A folding conveyor belt anti-infringement control method and system for a whole vehicle used on railways

Abstract

The present invention discloses a folding conveyor belt anti-invasion limit control method and system for a whole vehicle used on railways, which relates to the technical field of fault monitoring and control. It includes installing vibration sensors and tension sensors on the folding conveyor belt to collect conveyor belt vibration signals and tension signals, determining the direction based on the folding conveyor belt, analyzing the vibration characteristics of the conveyor belt to construct a conveyor belt stiffness matrix, synchronously extracting vibration signal features through fast kurtosis calculation, constructing a DBN model and optimizing it using the sparrow search algorithm combined with the firefly perturbation formula, and obtaining the conveyor belt fault monitoring results through the DBN model; based on the conveyor belt fault monitoring results, an adversarial generative network is used to generate a fault handling plan and the convolutional neural network is used to evaluate it, and then the fault handling plan is implemented. The present invention greatly improves the abnormal detection ability and detection accuracy of the conveyor belt, and realizes the automatic generation and real-time feedback of the fault plan.

Description

Technical Field

[0001] The present invention relates to the technical field of fault monitoring and control, and particularly to an anti-infringement limit control method and system for a folding conveyor belt used in a whole vehicle railway. Background Art

[0002] With the rapid development of the railway transportation industry, the automation and intelligence levels of railway equipment have been gradually improved. Especially in the whole vehicle railway transportation, the demands for cargo transportation and equipment maintenance are increasing continuously. As an important equipment for transferring goods in railway transportation, the folding conveyor belt has the advantages of compact structure, convenient installation and operation, and is widely used in the railway transportation system. However, with the increase in the usage frequency, various fault problems inevitably occur during the actual operation of the folding conveyor belt. These faults not only affect the transportation efficiency but also may cause potential safety hazards. Traditional methods usually adopt time-domain or frequency-domain analysis of vibration signals, and judge the fault types based on the characteristics such as the amplitude change and frequency components of the signals. However, these methods often ignore the high-order characteristics of the vibration signals, resulting in limitations in the accuracy and timeliness of fault diagnosis, making it difficult to accurately extract the deep-level information related to the faults, and the comprehensive analysis ability and fault handling ability are still insufficient. Summary of the Invention

[0003] In view of the above existing problems, the present invention is proposed.

[0004] Therefore, the present invention provides an anti-infringement limit control method and system for a folding conveyor belt used in a whole vehicle railway, which solves the problems that the prior art ignores the high-order characteristics of vibration signals, resulting in limitations in the accuracy and timeliness of fault diagnosis, making it difficult to accurately extract the deep-level information related to the faults, and the comprehensive analysis ability and fault handling ability are still insufficient.

[0005] To solve the above technical problems, the present invention provides the following technical solutions:

[0006] In the first aspect, the present invention provides an anti-infringement limit control method for a folding conveyor belt used in a whole vehicle railway, which includes

[0007] installing vibration sensors and tension sensors on the folding conveyor belt to collect the conveyor belt vibration signals and tension signals, determining the direction based on the folding conveyor belt, analyzing the vibration characteristics of the conveyor belt to construct a conveyor belt stiffness matrix, synchronously extracting the vibration signal features through fast kurtosis calculation, constructing a DBN model and optimizing it using the sparrow search algorithm combined with the firefly perturbation formula, and obtaining the conveyor belt fault monitoring result through the DBN model;

[0008] generating a fault handling plan based on the conveyor belt fault monitoring result using a generative adversarial network and implementing the fault handling plan after evaluating it through a convolutional neural network;

[0009] Visualize the fault handling solution and store the fault handling record in the database.

