Door machine key structure safety monitoring method and system based on physical data driving

Abstract

The invention discloses a door machine key structure safety monitoring method and system based on physical data driving, and the method comprises the following steps: collecting the strain data of all measurement points in a door machine key structure in real time, and carrying out the preprocessing of the strain data, and obtaining the preprocessed actual strain data; processing the actual strain data based on an inverse finite element incremental method, and gradually reconstructing the deformation and stress field of the structure to obtain a reconstruction result; the reconstruction result is compared with a preset safety threshold value, whether the structure is in a safe state or not is judged, and a machine learning algorithm is used for predicting the safety trend of the door machine key structure; and constructing a virtual model of the door machine key structure by using a VR technology, and displaying a reconstruction result in the virtual model in real time. By collecting and processing the strain data in real time, the safety state of the key structure of the portal crane can be reflected in real time. And an inverse finite element incremental method and an advanced data processing algorithm are adopted, so that the accuracy of safety assessment is improved.

Description

TECHNICAL FIELD

[0001] The application belongs to the technical field of port machinery safety monitoring, and particularly relates to a safety monitoring method and system for a key structure of a portal crane based on physical data driving. BACKGROUND

[0002] As indispensable key equipment in port loading and unloading operations, portal cranes bear heavy and frequent cargo loading and unloading tasks, and their running efficiency directly affects the logistics turnover speed and economic benefits of the port. However, the key structures of the portal cranes are subjected to complex stress environments in the long-term use process, such as cyclic loads caused by repeated lifting and unloading operations, and erosion of adverse natural environmental factors (such as sea wind corrosion, high temperature exposure, etc.), which are prone to safety hazards such as fatigue damage, cracks, deformation, etc. Once the key structures fail or fail, not only will it cause the interruption of port operations, causing huge economic losses, but also may cause serious safety accidents, endangering the safety of on-site operation personnel. Therefore, real-time and accurate safety monitoring of the key structures of the portal cranes, timely discovery of potential safety hazards and taking corresponding maintenance measures, have extremely important significance for ensuring the safety of port operations and improving the efficiency of port operations.

[0003] At present, the safety monitoring of the port portal cranes mainly relies on periodic manual inspection and traditional finite element analysis method. Manual inspection usually needs to be stopped, which not only consumes time and effort, but also is difficult to find early micro-damage and potential hazards, and cannot realize comprehensive grasp of the real-time state of the key structures of the portal cranes. Although the traditional finite element analysis method can simulate the stress condition of the portal crane structure to a certain extent, it needs accurate geometric model and material parameters, and has certain limitations in dealing with complex working conditions and large deformation problems, and is difficult to accurately reflect the dynamic changes of the structure in the actual operation. In addition, finite element analysis usually needs professional technicians to carry out complex modeling and calculation, which requires high professional quality of the operating personnel, and the calculation process consumes a long time, which is difficult to meet the real-time monitoring demand. Therefore, the traditional portal crane safety monitoring method has been difficult to adapt to the requirements of modern port efficient and safe operation, and an advanced, efficient and accurate safety monitoring technology is urgently needed.

[0004] With the rapid development of emerging technologies such as the Internet of Things, big data, artificial intelligence, and physical data-driven monitoring technology has gradually become an important development direction in the field of structural health monitoring. By installing various high-precision sensors such as strain sensors, displacement sensors, and acceleration sensors on the structure, various physical data of the structure during actual operation can be collected in real time. These data contain key information such as stress, strain, and deformation of the structure, and can truly reflect the stress state and damage of the structure. Using advanced data processing and analysis techniques such as machine learning, data mining, and signal processing, the collected large amount of data can be deeply mined and analyzed to extract the health status characteristics of the structure, thereby realizing real-time evaluation and prediction of the safety state of the structure. Physical data-driven monitoring technology has the advantages of strong real-time performance, high precision, and adaptability to complex working conditions, and can overcome the shortcomings of traditional monitoring methods, providing a new solution for the safety monitoring of key structures of the gantry crane. However, the application of this technology in the field of gantry cranes is still relatively small, and a mature technical system and application standard have not yet been formed, therefore, it is of great practical significance and broad application prospects to research and develop a physical data-driven safety monitoring method and system for key structures of a gantry crane. SUMMARY

[0005] The present application aims to solve the problems of the prior art and provides the following solution:

[0006] A physical data-driven safety monitoring method for key structures of a gantry crane, comprising the following steps:

[0007] S1. Real-time collection of strain data of all measurement points in the key structure of the gantry crane, and pre-processing of the strain data to obtain pre-processed actual strain data;

[0008] S2. Processing the actual strain data based on the inverse finite element incremental method, and gradually reconstructing the deformation and stress field of the structure to obtain a reconstruction result;

[0009] S3. Comparing the reconstruction result with a preset safety threshold to determine whether the structure is in a safe state, and using a machine learning algorithm to predict the safety trend of the key structure of the gantry crane;

[0010] S4. Using VR technology to construct a virtual model of the key structure of the gantry crane, and displaying the reconstruction result in real time in the virtual model.

[0011] Preferably, the pre-processing method comprises:

[0012] Filtering the strain data using a wavelet transform method to remove high-frequency noise in the strain data and obtain filtered data;

[0013] The filtered data is smoothed using the Kalman filter method to obtain the preprocessed actual strain data.

[0014] Preferably, S2 includes:

[0015] Theoretical strain data for all measurement points were obtained based on the Mindlin-Reissner plate theory. An error function was then calculated based on the theoretical strain data and the actual strain data.

[0016]

[0017] Where φ represents the error function, ω i ε represents the dimensionless weighting coefficients. theo,i Let ε represent the theoretical strain data at the i-th measurement point. meas,i This represents the actual strain data at the i-th measurement point, and n represents the number of measurement points.

[0018] By minimizing the error function, the displacement field U is obtained. n ;

[0019] Based on the displacement field U n Calculate the displacement increment ΔU n+1 :

[0020] K(U n )·ΔU n+1 =ΔF n ,

[0021] ΔF n =F n+1 -F n ,

[0022] Among them, K(U) n ) represents the rigidity matrix of the displacement field, ΔF n F represents the increment of external force in the current step. n+1 and F n Both represent external force vectors;

[0023] And based on the displacement increment ΔU n+1 The displacement field, stiffness matrix, and external force vector are updated to obtain the reconstruction result.

[0024] Preferably, S3 includes:

[0025] The reconstruction result is compared with a preset safety threshold to determine whether the structure is in a safe state. If the structure is not in a safe state, a warning signal is issued.

[0026] By combining historical and real-time data, machine learning algorithms are used to predict the safety trends of key structures of the gantry crane.

[0027] The present invention also provides a safety monitoring system for key structures of gantry cranes based on physical data, wherein the system applies the above-mentioned method and includes: a data processing module, a data reconstruction module, a monitoring and prediction module, and a visualization module;

[0028] The data processing module is used to collect strain data of all measurement points in the key structure of the gantry crane in real time, and to preprocess the strain data to obtain the preprocessed actual strain data.

[0029] The data reconstruction module processes the actual strain data based on the inverse finite element incremental method, and gradually reconstructs the deformation and stress field of the structure to obtain the reconstruction result;

[0030] The monitoring and prediction module is used to compare the reconstruction result with a preset safety threshold to determine whether the structure is in a safe state, and to use machine learning algorithms to predict the safety trend of the key structure of the gantry crane.

[0031] The visualization module uses VR technology to construct a virtual model of the key structure of the gantry crane and displays the reconstruction results in real time in the virtual model.

[0032] Preferably, the workflow of the data processing module includes:

[0033] The strain data is filtered using wavelet transform to remove high-frequency noise and obtain filtered data.

[0034] The filtered data is smoothed using the Kalman filter method to obtain the preprocessed actual strain data.

[0035] Preferably, the workflow of the data reconstruction module includes:

[0036] Theoretical strain data for all measurement points were obtained based on the Mindlin-Reissner plate theory. An error function was then calculated based on the theoretical strain data and the actual strain data.

[0037]

[0038] Where φ represents the error function, ω i ε represents the dimensionless weighting coefficients. theo，i Let ε represent the theoretical strain data at the i-th measurement point. meas，i This represents the actual strain data at the i-th measurement point, and n represents the number of measurement points.

