a fault detail category and a fault level category;

[0124] The detailed fault categories include mechanical faults, electrical faults and thermal faults;

[0125] The fault level categories specifically include minor faults, moderate faults and severe faults.

[0126] By performing the above operations, in view of the technical problem that in the existing equipment fault detection methods, there is an existing intelligent method for equipment fault detection, because equipment fault detection involves both complex numerical data features and requires clear classification categories, levels and detailed identification and analysis, and therefore traditional artificial intelligence methods are difficult to meet the needs of complex output conditions, this solution creatively adopts a convolutional long short-term memory network combined with an improved optimization algorithm to perform equipment fault detection, and by designing an improved sparrow search algorithm, the ability to solve parameter combinations for complex input data and detailed output data is optimized from the entire search process, and is applied to both feature optimization and model parameter levels, thereby improving the output dimension of equipment fault detection and the availability of results.

[0127] Example 5, see Figure 1 , Figure 2 and Figure 6 This embodiment is based on the above embodiment. In step S4, the maintenance plan optimization is used to automatically generate the optimal maintenance plan according to the result of fault detection. Specifically, based on the equipment fault detection data and the equipment fault maintenance optimization data, a weighted method of scheme optimization combining an improved optimization algorithm and reinforcement learning is used to optimize the maintenance plan to obtain maintenance plan optimization reference data. Specifically, the following steps are included:

[0128] Step S41: constructing an improved optimization algorithm, specifically executing the operation of step S31 to construct an improved optimization algorithm for optimizing the maintenance plan;

[0129] Step S42: reinforcement learning optimization of the maintenance plan, specifically, by constructing a maintenance plan parameter environment and a maintenance plan intelligent agent, and adopting a reinforcement learning method combined with an improved optimization algorithm to perform reinforcement learning optimization of the maintenance plan, and obtain reinforcement learning optimized maintenance plan reference data;

[0130] The maintenance plan parameter environment specifically includes equipment status parameters, maintenance action parameters and result reward function;

[0131] The equipment status parameters specifically include equipment current fault status parameters, equipment remaining operation time parameters, equipment current maintenance progress parameters and equipment available resource status parameters;

[0132] The maintenance action parameters specifically include resource allocation action, maintenance priority setting action and maintenance strategy adjustment action;

[0133] The result reward function is specifically composed of maintenance time, maintenance cost, fault recovery degree and resource consumption;

[0134] The maintenance plan agent is used to select maintenance action parameters according to the current equipment status parameters, and adjust and optimize the plan according to the result reward function;

[0135] The maintenance plan is optimized by reinforcement learning, and the calculation formula is:

[0136] ;

[0137] In the formula, is the output parameter combination of the maintenance plan reinforcement learning of the t+1th iteration, which is used to represent the reference data of the maintenance plan optimized by reinforcement learning. t is the iteration number index. is the learning rate parameter, iSA(·) is the improved optimization algorithm function, S opt It is the optimization search space, which specifically includes the parameter space of the Q-value function and the Q-value update hyperparameter space;

[0138] Step S43: optimizing the maintenance scheme optimization algorithm, specifically, searching and solving the maintenance scheme using the improved optimization algorithm according to the equipment fault detection data and the equipment fault maintenance optimization data, and obtaining the maintenance scheme reference data of the optimization algorithm;

[0139] Step S44: weighting the output of the solution, specifically, performing weighted average summation of the optimized maintenance time, optimized maintenance cost and optimized resource allocation solution in the reinforcement learning optimized maintenance solution reference data and the optimized algorithm maintenance solution reference data to obtain weighted output maintenance solution data;

[0140] Step S45: optimizing the maintenance plan, specifically optimizing the maintenance plan by constructing the improved optimization algorithm, optimizing the maintenance plan through reinforcement learning, optimizing the maintenance plan optimization algorithm, and weighting the plan output, to obtain maintenance plan optimization reference data;

[0141] The maintenance plan optimization reference data specifically includes optimizing maintenance time, optimizing maintenance costs, optimizing resource allocation plans, and optimizing fault recovery levels.