[0010] As a preferred solution of the folding conveyor belt anti-invasion control method for the whole vehicle railway described in the present invention, wherein: determining the direction based on the folding conveyor belt, analyzing the vibration characteristics of the conveyor belt to construct the conveyor belt stiffness matrix, synchronously extracting the vibration signal characteristics through fast kurtosis calculation, constructing a DBN model and optimizing it using the sparrow search algorithm combined with the firefly perturbation formula, and obtaining the conveyor belt fault monitoring result through the DBN model means defining the material transportation direction of the folding conveyor belt as the longitudinal direction, and defining the plane perpendicular direction of the material transportation direction as the transverse direction, obtaining the material parameters of the folding conveyor belt, respectively constructing the longitudinal vibration equation and the transverse vibration equation of the folding conveyor belt, and respectively obtaining the longitudinal displacement response of the folding conveyor belt by solving through the finite difference method and the transverse displacement response ;

[0011] According to the longitudinal displacement response of the folding conveyor belt and the transverse displacement response Construct the stiffness matrix K of the folding conveyor belt;

[0012] Use the short-time Fourier transform to extract the vibration signal in the complex frequency domain representation , calculate the amplitude envelope of the complex frequency domain representation ;

[0013] Through the amplitude envelope Calculate the spectral kurtosis of the vibration signal and combine the spectral kurtosis of the vibration signals at all frequencies b as the vibration signal characteristics :

[0014] Use principal component analysis to reduce the dimension of the vibration signal characteristics and construct a DBN model, and update the DBN model parameters through the sparrow search algorithm:

[0015]

[0016] where is the j-th parameter of the i-th sparrow at time t, including each DBN model parameter, is the sparrow position at time t + 1, is a random number, ST is a set threshold, when , the sparrow search algorithm performs local search, and when it performs global search, is an adjustment factor, h is the maximum number of iterations, Q is the migration factor, and W is the neighborhood range;

[0017] Synchronously calculate the sparrow position at time t+1 using the firefly perturbation formula:

[0018]

[0019] Where is the initial attractiveness, is the attenuation factor, r is the distance between the current position and the target solution, is the perturbation amplitude factor, is the random perturbation term;

[0020] Comprehensively average the sparrow position at time t+1 calculated by the firefly perturbation formula and the sparrow position at time t+1 obtained by the sparrow search algorithm as the final sparrow position and use it as the next iteration position to repeat the iteration;

[0021] After the iteration is completed, extract the optimal DBN model parameters from the sparrow positions and apply them to the DBN model. Input the stiffness matrix K and the vibration signal characteristics into the DBN model to obtain the conveyor belt fault monitoring result.

[0022] As a preferred solution of the folding conveyor belt anti-invasion control method for the whole vehicle railway described in the present invention, wherein: the step of generating a fault handling solution using a generative adversarial network based on the conveyor belt fault monitoring result and evaluating it through a convolutional neural network and then implementing the fault handling solution means constructing a generative adversarial network and training it. Input the conveyor belt fault monitoring result into the generative adversarial network, and the generator generates a fault handling solution. Synchronously construct a convolutional neural network for training, input the fault handling solution into the trained convolutional neural network for evaluation and set an evaluation threshold. When the evaluation result is greater than the evaluation threshold, judge the fault handling solution as the implementation solution and implement it.

[0023] As a preferred solution of the folding conveyor belt anti-invasion control method for the whole vehicle railway described in the present invention, wherein: the step of installing vibration sensors and tension sensors on the folding conveyor belt to collect conveyor belt vibration signals and tension signals means installing vibration sensors and tension sensors under the folding conveyor belt, connecting the vibration sensors and tension sensors through a wireless network to form a sensor network, and respectively collecting vibration signals and tension signals through the vibration sensors and tension sensors and then performing preprocessing operations.

[0024] As a preferred solution of the anti-infringement control method for the folding conveyor belt of the whole vehicle for railway use according to the present invention, the following is provided: A horizontal conveyor belt is arranged in front of the folding conveyor belt. The folding conveyor belt and the horizontal conveyor belt are connected through a start-stop signal sensor. Materials are transported to the folding conveyor belt through the horizontal conveyor belt. When the start signal of the folding conveyor belt is received, the horizontal conveyor belt starts and transports the materials. When it is detected that the folding conveyor belt fails and needs to be shut down, the horizontal conveyor belt needs to be shut down to send a shutdown signal to the folding conveyor belt and then the folding conveyor belt is shut down.

[0025] As a preferred solution of the anti-infringement control method for the folding conveyor belt of the whole vehicle for railway use according to the present invention, the following is provided: The visual display of the fault handling solution refers to the display of the conveyor belt fault monitoring results and the finally obtained fault handling solution to the staff, and the synchronous display of vibration data and tension data to assist the staff in maintaining the folding conveyor belt. The formation of the fault handling record and storage in the database refers to the formation of the implemented fault handling solution and the corresponding fault monitoring results into a fault handling record and storage in the database. The database stores data with time stamps, and the stored data is uploaded to the cloud for backup.

