Method and system for predicting fatigue life of main shaft of high-parameter elevator traction machine and related device

Abstract

The invention provides a high-parameter elevator traction machine main shaft fatigue life prediction method and system and a related device, and the method comprises the steps: building a main shaft digital model based on the structure parameters and operation parameters of a traction machine main shaft; a laser displacement sensor is used for obtaining the actual bending deformation of the main shaft in at least one load state, the actual bending deformation is compared with the simulation bending deformation obtained through simulation of the main shaft digital model, and the digital model is verified and corrected; simulating and acquiring fatigue life data under different bending deformations by utilizing the corrected main shaft digital model; training a fatigue life prediction model constructed based on a random forest regression algorithm by using fatigue life data obtained by simulation; the actual bending deformation and the current load state of the main shaft of the traction machine are obtained in real time; and inputting the actual bending deformation and the current load state into a trained fatigue life prediction model, and outputting a prediction value of the residual fatigue life of the traction machine main shaft by the model.

Description

TECHNICAL FIELD

[0001] The present application relates to the field of high parameter elevator manufacturing and maintenance, and more particularly, to a high parameter elevator traction machine main shaft fatigue life prediction method and system and related device. BACKGROUND

[0002] High parameter elevator refers to two types of high speed elevator and large load elevator. The complex dynamic characteristics exhibited during the operation of the high parameter elevator make the main shaft of the traction machine in a high speed variable load condition for a long time, which causes the main shaft of the traction machine to wear or bend and deform. This has a great impact on the fatigue life of the main shaft of the traction machine, and endangers the safe operation of the elevator.

[0003] Effective prediction of the fatigue life of the main shaft is an important means to ensure the safe and reliable operation of the high parameter elevator. The existing analysis and prediction of the fatigue life of the main shaft of the traction machine includes: (1) disassembling the main shaft of the traction machine on site, then detecting and analyzing the functional state of each main component to evaluate the fatigue life of the main shaft. The entire process is relatively complicated; (2) analyzing and judging the fatigue life of the main shaft by monitoring the vibration, friction heat and other related operating data of the bearings and other main shaft accessory parts. It is necessary to ensure that the bearings and other accessory parts are operating normally and within the reliable operating life.

[0004] According to the above, in the prediction of the fatigue life of the main shaft of the elevator traction machine, the existing detection technology is relatively complicated, generally needs to shut down the elevator to disassemble and detect the main shaft, or needs to analyze by monitoring the operating state of the accessory parts, which reduces the detection efficiency and accuracy of the fatigue life of the main shaft of the elevator. SUMMARY

[0005] The present application provides a high parameter elevator traction machine main shaft fatigue life prediction method, system and related device, which uses a fatigue life prediction model constructed based on a random forest regression algorithm to predict the fatigue life of the main shaft as the direct object, solving the technical problem of low detection efficiency and accuracy of the fatigue life of the existing elevator main shaft.

[0006] In one aspect, the present application provides a high-parameter elevator traction machine main shaft fatigue life prediction method, which comprises: constructing a main shaft digital model based on the structural parameters and operating parameters of the traction machine main shaft; using a laser displacement sensor installed on the upper end cover of the traction machine to obtain the actual bending deformation of the main shaft under at least one load state, and comparing the actual bending deformation with the simulation bending deformation obtained by simulating the main shaft digital model to verify and correct the main shaft digital model; using the corrected main shaft digital model to simulate and obtain the fatigue life data of the traction machine main shaft under different bending deformations; training a fatigue life prediction model constructed based on a random forest regression algorithm using the simulated fatigue life data; using the laser displacement sensor to obtain the actual bending deformation and the current load state of the traction machine main shaft in real time; inputting the actual bending deformation and the current load state obtained in real time into the trained fatigue life prediction model to output the predicted value of the remaining fatigue life of the traction machine main shaft.