[0039] By minimizing the error function, the displacement field U is obtained. n ;

[0040] Based on the displacement field Un Calculate the displacement increment ΔU n+1 :

[0041] K(U n )·ΔU n+1 =ΔF n ,

[0042] ΔF n =F n+1 -F n ,

[0043] Among them, K(U) n ) represents the rigidity matrix of the displacement field, ΔF n F represents the increment of external force in the current step. n+1 and F n Both represent external force vectors;

[0044] And based on the displacement increment ΔU n+1 The displacement field, stiffness matrix, and external force vector are updated to obtain the reconstruction result.

[0045] Preferably, the workflow of the monitoring and prediction module includes:

[0046] The reconstruction result is compared with a preset safety threshold to determine whether the structure is in a safe state. If the structure is not in a safe state, a warning signal is issued.

[0047] By combining historical and real-time data, machine learning algorithms are used to predict the safety trends of key structures of the gantry crane.

[0048] Compared with the prior art, the beneficial effects of the present invention are as follows:

[0049] This invention, through real-time acquisition and processing of strain data, can reflect the safety status of critical structures of the gantry crane in real time. Employing the inverse finite element incremental method and advanced data processing algorithms improves the accuracy of safety assessments, particularly when dealing with large deformations and nonlinear problems. Furthermore, visualization through virtual reality technology allows users to intuitively understand the structure's safety status, facilitating management and decision-making. Attached Figure Description

[0050] To more clearly illustrate the technical solution of the present invention, the drawings used in the embodiments are briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0051] Figure 1 This is a schematic diagram of the method flow according to an embodiment of the present invention. Detailed Implementation

[0052] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0053] To make the above-mentioned objects, features and advantages of the present invention more apparent and understandable, the present invention will be further described in detail below with reference to the accompanying drawings and specific embodiments.

[0054] Example 1

[0055] In this embodiment, as Figure 1 As shown, a method for safety monitoring of key structures of a gantry crane based on physical data includes the following steps:

[0056] S1. Real-time acquisition of strain data at all measurement points in the key structure of the gantry crane, and preprocessing of the strain data to obtain the preprocessed actual strain data.

[0057] The preprocessing methods include: using wavelet transform to filter the strain data to remove high-frequency noise and obtain filtered data; and using Kalman filtering to smooth the filtered data to obtain preprocessed actual strain data.

[0058] In this embodiment, high-precision strain sensors are used to collect strain data. These sensors are installed on stress-bearing parts of the gantry crane's key structures, such as the boom, tie rods, and balance beams. The collected raw strain data is then filtered and noise-reduced. A wavelet transform algorithm is used to remove high-frequency noise from the data, resulting in filtered data. Then, a Kalman filter algorithm is used to smooth the data, improving its accuracy and yielding preprocessed actual strain data.

[0059] S2. Based on the inverse finite element incremental method, the actual strain data is processed to gradually reconstruct the deformation and stress field of the structure, and the reconstruction results are obtained.

[0060] S2 includes:

[0061] Theoretical strain data ε at all measurement points were obtained based on the Mindlin-Reissner plate theory. theo The theoretical strain data is a function of the displacement field U, expressed as ε. theo(U); In each step of the calculation, the nonlinear characteristics of the structure need to be considered, especially the effects of large deformation and nonlinear strain. Based on Green-Lagrange strain theory, for a four-node inverse shell element, the theoretical strain-displacement relationship can be expressed as:

[0062] ε theo =ε theo (U);

[0063] Based on theoretical strain data and actual strain data, the error function is calculated as follows:

[0064]

[0065] Where φ represents the error function, ω i ε represents a dimensionless weighting coefficient used to adjust the contribution of different strain measurement points. theo，i Let ε represent the theoretical strain data at the i-th measurement point. meas，i Let represent the actual strain data at the i-th measurement point, and n represent the number of measurement points; by minimizing the error function, the displacement field U is obtained. n Based on displacement field U n Calculate the displacement increment ΔU n+1 :

[0066] K(U n )·ΔUn +1 =ΔF n ,

[0067] ΔF n =F n+1 -F n ,

[0068] Among them, K(U) n ) represents the rigidity matrix of the displacement field, ΔF n F represents the increment of external force in the current step. n+1 F represents the external force vector at step n+1. n This represents the external force vector at step n; and is based on the displacement increment ΔU. n+1 Update displacement field U n+1 :

[0069] U n+1 =U n +ΔU n+1 ,

[0070] Update stiffness matrix K(U) n+1 ) and external force vector F n+1 Repeat the above steps until the final state is reached and the reconstruction result is obtained.