[0026] As a preferred solution of the anti-infringement control method for the folding conveyor belt of the whole vehicle for railway use according to the present invention, the following is provided:

[0027] Second, the present invention provides an anti-infringement control system for the folding conveyor belt of the whole vehicle for railway use, including

[0028] A signal collection module, which is used to install vibration sensors and tension sensors on the folding conveyor belt to collect conveyor belt vibration signals and tension signals for preprocessing;

[0029] A fault monitoring module, which is used to extract the conveyor belt stiffness matrix and vibration signal characteristics, and construct a DBN model optimized by combining the sparrow search algorithm with the firefly perturbation formula, and obtain the conveyor belt fault monitoring results through the DBN model;

[0030] A fault handling module, which is used to generate a fault handling solution by using a generative adversarial network and implement the fault handling solution after evaluation by a convolutional neural network;

[0031] A display and storage module, which is used to visually display the fault handling solution and form a fault handling record and store it in the database.

[0032] Third, the present invention provides a computer device, including a memory and a processor. The memory stores a computer program, and the following is provided: When the computer program is executed by the processor, any step of the anti-infringement control method for the folding conveyor belt of the whole vehicle for railway use as described in the first aspect of the present invention is implemented.

[0033] Fourthly, the present invention provides a computer-readable storage medium, on which a computer program is stored, wherein: when the computer program is executed by a processor, any step of the anti-invasion limit control method for the folding conveyor belt for the whole vehicle railway as described in the first aspect of the present invention is implemented.

[0034] The beneficial effects of the present invention are as follows: By extracting the stiffness matrix and vibration signal characteristics of the conveyor belt, constructing a DBN model and optimizing it by combining the sparrow search algorithm with the firefly perturbation formula, and finally obtaining the conveyor belt fault monitoring result through the DBN model, the abnormal detection ability and detection accuracy of the conveyor belt are greatly improved, effectively overcoming the deficiencies of traditional methods in complex vibration environments. Moreover, by combining the adversarial generation network with the convolutional neural network to generate a fault handling scheme, the automatic generation and real-time feedback of the fault scheme are realized, improving the automation degree and intelligent level of the conveyor belt. Description of the Drawings

[0035] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings required for the description of the embodiments will be briefly introduced below. Obviously, the drawings in the following description are only some embodiments of the present invention, and those of ordinary skill in the art can obtain other drawings without creative efforts based on these drawings.

[0036] Figure 1 It is a flowchart of the anti-invasion limit control method for the folding conveyor belt for the whole vehicle railway in Embodiment 1;

[0037] Figure 2 It is a structural diagram of the anti-invasion limit control system for the folding conveyor belt for the whole vehicle railway in Embodiment 1;

[0038] Figure 3 It is a schematic diagram of the positions of the folding conveyor belt and the horizontal conveyor belt in Embodiment 1. Detailed Embodiments

[0039] To make the above objects, features and advantages of the present invention more obvious and understandable, the specific embodiments of the present invention will be described in detail below with reference to the accompanying drawings of the specification.

[0040] Many specific details are set forth in the following description in order to provide a thorough understanding of the present invention, but the present invention may be practiced in other ways different from those described herein. Those skilled in the art can make similar extensions without departing from the connotation of the present invention, so the present invention is not limited by the specific embodiments disclosed below.

[0041] Second, the "one embodiment" or "embodiment" referred to herein means a specific feature, structure, or characteristic that may be included in at least one implementation manner of the present invention. The "in one embodiment" that appears in different places in this specification does not all refer to the same embodiment, nor is it a separate or alternative embodiment that is mutually exclusive with other embodiments.

