A method and system for obtaining stress history of a bridge crane

Abstract

The application discloses a kind of bridge crane stress history acquisition method and system, the method includes obtaining the characteristic data of crane actual working state under preset time;The hoisting load data of preset time is expanded by Latin hypercube sampling technique, obtain the hoisting load data of crane inspection cycle, the hoisting load data of inspection cycle is input into the pre-trained tentacle-least squares support vector machine prediction model, and the predicted working cycle number is obtained;The hoisting load data of inspection cycle and the predicted working cycle number are converted into the trolley wheel pressure of equal sample size;The finite element model of crane girder is established, and the trolley is simulated with random load at rated speed from the side span of girder to the other side span, after traversing all trolley wheel pressure load, the stress history of crane girder at any position in inspection cycle is obtained.The application provides high-precision stress characteristic data for crane structure fatigue life assessment.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of crane stress acquisition, in particular to a bridge crane stress history acquisition method and system. BACKGROUND

[0002] The bridge crane is a kind of special equipment, which is widely used in various production workshops and related industrial scenes. Due to the randomness, intermittence and contingency of the working load, the fatigue performance of the main beam will gradually degrade, and the structure will be broken suddenly. In order to prevent such accidents, accurately acquiring the stress history of the key position of the structure is the key to fatigue life assessment and early prevention.

[0003] The existing method only calculates the stress change of the specific dangerous section and dangerous point under a certain or limited working cycle, and ignores the working characteristics of the crane trolley under random load for a long time, and also ignores the influence of the irregular stress change of the trolley during the load running process on the main beam structure, which leads to the increase of stress cycle number under single working cycle. Therefore, it is unreasonable to analyze the stress change of the crane structure by only bearing a specific load for a short time, which will cause great error in the later fatigue life assessment. SUMMARY

[0004] The purpose of the present application is to provide a bridge crane stress history acquisition method and system to solve the problem of poor accuracy of stress data used in the fatigue life assessment of the crane in the prior art.

[0005] To achieve the above purpose, the present application adopts the following technical scheme:

[0006] In a first aspect, the present application discloses a bridge crane stress history acquisition method, comprising:

[0007] Obtaining characteristic data of a preset time under the actual working state of the crane, wherein the characteristic data includes lifting load data and working cycle number data, and the preset time is less than the inspection period;

[0008] Extending the lifting load data of the preset time by Latin hypercube sampling technology to obtain the lifting load data of the inspection period of the crane, and inputting the lifting load data of the inspection period into the pre-trained tentacle-least squares support vector machine prediction model to obtain the predicted working cycle number;

[0009] Converting the lifting load data and the predicted working cycle number of the inspection period into trolley wheel pressure with equal sample size;

[0010] A finite element model of the main girder of the crane is established, a macro command is written, and the transient analysis is simulated by combining the random load of the trolley to run at the rated speed from one side of the main girder to the other side of the main girder, and after all the trolley wheel pressure loads are traversed, the stress history of the main girder of the crane at any position in the periodic inspection cycle is obtained.

[0011] Further, the characteristic data of the crane in the actual working state within the preset time comprises:

[0012] The lifting load data of the crane for 30 days is collected through the load sensor arranged at the support position of the lifting drum of the trolley of the crane.

[0013] The working cycle number data of the crane for 30 days is obtained, wherein the reading of the load sensor starts from 0 to 0 again for one working cycle.

[0014] Further, the training method of the tentacle-least squares support vector machine prediction model comprises:

[0015] The hyperparameter selection in the least squares support vector machine model is optimized through the tentacle search optimization algorithm, and an original tentacle-least squares support vector machine prediction model is built.

[0016] The characteristic data within the preset time is divided into training set data and test set data;

[0017] The training set data is input into the original tentacle-least squares support vector machine prediction model, and an intermediate tentacle-least squares support vector machine prediction model is obtained.

[0018] The intermediate tentacle-least squares support vector machine prediction model is optimized and corrected through the test set, and a final tentacle-least squares support vector machine prediction model is obtained.