[0142] By performing the above operations, it is found that in the existing maintenance plan optimization methods, the existing maintenance plans often rely on manual experience to make judgments. However, this method is difficult to provide more macro maintenance support when combined with augmented reality technology. Since the maintenance of equipment usually involves multiple problems of multiple equipment, an optimal plan that takes into account both overall scheduling and micro-actual maintenance should be used as the main technical support for augmented reality technology maintenance guidance. However, both the existing intelligent methods and manual methods are difficult to achieve the technical problem of combining overall scheduling with specific maintenance. This solution creatively adopts a weighted method for scheme optimization that combines improved optimization algorithms and reinforcement learning to optimize maintenance plans. By combining the improved search algorithm in the fault detection stage, it provides global decision support for maintenance plans. At the same time, combined with the scheme optimization performance of reinforcement learning technology, it provides suggestions for specific maintenance plans. Through the weighted combination of the two, the scheduling capabilities of the two levels of intelligent maintenance plans are effectively optimized, and a practical attempt is also provided for the effective combination with augmented reality technology.

[0143] Example 6, see Figure 1 and Figure 2 This embodiment is based on the above embodiment. In step S5, the maintenance enhancement management is used to enhance the equipment maintenance management by combining the equipment fault detection results and the maintenance optimization plan through augmented reality technology and intelligent feedback system. Specifically, the augmented reality device is used to provide equipment maintenance reference for equipment maintenance personnel, and intelligent maintenance management is performed by combining the equipment fault detection data and the maintenance plan optimization reference data to obtain equipment maintenance management enhancement data, and equipment maintenance enhancement management is performed according to the equipment maintenance management enhancement data;

[0144] The equipment maintenance enhanced management specifically includes equipment maintenance operation guidance, equipment fault location marking and equipment maintenance step demonstration;

[0145] The use of augmented reality equipment provides equipment maintenance personnel with equipment maintenance references, specifically by superimposing equipment fault information and optimal maintenance solutions based on the equipment fault detection data and the maintenance solution optimization reference data in the control interface of the augmented reality equipment, and providing equipment maintenance references by introducing three-dimensional animation demonstration tools and maintenance tool prompts, and recording, analyzing and maintaining maintenance data through an intelligent data management system.

[0146] Embodiment 7, see Figure 1 and Figure 2 , this embodiment is based on the above embodiment, and the present invention provides an equipment fault repair management system based on augmented reality technology, including a data management module, an equipment fault repair module, an augmented reality guidance module and a hardware management module;

[0147] The data management module is used for data collection and data preprocessing, and obtains equipment fault repair optimization data through data collection and data preprocessing, and sends the equipment fault repair optimization data to the equipment fault repair module;

[0148] The equipment fault repair module is used for equipment fault detection and repair scheme optimization. Through equipment fault detection and repair scheme optimization, equipment fault detection data and repair scheme optimization reference data are obtained, and the equipment fault detection data and repair scheme optimization reference data are sent to the augmented reality guidance module;

[0149] The augmented reality guidance module is used for maintenance enhancement management, through which equipment maintenance management enhancement data is obtained, and the equipment maintenance management enhancement data is sent to the hardware management module;

[0150] The hardware management module is used for hardware management. Through hardware management, augmented reality fault marking and maintenance operation guidance are performed in the maintenance scene, and fault maintenance confirmation is confirmed through human-computer interaction to obtain fault maintenance management data records. Based on the fault maintenance management data records, management and replacement of augmented reality hardware equipment and maintenance equipment are performed.

[0151] It should be noted that, in this article, relational terms such as first and second, etc. are only used to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Moreover, the terms "include", "comprise" or any other variants thereof are intended to cover non-exclusive inclusion, so that a process, method, article or device including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, article or device.

[0152] While the embodiments of the present invention have been shown and described, it will be appreciated by those skilled in the art that many changes, modifications, substitutions and variations can be made to the embodiments without departing from the principles and spirit of the invention.

[0153] The present invention and its embodiments are described above, and such description is not restrictive. The drawings show only one embodiment of the present invention, and the actual structure is not limited thereto. In short, if ordinary technicians in the field are inspired by it, without departing from the purpose of the invention, they can design a structure and embodiment similar to the technical solution without creativity, which should belong to the protection scope of the present invention.

