Training Method, Device, Equipment and Storage Medium for Tower Crane Health Monitoring Model

Abstract

The present application discloses a training method, device, electronic device and storage medium for a tower crane health monitoring model. The method includes: obtaining first measured data and second measured data of the tower crane; simulating the tower crane to obtain a first simulated tower crane, and obtaining first simulation data of the first simulated tower crane in a healthy state; determining whether to perform simulation adjustment on the first simulated tower crane according to the first measured data and the first simulation data; when it is determined not to perform simulation adjustment on the first simulated tower crane, performing simulation adjustment on the first simulated tower crane based on the first fault parameter to obtain a second simulated tower crane, and performing simulated operation on the second simulated tower crane to obtain second simulation data; training a pre-constructed tower crane health monitoring model according to the first measured data, the second measured data, the first simulation data and the second simulation data, and obtaining a trained tower crane health monitoring model when the training is completed. The convenience and accuracy of tower crane health monitoring are improved.

Description

Technical Field

[0001] The present application relates to the technical field of safety monitoring of tower cranes, and particularly relates to a method, device, electronic device and storage medium for training a tower crane health monitoring model. Background Art

[0002] With continuous development, tower cranes are widely used in the construction of high-rise buildings, bridges and other buildings. The safe operation of tower cranes is an extremely important part, such as the operation safety of operators and the application safety of tower cranes. Among them, the operation safety of operators is reflected in the standardization of operating the tower crane, and the application safety of the tower crane is reflected in the safety of the tower crane's own structure. Therefore, the health and safety monitoring of the tower crane structure is extremely important.

[0003] Currently, when monitoring the health and safety of the tower crane structure, in addition to the traditional manual inspection method, it also includes diagnosing the steel structure of the tower body through certain measurement means, such as comparing the monitoring data with the safety data. However, there are certain deficiencies, such as the time-consuming and laborious manual inspection method, and the low accuracy of comparing based on specific monitoring data. Summary of the Invention

[0004] The purpose of the embodiments of the present application is to provide a method, device, electronic device and storage medium for training a tower crane health monitoring model to solve the technical problems of low accuracy and convenience in the health and safety monitoring of the tower crane structure in the related art.

[0005] In a first aspect, the embodiments of the present application provide a method for training a tower crane health monitoring model, including:

[0006] Obtain the first measured data and the second measured data of the tower crane, where the first measured data is the health status data collected based on a plurality of sensors arranged on the tower crane, and the second measured data is the fault status data collected based on a plurality of sensors arranged on the tower crane;

[0007] Simulate the tower crane to obtain a first simulated tower crane, and simulate the operation of the first simulated tower crane to obtain first simulation data of the first simulated tower crane in a healthy state;

[0008] Determine whether to perform simulation adjustment on the first simulated tower crane according to the first measured data and the first simulation data;

[0009] When it is determined not to perform simulation adjustment on the first simulated tower crane, perform simulation adjustment on the first simulated tower crane based on the first fault parameter to obtain a second simulated tower crane, and simulate the operation of the second simulated tower crane to obtain second simulation data;

[0010] Train a pre - constructed tower crane health monitoring model based on the first measured data, the second measured data, the first simulation data, and the second simulation data, and obtain a trained tower crane health monitoring model when the training is completed.

[0011] Optionally, determining whether to perform simulation adjustment on the first simulated tower crane according to the first measured data and the first simulation data includes:

[0012] Perform filtering transformation pre - processing on the first measured data and the first simulation data to obtain the processed first measured data and the processed first simulation data;

[0013] Calculate the data similarity between the processed first measured data and the processed first simulation data, and compare the data similarity with a similarity threshold;

[0014] Determine whether to adjust the first simulated tower crane according to the comparison result obtained from the comparison;

[0015] If the data similarity is less than the similarity threshold, determine to perform simulation adjustment on the first simulated tower crane;

[0016] If the data similarity is greater than or equal to the similarity threshold, determine not to perform simulation adjustment on the first simulated tower crane.

[0017] Optionally, when it is determined not to perform simulation adjustment on the first simulated tower crane, performing simulation adjustment on the first simulated tower crane based on the first fault parameter to obtain a second simulated tower crane, and performing simulated operation on the second simulated tower crane to obtain second simulation data includes:

[0018] When it is determined not to perform simulation adjustment on the first simulated tower crane, load a fault parameter generation model to output the first fault parameter corresponding to the tower crane;

[0019] Perform simulation adjustment on the first simulated tower crane according to the first fault parameter to obtain a second simulated tower crane, and perform simulated operation on the second simulated tower crane to obtain second simulation data under the first fault parameter.

[0020] Optionally, training a pre - constructed tower crane health monitoring model based on the first measured data, the second measured data, the first simulation data, and the second simulation data, and obtaining a trained tower crane health monitoring model when the training is completed includes:

[0021] Perform filtering transformation pre - processing on the second measured data and the second simulation data to obtain the processed second measured data and the processed second simulation data;

[0022] Load the pre-built tower crane health monitoring model, train it using the processed second simulation data, and obtain the first tower crane health monitoring model after training;

[0023] Use the processed second measured data to conduct an effectiveness test on the first tower crane health monitoring model to obtain corresponding test results, where the test results include passing the test and failing the test;

[0024] When it is determined that the test result is passing the test, perform data fusion processing on the first test data, the second measured data, the first simulation data, and the second simulation data to obtain corresponding fusion data, and perform model fine-tuning processing on the first tower crane health monitoring model based on the fusion data to obtain the tower crane health monitoring model.

[0025] Optionally, the training using the processed second simulation data and obtaining the first tower crane health monitoring model after training includes:

[0026] Divide the processed second simulation data into training set data and validation set data, and train the pre-built tower crane health monitoring model based on the training set data to obtain an intermediate model;

[0027] Use the validation set data to perform model fine-tuning on the intermediate model and obtain the first tower crane health monitoring model when the fine-tuning is completed.

[0028] Optionally, the obtaining the corresponding test results by using the processed second measured data to conduct an effectiveness test on the first tower crane health monitoring model includes:

[0029] Input the fault data included in the processed second measured data into the first tower crane health monitoring model and output the predicted data;

[0030] Compare the predicted data with the standard data included in the processed second measured data to determine the prediction effect of the first tower crane health monitoring model;

[0031] Obtain the test results of the first tower crane health monitoring model according to the test effect;

[0032] If the test accuracy rate in the test effect is greater than or equal to the preset threshold, it is determined that the test is passed;

[0033] If the test accuracy rate in the test effect is less than the preset threshold, it is determined that the test is not passed.

[0034] Optionally, the method further includes:

[0035] When it is determined that the test result fails the test, obtain the second fault parameter, and perform simulation adjustment on the first simulated tower crane based on the second fault parameter to obtain a third simulated tower crane;

[0036] Perform simulated operation on the third simulated tower crane to obtain third simulation data under the second fault parameter;

[0037] Train the pre-constructed tower crane health monitoring model according to the first measured data, the second measured data, the first simulation data, and the third simulation data, and obtain a trained tower crane health monitoring model when the training is completed.