[0007] In some embodiments, the present application distinguishes between the elastic deformation stage and the plastic deformation stage of the traction machine main shaft to construct the main shaft digital model, wherein: For the elastic deformation stage, the main shaft fatigue life is represented by the Basquin equation as follows:

[0008] N f represents the main shaft fatigue life, is the material fatigue strength coefficient, b is the Basquin index, L represents the main shaft span, d represents the main shaft diameter, δ represents the bending deformation (also referred to as "deflection") of the main shaft, and E is the elastic modulus; For the plastic deformation stage, the main shaft fatigue life is represented by the Manson-Coffin equation as follows:

[0009] N f represents the main shaft fatigue life, δ p represents the permanent bending deformation of the main shaft, is the fatigue ductility coefficient, c is the fatigue ductility index, and L represents the main shaft span and d represents the main shaft diameter.

[0010] In some embodiments, the laser displacement sensor comprises: a first laser displacement sensor and a second laser displacement sensor installed respectively near the bearing support points of the main shaft on the upper end cover of the traction machine, and a third laser displacement sensor installed at the center of the upper end cover of the traction machine. Wherein, the actual bending deformation of the main shaft is obtained by using a three-point test method, which comprises: The first laser displacement sensor, the second laser displacement sensor and the third laser displacement sensor are used to respectively measure displacement initial values in an empty state of the elevator during initial use and displacement values in subsequent detection; The bending deformation of the main shaft is calculated by the following formula: δ= Δb-0.5× (Δa+Δc) Wherein, δ is the bending deformation of the main shaft, used to represent the bending deformation of the main shaft, Δa represents the difference between the displacement value currently detected by the first laser displacement sensor and the corresponding displacement initial value, Δb represents the difference between the displacement value currently detected by the second laser displacement sensor and the corresponding displacement initial value, and Δc represents the difference between the displacement value currently detected by the third laser displacement sensor and the corresponding displacement initial value.

[0011] In some embodiments, the simulation of obtaining the fatigue life data of the main shaft of the traction machine under different bending deformations by using the corrected digital model of the main shaft includes: the different bending deformations take values in a predetermined range with a preset increment, and the fatigue life data covering the elastic deformation and plastic deformation stages are obtained by the digital model of the main shaft.

[0012] In some embodiments, the training of the fatigue life prediction model based on the random forest regression algorithm by using the fatigue life data obtained by simulation includes: analyzing the fatigue life of the main shaft of the traction machine under each bending deformation in the empty and full load states of the elevator by using the fatigue life data obtained by simulation, and corresponding the analyzed fatigue life values to each bending deformation to form a data set; using the data set, taking the bending deformation and the load state as input features, and taking the fatigue life as output, training the fatigue life prediction model.

[0013] In some embodiments, the corresponding of the analyzed fatigue life values to each bending deformation to form a data set includes: classifying the fatigue life values into initial, intermediate and terminal stages, setting labels corresponding to the bending deformations and load states, and collectively forming a data set.

[0014] In some embodiments, the method further includes: dividing the data set into a training set and a test set according to a predetermined proportion, and using stratified sampling to maintain data distribution consistency.

[0015] In some embodiments, training the fatigue life prediction model using the dataset comprises: drawing multiple training subsets from the training set with replacement by a bootstrap method, each decision tree is generated based on one training subset; when splitting at each node of a single decision tree, randomly selecting a predetermined number of candidate features from all input features, and selecting the best split from the predetermined number of candidate features to minimize the squared error; and evaluating the fatigue life prediction model by an out-of-bag coefficient of determination and a root mean square error based on data of the test set to determine the effectiveness of the prediction model.

[0016] In some embodiments, the method further comprises: performing hyperparameter optimization by a grid search method during the training process, and the optimized hyperparameters at least include: a number of decision trees (n_estimators), a maximum number of features (max_feature), a maximum depth of a tree (max_depth), a minimum number of samples required to split a node (min_samples_split), and a minimum number of samples required to be at a leaf node (min_samples_leaf).