[0042] Embodiment 1, referring to Figures 1 to 3 , is the first embodiment of the present invention. This embodiment provides a folding conveyor belt anti-invasion limit control method for a whole vehicle railway, including the following steps:

[0043] S1. Install vibration sensors and tension sensors on the folding conveyor belt to collect conveyor belt vibration signals and tension signals. Determine the direction based on the folding conveyor belt, analyze the vibration characteristics of the conveyor belt to construct a conveyor belt stiffness matrix, synchronously extract vibration signal features through fast kurtosis calculation, construct a DBN model, and optimize it using the sparrow search algorithm combined with the firefly perturbation formula. Obtain the conveyor belt fault monitoring result through the DBN model;

[0044] Specifically, installing vibration sensors and tension sensors on the folding conveyor belt to collect conveyor belt vibration signals and tension signals means installing vibration sensors and tension sensors under the folding conveyor belt, and connecting the vibration sensors and tension sensors through a wireless network to form a sensor network. After collecting vibration signals and tension signals through the vibration sensors and tension sensors respectively, perform preprocessing operations.

[0045] By installing vibration sensors and tension sensors on the folding conveyor belt, the operating state of the conveyor belt can be monitored in real time, solving the deficiencies of the lack of real-time monitoring and status feedback in traditional conveyor belt monitoring methods. By accurately obtaining the vibration and tension signals of the conveyor belt, potential fault problems of the equipment, especially abnormal vibration or tension problems, can be discovered in time, and early warnings can be given to avoid equipment damage and production stagnation. The introduction of the wireless network makes the sensor system no longer limited by the wiring space and can be flexibly arranged in various working environments, especially suitable for applications in dynamic, narrow, or complex environments. The sensors transmit data to the central processing unit wirelessly, ensuring the real-time nature and transmission stability of the data.

[0046] Furthermore, determine the direction based on the folded conveyor belt, analyze the vibration characteristics of the conveyor belt to construct the conveyor belt stiffness matrix, simultaneously extract the vibration signal characteristics through fast spectral kurtosis calculation, construct a DBN model and optimize it using the sparrow search algorithm combined with the firefly perturbation formula, and obtain the conveyor belt fault monitoring result through the DBN model. Define the material transportation direction of the folded conveyor belt as the longitudinal direction, and define the plane perpendicular to the material transportation direction as the transverse direction. Obtain the material parameters of the folded conveyor belt, construct the longitudinal vibration equation and the transverse vibration equation of the folded conveyor belt respectively, and obtain the longitudinal displacement response of the folded conveyor belt by solving through the finite difference method and the transverse displacement response ;

[0047] The transverse vibration equation is as follows:

[0048]

[0049] where E is the Young's modulus of the folded conveyor belt, I is the second moment of the cross-section, is the material density, A is the cross-sectional area of the folded conveyor belt, is the external force at the position x of the folded conveyor belt at time t, extracted through the tension signal, and the position x of the folded conveyor belt is defined by transverse division;

[0050] The longitudinal vibration equation is as follows:

[0051]

[0052] where L is the length of the folded conveyor belt;

[0053] According to the longitudinal displacement response and the transverse displacement response of the folded conveyor belt, construct the stiffness matrix K of the folded conveyor belt;

[0054] Use the short-time Fourier transform to extract the complex frequency domain representation of the vibration signal , and calculate the amplitude envelope of the complex frequency domain representation:

[0055]

[0056] where and respectively represent the real part and the imaginary part of the complex frequency domain representation ;

[0057] Calculate the spectral kurtosis of the vibration signal through the amplitude envelope and combine the spectral kurtosis of the vibration signals at all frequencies b as the vibration signal characteristics :

[0058]

[0059] Use principal component analysis for vibration signal features to perform dimensionality reduction, construct a DBN model, and update the DBN model parameters through the sparrow search algorithm:

[0060]

[0061] where is the jth parameter of the ith sparrow at time t, including each DBN model parameter, is the sparrow position at time t + 1, is a random number between 0 and 1, ST is the set threshold, when , the sparrow search algorithm performs local search, when it performs global search, is the adjustment factor, h is the maximum number of iterations, Q is the migration factor, controlling the amplitude of the sparrow's position change, and W is the neighborhood range, representing the area where the sparrow jumps;

[0062] Synchronously use the firefly perturbation formula to calculate the sparrow position at time t + 1:

[0063]

[0064] where is the initial attractiveness, controlling the intensity of the firefly perturbation, is the decay factor, r is the distance between the current position and the target solution, is the perturbation amplitude factor, controlling the size of the perturbation, is the random perturbation term, usually a randomly generated decimal, representing noise or random perturbation, ensuring randomness and global exploration in the search process;

[0065] Perform a comprehensive average calculation on the sparrow position at time t + 1 obtained by the firefly perturbation formula and the sparrow position at time t + 1 obtained by the sparrow search algorithm as the final sparrow position and use it as the next iteration position to repeat the iteration;

[0066] After the iteration is completed, extract the optimal DBN model parameters from the sparrow positions and apply them to the DBN model, and input the stiffness matrix K and the vibration signal features into the DBN model to obtain the conveyor belt fault monitoring results.