[0071] S3. Compare the reconstruction results with the preset safety threshold to determine whether the structure is in a safe state, and use machine learning algorithms to predict the safety trend of the key structure of the gantry crane.

[0072] S3 includes: comparing the reconstruction results with preset safety thresholds to determine whether the structure is in a safe state; if the structure is not in a safe state, issuing an early warning signal; and combining historical and real-time data to use machine learning algorithms to predict the safety trend of the gantry crane's key structures, providing decision support suggestions for port management personnel.

[0073] S4. Use VR technology to construct a virtual model of the key structure of the gantry crane, and display the reconstruction results in real time in the virtual model.

[0074] In this embodiment, virtual reality (VR) technology is used to construct a virtual model of the key structure of the gantry crane and display the deformation and stress distribution information of the structure in real time. Users can intuitively understand the safety status of the structure through VR devices or computer screens.

[0075] Example 2

[0076] In this embodiment, a gantry crane key structure safety monitoring system based on physical data driving includes: a data processing module, a data reconstruction module, a monitoring and prediction module, and a visualization module.

[0077] The data processing module is used to collect strain data from all measurement points in the key structure of the gantry crane in real time, and to preprocess the strain data to obtain the preprocessed actual strain data.

[0078] The workflow of the data processing module includes: filtering the strain data using wavelet transform to remove high-frequency noise and obtain filtered data; and smoothing the filtered data using Kalman filtering to obtain preprocessed actual strain data.

[0079] The data reconstruction module processes actual strain data based on the inverse finite element incremental method, and gradually reconstructs the deformation and stress field of the structure to obtain the reconstruction result.

[0080] The workflow of the data reconstruction module includes: acquiring theoretical strain data for all measurement points based on the Mindlin-Reissner plate theory; and calculating the error function based on the theoretical strain data and the actual strain data.

[0081]

[0082] Where φ represents the error function, ω i ε represents the dimensionless weighting coefficients. theo，i Let ε represent the theoretical strain data at the i-th measurement point. meas，iLet represent the actual strain data at the i-th measurement point, and n represent the number of measurement points; by minimizing the error function, the displacement field U is obtained. n Based on displacement field U n Calculate the displacement increment ΔU n+1 :

[0083] K(U n )·ΔU n+1 =ΔF n ,

[0084] ΔF n =F n+1 -F n ,

[0085] Among them, K(U) n ) represents the rigidity matrix of the displacement field, ΔF n F represents the increment of external force in the current step. n+1 and F n Both represent external force vectors; and are based on displacement increment ΔU n+1 Update the displacement field, stiffness matrix, and external force vector to obtain the reconstruction result.

[0086] The monitoring and prediction module is used to compare the reconstruction results with preset safety thresholds to determine whether the structure is in a safe state, and to use machine learning algorithms to predict the safety trend of the key structure of the gantry crane.

[0087] The monitoring and prediction module's workflow includes: comparing the reconstruction results with preset safety thresholds to determine whether the structure is in a safe state; if the structure is not in a safe state, issuing an early warning signal; and combining historical and real-time data to use machine learning algorithms to predict the safety trend of the gantry crane's key structures.

[0088] The visualization module uses VR technology to construct a virtual model of the key structure of the gantry crane and displays the reconstruction results in real time within the virtual model.

[0089] The embodiments described above are merely preferred embodiments of the present invention and are not intended to limit the scope of the present invention. Various modifications and improvements made to the technical solutions of the present invention by those skilled in the art without departing from the spirit of the present invention should fall within the protection scope defined by the claims of the present invention.

Claims

1. A method for safety monitoring of key structures of a gantry crane based on physical data, characterized in that, Includes the following steps: S1. Real-time acquisition of strain data at all measurement points in the key structure of the gantry crane, and preprocessing of the strain data to obtain preprocessed actual strain data; S2. The actual strain data is processed based on the inverse finite element incremental method to gradually reconstruct the deformation and stress field of the structure and obtain the reconstruction result; S3. Compare the reconstruction result with a preset safety threshold to determine whether the structure is in a safe state, and use machine learning algorithms to predict the safety trend of the key structure of the gantry crane; S4. Use VR technology to construct a virtual model of the key structure of the gantry crane, and display the reconstruction results in the virtual model in real time.