[0017] In another aspect, the embodiments of the present application provide a high-parameter elevator traction machine main shaft fatigue life prediction system for implementing the method of any one of the embodiments of the present application, which comprises: a digital modeling module for constructing a main shaft digital model based on structural parameters and operating parameters of the main shaft; a sensor detection module comprising a laser displacement sensor installed on an upper end cover of the traction machine, for detecting actual bending deformation of the main shaft under at least one load state; a digital model verification module for comparing the actual bending deformation with a simulated bending deformation obtained by simulating the main shaft digital model, to verify and correct the main shaft digital model; a data analysis module for simulating and obtaining fatigue life data of the main shaft under different bending deformations by using the corrected main shaft digital model; a prediction model training module for training a fatigue life prediction model based on a random forest regression algorithm by using the simulated fatigue life data; and a real-time prediction module for inputting the actual bending deformation and the current load state obtained by the sensor detection module into the trained fatigue life prediction model, and outputting a prediction value of the remaining fatigue life of the main shaft from the fatigue life prediction model.

[0018] Furthermore, the embodiments of the present application provide a computer program product comprising a computer program which, when executed by a processor, implements the steps of the method according to any one of the embodiments of the present application. The embodiments of the present application also provide a computer-readable storage medium having stored thereon a computer program which, when executed by a processor, implements the steps of the method according to any one of the embodiments of the present application.

[0019] In addition, the embodiments of the present application also provide a computer device comprising: a memory having stored thereon a computer program; and a processor which, when executing the computer program, implements the steps of the method according to any one of the embodiments of the present application.

[0020] According to the above embodiments, the present application has the following technical effects: According to the above embodiments, the present application has the following technical effects:

[0021] Various aspects, features, advantages of the embodiments of the present application will be described in detail below in conjunction with the accompanying drawings. BRIEF DESCRIPTION OF DRAWINGS

[0022] Figure 1 is a schematic diagram showing the main flow of the method for predicting the fatigue life of the main shaft of the high-parameter elevator traction machine according to the embodiments of the present application.

[0023] Figure 2 is a schematic diagram showing the specific flow of the method for predicting the fatigue life of the main shaft of the high-parameter elevator traction machine according to the embodiments of the present application.

[0024] Figure 3 is a schematic diagram showing the flow of the random forest regression algorithm according to the embodiments of the present application.

[0025] Figure 4is a schematic diagram showing a hardware configuration on which a three-point test method according to an embodiment of the present application is based.

[0026] Figure 5 is a block diagram showing a high-parameter elevator traction machine main shaft fatigue life prediction system according to an embodiment of the present application.

[0027] Figure 6 is a structural block diagram of a computer device according to an embodiment of the present application. DETAILED DESCRIPTION

[0028] Exemplary embodiments of the present application will be described in detail with reference to the accompanying drawings. It should be understood that the present application can be implemented in various forms and is not limited to the specific embodiments described herein or shown in the drawings.

[0029] The terms "comprise", "include" and "have" are used herein to indicate the presence of certain features, steps, operations, elements and / or components, but do not exclude the presence or addition of other features, steps, operations, elements, components or combinations thereof. Unless the context clearly dictates otherwise, the terms "first", "second", and the like are not intended to denote any priority or order, but merely to distinguish different elements in the description.

[0030] Figure 1 The main flow of a high-parameter elevator traction machine main shaft fatigue life prediction method according to an embodiment of the present application is shown. In general, the high-parameter elevator traction machine main shaft fatigue life prediction method of the present embodiment includes: S10, constructing a main shaft digital model based on the structural parameters and operating parameters of the traction machine main shaft; S20, using a laser displacement sensor installed on the upper end cover of the traction machine to obtain the actual bending deformation of the main shaft under at least one load state; S30, comparing the actual bending deformation with the simulation bending deformation obtained by simulating the main shaft digital model to verify and correct the main shaft digital model; S40, using the corrected main shaft digital model to simulate and obtain fatigue life data of the traction machine main shaft under different bending deformations; S50, using the simulated fatigue life data to train a fatigue life prediction model constructed based on a random forest regression algorithm; S60, using the laser displacement sensor to obtain the actual bending deformation and the current load state of the traction machine main shaft in real time; S70, inputting the real-time obtained actual bending deformation and current load state into the trained fatigue life prediction model, and outputting a predicted value of the remaining fatigue life of the traction machine main shaft from the fatigue life prediction model.