[0067] By clearly defining the material transportation direction of the folding conveyor belt as longitudinal and the plane perpendicular direction of the material transportation direction as transverse, this clear division of directions provides a clear physical basis for subsequent vibration analysis. In vibration analysis, longitudinal and transverse vibrations are treated separately, effectively avoiding the errors that may be brought about by mixing the two for analysis. This vibration analysis method based on direction enhances the accuracy of the model in dynamic response. Especially when there may be significant differences in vibration characteristics in different directions, it provides higher diagnostic flexibility and accuracy. By establishing longitudinal and transverse vibration equations for the folding conveyor belt respectively and using the finite difference method for numerical solution, the longitudinal and transverse displacement responses of the system are obtained. The finite difference method can accurately simulate and solve the displacement response in a dynamic system. Especially when considering complex boundary conditions and external load effects, more accurate vibration response data can be obtained. By combining the longitudinal and transverse displacement responses, the calculated stiffness matrix can accurately reflect the mechanical behavior of the folding conveyor belt in actual operation. The stiffness matrix is the core of analyzing the dynamic response of the system, which reflects the deformation and reaction ability of the conveyor belt under external forces and constraint conditions. Through fast kurtosis calculation, the high-order non-stationary characteristics in the vibration signal can be effectively identified, which is very useful for capturing abnormal phenomena such as transients, shocks or imbalances caused by equipment failures. The feature extraction of kurtosis not only enhances the fault sensitivity of the signal, but also provides more accurate feature data input to help the subsequent fault detection model make better judgments. By constructing a deep belief network (DBN) and combining the sparrow search algorithm (SSA) with firefly perturbation optimization, the training process of the fault monitoring model is optimized. The sparrow search algorithm can efficiently adjust the structure and hyperparameters of the model, enabling the DBN to adapt to complex vibration characteristics and fault modes, thereby improving the robustness and accuracy of the fault diagnosis model. The firefly perturbation further enhances the global search ability, avoiding falling into local optimal solutions and ensuring the stability and globality of the optimal solution. The fault monitoring results obtained through the optimized DBN model can monitor and diagnose the health status of the folding conveyor belt in real time. This result not only provides a scientific basis for fault warning, but also provides accurate data support for subsequent fault handling.

[0068] Furthermore, a horizontal conveyor belt is provided in front of the folding conveyor belt. The folding conveyor belt and the horizontal conveyor belt are connected through a start-stop signal sensor. Materials are transported to the folding conveyor belt through the horizontal conveyor belt. When the start signal of the folding conveyor belt is received, the horizontal conveyor belt starts and transports materials. When it is detected that the folding conveyor belt fails and needs to be shut down, the horizontal conveyor belt needs to be shut down and a shutdown signal is transmitted to the folding conveyor belt before the folding conveyor belt is shut down.

[0069] The folding conveyor belt and the horizontal conveyor belt are linked through the start-stop signal sensor. This linkage mechanism can ensure the coordinated work of the two conveyor belts during operation, especially when the material is transported to the folding conveyor belt, the horizontal conveyor belt can be automatically started according to the start signal of the folding conveyor belt. In this way, the automation level of the system is improved, reducing the possibility of manual intervention and operational errors. By automatically linking the horizontal conveyor belt with the start and stop signals of the folding conveyor belt, the horizontal conveyor belt can automatically start when the folding conveyor belt starts, and automatically stop when the folding conveyor belt fails or stops. This automated control reduces manual intervention and improves the efficiency and safety of the system. When the folding conveyor belt detects a fault, the system will automatically send a shutdown signal to the horizontal conveyor belt to ensure that materials will not be accumulated or cause other problems due to the fault during transportation. This mechanism not only improves the timeliness of fault response, but also prevents further transmission of materials when equipment fails, thereby avoiding potential material damage and safety hazards.