[0038] In a second aspect, an embodiment of the present application provides a training device for a tower crane health monitoring model, including:

[0039] A data acquisition module, configured to acquire first measured data and second measured data of a tower crane, where the first measured data is health status data collected based on a plurality of sensors disposed on the tower crane, and the second measured data is fault status data collected based on a plurality of sensors disposed on the tower crane;

[0040] A first simulation module, configured to simulate the tower crane to obtain a first simulated tower crane, and perform simulated operation on the first simulated tower crane to obtain first simulation data of the first simulated tower crane in a healthy state;

[0041] A simulation adjustment module, configured to determine whether to perform simulation adjustment on the first simulated tower crane according to the first measured data and the first simulation data;

[0042] A second simulation module, configured to, when it is determined not to perform simulation adjustment on the first simulated tower crane, perform simulation adjustment on the first simulated tower crane based on a first fault parameter to obtain a second simulated tower crane, and perform simulated operation on the second simulated tower crane to obtain second simulation data;

[0043] A model training module, configured to train the pre-constructed tower crane health monitoring model according to the first measured data, the second measured data, the first simulation data, and the second simulation data, and obtain a trained tower crane health monitoring model when the training is completed.

[0044] In a third aspect, an embodiment of the present application provides an electronic device, which includes a processor, a memory, and a computer program stored in the memory and executable on the processor. When the processor executes the computer program, the steps in the training method of the tower crane health monitoring model described in any one of the above are implemented.

[0045] Fourthly, an embodiment of the present application provides a computer-readable storage medium storing a computer program, and when the computer program is executed by a processor, the steps in the training method of the tower crane health monitoring model described in any one of the above are implemented.

[0046] An embodiment of the present application provides a training method, device, electronic device, and storage medium for a tower crane health monitoring model. When training the model, first obtain the first measured data of the tower crane in a fault-free state and the second measured data in a faulty state, and perform simulation processing on the tower crane to obtain a first simulated tower crane. At the same time, perform simulation operation to obtain the first simulation data of the first simulated tower crane. Then, based on the first simulation data and the first measured data, determine the rationality of the first simulated tower crane, that is, determine whether the first simulated tower crane needs to be corrected and adjusted. When no correction and adjustment are required, set corresponding first fault parameters on the first simulated tower crane to simulate the operation of the tower crane in a faulty state and obtain second simulation data. Finally, train the constructed tower crane health monitoring model based on the first measured data, the second measured data, the first simulation data, and the second simulation data. Then, when the training is completed, a trained tower crane health monitoring model is obtained. During the training process, a large number of training samples for training are obtained by simulating the tower crane, and then the effectiveness of the model is judged using the measured data after training based on the training samples. This enables the model to more closely approximate the real tower crane for fault identification. By combining the training of the measured data and simulation data in the fault-free state and the faulty state, fault localization can be achieved. Therefore, during use, the existing faults can be identified and located only by obtaining sensor data, improving the convenience and accuracy of tower crane health monitoring. BRIEF DESCRIPTION OF THE DRAWINGS

[0047] Figure 1 is a flowchart showing the steps of the training method for the tower crane health monitoring model provided by an embodiment of the present application;

[0048] Figure 2 is a flowchart showing the steps of obtaining the tower crane health monitoring model provided by an embodiment of the present application;

[0049] Figure 3 is a flowchart showing the steps of obtaining the test result provided by an embodiment of the present application;

[0050] Figure 4 is another flowchart showing the steps of the training method for the tower crane health monitoring model provided by an embodiment of the present application;

[0051] Figure 5 is a structural diagram of the training device for the tower crane health monitoring model provided by an embodiment of the present application;

[0052] Figure 6 It is a schematic structural diagram of an electronic device provided by an embodiment of the present application;

[0053] Figure 7 It is another schematic structural diagram of an electronic device provided by an embodiment of the present application. Specific implementation manners

[0054] Next, the technical solutions in the embodiments of the present application will be clearly and completely described in conjunction with the accompanying drawings in the embodiments of the present application. Obviously, the described embodiments are only a part of the embodiments of the present application, rather than all the embodiments. All other embodiments obtained by those of ordinary skill in the art based on the embodiments of the present application without making creative efforts shall fall within the protection scope of the present application.

[0055] It should be understood that the steps recorded in the method implementation manners disclosed in the present application can be executed in different orders and / or executed in parallel. In addition, the method implementation manners may include additional steps and / or omit the shown steps. The scope disclosed in the present application is not limited in this regard.

[0056] The term "including" and its variants used herein are open-ended, that is, "including but not limited to". The term "based on" is "at least partially based on". The term "one embodiment" means "at least one embodiment"; the term "another embodiment" means "at least one additional embodiment"; the term "some embodiments" means "at least some embodiments". The relevant definitions of other terms will be given in the following description.

[0057] In the related art, when monitoring the structure of a tower crane for health and safety, in addition to the traditional manual inspection method, it also includes diagnosing the steel structure of the tower body through certain measurement means, such as comparing the monitoring data with the safety data, but there are certain deficiencies, such as the time-consuming and laborious manual inspection method, and the low adaptability and accuracy of comparing based on specific monitoring data.

[0058] To solve the technical problems existing in the related art, an embodiment of the present application provides a method for training a tower crane health monitoring model. Please refer to Figure 1 , Figure 1 It is a schematic flow diagram of the steps of the method for training a tower crane health monitoring model provided by an embodiment of the present application. The method includes steps 101 to 105.

[0059] Step 101, obtain the first measured data and the second measured data of the tower crane, where the first measured data is the health status data collected based on a plurality of sensors arranged on the tower crane, and the second measured data is the fault status data collected based on a plurality of sensors arranged on the tower crane.

[0060] In one embodiment, when training the constructed tower crane health monitoring model, it is trained with simulation data and the verification process of measured data is carried out. Then, after completing the training and passing the verification, a tower crane health monitoring model with good prediction effect is obtained. Therefore, after training, it is necessary to obtain the measured data of the tower crane, specifically including the first measured data collected by each sensor set in the healthy state and the second measured data collected by each sensor set in the fault state. In particular, for the second measured data, relevant fault data are also included, such as the specific fault location, fault value, etc.

[0061] Exemplarily, when training the tower crane health monitoring model, it is trained for the same type of tower crane, that is, the trained tower crane health monitoring model is applicable to the health monitoring of the same type of tower crane. Therefore, when obtaining the measured data, the monitoring data of the same model of tower crane during actual operation can be obtained, and for the health state of the tower crane, it includes faults and health.

[0062] In addition, in order to improve the applicability of the tower crane health monitoring model, after obtaining the tower crane health monitoring model for a tower crane, the tower crane health monitoring model can be fine-tuned based on the differences between tower cranes. The specific adjustment basis and adjustment method can be obtained through analysis and processing based on the structural differences of the tower crane, the differences between the structural differences and the measured data, etc.

[0063] Step 102: Simulate the tower crane to obtain the first simulated tower crane, and perform simulated operation on the first simulated tower crane to obtain the first simulation data of the first simulated tower crane in the healthy state.

[0064] In one embodiment, when training the model, the training data used includes corresponding simulation data. Therefore, in addition to obtaining the measured data of the tower crane, relevant simulation software will also be used to simulate the tower crane, and the corresponding simulation data will be obtained using the simulated tower crane obtained by the simulation. Specifically, the tower crane is simulated using the simulation software to obtain the corresponding first simulated tower crane, and at the same time, the obtained first simulated tower crane is simulated to obtain the first simulation data of the first simulated tower crane in the healthy state.