[0019] Further, the method for obtaining the equal sample size of the trolley wheel pressure comprises:

[0020] The lifting load data in the periodic inspection cycle and the predicted working cycle number are randomized by using matlab;

[0021] The equal sample size of the trolley wheel pressure is obtained through the lifting load data after randomization and the self weight of the trolley.

[0022] Further, the stress history of the main girder of the crane at any position in the periodic inspection cycle comprises the following steps:

[0023] A finite element model of the main girder of the crane is established, and a macro command for reading the wheel pressure, running speed, cyclic loading and damping setting is written;

[0024] Based on the finite element model of the main girder of the crane, the FULL complete method is selected for solving, the Rayleigh damping parameter and the step loading mode are defined, and the transient analysis setting is realized.

[0025] The wheel pressure data of the inspection cycle is input into the finite element model of the main girder of the crane by using the macro command of reading wheel pressure, the initial position of the trolley is selected, the first wheel pressure load is applied, the running speed macro command is used to control the trolley to run at the rated speed to the other side span end of the main girder, and one working cycle is completed;

[0026] The finite element simulation of the random load running process of the crane trolley in the inspection cycle is completed by using the loop loading macro command to loop through all the wheel pressure loads;

[0027] According to the fatigue position coordinates to be analyzed, the stress history data of the corresponding nodes in the inspection cycle are extracted by combining the node selection command.

[0028] Further, the reading wheel pressure macro command is written as: defining the wheel pressure matrix size, and automatically writing specific wheel pressure values by combining the loop command.

[0029] Further, the running speed macro command is written as: according to the running speed of the trolley and the grid size, the number of grids advanced per second is calculated, and the node position loaded with wheel pressure per second is controlled by using the loop command.

[0030] Further, the loop loading macro command is written as: the trolley is controlled to run from one side span end of the main girder to the other side span end in seconds by using the time step and the node selection command, and all wheel pressure data in the inspection cycle is applied by combining loop traversal.

[0031] Further, the macro command of damping setting is: according to the maximum and minimum frequency values of the modal analysis result, the Rayleigh damping two parameter values are calculated and set in the transient analysis.

[0032] Secondly, the application discloses a bridge crane stress history acquisition system, comprising a processor and a storage medium.

[0033] The storage medium is used for storing instructions.

[0034] The processor is used for operating according to the storage instructions to execute the steps of the method of any one of the first aspect.

[0035] According to the above technical solution, the application has the following beneficial effects:

[0036] The application is based on the characteristic data of the collected small sample in the actual working state, is expanded through the Latin hypercube sampling technology, is combined with the trained tentacle-least square support vector machine prediction model, is obtained through the collection, expansion and prediction forms, and is obtained the crane periodic inspection cycle service information conforming to the actual working condition, the stress change condition of the structure at any position in the process that the crane trolley is randomly loaded at the rated speed from one side of the girder to the other side of the girder is simulated through the establishment of the finite element model, the relevant macro command is combined, the high-precision stress characteristic data is provided for the crane structure fatigue life assessment, and at the same time, the high-precision and low-cost technical reference is provided for the high-cycle stress history acquisition of the same type of mechanical equipment. BRIEF DESCRIPTION OF DRAWINGS

[0037] Figure 1 It is the overall flow block diagram of the stress acquisition method of the application;

[0038] Figure 2 It is the training flow chart of the tentacle-least square support vector machine prediction model in the application;

[0039] Figure 3 It is the extension and prediction method schematic diagram of the bridge crane periodic inspection cycle service information in the application;

[0040] Figure 4 It is the bridge crane periodic inspection cycle lifting load and working cycle number data graph in the application;

[0041] Figure 5 It is the comparison graph of the bridge crane service information before and after randomization in the application;

[0042] Figure 6 It is the schematic diagram of the simulation of the crane trolley under the single working cycle with load in the application;

[0043] Figure 7 It is the wheel pressure load cycle traversal flow chart in the application;