[0070] S2, based on the conveyor belt fault monitoring results, a fault handling solution is generated by using a generative adversarial network and implemented after evaluation through a convolutional neural network;

[0071] Specifically, based on the conveyor belt fault monitoring results, a fault handling plan is generated by using an adversarial generative network and implemented after being evaluated by a convolutional neural network. This means constructing an adversarial generative network and training it, inputting the conveyor belt fault monitoring results into the adversarial generative network to generate a fault handling plan through a generator, and simultaneously constructing a convolutional neural network for training, inputting the fault handling plan into the trained convolutional neural network for evaluation and setting an evaluation threshold, and when the evaluation result is greater than the evaluation threshold, the fault handling plan is judged as an implementation plan and implemented.

[0072] GAN can generate personalized and targeted fault handling solutions based on the real-time monitoring data of the conveyor belt (such as vibration, tension signals, etc.). It not only takes into account the current fault status of the equipment, but also continuously optimizes the quality of the solution through the "generation-discrimination" mechanism to ensure that the generated solution is more adaptable to complex and changing fault situations. The CNN model can learn how to evaluate and verify the effectiveness of the fault handling solution through a large number of historical faults and handling cases during the training process. This approach not only improves the degree of automation of the evaluation process, but also ensures that the implementation of the fault handling solution has a high degree of reliability. CNN can identify the advantages and potential problems of the handling solution through its powerful pattern recognition ability. By setting the evaluation threshold, the implementation of the fault handling solution can be effectively controlled. The evaluation threshold ensures that the solution will only be executed when it meets certain standards, avoiding potential damage to the equipment caused by incorrect or inappropriate handling solutions.

[0073] S3. Visualize the fault handling solution and form a fault handling record to be stored in the database;

[0074] Specifically, visualizing the fault handling solution means presenting the fault monitoring results of the conveyor belt and the finally obtained fault handling solution to the staff, and synchronously presenting the vibration data and tension data to assist the staff in maintaining the folding conveyor belt.

[0075] Furthermore, forming a fault handling record and storing it in the database means forming a fault handling record from the implemented fault handling solution and the corresponding fault monitoring results and storing it in the database. The database stores data with timestamps marked and uploads the stored data to the cloud for backup.

[0076] This embodiment also provides a folding conveyor belt anti-invasion control system for a whole vehicle on railway, including:

[0077] A signal collection module, used to install vibration sensors and tension sensors on the folding conveyor belt to collect conveyor belt vibration signals and tension signals for preprocessing;

[0078] A fault monitoring module, used to extract the conveyor belt stiffness matrix and vibration signal features, construct a DBN model and optimize it using the sparrow search algorithm combined with the firefly perturbation formula, and obtain the conveyor belt fault monitoring results through the DBN model;

[0079] A fault handling module, used to generate a fault handling solution using a generative adversarial network and implement the fault handling solution after evaluation through a convolutional neural network;

[0080] A display and storage module, used to visualize the fault handling solution and form a fault handling record to be stored in the database.

[0081] This embodiment also provides a computer device applicable to the folding conveyor belt anti-invasion control method for a whole vehicle on railway, including: a memory and a processor; the memory is used to store computer-executable instructions, and the processor is used to execute the computer-executable instructions to implement the folding conveyor belt anti-invasion control method for a whole vehicle on railway as proposed in the above embodiment.

[0082] The computer device can be a terminal, which includes a processor, a memory, a communication interface, a display screen, and an input device connected via a system bus. Among them, the processor of the computer device is used to provide computing and control capabilities. The memory of the computer device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system and computer programs. The internal memory provides an environment for the operation of the operating system and computer programs in the non-volatile storage medium. The communication interface of the computer device is used to communicate with external terminals in a wired or wireless manner. The wireless manner can be implemented through WIFI, a carrier network, NFC (Near Field Communication), or other technologies. The display screen of the computer device can be a liquid crystal display screen or an electronic ink display screen. The input device of the computer device can be a touch layer covering the display screen, or buttons, a trackball, or a touchpad provided on the housing of the computer device, or an external keyboard, touchpad, or mouse, etc.