[0065] Exemplarily, when performing the simulation process of the tower crane, a finite element model of the tower crane can be constructed based on the ANSYS software. For the simulated tower crane obtained by the simulation, it is necessary to maintain the consistency with the real tower crane in terms of structure and materials. Then, when performing the simulated operation, the spatial beam element can be selected as the element type to obtain the first simulation data of the simulated tower crane in the healthy state.

[0066] In the actual application process, when constructing a simulated tower crane, since it is impossible to ensure that every tower crane of the same type is exactly the same in terms of structural parameters or structural dimensions, when constructing a simulated tower crane, for data with intervals such as structural parameters or structural dimensions, random selection is made. This may result in the simulated results or simulated data obtained during simulation being somewhat different from the real data. When the difference is small, it can be accepted, while when the difference is too large, it will lead to a large error in model training. Therefore, the first simulated data obtained based on the first simulated tower crane cannot be directly used for model training, but rather requires certain analysis and judgment processing.

[0067] Step 103, determine whether to perform simulation adjustment on the first simulated tower crane according to the first measured data and the first simulated data.

[0068] In one embodiment, when performing simulation processing on a tower crane, it is necessary to ensure that the similarity between the simulated tower crane for simulation processing and the real tower crane is as high as possible. Only in this way can the simulated tower crane be used to simulate realistic accident data. Therefore, after obtaining the first simulated data through simulation, it will be determined whether to perform simulation adjustment on the first simulated tower crane according to the first measured data and the first simulated data.

[0069] Exemplarily, when determining whether to perform simulation adjustment on the first simulated tower crane, it is necessary to determine whether the first simulated data is close to the first measured data. The closer the two are, the more the simulated tower crane fits the real tower crane. When determining whether the first simulated data and the first real data are close, it can be determined by calculating the distance between the data. For example, calculating the similarity between the data. The higher the similarity, the closer the distance. And there are various ways to calculate similarity, such as cosine similarity, Euclidean distance, etc.

[0070] Therefore, when determining whether to perform simulation adjustment on the first simulated tower crane according to the first simulated data and the first measured data, it can be determined through similarity calculation, specifically including: performing filter transformation preprocessing on the first measured data and the first simulated data to obtain the processed first measured data and the processed first simulated data; calculating the data similarity between the processed first measured data and the processed first simulated data, and comparing the data similarity with the similarity threshold; determining whether to adjust the first simulated tower crane according to the comparison result obtained from the comparison; if the data similarity is less than the similarity threshold, determine to perform simulation adjustment on the first simulated tower crane; if the data similarity is greater than or equal to the similarity threshold, determine not to perform simulation adjustment on the first simulated tower crane.

[0071] When calculating the similarity between the first measured data and the first simulation data, by calculating the similarity between each set of measured data and the corresponding set of simulation data, first through filter transformation preprocessing, the first measured data and the first simulation data, which are sensor data, are converted into a data form that can be calculated, such as being converted into the corresponding vector form, and then the data similarity between the data after the conversion is calculated, such as calculating the distance between two vectors (the smaller the distance, the higher the similarity), and by comparing the calculated data similarity with the preset similarity threshold. Since the higher the similarity indicates that the data is closer, then when comparing, if the data similarity is less than the similarity threshold, it means that the data difference is large, and the first simulated tower crane needs to be simulated and optimized. On the contrary, if the data similarity is greater than or equal to the similarity threshold, it means that the data difference is small, and at this time, further processing can be carried out without simulating and optimizing the first simulated tower crane.

[0072] Furthermore, when the calculated data similarity is less than the set similarity threshold, the first simulated tower crane needs to be subjected to corresponding simulation and optimization processing. In fact, when the data similarity is less than the similarity threshold, it means that there are differences in characteristics between the constructed first simulated tower crane and the real tower crane, and the first simulated tower crane cannot be used to simulate the operating data of the tower crane in a fault state. Therefore, it is necessary to re - simulate, specifically, by correcting the parameters of the tower crane, such as correcting and adjusting the structural dimensions of the tower crane, to obtain a new first simulated tower crane. For the new first simulated tower crane, it is also necessary to carry out a simulation operation and compare the obtained simulation data with the first measured data again until there is no need to correct the simulated tower crane.

[0073] Step 104, when it is determined not to perform simulation adjustment on the first simulated tower crane, perform simulation adjustment on the first simulated tower crane based on the first fault parameter to obtain a second simulated tower crane, and perform a simulation operation on the second simulated tower crane to obtain second simulation data.

[0074] In one embodiment, when determining whether to perform simulation adjustment on the first simulated tower crane based on the first measured data and the first simulation data, if it is determined that no simulation adjustment is required, further processing can be carried out based on the first simulated tower crane, specifically including performing a simulation on the first simulated tower crane to obtain relevant fault data. Specifically, when it is determined not to perform simulation adjustment on the first simulated tower crane, obtain relevant first fault parameters to adjust the first simulated tower crane. Among them, the added first fault parameters at least include the damage position, connection stiffness, and material anti - interference coefficient, etc. Then, after the adjustment is completed, a second simulated tower crane is obtained, and a simulation operation is performed on the second simulated tower crane to obtain second simulation data.

[0075] Exemplarily, the second simulated tower crane is a simulated tower crane after fault settings are made for the tower crane, used to simulate the operation of the tower crane in a fault state, and thus the operation data of the tower crane in the presence of faults can be simulated, that is, the second simulation data.

[0076] Furthermore, the first fault parameters added are obtained based on the characteristics of the tower crane itself. The settings for the tower crane itself at least include fault parameters such as the damage location, connection stiffness, and material anti-interference coefficient, which are input into the first simulated tower crane to obtain a second simulated tower crane with certain faults. Since the fault factors and factors existing in different types of tower cranes may be different, therefore, for the first fault parameters, they can be obtained by using a pre-trained fault parameter generation model. By inputting the characteristics of the simulated tower crane, more suitable first fault parameters for itself can be obtained, and then the second simulation data can be obtained.

[0077] Specifically, when obtaining the second simulation data, it includes: when it is determined that no simulation adjustment is made to the first simulated tower crane, load the fault parameter generation model to output the first fault parameters corresponding to the tower crane; perform simulation adjustment on the first simulated tower crane according to the first fault parameters to obtain a second simulated tower crane, and perform simulated operation on the second simulated tower crane to obtain the second simulation data under the first fault parameters.

[0078] In practical applications, the first fault parameters added to the first simulated tower crane can be set by the operator based on experience, or can be automatically generated based on the tower crane. For automatically generating the corresponding fault parameters, a corresponding fault parameter generation model can be pre-prepared. By inputting the tower crane characteristics of the simulated tower crane into the fault parameter generation model, one or more groups of fault parameters can be obtained by random generation. Among them, the tower crane characteristics include but are not limited to tower crane size information, tower crane load-bearing information, tower crane operation information, and tower crane structure characteristics, etc. Then, randomly select a group of fault parameters as the first fault parameters for adjusting the first simulated tower crane to obtain a second simulated tower crane with certain faults.

[0079] When obtaining the second simulation data, it is the same as the way of obtaining the first simulation data. By performing simulated operation on the second simulated tower crane, the second simulation data generated during operation in the fault state can be obtained.