Claims

1. A method for equipment fault inspection and maintenance management based on augmented reality technology, characterized in that: The method comprises the following steps: Step S1: Data collection, obtaining the original data of equipment fault inspection and maintenance management and optimizing the real-time data collection through industrial Internet of Things technology; Step S2: Data preprocessing to obtain equipment fault maintenance optimization data; Step S3: equipment fault detection, used to intelligently analyze whether the equipment has a fault or a potential fault. Specifically, based on the equipment fault maintenance optimization data, a convolutional long short-term memory network combined with an improved optimization algorithm is used to perform equipment fault detection to obtain equipment fault detection data. Specifically, the following steps are included: step S31: constructing an improved optimization algorithm; step S32: constructing a convolutional long short-term memory network; step S33: parallel optimization design; step S34: equipment fault detection model training; step S35: equipment fault detection; The construction of the improved optimization algorithm specifically includes the following steps: step S311: population initialization improvement; step S312: adaptive update rule improvement; step S313: imitation learning improvement; step S314: dynamic adjustment fitness calculation improvement; step S315: improved optimization algorithm construction; Step S4: Maintenance scheme optimization, which is used to automatically generate the best maintenance scheme according to the result of fault detection. Specifically, based on the equipment fault detection data and the equipment fault maintenance optimization data, a scheme optimization weighting method combining an improved optimization algorithm and reinforcement learning is used to optimize the maintenance scheme to obtain maintenance scheme optimization reference data, which specifically includes the following steps: Step S41: constructing an improved optimization algorithm; Step S42: maintenance scheme reinforcement learning optimization; Step S43: maintenance scheme optimization algorithm optimization; Step S44: scheme output weighting; Step S45: maintenance scheme optimization; Step S5: Enhanced maintenance management: perform intelligent maintenance management, obtain equipment maintenance management enhanced data, and perform enhanced maintenance management of equipment based on the equipment maintenance management enhanced data.

2. The method for equipment fault inspection and maintenance management based on augmented reality technology according to claim 1, characterized in that: In step S1, the data collection is used to collect the original data set required for equipment fault troubleshooting and maintenance management, specifically, by collecting data from industrial machinery equipment, obtaining the original data for equipment fault troubleshooting and maintenance management, and optimizing the real-time data collection through industrial Internet of Things technology; The equipment fault inspection and maintenance management original data specifically includes mechanical vibration data, temperature data, electrical data, shaft speed data and lubrication status data.

3. The method for equipment fault inspection and maintenance management based on augmented reality technology according to claim 2, characterized in that: In step S2, the data preprocessing is used to enhance and optimize the original data, specifically, based on the original data of the equipment fault maintenance management, the data content and characteristics are optimized, and the data is stored to obtain the equipment fault maintenance optimization data.

4. The method for equipment fault inspection and maintenance management based on augmented reality technology according to claim 3 is characterized in that: In step S3, the equipment fault detection is used to intelligently analyze whether the equipment has a fault or a potential fault. Specifically, based on the equipment fault maintenance optimization data, a convolutional long short-term memory network combined with an improved optimization algorithm is used to perform equipment fault detection to obtain equipment fault detection data, which specifically includes the following steps: In step S3, the equipment fault detection is used to intelligently analyze whether the equipment has a fault or a potential fault. Specifically, based on the equipment fault repair optimization data, a convolutional long short-term memory network combined with an improved optimization algorithm is used to perform equipment fault detection to obtain equipment fault detection data; Step S32: constructing a convolutional long short-term memory network, specifically constructing a standard convolutional long short-term memory neural network including a convolutional layer, a pooling layer and a long short-term memory layer, for performing temporal and spatial feature extraction optimization for equipment fault detection, and obtaining the convolutional long short-term memory network; Step S33: parallel optimization design, specifically, by executing the operation of step S31, using the improved optimization algorithm to respectively perform feature selection optimization in the equipment fault maintenance optimization data and model parameter optimization of the convolutional long short-term memory network, and optimizing the performance of equipment fault detection in parallel; Step S34: Equipment fault detection model training, specifically, training the equipment fault detection model through the improved optimization algorithm, the convolutional long short-term memory network and the parallel optimization design to obtain the equipment fault detection model Model SC ; Step S35: Device fault detection, specifically using the device fault detection model Model SC , performing equipment fault detection according to the equipment fault repair optimization data to obtain equipment fault detection data; The equipment fault detection data specifically includes numerical detection data and categorical detection data; The numerical detection data specifically includes vibration fault detection data, temperature fault detection data, current fault detection data, voltage fault detection data, shaft speed fault detection data and lubrication state fault detection data; The categorized detection data specifically includes normal equipment operation category and equipment operation failure category; The equipment operation fault category specifically includes a fault detail category and a fault level category; The detailed fault categories include mechanical faults, electrical faults and thermal faults; The fault level categories specifically include minor faults, moderate faults and severe faults.