[0044] Figure 8 It is the schematic diagram of the dangerous point of the crane girder structure in the middle and the end of the span in the application;

[0045] Figure 9 It is the high-precision stress history graph of the dangerous point of the crane girder structure in the application. DETAILED DESCRIPTION

[0046] In order to make the technical means, creative features, purposes and effects realized by the application easy to understand, the application is further described below in combination with specific embodiments. Embodiment 1

[0047] As Figures 1 to 9As shown, the application discloses a kind of bridge crane stress history acquisition method, comprising: obtaining the characteristic data of crane actual working state under preset time, wherein the characteristic data includes hoisting load data and working cycle number data, and preset time is less than fixed inspection period;The hoisting load data of preset time is extended by Latin hypercube sampling technique, and the hoisting load data of crane fixed inspection period is obtained, the hoisting load data of fixed inspection period is input into the pre-trained tentacle-least square support vector machine prediction model, and the predicted working cycle number is obtained;The hoisting load data of fixed inspection period and the predicted working cycle number are converted into trolley wheel pressure of equal sample size;Establish the finite element model of crane girder, write macro command, simulate trolley random load at rated speed from one side of girder to the other side of girder after traversing all trolley wheel load, and obtain the stress history of crane girder at any position in fixed inspection period.

[0048] The application is based on the characteristic data of small sample under actual working state, is extended by Latin hypercube sampling technique, is combined with the tentacle-least square support vector machine prediction model trained, is obtained by the form of collection, extension and prediction, and is obtained the crane fixed inspection period service information conforming to actual working condition, is obtained the stress change condition of structure at any position in the process that crane trolley is randomly loaded at rated speed from one side of girder to the other side of girder under fixed inspection period long range by establishing finite element model and combining relevant macro command, provides high-precision stress characteristic data for crane structure fatigue life assessment, and simultaneously provides high-precision and low-cost technical reference for high-cycle stress history acquisition of same type mechanical equipment.

[0049] Step 1, obtaining the characteristic data of crane actual working state under preset time, wherein the characteristic data includes hoisting load data and working cycle number data.

[0050] Specifically, the load spectrum acquisition system is used, the hoisting load is collected by the load sensor arranged at the support position of the hoisting drum of the crane trolley, the reading of the load sensor is recorded as a working cycle when the reading changes from 0 to 0 again, the hoisting load and the working cycle number under the actual working state of the crane for 30 days are obtained, and the small sample collection of crane service information is completed.

[0051] Step 2, the hoisting load data of preset time is extended by Latin hypercube sampling technique, and the hoisting load data of crane fixed inspection period is obtained, the hoisting load data of fixed inspection period is input into the pre-trained tentacle-least square support vector machine prediction model, and the predicted working cycle number is obtained;

[0052] Based on the 30-day lifting load sample, combined with the Latin hypercube sampling technique, the lifting load data of the crane is extended to 2 years of inspection cycle, which is input into the trained BAS-LSSVM prediction model, and the corresponding working cycle number is output to form the 2-year crane load spectrum data, that is, the service information.

[0053] The model training process of BAS-LSSVM includes:

[0054] (a) The prediction performance of the LSSVM model is determined by the selection of two hyperparameters in the model. The Broomstick search (BAS) optimization algorithm is used to optimize the selection of hyperparameters 、 In the LSSVM model to improve the accuracy of LSSVM prediction, and complete the establishment of BAS-LSSVM prediction model.

[0055] (b) Divide the 30-day crane service information small sample into training set and test set, use BAS-LSSVM prediction model to learn the implicit mapping relationship between lifting load and working cycle number in the training set, get the trained BAS-LSSVM prediction model, and use the samples in the test set to verify the model performance, ensure that the trained BAS-LSSVM prediction model can predict the crane high-precision working characteristic parameters with high fitting degree with the actual use condition.

[0056] Step 3, convert the lifting load data of the inspection cycle and the predicted working cycle number into equal sample amount of trolley wheel pressure.