[0031] For ease of illustration, the traction machine main shaft is simplified as a simply supported beam subjected to concentrated load at the midpoint for illustration. As shown in FIG. 2, the traction machine main shaft is subjected to a concentrated load at the midpoint, and the bending deformation of the traction machine main shaft under the load is shown. Figure 4As shown, the high-parameter elevator traction machine mainly comprises a main shaft 401 and a traction sheave 402 mounted on the main shaft 401. The main shaft 401 is rotatably supported on a fixed base 405 and an upper end cover 406 through a first bearing 403 and a second bearing 404.

[0032] The main shaft is set as a linear elastic, homogeneous and isotropic circular cross-section beam with a diameter of d (m) and a span (the length between the first bearing 403 and the second bearing 404) of L (m). Analysis shows that the load F (N) mainly acts on the midspan, generating a midspan bending deformation δ (m); the load cycle is a symmetric cycle with R = -1 (rotational bending).

[0033] When the main shaft is elastically deformed, the bending deformation and the load have the following relationship: (1) (2) In formulas (1) and (2), δ is the bending deformation of the main shaft of the traction machine, F is the load borne by the main shaft, E is the elastic modulus, I is the cross-sectional moment of inertia, and d is the cross-sectional diameter of the main shaft.

[0034] It can be known from formulas (1) and (2) that the load F value of the main shaft is: (3) Since the bending moment of the main shaft of the traction machine at the midspan is: (4) The stress at the outer edge of the cross section is: (5) In formula (5), W is the torsional section modulus: (6) Substitute F with δ to obtain the stress-bending deformation relationship as follows: (7) (8) In formulas (7) and (8), is a stiffness stress conversion coefficient, and therefore the symmetric cycle stress amplitude Thus, the fatigue life of the main shaft is obtained from the Basquin equation as follows: (9) In formula (9) is a material fatigue strength coefficient, and b is the Basquin index.

[0035] It can be known from formula (9) that, when the main shaft is elastically deformed, the fatigue life and the bending deformation of the main shaft are inversely proportional in a power law, and the slenderness ratio L2 The greater the value of / d, the more sensitive the life is to the amount of bending deformation.

[0036] When the traction machine main shaft occurs plastic deformation, the main shaft has low cycle fatigue and stress relaxation / redistribution, and strain-life method must be used for analysis: From the Manson-Coffin basic formula: (10) In formula (10), is the plastic strain amplitude (single side), is the fatigue ductility coefficient, and c is the fatigue ductility index.

[0037] Permanent bending deformation amount δ of the main shaft p The relationship with the maximum curvature in the beam is as follows: (11) Then the maximum plastic strain is: (12) For symmetric cyclic loading, the plastic strain amplitude is: (13) Thus, the relationship between the plastic deformation of the traction machine main shaft and the life of the main shaft is: (14) Therefore, the relationship between the plastic deformation of the traction machine main shaft and the life of the main shaft is: (15) It can be seen from formula (15) that the plastic deformation and the fatigue life are inversely proportional; in the Manson-Coffin framework, δ p Each increase of 1 times, the life is reduced to the original value (1 / 2) 1 / c times (about 34% when c = -0.6).

[0038] As can be seen from the above, when the traction machine main shaft occurs elastic deformation or plastic deformation, the deformation amount of the main shaft is inversely proportional to the fatigue life of the main shaft. Therefore, in the embodiments of the present application, the elastic deformation amount of the main shaft is measured when the high-parameter elevator is fully loaded, the fatigue life of the main shaft in this state is predicted, the plastic deformation amount of the main shaft is measured when the high-parameter elevator is unloaded, and the fatigue life of the main shaft in this state is predicted, thereby realizing prediction of the fatigue life of the main shaft in different states. Therefore, in the example embodiments, in S10, the digital model of the main shaft is constructed by distinguishing whether the traction machine main shaft is in an elastic deformation stage or a plastic deformation stage: for the elastic deformation stage, the fatigue life of the main shaft is represented by formula (9); for the plastic deformation stage, the fatigue life of the main shaft is represented by formula (15).