2. The method for safety monitoring of key structures of gantry cranes based on physical data as described in claim 1, characterized in that, The preprocessing method includes: The strain data is filtered using wavelet transform to remove high-frequency noise and obtain filtered data. The filtered data is smoothed using the Kalman filter method to obtain the preprocessed actual strain data.

3. The method for safety monitoring of key structures of a gantry crane based on physical data as described in claim 1, characterized in that, S2 includes: Theoretical strain data for all measurement points were obtained based on the Mindlin-Reissner plate theory. An error function was then calculated based on the theoretical strain data and the actual strain data. Where φ represents the error function, ω i ε represents the dimensionless weighting coefficients. theo，i Let ε represent the theoretical strain data at the i-th measurement point. meas，i This represents the actual strain data at the i-th measurement point, and n represents the number of measurement points. By minimizing the error function, the displacement field U is obtained. n ; Based on the displacement field U n Calculate the displacement increment ΔU n+1 : K(U n )·ΔU n+1 =ΔF n , ΔF n =F n+1 -F n , Among them, K(U) n ) represents the rigidity matrix of the displacement field, ΔF n F represents the increment of external force in the current step. n+1 and F n Both represent external force vectors; And based on the displacement increment ΔU n+1 The displacement field, stiffness matrix, and external force vector are updated to obtain the reconstruction result.

4. The method for safety monitoring of key structures of a gantry crane based on physical data as described in claim 1, characterized in that, S3 includes: The reconstruction result is compared with a preset safety threshold to determine whether the structure is in a safe state. If the structure is not in a safe state, a warning signal is issued. By combining historical and real-time data, machine learning algorithms are used to predict the safety trends of key structures of the gantry crane.

5. A physical data-driven safety monitoring system for critical structures of a gantry crane, wherein the system employs the method described in any one of claims 1-4, characterized in that, include: Data processing module, data reconstruction module, monitoring and prediction module, and visualization module; The data processing module is used to collect strain data of all measurement points in the key structure of the gantry crane in real time, and to preprocess the strain data to obtain the preprocessed actual strain data. The data reconstruction module processes the actual strain data based on the inverse finite element incremental method, and gradually reconstructs the deformation and stress field of the structure to obtain the reconstruction result; The monitoring and prediction module is used to compare the reconstruction result with a preset safety threshold to determine whether the structure is in a safe state, and to use machine learning algorithms to predict the safety trend of the key structure of the gantry crane. The visualization module uses VR technology to construct a virtual model of the key structure of the gantry crane and displays the reconstruction results in real time in the virtual model.

6. The gantry crane key structure safety monitoring system based on physical data driving according to claim 5, characterized in that, The workflow of the data processing module includes: The strain data is filtered using wavelet transform to remove high-frequency noise and obtain filtered data. The filtered data is smoothed using the Kalman filter method to obtain the preprocessed actual strain data.

7. The gantry crane key structure safety monitoring system based on physical data driving according to claim 5, characterized in that, The workflow of the data reconstruction module includes: Theoretical strain data for all measurement points were obtained based on the Mindlin-Reissner plate theory. An error function was then calculated based on the theoretical strain data and the actual strain data. Where φ represents the error function, ω i ε represents the dimensionless weighting coefficients. theo,i Let ε represent the theoretical strain data at the i-th measurement point. meas,i This represents the actual strain data at the i-th measurement point, and n represents the number of measurement points. By minimizing the error function, the displacement field U is obtained. n ; Based on the displacement field U n Calculate the displacement increment ΔU n+1 : K(U n )·ΔU n+1 =ΔF n , ΔF n =F n+1 -F n , Among them, K(U) n ) represents the rigidity matrix of the displacement field, ΔF n F represents the increment of external force in the current step. n+1 and F n Both represent external force vectors; And based on the displacement increment ΔU n+1 The displacement field, stiffness matrix, and external force vector are updated to obtain the reconstruction result.

8. The gantry crane key structure safety monitoring system based on physical data driven according to claim 5, characterized in that, The workflow of the monitoring and prediction module includes: The reconstruction result is compared with a preset safety threshold to determine whether the structure is in a safe state. If the structure is not in a safe state, a warning signal is issued. By combining historical and real-time data, machine learning algorithms are used to predict the safety trends of key structures of the gantry crane.

Images