[0080] Step 105, train the pre-constructed tower crane health monitoring model according to the first measured data, the second measured data, the first simulation data, and the second simulation data, and obtain a trained tower crane health monitoring model when the training is completed.

[0081] In one embodiment, after obtaining the measured data and simulation data, the obtained measured data and simulation data will be used for training the model, and then a tower crane health monitoring model that can be used for health monitoring of the tower crane will be obtained when the training is completed. Specifically, after obtaining the measured data and simulation data, the pre-constructed tower crane health monitoring model will be trained and optimized according to the obtained first measured data, second measured data, first simulation data, and second simulation data, and then a trained tower crane health monitoring model will be obtained after the training and optimization are completed.

[0082] Exemplarily, during training, it is necessary to train the tower crane in both the healthy state and the faulty state to ensure that during the use of the model, it can accurately identify and analyze whether the tower crane is in a healthy or faulty state. At the same time, when the tower crane has a fault, the location and fault level of the fault can be determined through the analysis of the operation data during the operation of the tower crane.

[0083] In practical applications, several sensors will be set on the tower crane to obtain data of the tower crane in different states, and different data represent the fault conditions of the corresponding structural parts on the tower crane, such as whether there is a fault and the fault level. Therefore, for the obtained data, including measured data and simulation data, data marking processing can be performed. By marking the data, the specific fault can be located, and thus the fault can be identified and located during the use process. In order to accurately determine whether there is a fault and locate the fault, it is necessary to combine the simulation data and the measured data for training and optimization processing, which can improve the accuracy of the tower crane health monitoring model in predicting faults and accurately locate the faults at the same time.

[0084] Further, when training the model, refer to Figure 2 , Figure 2 FIG. is a schematic flowchart of a step for obtaining the tower crane health monitoring model provided by an embodiment of the present application, where this step includes steps 201 to 204.

[0085] Step 201: Perform filtering transformation preprocessing on the second measured data and the second simulation data to obtain the processed second measured data and the processed second simulation data;

[0086] Step 202: Load the pre-constructed tower crane health monitoring model, use the processed second simulation data for training, and obtain the first tower crane health monitoring model after training;

[0087] Step 203: Use the processed second measured data to perform an effect test on the first tower crane health monitoring model to obtain the corresponding test results, where the test results include passing the test and failing the test;

[0088] Step 204, when it is determined that the test result is a pass, perform data fusion processing on the first test data, the second measured data, the first simulation data, and the second simulation data to obtain corresponding fusion data, and perform model fine-tuning processing on the first tower crane health monitoring model based on the fusion data to obtain the tower crane health monitoring model.

[0089] Specifically, during training, first use the second measured data and the second simulation data to perform training and testing processing on the constructed tower crane health monitoring model. After the test passes, then use the fusion data obtained by performing fusion processing on the first measured data, the second measured data, the first simulation data, and the second simulation data to perform model fine-tuning and optimization processing, and then obtain a usable tower crane health monitoring model after the fine-tuning and optimization processing.

[0090] Exemplarily, when using the obtained relevant data for model training, it is necessary to ensure the consistency and accuracy of the data. Through filtering and transformation processing on the second measured data and the second simulation data, obtain the processed second measured data and the processed second simulation data. Then use the processed second simulation data to train the constructed tower crane health monitoring model, and obtain the first tower crane health monitoring model after training. Then use the second measured data to perform an effect test on the first tower crane health monitoring model, and obtain the second tower crane health monitoring model after passing the effect test. At the same time, perform fusion processing on the first measured data, the second measured data, the first simulation data, and the second simulation data based on the differences between the data, and perform model fine-tuning and optimization processing on the first tower crane health monitoring model based on the fusion data obtained by the fusion processing to obtain a usable tower crane health monitoring model.

[0091] Based on the above description, the entire training process is divided into three stages, including the first stage of training based on the second simulation data, the second stage of effect verification based on the second measured data, and the third stage of model fine-tuning and optimization based on the fusion data. Among them, the first stage is to achieve the identification of fault states, the second stage is to identify faults closer to the actual situation, and the third stage is to achieve the localization of faults.

[0092] In practical applications, due to the lack or small amount of measured data in case of faults, the accuracy of training cannot be guaranteed. However, there are many combinations of fault parameters, so there will be many simulated fault conditions during the simulation process, and thus a large amount of second simulated data (the operating data of the tower crane in the fault state) will be obtained. This can better and more completely train the model and improve the accuracy of the model. However, since the training is based on a large amount of simulated data, it is necessary to combine the measured data to verify the effectiveness of the model. Therefore, a second-stage processing is required. The model obtained in the first stage is verified for effectiveness through the second measured data, so that the verified model has better accuracy and is more in line with the actual tower crane. Finally, since the faults of the tower crane are reflected in the data differences, that is, the location of the faults can be located by reading and analyzing the data. Therefore, by fusing the measured data (including the first measured data and the second measured data) and the simulated data (including the first simulated data and the second simulated data), and combining the differential features between the data, the fault location function of the model can be realized.

[0093] Further, in the process of the first-stage processing, that is, training the first tower crane health monitoring model based on the second simulated data, the specific process includes: dividing the processed second simulated data into training set data and validation set data, and training the pre-constructed tower crane health monitoring model based on the training set data to obtain an intermediate model; using the validation set data to fine-tune the intermediate model, and obtaining the first tower crane health monitoring model when the fine-tuning is completed.

[0094] When training based on the second simulated data, the processed second simulated data is divided into training set data and validation set data, and then the training set data is used for training and the validation set data is used for validation processing. The difference from the traditional model training process in this process is that in the subsequent test process, the second measured data obtained from the actual test is used. Therefore, for the processed second simulated data, it is only divided into training set data and validation set data, and the specific ratio can be 9:1.

[0095] Further, when using the second measured data to test the effect of the model, reference can be made to Figure 3 , Figure 3 which is a schematic flow chart of the steps for obtaining the test results provided by the embodiment of the present application, specifically including steps 301 to 303.

[0096] Step 301, input the fault data contained in the processed second measured data into the first tower crane health monitoring model, and output the predicted data;

[0097] Step 302: Compare the predicted data with the standard data included in the processed second measured data to determine the prediction effect of the first tower crane health monitoring model;

[0098] Step 303: Obtain the test result of the first tower crane health monitoring model according to the test effect;

[0099] Among them, if the test accuracy rate in the test effect is greater than or equal to the preset threshold, it is determined that the test passes; if the test accuracy rate in the test effect is less than the preset threshold, it is determined that the test fails.

[0100] Specifically, when conducting the effect test, it is to determine whether the trained first tower crane health monitoring model is close to the actual usage of the tower crane. Therefore, the second measured data is used to conduct the effect test on the first tower crane health monitoring model, and then whether to re-train and optimize the model is determined according to the effect test result.

[0101] Exemplarily, when using the second measured data to conduct the effect test on the first tower crane health monitoring model, the fault data included in the second measured data is input into the first tower crane health monitoring model, and the predicted data corresponding to the fault data is output. At the same time, the fault data corresponds to a standard data in the second measured data. Therefore, by comparing the predicted data with the standard data, the prediction effect of the first tower crane health monitoring model is obtained, and the test effect is used to determine the test result of the first tower crane health monitoring model, including passing the test and failing the test.