5. The method for equipment fault inspection and maintenance management based on augmented reality technology according to claim 4, characterized in that: In step S31, the construction of the improved optimization algorithm specifically includes the following steps: Step S311: Improvement of population initialization, which is used to adapt to the multi-feature and multi-parameter requirements of equipment fault detection and expand the search range of the optimization algorithm. Specifically, the population initialization stage is improved by combining Latin hypercube sampling and distribution adaptive adjustment method. Specifically, the initial population of the standard sparrow algorithm is generated by Latin hypercube sampling, and the calculation of the initial population position is improved by adding a dynamic adjustment factor. The calculation formula is: ； In the formula, is the population sampling sample value corresponding to the total sample size of N and the sampling interval of m, which is used to represent the specific initial value to be optimized. In step S3, it is specifically used to represent the initial value to be optimized of the original features in the equipment fault maintenance optimization data and the model parameters of the equipment fault detection model. K is the total number of samples. is a random weight with a value range of [0,1], m is the sampling interval of the input data, which is used to indicate the frequency of data collection; M Nm It is the initial value of the sampling data point corresponding to the total sample size of N and the sampling interval of m, which is used to represent the specific initial value of the sparrow search algorithm population. is the inverse function of the cumulative probability distribution function, is the population expansion factor, the initial value is set to 0.1, rand(0,1) is a fine-tuned random number randomly sampled from a normal distribution; Step S312: Adaptive update rule improvement, used to introduce global optimal information enhancement features and model parameter optimization capabilities, specifically, adaptive update improvement of forager parameter update rules and follower parameter update rules in the sparrow search algorithm, and adaptive update based on the improved forager parameter update rules and follower parameter update rules; The improved calculation formula of the forager parameter update rule is: ； In the formula, is the sparrow position of the i-th forager individual at the t+1-th iteration, and the sparrow position is used to represent the search result of the sparrow algorithm. is the sparrow position of the i-th forager individual at the t-th iteration, where i is the forager individual index, which is used to represent the forager individual that performs the global optimal search in the sparrow algorithm, t is the iteration number index, which is used to control the running termination condition of the improved optimization algorithm, and T is the maximum number of iterations. is an exponential decay factor, which is used to control the update amplitude of the update rule. is the guiding factor, which is used to control the speed at which individuals search for the global optimal solution. is the optimal sparrow position at the tth iteration, used to represent the optimal solution, is the first random perturbation factor, r1 is a random update control parameter with a value range of [0,1], P is the forager environment control parameter, is the second random perturbation factor, is the average sparrow position at the tth iteration; The improved calculation formula of the follower parameter update rule is: ； In the formula, is the sparrow position of the jth follower individual at the t+1th iteration, is the sparrow position of the jth follower individual at the tth iteration, where j is the follower individual index, which is used to represent the follower individual performing the local optimal search in the sparrow algorithm. is a balancing factor used to control the scope of local search; Step S313: Imitation learning improvement is used to optimize the diversity of the search process. Specifically, by introducing the imitation learning of the best individual and the reverse incentive of the worst individual, the imitation learning improvement of the standard warning mechanism is performed. The calculation formula is: ； In the formula, is the sparrow position of the kth alert individual at the t+1th iteration, is the sparrow position of the kth alert individual at the tth iteration, where k is the alert individual index, which is used to represent the alert individual for optimizing the local optimal solution in the sparrow algorithm. is a random optimal individual weight in the range [0,1], is the random worst individual weight in the range [0,1], is the worst sparrow position at the tth iteration; Step S314: Dynamically adjust the fitness calculation and improve it, specifically by introducing a gradient penalty term and dynamically adjusting the weight to improve the multi-objective fitness function. The calculation formula is: ； Where f(·) is the dynamically adjusted multi-objective improved fitness function, It's the sparrow position. is the sparrow position index, w is the balance factor, f org (·) is the original fitness function, g(·) is the gradient penalty term; Step S315: improving the optimization algorithm construction, specifically, improving the standard sparrow search algorithm to adapt to the needs of equipment fault detection through the population initialization improvement, the adaptive update rule improvement, the imitation learning improvement and the dynamic adjustment fitness calculation improvement, and optimizing the parameter target by setting the operation termination condition of the improved optimization algorithm; The parameter target optimization specifically includes feature selection optimization of equipment fault detection and model parameter optimization of equipment fault detection.