[0039] In some embodiments, a digital model of the hoisting machine main shaft is generated by obtaining the structural parameters (e.g., diameter, span, etc.) and operating parameters (e.g., load range, starting speed, and operating speed) of the hoisting machine main shaft and combining them with formulas (9) and (15). The digital model of the hoisting machine main shaft is used to analyze the different bending deformations of the hoisting machine main shaft under no-load and full-load conditions. The obtained results are analyzed and compared with the actual bending deformations of the main shaft measured under the same working conditions, and the constructed digital model is corrected to improve the accuracy of its analysis.

[0040] In some embodiments, in S20, a laser displacement sensor installed on the upper end cover of the traction machine is used to obtain the actual bending deformation of the main shaft under at least one load state based on a three-point test method. Figure 4 As shown, the laser displacement sensors include: laser displacement sensors A and B, mounted on the hoisting machine upper end cover 406 at the fulcrums near the two main shaft bearings (i.e., the first bearing 403 and the second bearing 404), and laser displacement sensor C, mounted at the center of the hoisting machine upper end cover. The three-point testing method involves measuring the initial displacement value and subsequent displacement values ​​during initial use when the elevator is unloaded, using laser displacement sensors A, B, and C; and calculating the bending deformation of the main shaft based on the initial and subsequent displacement values.

[0041] Since sensors A and B are installed on both sides of the main shaft near the two main shaft bearings, and since this location is close to the main shaft fulcrum, the main shaft deformation generated by them is small. Therefore, the main shaft points a and b measured by sensors A and B are set as zero points. Sensor C is installed at the center of the upper end cover. Because the position it measures is at the center between the two fulcrums of the traction machine main shaft, the main shaft deformation generated is the largest. Therefore, the corresponding main shaft point c is the point of maximum bending deformation. Among them, laser displacement sensor C measures the displacement of the traction wheel 402. Since the traction wheel 402 is installed on the main shaft 401, the displacement of point c actually reflects the displacement of the main shaft and can be regarded as the displacement of the main shaft 401. The specific test process is as follows: First, when the traction machine main shaft is used for the first time, in an unloaded high-parameter elevator, the vertical displacement values ​​at points a, b, and c are measured. These initial displacement values ​​are a0, b0, and c0, respectively. In some embodiments, during testing, the main shaft is rotated one revolution so that sensors A, B, and C measure the maximum vertical displacement corresponding to points a, b, and c. This is repeated three times, and the average of these maximum values ​​is calculated to yield the values ​​a0, b0, and c0, respectively.

[0042] Secondly, in subsequent elevator use, the displacement values generated by the three points a, b, c under the main shaft no-load and full load are detected at random, and the displacement values under no-load are measured as a1, b1, c1, and the displacement values under full load are measured as a2, b2, c2. In some embodiments, each detection makes the main shaft rotate one circle, so that the A, B, C three sensors measure the maximum values of the vertical direction displacement of the corresponding a, b, c three points, and the average value of each maximum value is taken as the final detection value.

[0043] Finally, the corresponding difference values are calculated through the first measured displacement initial value and the subsequently measured displacement value, wherein the displacement difference values under no-load are respectively: Δa1=a1-a0, Δb1=b1-b0, Δc1=c1-c0 (16) The displacement difference values under full load are respectively: Δa2=a2-a0, Δb2=b2-b0, Δc2=c2-c0。 (17) Further, the main shaft crosswise bending deformation amounts under no-load and full load are respectively calculated as: Under no-load: δ1=Δb1-0.5╳(Δa1+Δc1) (18) Under full load: δ2=Δb2-0.5╳(Δa2+Δc2) (19) Thus, the bending deformation amounts of the main shaft under no-load and full load can be obtained.

[0044] In the embodiments of the present application, on the one hand, the main shaft bending deformation amount obtained by the laser displacement sensor and the three-point test method is used to verify and correct the main shaft digital model at S30. On the other hand, when predicting the fatigue life of the main shaft, the main shaft bending deformation amount obtained in real time by the laser displacement sensor and the three-point test method is input into the fatigue life prediction model constructed based on the random forest regression algorithm as described below, and the prediction value of the remaining fatigue life of the traction machine main shaft is output by the prediction model.