[0102] In practical applications, when it is determined that the test passes, it indicates that the fault monitoring and analysis performed by the obtained first tower crane health monitoring model conform to the actual operation of the tower crane. Then, at this time, the third stage will be entered, that is, to fine-tune and optimize the first tower crane health monitoring model so that the fine-tuned and optimized first tower crane health model can achieve fault location.

[0103] Further, when conducting the effect test on the first tower crane health monitoring model, if the test fails, relevant processes such as re-training need to be carried out. Specifically, Figure 4 , Figure 4 is another schematic flowchart of the steps of the training method of the tower crane health monitoring model provided by the embodiments of the present application, where the steps include Step 401 to Step 403.

[0104] Step 401: When it is determined that the test fails, obtain the second fault parameter, and perform simulation adjustment on the first simulation tower crane based on the second fault parameter to obtain the third simulation tower crane;

[0105] Step 402: Simulate the operation of the third simulation tower crane to obtain the third simulation data under the second fault parameter;

[0106] Step 403: Train the pre-constructed tower crane health monitoring model based on the first measured data, the second measured data, the first simulation data, and the third simulation data, and obtain the trained tower crane health monitoring model when the training is completed.

[0107] In one embodiment, during the test, if the test result is passed, it is determined that the obtained first tower crane health monitoring model can closely monitor and predict the tower crane under real faults. When the test result is not passed, it indicates that the obtained first tower crane health monitoring model cannot fit the actual usage situation to monitor and predict faults. Therefore, corresponding adjustment processing is required to obtain a tower crane health monitoring model that can fit the actual usage situation of the tower crane for fault monitoring and prediction.

[0108] Exemplarily, when it is determined that the test is not passed, the method used at this time is to retrain the model to obtain a new first tower crane health monitoring model. To obtain the new first tower crane health monitoring model, a new set of fault parameters will be used to set up the simulated tower crane at this time. Specifically, obtain the second fault parameters, and based on the second fault parameters, perform simulation adjustment on the first simulated tower crane, that is, simulation setting, to obtain the third simulated tower crane. The second fault parameters are different from the first fault parameters. Then, simulate the operation of the third simulated tower crane to obtain the third simulation data under the second fault parameters. Furthermore, based on the newly obtained third simulation data, retrain the model, and then obtain the trained tower crane health monitoring model when the training is completed.

[0109] Based on the above description, when the test is not passed, by resetting the fault parameters (second fault parameters) for the first simulated tower crane and performing simulated operation, the simulation data under the new fault parameters is obtained for model training. When training the model, the difference from step 105 is that the second simulation data used in step 105 is replaced by the third simulation data obtained in step 402, and the other data used for training is the same, and the training process is also the same. That is, the entire training process is a cyclic process. Only when the test effect is good can the training in the third stage described above be carried out. Otherwise, it is necessary to continuously adjust the fault parameters to achieve a good test effect.

[0110] It should be noted that since the training processes of step 403 and step 105 are the same, they will not be elaborated here. Specifically, refer to the description of the relevant embodiments in step 105.

[0111] Furthermore, when processing based on the above-described tower crane health monitoring model training method, the following steps are included:

[0112] Step 1: Data acquisition;

[0113] Install corresponding sensor devices on each structural part of the real tower crane, and obtain the measured data of the vibration displacement of tower crane A, etc. by collecting sensor data, including the first measured data in the fault-free state and the second measured data in the fault state.

[0114] Step 2: Simulate the tower crane to obtain the first simulated tower crane and obtain the first simulation data of the first simulated tower crane;

[0115] Use simulation software such as ANSYS to simulate the real tower crane, and construct a finite element model of the real tower crane. Among them, the constructed simulation model needs to maintain consistency with the real tower crane in terms of structure and materials. At the same time, the element type can be set to select the space beam element, and the first simulation data of the simulated tower crane in the fault-free state can be obtained through simulation operation.

[0116] Step 3: Data preprocessing;

[0117] Perform filtering transformation preprocessing on the measured data and the first simulation data.

[0118] Step 4: Calculate the similarity between the first simulated tower crane and the real tower crane, and determine the difference between the first simulated tower crane and the real tower crane according to the similarity;

[0119] Calculate the similarity value between the first measured data and the first simulation data after filtering transformation and processing. If the similarity value is greater than or equal to the similarity threshold, it is determined that the difference is small. If the similarity value is less than the similarity threshold, it is determined that the difference is large.

[0120] Step 5: If the difference is large, correct the first simulated tower crane and execute Step 2;

[0121] Step 6: If the difference is small, add the first fault parameters to the first simulated tower crane, including the damage location, connection stiffness, and material anti-interference coefficient, etc., and perform simulation to obtain the second simulation data;

[0122] Step 7: Perform filtering transformation processing on the first simulation data;

[0123] Step 8: Train the pre-constructed tower crane health monitoring model based on the first simulation data;

[0124] Divide the first simulation data after filtering transformation processing into a training set and a validation set, and use the training set to train the pre-constructed tower crane health monitoring model, and use the validation set for validation processing;

[0125] Step 9: Effect test of the model;

[0126] When the training is completed and the verification is passed, the trained first tower crane health monitoring model is tested using the second test data to obtain a test result, where the test result includes passing the test and failing the test.

[0127] Step 10: If the test result is failing the test, update the first fault parameter and execute Step 6;

[0128] Step 11: If the test result is passing the test, fuse the first test data, the second test data, the first simulation data, and the second simulation data;

[0129] Step 12: Use the fused data obtained by the fusion to fine-tune and optimize the first tower crane health monitoring model, so that the tower crane health monitoring model obtained after the fine-tuning and optimization can identify and locate faults.

[0130] In summary, the present application discloses a method for training a tower crane health monitoring model. A number of sensors are provided on the tower crane to monitor and collect data on various structural parts of the tower crane. When training the model, first obtain the first measured data in the fault-free state and the second measured data in the faulty state of the tower crane, and perform simulation processing on the tower crane to obtain the first simulated tower crane. At the same time, perform simulated operation to obtain the first simulation data of the simulated tower crane. Then, according to the first simulation data and the first measured data, determine the rationality of the first simulated tower crane, that is, determine whether it is necessary to correct and adjust the first simulated tower crane. When no correction and adjustment are required, set the corresponding first fault parameter on the first simulated tower crane to simulate the operation of the tower crane in the faulty state to obtain the second simulation data. Finally, train the constructed tower crane health monitoring model according to the first measured data, the second measured data, the first simulation data, and the second simulation data, and then obtain the trained tower crane health monitoring model when the training is completed. It is realized that during the training process, a large number of training samples for training are obtained through the simulation processing of the tower crane, and then the effectiveness of the model is judged using the measured data after training based on the training samples, so that the model can identify faults more closely to the real tower crane. By combining the training of the measured data and the simulation data in the fault-free state and the faulty state, the location of the fault can be realized. Furthermore, during the use process, the existing faults can be identified and located only by obtaining the sensor data, improving the convenience and accuracy of the tower crane health monitoring.

[0131] According to the method described in the above embodiments, this embodiment will further describe from the perspective of the training device of the tower crane health monitoring model. The training device of the tower crane health monitoring model can be specifically implemented as an independent entity, or can be integrated in an electronic device, such as a terminal. The terminal can include a mobile phone, a tablet computer, etc.