[0057] Use matlab to randomize the service information working sequence, add the trolley weight and lifting load to calculate the trolley wheel pressure, and form the equal sample amount of trolley wheel pressure.

[0058] Step 4, establish the finite element model of the main beam of the crane, write macro command, combine transient analysis to simulate the random load trolley running at rated speed from one side of the main beam to the other side of the main beam, and after circulating all trolley wheel pressure loads, get the stress history of the main beam of the crane at any position in the inspection cycle.

[0059] Step 41, use APDL to establish the finite element model of the main beam of the crane, write macro command to prepare for the work of transient analysis module, including reading wheel pressure, running speed, cycle loading and damping setting four parts.

[0060] a. Read the wheel pressure: use the *dim command to define the wheel pressure matrix size, and combine the *do loop command to automatically write the specific wheel pressure value.

[0061] b. Running speed: according to the trolley running speed and grid size, calculate the number of grids advancing per second, use the *do loop command to control the node position of loading wheel pressure per second.

[0062] c. Loop loading: Using the time step time and node selection nsel command, control the trolley to run from one side of the girder to the other side of the cross section in seconds, combined with the do loop to apply all wheel pressure data for 2 years.

[0063] d. Damping setting: According to the maximum and minimum frequency values of the modal analysis results, calculate and set the Rayleigh damping two parameter values alphad and betad in the transient module.

[0064] Step 42, based on the finite element model of the crane girder, complete the material definition, meshing and boundary constraint, enter the transient analysis setting, select FULL complete method for solution, define Rayleigh damping parameters and step loading mode, complete the pre-processing.

[0065] Step 43, use the macro command to read the wheel pressure to input the trolley wheel pressure data of 2 years into the APDL program, and select the node of the side of the girder span, that is, the initial position of the trolley, to apply the first wheel pressure load, and then use the running speed macro command to control the trolley to run at the rated speed to the other side of the girder, complete a working cycle.

[0066] Step 44, use the loop loading macro command to loop through all the wheel pressure loads, as described in step 6, complete the finite element simulation of the random load running process of the crane trolley for 2 years.

[0067] Step 45, according to the fatigue position coordinates to be analyzed, combined with the node selection nsel command, extract the stress history data of the corresponding nodes for 2 years in the POST26 post-processing.

[0068] The application will be further described in detail below with reference to the accompanying drawings.

[0069] 1) Taking a 32-ton bridge crane as the target, using the load spectrum acquisition system, obtaining its service information under the actual working condition for 30 days, as shown in Table 1;

[0070] Table 1, crane 30-day working service information

[0071] Sample No. Rated Lifting Load Actual Lifting Load Number of Work Cycles Sample No. Rated Lifting Load Actual Lifting Load Number of Work Cycles 1 32 2.8 1 31 32 17.9 42 2 32 4.1 2 32 32 18.2 44 3 32 4.5 2 33 32 18.9 45 4 32 5.3 4 34 32 19.2 49 5 32 5.8 5 35 32 19.6 51 6 32 6.6 5 36 32 20 54 7 32 6.6 7 37 32 20.6 59 8 32 7.4 7 38 32 20.7 61 9 32 7.6 8 39 32 21 63 10 32 9 9 40 32 22.8 67 11 32 9.1 9 41 32 23.1 65 12 32 10.2 10 42 32 23.2 62 13 32 10.2 10 43 32 23.5 59 14 32 10.2 11 44 32 24.1 55 15 32 10.3 11 45 32 24.2 52 16 32 11.3 11 46 32 24.8 51 17 32 11.6 12 47 32 25.3 49 18 32 13 12 48 32 25.9 44 19 32 13.3 16 49 32 26 42 20 32 14 16 50 32 26.2 41 21 32 14.8 19 51 32 26.6 39 22 32 15.4 20 52 32 26.9 38 23 32 15.6 23 53 32 27.2 34 24 32 15.8 24 54 32 27.3 33 25 32 16 29 55 32 27.4 29 26 32 16.5 31 56 32 27.7 27 27 32 16.9 33 57 32 28 24 28 32 17 36 58 32 28.8 23 29 32 17.3 39 59 32 29.1 20 30 32 17.9 39 60 32 30.4 17