[0045] In some embodiments, at S40, the simulation of obtaining the fatigue life data of the traction machine main shaft under different bending deformation amounts by using the corrected main shaft digital model includes that the different bending deformation amounts take values in a predetermined range with a preset increment, and the fatigue life data covering the elastic deformation and plastic deformation stages are obtained by the main shaft digital model.

[0046] In some embodiments, training the fatigue life prediction model based on a random forest regression algorithm using the fatigue life data obtained from the simulation at S50 comprises: analyzing fatigue life of each bending deformation of the traction machine main shaft in the elevator in the empty and full load states using the fatigue life data obtained from the simulation, and corresponding the analyzed fatigue life values to each bending deformation to form a data set; and training the fatigue life prediction model using the data set, with the bending deformation and the load state as input features and the fatigue life as output.

[0047] In some embodiments, corresponding the analyzed fatigue life values to each bending deformation to form a data set comprises: classifying the fatigue life values into initial, intermediate, and terminal stages, and setting the labels corresponding to the bending deformations and the load states to form the data set.

[0048] In some embodiments, the method further comprises: dividing the data set into a training set and a test set according to a predetermined proportion, and using stratified sampling to maintain data distribution consistency.

[0049] In some embodiments, training the fatigue life prediction model using the data set comprises: extracting multiple training subsets from the training set with replacement by a Bootstrap method, and generating each decision tree based on a training subset; randomly selecting a predetermined number of candidate features from all input features and selecting the best split from the predetermined number of candidate features to minimize the squared error when splitting each node of a single decision tree; and evaluating the fatigue life prediction model by an out-of-bag determination coefficient and a root mean square error according to the data of the test set to determine the effectiveness of the prediction model. In some embodiments, the predetermined number is the maximum number of features (max_feature).

[0050] In some embodiments, the method further comprises: performing hyperparameter optimization using a grid search method during the training process, and the optimized hyperparameters at least include: the number of decision trees (n_estimators), the maximum number of features (max_feature), the maximum depth of the tree (max_depth), the minimum number of samples required for node splitting (min_samples_split), and the minimum number of samples required for leaf nodes (min_samoles_leaf).

[0051] Specifically, as Figure 2As shown, the high-parameter elevator traction machine main shaft fatigue life prediction method of the present application comprises: inputting the structural parameters and operating parameters of the traction machine main shaft obtained by actual measurement, generating a traction machine main shaft digital model; analyzing the different bending deformation amounts of the traction machine main shaft under no load and full load through the traction machine main shaft digital model, and comparing the obtained results with the actual measured main shaft bending deformation amounts under the same working conditions obtained by the laser displacement sensor and the three-point test method, to analyze the accuracy of the main shaft digital model; when it is analyzed that the main shaft digital model is not accurate, adjusting the related parameters of the model to correct the main shaft digital model; when it is analyzed that the main shaft digital model is accurate, using the main shaft digital model to perform main shaft fatigue life analysis of different main shaft deformation amounts under no load of the elevator, and to perform main shaft fatigue life analysis of different main shaft deformation amounts under full load of the elevator, to obtain main shaft fatigue life data under different main shaft deformation amounts; using the data to generate training data with main shaft deformation amount as the label; adopting a random forest regression algorithm to construct a main shaft fatigue life prediction model based on the main shaft deformation amount, and using the training data to train the fatigue life prediction model; obtaining the real-time deformation amount of the main shaft in real time through the laser displacement sensor and the three-point test method, and inputting it into the trained fatigue life prediction model, to realize rapid and accurate prediction of the fatigue life of the traction machine main shaft through the prediction model, and output the prediction value of the remaining fatigue life of the traction machine main shaft.

[0052] In some embodiments, the training of the traction machine main shaft fatigue life prediction model based on the random forest regression analysis algorithm at least includes data preparation, generation of single decision regression tree, integrated prediction, and model evaluation.