[0132] Please refer to Figure 5 , Figure 5 which is a schematic structural diagram of a training device for a tower crane health monitoring model provided by an embodiment of the present application. As Figure 5 shown, the training device 500 for the tower crane health monitoring model provided by the embodiment of the present application includes:

[0133] A data acquisition module 501, configured to acquire first measured data and second measured data of the tower crane. Among them, the first measured data is health status data collected based on a plurality of sensors arranged on the tower crane, and the second measured data is fault status data collected based on a plurality of sensors arranged on the tower crane;

[0134] A first simulation module 502, configured to simulate the tower crane to obtain a first simulated tower crane, and perform a simulated operation on the first simulated tower crane to obtain first simulation data of the first simulated tower crane in a healthy state;

[0135] A simulation adjustment module 503, configured to determine whether to perform simulation adjustment on the first simulated tower crane according to the first measured data and the first simulation data;

[0136] A second simulation module 504, configured to, when it is determined not to perform simulation adjustment on the first simulated tower crane, perform simulation adjustment on the first simulated tower crane based on the first fault parameter to obtain a second simulated tower crane, and perform a simulated operation on the second simulated tower crane to obtain second simulation data;

[0137] A model training module 505, configured to train a pre-constructed tower crane health monitoring model according to the first measured data, the second measured data, the first simulation data, and the second simulation data, and obtain a trained tower crane health monitoring model when the training is completed.

[0138] In an embodiment, the simulation adjustment module 503 is further configured to:

[0139] Perform filtering transformation preprocessing on the first measured data and the first simulation data to obtain processed first measured data and processed first simulation data;

[0140] Calculate the data similarity between the processed first measured data and the processed first simulation data, and compare the data similarity with a similarity threshold;

[0141] Determine whether to adjust the first simulated tower crane according to the comparison result obtained from the comparison;

[0142] If the data similarity is less than the similarity threshold, it is determined to perform simulation adjustment on the first simulated tower crane;

[0143] If the data similarity is greater than or equal to the similarity threshold, it is determined not to perform simulation adjustment on the first simulated tower crane.

[0144] In one embodiment, the second simulation module 504 is further configured to:

[0145] When it is determined that no simulation adjustment is to be made to the first simulated tower crane, load a fault parameter generation model to output first fault parameters corresponding to the tower crane;

[0146] Perform simulation adjustment on the first simulated tower crane according to the first fault parameters to obtain a second simulated tower crane, and perform simulated operation on the second simulated tower crane to obtain second simulation data under the first fault parameters.

[0147] In one embodiment, the model training module 505 is further configured to:

[0148] Perform filtering transformation preprocessing on the second measured data and the second simulation data to obtain processed second measured data and processed second simulation data;

[0149] Load a pre-constructed tower crane health monitoring model, use the processed second simulation data for training, and obtain a first tower crane health monitoring model after training;

[0150] Use the processed second measured data to perform an effect test on the first tower crane health monitoring model to obtain corresponding test results, where the test results include passing the test and failing the test;

[0151] When it is determined that the test result is passing the test, perform data fusion processing on the first test data, the second measured data, the first simulation data, and the second simulation data to obtain corresponding fusion data, and perform model fine-tuning processing on the first tower crane health monitoring model based on the fusion data to obtain a tower crane health monitoring model.

[0152] In one embodiment, the model training module 505 is further configured to:

[0153] Divide the processed second simulation data into training set data and validation set data, and train a pre-constructed tower crane health monitoring model based on the training set data to obtain an intermediate model;

[0154] Use the validation set data to perform model fine-tuning on the intermediate model, and obtain a first tower crane health monitoring model when the fine-tuning is completed.

[0155] In one embodiment, the model training module 505 is further configured to:

[0156] Input the fault data included in the processed second measured data into the first tower crane health monitoring model, and output prediction data;

[0157] Compare the prediction data with the standard data included in the processed second measured data to determine the prediction effect of the first tower crane health monitoring model;

[0158] According to the test results, the test results of the first tower crane health monitoring model are obtained;

[0159] If the test accuracy rate in the test results is greater than or equal to the preset threshold, it is determined that the test passes;

[0160] If the test accuracy rate in the test results is less than the preset threshold, it is determined that the test fails.

[0161] In one embodiment, the model training module 505 is further configured to:

[0162] When it is determined that the test result is that the test fails, obtain the second fault parameter, and perform simulation adjustment on the first simulated tower crane based on the second fault parameter to obtain the third simulated tower crane;

[0163] Perform simulated operation on the third simulated tower crane to obtain the third simulated data under the second fault parameter;

[0164] Train the pre-constructed tower crane health monitoring model according to the first measured data, the second measured data, the first simulated data and the third simulated data, and obtain the trained tower crane health monitoring model when the training is completed.

[0165] In addition, please refer to Figure 6 , Figure 6 which is a schematic structural diagram of an electronic device provided by an embodiment of the present application. The electronic device may be a mobile terminal such as a smart phone, a tablet computer, etc. As Figure 6 shown, the electronic device 600 includes a processor 601 and a memory 602. Among them, the processor 601 is electrically connected to the memory 602.

[0166] The processor 601 is the control center of the electronic device 600, connects various parts of the entire electronic device through various interfaces and lines, executes various functions of the electronic device 600 and processes data by running or loading application programs stored in the memory 602, and calling data stored in the memory 602, so as to perform overall monitoring on the electronic device 600.

[0167] In this embodiment, the processor 601 in the electronic device 600 will load the instructions corresponding to the processes of one or more application programs into the memory 602 according to the following steps, and the processor 601 will run the application programs stored in the memory 602, so as to implement any step in the tower crane health monitoring model training method provided in the above embodiment.

[0168] The electronic device 600 can implement the steps in any of the embodiments of the training method of the tower crane health monitoring model provided in the embodiments of the present application. Therefore, it can achieve the beneficial effects that can be achieved by any of the training methods of the tower crane health monitoring model provided in the embodiments of the present application. For details, please refer to the previous embodiments and will not be repeated here.

[0169] Please refer to Figure 7 , Figure 7 which is another structural schematic diagram of the electronic device provided in the embodiments of the present application. As Figure 7 shown, Figure 7 it shows the specific structural block diagram of the electronic device provided in the embodiments of the present application. The electronic device 700 can be used to implement the training method of the tower crane health monitoring model provided in the above embodiments. The electronic device 700 can be a mobile terminal such as a smart phone or a laptop computer, etc.

[0170] The RF circuit 710 is used to receive and transmit electromagnetic waves, realizing the mutual conversion between electromagnetic waves and electrical signals, so as to communicate with a communication network or other devices. The RF circuit 710 may include various existing circuit elements for performing these functions. For example, an antenna, a radio frequency transceiver, a digital signal processor, an encryption / decryption chip, a subscriber identity module (SIM) card, a memory, and so on. The RF circuit 710 can communicate with various networks such as the Internet, an enterprise intranet, a wireless network or communicate with other devices through a wireless network. The above-mentioned wireless network may include a cellular phone network, a wireless local area network or a metropolitan area network. The above-mentioned wireless network can use various communication standards, protocols and technologies, including but not limited to Global System for Mobile Communication (GSM), Enhanced Data GSM Environment (EDGE), Wideband Code Division Multiple Access (WCDMA), Code Division Access (CDMA), Time Division Multiple Access (TDMA), Wireless Fidelity (Wi-Fi) (such as Institute of Electrical and Electronics Engineers standards IEEE 802.11a, IEEE 802.11b, IEEE 802.11g and / or IEEE 802.11n), Voice over Internet Protocol (VoIP), Worldwide Interoperability for Microwave Access (Wi-Max), other protocols for email, instant messaging and short messages, and any other suitable communication protocols, and may even include those protocols that have not yet been developed currently.