[0072] 2) As Figure 2As shown, the working service information data in Table 1 is divided into a training set and a test set, the test set is input into the LSSVM prediction model, the selection of the super parameter in the LSSVM is optimized in combination with the Bolzano Algorithm Search (BAS), the model training is carried out, the implicit law between the lifting load and the working cycle number in the training set is mined and the mapping relationship is formed, and the trained BAS-LSSVM prediction model is obtained; the lifting load in the test set is input into the trained BAS-LSSVM prediction model, the gap between the model prediction result and the measured value is compared, and it is verified that the trained BAS-LSSVM prediction model has excellent performance in the prediction of the crane service information.

[0073] 3) As shown in Figure 3 , based on the 30-day lifting load collection data, in combination with the Latin hypercube sampling technology, the lifting load data of the crane for 2 years of inspection cycle is expanded and obtained, which is input into the trained BAS-LSSVM prediction model, and the corresponding working cycle number is output, so as to form the 2-year high-precision working service information of the crane as shown in Figure 4 .

[0074] 4) Since the service information in Table 1 is a regular load spectrum arranged in the order of "from small to large" according to the load size, but the load size is randomly changed in actual situation, in order to make the simulation more close to the actual situation, the load sequence is "randomized" by programming under the premise that the cycle number corresponding to each load is unchanged, the regular lifting weight is converted into the random lifting weight, and the result is shown in Figure 5 .

[0075] 5) The trolley wheel pressure is calculated in the form of the trolley self weight and the lifting load, and the trolley wheel pressure with equal sample size is formed.

[0076] 6) The APDL is used to establish the finite element model of the crane girder, and the material definition, mesh division and boundary constraint are completed, the transient analysis setting is entered, the FULL complete method is selected for solving, the Rayleigh damping parameters and the step loading mode are defined, wherein:

[0077] The calculation formula of the Rayleigh damping parameters alphad and betad is , the viscous damping coefficient is 0.03, the upper limit and the lower limit of the frequency are selected according to the maximum and minimum values of the modal natural frequency.

[0078] Regarding the selection of the step loading method, since this invention assumes that the crane has completed the lifting of the goods and the trolley is running on the main beam at the rated operating speed, the load loading method is selected as stepped load, ignoring the process of the load gradually increasing during the lifting stage.

[0079] 7) For example Figure 6 As shown, taking one work cycle as an example, the finite element simulation of the crane trolley running under load is illustrated. Ignoring the lifting and unloading stages, during the trolley's operation at a speed of 30 m / min (500 mm / s) under load, the magnitude of the load remains unchanged, but the load application position changes continuously over time. The random load is applied equivalently to 5 element nodes within the wheel pressure range. A step load is selected, and the distance between two wheel pressure applications is equal to the trolley wheel track of 2450 mm, which is 49 element lengths. The trolley travels 10 element lengths per second along the main beam direction until it reaches the end position of the track, thus completing one work cycle.

[0080] 8) The crane trolley shall be operated as follows: Figure 7 The illustrated wheel load cyclic traversal process simulates the operational service information of a crane with a 2-year scheduled maintenance cycle. The specific process is as follows:

[0081] a: Set initial parameters: time T=0, load cycle number Z=1.

[0082] b: Set the initial action node number J of the wheel pressure of the small car wheel to 50 at the beginning of each cycle.

[0083] c: Let time T = T+1, at which point time step = T. Based on J = 50, use the NSEL node selection command to select the 5 node numbers (J-49, J-48, J-47, J-46, J-45) of the first wheel's wheel pressure and the 5 node numbers (J, J+1, J+2, J+3, J+4) of the second wheel's wheel pressure. Apply the Zth load from the array SUIJIZAIHE to each node, perform transient analysis, and save the calculation results to the RST result file.