6. The method for equipment fault inspection and maintenance management based on augmented reality technology according to claim 5, characterized in that: In step S4, the maintenance scheme optimization is used to automatically generate the optimal maintenance scheme according to the result of fault detection. Specifically, based on the equipment fault detection data and the equipment fault maintenance optimization data, a scheme optimization weighting method combining an improved optimization algorithm and reinforcement learning is used to optimize the maintenance scheme to obtain maintenance scheme optimization reference data, which specifically includes the following steps: Step S41: constructing an improved optimization algorithm, specifically executing the operation of step S31 to construct an improved optimization algorithm for optimizing the maintenance plan; Step S42: reinforcement learning optimization of the maintenance plan, specifically, by constructing a maintenance plan parameter environment and a maintenance plan intelligent agent, and adopting a reinforcement learning method combined with an improved optimization algorithm to perform reinforcement learning optimization of the maintenance plan, and obtain reinforcement learning optimized maintenance plan reference data; Step S43: optimizing the maintenance scheme optimization algorithm, specifically, searching and solving the maintenance scheme using the improved optimization algorithm according to the equipment fault detection data and the equipment fault maintenance optimization data, and obtaining the maintenance scheme reference data of the optimization algorithm; Step S44: weighting the output of the solution, specifically, performing weighted average summation of the optimized maintenance time, optimized maintenance cost and optimized resource allocation solution in the reinforcement learning optimized maintenance solution reference data and the optimized algorithm maintenance solution reference data to obtain weighted output maintenance solution data; Step S45: optimizing the maintenance plan, specifically optimizing the maintenance plan by constructing the improved optimization algorithm, optimizing the maintenance plan through reinforcement learning, optimizing the maintenance plan optimization algorithm, and weighting the plan output, to obtain maintenance plan optimization reference data; The maintenance plan optimization reference data specifically includes optimizing maintenance time, optimizing maintenance costs, optimizing resource allocation plans, and optimizing fault recovery levels.

7. The method for equipment fault inspection and maintenance management based on augmented reality technology according to claim 6, characterized in that: In step S5, the maintenance enhancement management is used to enhance the equipment maintenance management by combining the equipment fault detection results and the maintenance optimization plan through augmented reality technology and intelligent feedback system, specifically, by using augmented reality equipment, providing equipment maintenance reference for equipment maintenance personnel, and by combining the equipment fault detection data and the maintenance plan optimization reference data, intelligent maintenance management is performed to obtain equipment maintenance management enhancement data, and equipment maintenance enhancement management is performed according to the equipment maintenance management enhancement data; The equipment maintenance enhanced management specifically includes equipment maintenance operation guidance, equipment fault location marking and equipment maintenance step demonstration; The use of augmented reality equipment provides equipment maintenance personnel with equipment maintenance references, specifically by superimposing equipment fault information and optimal maintenance solutions based on the equipment fault detection data and the maintenance solution optimization reference data in the control interface of the augmented reality equipment, and providing equipment maintenance references by introducing three-dimensional animation demonstration tools and maintenance tool prompts, and recording, analyzing and maintaining maintenance data through an intelligent data management system.

8. An equipment fault maintenance management system based on augmented reality technology, used to implement an equipment fault maintenance management method based on augmented reality technology as claimed in any one of claims 1 to 7, characterized in that: It includes data management module, equipment fault troubleshooting module, augmented reality guidance module and hardware management module.

9. The equipment fault inspection and maintenance management system based on augmented reality technology according to claim 8, characterized in that: The data management module is used for data collection and data preprocessing, and obtains equipment fault repair optimization data through data collection and data preprocessing, and sends the equipment fault repair optimization data to the equipment fault repair module; The equipment fault repair module is used for equipment fault detection and repair scheme optimization. Through equipment fault detection and repair scheme optimization, equipment fault detection data and repair scheme optimization reference data are obtained, and the equipment fault detection data and repair scheme optimization reference data are sent to the augmented reality guidance module; The augmented reality guidance module is used for maintenance enhancement management, through which equipment maintenance management enhancement data is obtained, and the equipment maintenance management enhancement data is sent to the hardware management module; The hardware management module is used for hardware management. Through hardware management, augmented reality fault marking and maintenance operation guidance are performed in the maintenance scene, and fault maintenance confirmation is confirmed through human-computer interaction to obtain fault maintenance management data records. Based on the fault maintenance management data records, management and replacement of augmented reality hardware equipment and maintenance equipment are performed.

Images