[0171] The memory 720 can be used to store software programs and modules, such as the program instructions / modules corresponding to the training method of the tower crane health monitoring model in the above embodiments. The processor 780 executes various functional applications and the training method of the tower crane health monitoring model by running the software programs and modules stored in the memory 720.

[0172] The memory 720 may include high-speed random access memory and may also include non-volatile memory, such as one or more magnetic storage devices, flash memory, or other non-volatile solid-state memory. In some examples, the memory 720 may further include a memory remotely located with respect to the processor 780, and these remote memories may be connected to the electronic device 700 through a network. Examples of the above network include but are not limited to the Internet, an intranet, a local area network, a mobile communication network, and combinations thereof.

[0173] The input unit 730 can be used to receive uploaded digital or character information and generate keyboard, mouse, joystick, optical, or trackball signal inputs related to user settings and function controls. Specifically, the input unit 730 may include a touch-sensitive surface 731 and other input devices 732. The touch-sensitive surface 731, also known as a touch display screen or a touchpad, can collect touch operations of a user thereon or nearby (such as operations of the user using a finger, a stylus, or any suitable object or accessory on or near the touch-sensitive surface 731), and drive a corresponding connection device according to a pre-set program. Optionally, the touch-sensitive surface 731 may include two parts: a touch detection device and a touch controller. Among them, the touch detection device detects the touch orientation of the user and detects the signal brought by the touch operation, and transmits the signal to the touch controller; the touch controller receives the touch information from the touch detection device, converts it into contact coordinates, and then sends it to the processor 780, and can also receive commands sent by the processor 780 and execute them. In addition, various types such as resistive, capacitive, infrared, and surface acoustic wave can be used to implement the touch-sensitive surface 731. In addition to the touch-sensitive surface 731, the input unit 730 may further include other input devices 732. Specifically, the other input devices 732 may include but are not limited to one or more of a physical keyboard, function keys (such as volume control keys, switch keys, etc.), a trackball, a mouse, a joystick, etc.

[0174] The display unit 740 can be used to display information input by the user or information provided to the user, as well as various graphical user interfaces of the electronic device 700. These graphical user interfaces can be composed of graphics, text, icons, videos, and any combination thereof. The display unit 740 may include a display panel 741. Optionally, the display panel 741 can be configured in the form of an LCD (Liquid Crystal Display) or an OLED (Organic Light-Emitting Diode). Further, the touch-sensitive surface 731 can cover the display panel 741. When the touch-sensitive surface 731 detects a touch operation on or near it, it is transmitted to the processor 780 to determine the type of touch event. Subsequently, the processor 780 provides a corresponding visual output on the display panel 741 according to the type of touch event. Although in the figure, the touch-sensitive surface 731 and the display panel 741 are implemented as two independent components to perform input and output functions, in some embodiments, the touch-sensitive surface 731 and the display panel 741 can be integrated to implement input and output functions.

[0175] The electronic device 700 may further include at least one sensor 750, such as a light sensor, a motion sensor, and other sensors. Specifically, the light sensor may include an ambient light sensor and a proximity sensor. Among them, the ambient light sensor can adjust the brightness of the display panel 741 according to the brightness of the ambient light, and the proximity sensor can generate an interruption when the flip cover is closed or opened. As a kind of motion sensor, the gravity acceleration sensor can detect the magnitude of acceleration in all directions (generally three axes). When stationary, it can detect the magnitude and direction of gravity, and can be used in applications for identifying the posture of the mobile phone (such as horizontal and vertical screen switching, related games, magnetometer posture calibration), vibration recognition related functions (such as pedometer, tapping), etc. As for other sensors such as gyroscopes, barometers, hygrometers, thermometers, and infrared sensors that the electronic device 700 can also be configured with, they will not be elaborated here.

[0176] The audio circuit 760, the speaker 761, and the microphone 762 can provide an audio interface between the user and the electronic device 700. The audio circuit 760 can transmit the electrical signal converted from the received audio data to the speaker 761, and the speaker 761 converts it into a sound signal for output. On the other hand, the microphone 762 converts the collected sound signal into an electrical signal, which is received by the audio circuit 760 and then converted into audio data. After the audio data is output to the processor 780 for processing, it is sent through the RF circuit 710 to, for example, another terminal, or the audio data is output to the memory 720 for further processing. The audio circuit 760 may also include an earphone jack to provide communication between the peripheral earphone and the electronic device 700.

[0177] The electronic device 700 can assist the user in receiving requests, sending information, etc. through a transmission module 770 (such as a Wi-Fi module), providing the user with wireless broadband Internet access. Although the transmission module 770 is shown in the figure, it can be understood that it does not belong to the essential components of the electronic device 700 and can be omitted entirely within the scope of not changing the essence of the invention as needed.

[0178] The processor 780 is the control center of the electronic device 700, connecting various parts of the entire mobile phone through various interfaces and circuits. By running or executing software programs and / or modules stored in the memory 720, and by invoking data stored in the memory 720, it executes various functions of the electronic device 700 and processes data, thereby monitoring the electronic device as a whole. Optionally, the processor 780 may include one or more processing cores; in some embodiments, the processor 780 may integrate an application processor and a modem processor. Among them, the application processor mainly processes the operating system, user interface, application programs, etc., and the modem processor mainly processes wireless communication. It can be understood that the above-mentioned modem processor may not be integrated into the processor 780 either.

[0179] The electronic device 700 also includes a power source 790 (such as a battery) that powers each component. In some embodiments, the power source can be logically connected to the processor 780 through a power management system, thereby implementing functions such as management of charging, discharging, and power consumption management through the power management system. The power source 790 may also include any components such as one or more DC or AC power sources, a recharge system, a power failure detection circuit, a power converter or inverter, and a power status indicator.

[0180] Although not shown, the electronic device 700 also includes a camera (such as a front camera and a rear camera), a Bluetooth module, etc., which will not be elaborated here. Specifically, in this embodiment, the display unit of the electronic device is a touch screen display, and the mobile terminal also includes a memory and one or more programs, where one or more programs are stored in the memory and are configured to be executed by one or more processors to implement any step in the training method of the tower crane health monitoring model provided in the above embodiment.

[0181] During specific implementation, the above-mentioned modules can be implemented as independent entities, or can be combined arbitrarily and implemented as the same or several entities. For the specific implementation of the above-mentioned modules, reference can be made to the method embodiments described above, which will not be elaborated here.

[0182] Those of ordinary skill in the art can understand that all or part of the steps in the various methods of the above embodiments can be completed by instructions or by controlling relevant hardware through instructions. The instructions can be stored in a computer-readable storage medium and loaded and executed by a processor. For this purpose, an embodiment of the present application provides a storage medium in which multiple instructions are stored. When the instructions are executed by a processor, any step in the training method of the tower crane health monitoring model provided in the above embodiments can be implemented.