[0084] d: Using the FDELE,ALL command, delete all loads applied in the previous time step. Let J = J + 10, which means the trolley moves forward at a speed of 500 mm / s for 1 second, advancing 10 units. Determine if J is greater than 545. If not, it means the trolley has not reached the end of the track, and jump to step 3 to continue the loop iteration. If yes, it means the trolley has reached the end of the track, and the random load needs to be updated to start running again from the initial position of the track, and proceed to step 5.

[0085] e: Let Z = Z + 1, determine whether Z is greater than the maximum number of cycles 1800, if not, it means the trolley has not completed the cyclic loading process of all loads, and needs to jump to step 2 for continuous iteration, if yes, go to step 6.

[0086] f: complete the cyclic loading process of the trolley, and end the cycle.

[0087] 9) Using the POST26 time history post-processor, combined with the APDL selection command, select the nodes on the elements where the key positions (midspan and end span) are located, extract the Von Mises stress of the dangerous point ① at the connection position of the midspan lower flange plate and the web of the main beam structure, and the shear stress of the dangerous point ① in the middle of the end span web, etc. The results are shown in FIG. 8, if the stress change of other position points is needed, only the corresponding position nodes need to be selected to complete the extraction of the stress history of the dangerous point. Figure 9 Example 2

[0088] Those skilled in the art should understand that the embodiments of the present application can be provided as a method, a system, or a computer program product. Therefore, the present application can take the form of an entirely hardware embodiment, an entirely software embodiment, or an embodiment combining software and hardware aspects. Moreover, the present application can take the form of a computer program product implemented on one or more computer-usable storage media (including, but not limited to, disk storage, CD-ROMs, optical storage, etc.) containing computer-usable program code.

[0089] The present application is described with reference to flowcharts and / or block diagrams according to the methods, devices (systems), and computer program products of the embodiments of the present application. It should be understood that each flow and / or block in the flowcharts and / or block diagrams, and the combination of flows and / or blocks in the flowcharts and / or block diagrams can be implemented by computer program instructions. These computer program instructions can be provided to a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing apparatus to produce a machine, so that the instructions executed by the computer or other programmable data processing apparatus produce a device that implements the functions specified in the flowcharts and / or block diagrams. Figure 1 one or more flows and / or blocks Figure 1 an apparatus that carries out the functions specified in one or more flows and / or blocks.

[0090] These computer program instructions can also be stored in a computer-readable memory that can direct the computer or other programmable data processing apparatus to work in a specific manner, so that the instructions stored in the computer-readable memory produce a manufactured product including instruction apparatus, which implements the functions specified in the flowcharts and / or block diagrams. Figure 1 one or more flows and / or blocks Figure 1 an apparatus that carries out the functions specified in one or more flows and / or blocks.

[0091] ​These computer program instructions can also be loaded into a computer or other programmable data processing devices, so that a series of operational steps are performed on the computer or other programmable data processing devices to generate a computer implemented process, so that the instructions executed on the computer or other programmable data processing devices provide a process for implementing the functions specified in the flowchart Figure 1 one flow or multiple flows and / or blocks Figure 1 one block or multiple blocks.

[0092] It is apparent that the present application can be carried out in other specific ways than those set forth herein without departing from the spirit and essential characteristics of the present application. Thus, the above disclosed embodiments are merely exemplary and it is not intended to limit the present application to them. All changes and modifications that come within the meaning and range of equivalents of the present application are to be embraced within the inventive concept and are to be embraced within the scope of the following claims.