[0083] This embodiment also provides a storage medium, on which a computer program is stored. When the program is executed by a processor, it implements the method for preventing intrusion and controlling the folding conveyor belt for railway vehicles as proposed in the above embodiment; the storage medium can be implemented by any type of volatile or non-volatile storage device or a combination thereof, such as Static Random Access Memory (SRAM for short), Electrically Erasable Programmable Read-Only Memory (EEPROM for short), Erasable Programmable Read-Only Memory (EPROM for short), Programmable Read-Only Memory (PROM for short), Read-Only Memory (ROM for short), magnetic memory, flash memory, a magnetic disk, or an optical disc.

[0084] In summary, the present invention extracts the stiffness matrix and vibration signal characteristics of the conveyor belt, constructs a DBN model, optimizes it by combining the sparrow search algorithm with the firefly perturbation formula, and finally obtains the conveyor belt fault monitoring result through the DBN model, greatly improving the abnormal detection ability and detection accuracy of the conveyor belt, effectively overcoming the deficiencies of traditional methods in a complex vibration environment, and realizing the automatic generation and real-time feedback of fault solutions by combining a generative adversarial network with a convolutional neural network to generate fault handling solutions, improving the automation degree and intelligent level of the conveyor belt.

[0085] It should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and not to limit them. Although the present invention has been described in detail with reference to the preferred embodiments, those of ordinary skill in the art should understand that the technical solutions of the present invention can be modified or equivalently replaced without departing from the spirit and scope of the technical solutions of the present invention, and they should all be covered by the scope of the claims of the present invention.

Claims

1. A method for controlling the intrusion limit of a foldable conveyor belt for a whole vehicle of railway, characterized in that: include, Vibration sensors and tension sensors are installed on the folding conveyor belt to collect the vibration and tension signals of the conveyor belt. The direction is determined based on the folding conveyor belt, and the conveyor belt stiffness matrix is ​​constructed by analyzing the vibration characteristics of the conveyor belt. The vibration signal characteristics are extracted by fast spectral kurtosis calculation, and a DBN model is constructed and optimized using the sparrow search algorithm combined with the firefly disturbance formula. The conveyor belt fault monitoring results are obtained through the DBN model. Based on the conveyor belt fault monitoring results, a fault handling plan is generated by using a generative adversarial network and implemented after evaluation through a convolutional neural network; Visualize the fault handling solution and store the fault handling record in the database; The method determines the direction based on the folding conveyor belt, analyzes the vibration characteristics of the conveyor belt, constructs the conveyor belt stiffness matrix, simultaneously extracts the vibration signal characteristics through fast spectral peak calculation, constructs a DBN model and uses the sparrow search algorithm combined with the firefly disturbance formula for optimization, and obtains the conveyor belt fault monitoring result through the DBN model, which means that the material transportation direction of the folding conveyor belt is defined as the longitudinal direction, and the plane perpendicular to the material transportation direction is defined as the transverse direction, the material parameters of the folding conveyor belt are obtained, the longitudinal vibration equation of the folding conveyor belt is constructed respectively, and the transverse vibration equation is constructed with the tension signal, and the longitudinal displacement response of the folding conveyor belt is obtained respectively through the finite difference method. and lateral displacement response ; According to the longitudinal displacement response of the folded conveyor belt and lateral displacement response Construct the stiffness matrix K of the folding conveyor belt; Extracting vibration signals using short-time Fourier transform The complex frequency domain representation of , calculate the amplitude envelope represented in the complex frequency domain ; Through the amplitude envelope Calculate the spectral kurtosis of a vibration signal And combine the spectral kurtosis of the vibration signal of all frequencies b As a vibration signal feature ; Principal component analysis was used to characterize vibration signals Perform dimensionality reduction, build a DBN model, and update the DBN model parameters through the sparrow search algorithm: ； in is the jth parameter of the ith sparrow at time t, including each DBN model parameter, is the position of the sparrow at time t+1, is a random number, ST is the set threshold, when , the sparrow search algorithm performs local search, when Perform a global search when is the adjustment factor, h is the maximum number of iterations, Q is the migration factor, and W is the neighborhood range; The position of the sparrow at time t+1 is calculated synchronously using the firefly perturbation formula: ； in is the initial attraction, is the attenuation factor, r is the distance between the current position and the target solution, is the disturbance amplitude factor, is the random disturbance term; The sparrow position at time t+1 calculated by the firefly perturbation formula and the sparrow position at time t+1 obtained by the sparrow search algorithm are comprehensively averaged and used as the final sparrow position and repeated iteration as the next iteration position; After the iteration is completed, the optimal DBN model parameters are extracted from the sparrow position and applied to the DBN model, and the stiffness matrix K and vibration signal characteristics are combined. Input the DBN model to obtain the conveyor belt fault monitoring results.