[0183] Among them, the storage medium may include: read-only memory (ROM, Read Only Memory), random access memory (RAM, Random Access Memory), magnetic disk or optical disc, etc.

[0184] Since the instructions stored in the storage medium can execute the steps in any embodiment of the training method of the tower crane health monitoring model provided in the embodiments of the present application, the beneficial effects that can be achieved by any training method of the tower crane health monitoring model provided in the embodiments of the present application can be realized. For details, see the previous embodiments and will not be repeated here.

[0185] The above has introduced in detail a training method, device, electronic device and storage medium of a tower crane health monitoring model provided in the embodiments of the present application. Specific examples are used in this article to elaborate on the principle and implementation manner of the present application. The description of the above embodiments is only used to help understand the method and its core idea of the present application; at the same time, for those skilled in the art, according to the idea of the present application, there will be changes in the specific implementation manner and application scope. In summary, the content of this specification should not be construed as a limitation to the present application. Moreover, for those of ordinary skill in the technical field, without departing from the principle of the present application, several improvements and refinements can be made, and these improvements and refinements are also regarded as the protection scope of the present application.

Claims

1. A training method for a tower crane health monitoring model, characterized in that, Including: Obtain the first measured data and the second measured data of the tower crane. Among them, the first measured data is the health status data collected based on a number of sensors arranged on the tower crane, and the second measured data is the fault status data collected based on a number of sensors arranged on the tower crane; Simulate the tower crane to obtain a first simulated tower crane, and simulate the operation of the first simulated tower crane to obtain the first simulation data of the first simulated tower crane in a healthy state; Perform data differential calculation on the first measured data and the first simulation data, and determine whether to adjust the simulation parameters of the first simulated tower crane according to the calculated data differential result. Among them, the simulation parameters used for adjustment are the parameters when the tower crane is in a normal state; When it is determined not to adjust the simulation of the first simulated tower crane, perform simulation adjustment on the first simulated tower crane based on the first fault parameter to obtain a second simulated tower crane, and simulate the operation of the second simulated tower crane to obtain second simulation data; Train a pre-constructed tower crane health monitoring model according to the first measured data, the second measured data, the first simulation data and the second simulation data, and obtain a trained tower crane health monitoring model when the training is completed.

2. The method according to claim 1, wherein The determining whether to adjust the simulation of the first simulated tower crane according to the first measured data and the first simulation data includes: Perform filter transformation preprocessing on the first measured data and the first simulation data to obtain the processed first measured data and the processed first simulation data; Calculate the data similarity between the processed first measured data and the processed first simulation data, and compare the data similarity with a similarity threshold; Determine whether to adjust the first simulated tower crane according to the comparison result obtained from the comparison; If the data similarity is less than the similarity threshold, it is determined to adjust the simulation of the first simulated tower crane; If the data similarity is greater than or equal to the similarity threshold, it is determined not to adjust the simulation of the first simulated tower crane.

3. The method according to claim 1, wherein The performing simulation adjustment on the first simulated tower crane based on the first fault parameter to obtain a second simulated tower crane and simulating the operation of the second simulated tower crane to obtain second simulation data when it is determined not to adjust the simulation of the first simulated tower crane includes: When it is determined not to adjust the simulation of the first simulated tower crane, load a fault parameter generation model to output the first fault parameter corresponding to the tower crane; Perform simulation adjustment on the first simulated tower crane according to the first fault parameter to obtain a second simulated tower crane, and simulate the operation of the second simulated tower crane to obtain the second simulation data under the first fault parameter.

4. The method according to claim 1, wherein The training a pre-constructed tower crane health monitoring model according to the first measured data, the second measured data, the first simulation data and the second simulation data and obtaining a trained tower crane health monitoring model when the training is completed includes: Perform filtering transformation preprocessing on the second measured data and the second simulation data to obtain the processed second measured data and the processed second simulation data; Load the pre-constructed tower crane health monitoring model, use the processed second simulation data for training, and obtain the first tower crane health monitoring model after training; Use the processed second measured data to perform an effect test on the first tower crane health monitoring model to obtain corresponding test results, where the test results include passing the test and failing the test; When it is determined that the test result is passing the test, perform data fusion processing on the first measured data, the second measured data, the first simulation data, and the second simulation data to obtain corresponding fusion data, and perform model fine-tuning processing on the first tower crane health monitoring model based on the fusion data to obtain the tower crane health monitoring model.

5. The method according to claim 4, wherein The using the processed second simulation data for training and obtaining the first tower crane health monitoring model after training includes: Divide the processed second simulation data into training set data and validation set data, and train the pre-constructed tower crane health monitoring model based on the training set data to obtain an intermediate model; Use the validation set data to perform model fine-tuning on the intermediate model, and obtain the first tower crane health monitoring model when the fine-tuning is completed.

6. The method according to claim 4, characterized in that The using the processed second measured data to perform an effect test on the first tower crane health monitoring model and obtaining corresponding test results includes: Input the fault data included in the processed second measured data into the first tower crane health monitoring model, and output the predicted data; Compare the predicted data with the standard data included in the processed second measured data to determine the prediction effect of the first tower crane health monitoring model; Obtain the test result of the first tower crane health monitoring model according to the test effect; If the test accuracy rate in the test effect is greater than or equal to the preset threshold, it is determined that the test is passed; If the test accuracy rate in the test effect is less than the preset threshold, it is determined that the test is not passed.

7. The method according to claim 4, wherein The method further includes: When it is determined that the test result is not passing the test, obtain the second fault parameter, and perform simulation adjustment on the first simulated tower crane based on the second fault parameter to obtain the third simulated tower crane; Perform simulated operation on the third simulated tower crane to obtain the third simulation data under the second fault parameter; Train the pre-constructed tower crane health monitoring model according to the first measured data, the second measured data, the first simulation data, and the third simulation data, and obtain the trained tower crane health monitoring model when the training is completed.

8. A training device for a tower crane health monitoring model, characterized in that, including: A data acquisition module for acquiring the first measured data and the second measured data of the tower crane, where the first measured data is the health status data collected based on a plurality of sensors arranged on the tower crane, and the second measured data is the fault status data collected based on a plurality of sensors arranged on the tower crane; The first simulation module is used to simulate the tower crane to obtain a first simulated tower crane, and simulate the operation of the first simulated tower crane to obtain first simulation data of the first simulated tower crane in a healthy state; The simulation adjustment module is used to calculate the data difference between the first measured data and the first simulation data, and determine whether to adjust the simulation parameters of the first simulated tower crane according to the calculated data difference result. The simulation parameters used for adjustment are the parameters of the tower crane in a normal state; The second simulation module is used to, when it is determined not to adjust the simulation of the first simulated tower crane, perform simulation adjustment on the first simulated tower crane based on the first fault parameter to obtain a second simulated tower crane, and simulate the operation of the second simulated tower crane to obtain second simulation data; The model training module is used to train a pre-constructed tower crane health monitoring model according to the first measured data, the second measured data, the first simulation data, and the second simulation data, and obtain a trained tower crane health monitoring model when the training is completed.

9. An electronic device, characterized in that, The electronic device includes a processor, a memory, and a computer program stored in the memory and executable on the processor. When the processor executes the computer program, the steps in the method according to any one of claims 1 to 7 are implemented.

10. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, the steps in the method according to any one of claims 1 to 7 are implemented.

Images