Claims

1. A method for obtaining a stress history of a bridge crane, characterized by, The method comprises the following steps: Obtaining characteristic data of a crane in an actual working state for a preset time, wherein the characteristic data comprises hoisting load data and working cycle number data, and the preset time is less than a fixed inspection period; Expanding the hoisting load data for the preset time by using a Latin hypercube sampling technique to obtain hoisting load data for the fixed inspection period, inputting the hoisting load data for the fixed inspection period into a pre-trained tentacle-least squares support vector machine prediction model, and obtaining predicted working cycle numbers; Converting the hoisting load data for the fixed inspection period and the predicted working cycle numbers into trolley wheel pressures with equal sample sizes, wherein the method for obtaining the trolley wheel pressures with equal sample sizes comprises the following steps: randomizing the hoisting load data for the fixed inspection period and the predicted working cycle numbers by using MATLAB; and obtaining the trolley wheel pressures with equal sample sizes by using the hoisting load data after randomization and the self weight of the trolley; Establishing a finite element model of a main beam of the crane, writing macro commands, simulating random trolley loads at a rated speed from one side of the main beam to the other side of the main beam, and obtaining stress histories of the main beam of the crane at any position in the fixed inspection period after cyclically traversing all trolley wheel loadings.

2. The bridge crane stress history acquisition method according to claim 1, characterized in that, The method for obtaining the characteristic data of the crane in the actual working state for the preset time comprises the following steps: Collecting hoisting load data of the crane for 30 days by using load sensors arranged at hoisting drum support positions of trolleys of the crane; Obtaining working cycle number data of the crane for 30 days, wherein one working cycle is from the start of reading of the load sensors to the reading of the load sensors again.

3. The bridge crane stress history acquisition method of claim 1, wherein, The training method of the tentacle-least squares support vector machine prediction model comprises the following steps: Optimizing selection of hyperparameters in a least squares support vector machine model by using a tentacle search optimization algorithm, and building an original tentacle-least squares support vector machine prediction model; Dividing the characteristic data for the preset time into training set data and test set data; Inputting the training set data into the original tentacle-least squares support vector machine prediction model, and obtaining an intermediate tentacle-least squares support vector machine prediction model; Optimizing and correcting the intermediate tentacle-least squares support vector machine prediction model by using the test set, and obtaining a final tentacle-least squares support vector machine prediction model.

4. The bridge crane stress history acquisition method of claim 1, wherein, The method for obtaining the stress histories of the main beam of the crane at any position in the fixed inspection period comprises the following steps: Establishing a finite element model of the main beam of the crane, writing macro commands for reading wheel pressures, running speeds, cyclic loadings, and damping settings; Based on the finite element model of the main beam of the crane, selecting a FULL complete method for solving, defining Rayleigh damping parameters and a step loading mode, and realizing transient analysis settings; Inputting the trolley wheel pressure data in the fixed inspection period into the finite element model of the main beam of the crane by using the macro command for reading wheel pressures, selecting an initial position of the trolley, applying a first wheel pressure loading, controlling the trolley to run at a rated speed to the other side of the main beam by using the macro command for running speeds, and completing one working cycle; Cyclically traversing all wheel loadings by using the macro command for cyclic loadings, and completing finite element simulation of a random trolley load running process of the crane in the fixed inspection period; According to coordinates of a fatigue position to be analyzed, combining a node selection command, and extracting stress history data of the corresponding nodes in the fixed inspection period.

5. The bridge crane stress history acquisition method according to claim 4, characterized in that, The reading wheel pressure macro command is written as: defining the wheel pressure matrix size, and combining the loop command to automatically write the specific wheel pressure value.

6. The bridge crane stress history acquisition method according to claim 4, characterized in that, The running speed macro command is written as: calculating the number of grids advanced per second according to the trolley running speed and the grid size, and using the loop command to control the node position of the wheel pressure loaded per second.

7. The bridge crane stress history acquisition method according to claim 4, characterized in that, The loop loading macro command is written as: using the time step and node selection command to control the trolley to run from one side of the main beam to the other side of the span in seconds, and combining the loop to apply all wheel pressure data in the detection period.

8. The bridge crane stress history acquisition method according to claim 4, characterized in that, The macro command for damping setting is: calculating and setting the two parameter values of Rayleigh damping in the transient analysis according to the maximum and minimum frequency values of the modal analysis results.

9. A bridge crane stress history acquisition system characterized by, The processor and the storage medium are included. The storage medium is used for storing instructions. The processor is used for operating according to the storage instructions to execute the steps of the method according to any one of claims 1-8.

Images