2. The anti-intrusion control method for a foldable conveyor belt for a whole vehicle for railway use according to claim 1, characterized in that: The method of generating a fault handling solution based on the conveyor belt fault monitoring result by using a generative adversarial network and implementing the fault handling solution after evaluating it through a convolutional neural network refers to constructing a generative adversarial network and training it, inputting the conveyor belt fault monitoring result into the generative adversarial network to generate a fault handling solution through a generator, simultaneously constructing a convolutional neural network for training, inputting the fault handling solution into the trained convolutional neural network for evaluation and setting an evaluation threshold, and judging the fault handling solution as an implementation plan for implementation when the evaluation result is greater than the evaluation threshold.

3. The anti-intrusion control method for a foldable conveyor belt for railway of a whole vehicle according to claim 2, characterized in that: The method of installing a vibration sensor and a tension sensor on the folding conveyor belt to collect the vibration signal and tension signal of the conveyor belt refers to installing a vibration sensor and a tension sensor under the folding conveyor belt, and connecting the vibration sensor and the tension sensor through a wireless network to form a sensor network, and performing preprocessing operations after collecting the vibration signal and the tension signal respectively through the vibration sensor and the tension sensor.

4. The anti-intrusion control method for a foldable conveyor belt for a whole vehicle of railway according to claim 1, characterized in that: A horizontal conveyor belt is arranged in front of the folding conveyor belt. The folding conveyor belt and the horizontal conveyor belt are connected through a start-stop signal sensor. The material is transported to the folding conveyor belt through the horizontal conveyor belt. When the start signal of the folding conveyor belt is received, the horizontal conveyor belt starts and transports the material. When a folding conveyor belt failure is detected and the folding conveyor belt needs to be closed, the horizontal conveyor belt needs to be closed and the folding conveyor belt is closed after the closing signal is transmitted to the folding conveyor belt.

5. The anti-intrusion control method for a foldable conveyor belt for railway of a whole vehicle according to claim 3, characterized in that: The visual display of the fault handling solution refers to displaying the conveyor belt fault monitoring results and the final fault handling solution to the staff, and simultaneously displaying the vibration data and tension data to assist the staff in performing folding conveyor belt maintenance.

6. The anti-intrusion control method for a foldable conveyor belt for railway as claimed in claim 5, characterized in that: The forming of the fault handling record and storing it in the database refers to storing the implemented fault handling plan and the corresponding fault monitoring result into the fault handling record in the database, the database marks the timestamp for the stored data, and uploads the stored data to the cloud for backup.

7. A vehicle-wide railway folding conveyor belt anti-intrusion control system, based on the vehicle-wide railway folding conveyor belt anti-intrusion control method according to any one of claims 1 to 6, characterized in that: include, A signal collection module is used to install a vibration sensor and a tension sensor on the folding conveyor belt to collect the vibration signal and tension signal of the conveyor belt for pre-processing; The fault monitoring module is used to extract the stiffness matrix and vibration signal characteristics of the conveyor belt, and to construct a DBN model using the sparrow search algorithm combined with the firefly disturbance formula optimization to obtain the conveyor belt fault monitoring results through the DBN model; A fault handling module, used to generate a fault handling solution using a generative adversarial network and implement the fault handling solution after evaluation through a convolutional neural network; The display storage module is used to visually display the fault handling solution and store the fault handling record in the database.

8. A computer device comprising a memory and a processor, wherein the memory stores a computer program, wherein: When the processor executes the computer program, the steps of the anti-intrusion limit control method for the folding conveyor belt for railway of a whole vehicle as described in any one of claims 1 to 6 are implemented.

9. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by the processor, the steps of the anti-intrusion limit control method for the foldable conveyor belt for railway of a whole vehicle as described in any one of claims 1 to 6 are implemented.

Images