[0053] (1) Training and testing data preparation According to the main shaft fatigue life data obtained by analyzing the traction machine main shaft digital model, the useful training and testing data are obtained by cleaning the obtained fatigue life data. Each time the training data is input, the amount of data input is N, and each data contains an M-dimensional feature vector x and a corresponding main shaft fatigue life target y. In data sampling, for the bth tree (b e (1, 2, … B)), N is extracted from the original data with replacement by the Bootstrap method, obtaining the training subset D(b) for the bth tree. Therefore, for a prediction model with B decision trees, the Bootstrap method is used for sampling with replacement B times: (20) where S b is the sample index of the bth tree, x is the feature vector of the traction machine main shaft (including the main shaft load value and the main shaft bending deformation amount), and y is the fatigue life of the traction machine main shaft. At the same time, there are about 36.8% of out-of-bag (OOB) samples that do not participate in the training of the tree.

[0054] (2) Generation of single decision regression tree At the current node t, find the best split Minimize the squared error: (21) where R L , R R are the left and right sub-regions, is the sample mean, and the tree continues to split until the stopping condition is met. The node split stopping condition is set as: tree depth ≥ max_depth (None (natural growth), 10-30); node sample number ≤ min_samples_split (commonly used 2-20); leaf node sample number ≤ min_samples_leaf (commonly used 1-8). Among them, max_depth is the maximum depth of the tree, min_samples_split is the minimum number of node samples required for splitting, and min_samples_leaf is the minimum number of leaf node samples required for splitting.

[0055] At each node, only the best split is selected from the randomly selected max_feature candidate features to reduce the correlation between trees. For example, in the present embodiment, the number of feature vectors is 2 (including the main shaft load case and the main shaft deformation amount), so max_feature = 1 is taken.

[0056] (3) Integrated prediction A total of B trees are trained, and the larger B is, the lower the variance is, where the empirical value B ∈ [200, 1000], and for a new sample feature vector x, the forest outputs the average value of B trees as: (22) (4) Model evaluation According to the training and prediction results, the prediction model is evaluated by the out-of-bag coefficient of determination OOB-R2 and the root mean square error RMSE: (23) (24) In formulas (23) and (24), y i is the true value of the i-th sample, is the out-of-bag prediction value of the i-th sample, is the average value of all true values, and N is the total number of samples.

[0057] The exemplary embodiments of the present application will be described below in conjunction with Figure 3 .

[0058] (I) Construction of digital model The structural parameters and operating parameters of the main shaft of the traction machine are acquired in the initial operation stage of the high-parameter elevator, and a corresponding digital model of the main shaft of the traction machine is quickly generated through the acquired related parameters.

[0059] (II) Deformation detection of the main shaft of the traction machine based on a laser displacement sensor Through the laser displacement sensor installed on the end cover of the traction machine, the displacement initial values of each test point of the main shaft of the traction machine under no load and full load are recorded by detecting the test points of the main shaft of the traction machine under no load and full load in the initial operation stage of the high-parameter elevator. In the later operation of the elevator, the displacement values of each point under different states of the high-parameter elevator are obtained in real time through the laser displacement sensor. The bending deformation of the main shaft under the corresponding state is calculated through formulas (16)-(19).

[0060] (III) Data acquisition According to the generated digital model of the main shaft of the traction machine and formulas (9) and (15), the bending deformation of the main shaft of the traction machine under no load and full load is analyzed. For example, the deformation of the main shaft is set in the range of 0.001-0.1 mm, and 100 deformation values of the main shaft are set with an increment of 0.001 mm. The analysis and testing are performed under no load and full load of the main shaft of the traction machine, and 200 fatigue life data of the main shaft are obtained. Meanwhile, the fatigue life of the main shaft of the traction machine is divided into initial stage (use frequency ≥10 6 times), middle stage (10 3 ≤ use frequency <10 6 times), and final stage (use frequency <10 3 times). The fatigue life values obtained by analysis are classified, and are set as labels and corresponding deformation of the main shaft and loading conditions to form a data set. It should be understood that the specific values listed herein are only examples and do not limit the present application. According to specific applications, other values can be used in the present application.

[0061] (IV) Construction of the prediction model Through the obtained data set, the data set is divided into a training set and a test set, and is split in layers at a ratio of 7:3, i.e., 140 training sets and 60 test sets. The random seed is fixed as 2024 to ensure repeatability. It should be understood that the specific values listed herein are only examples and do not limit the present application. According to specific applications, other values can be used in the present application.

[0062] (V) Training of the prediction model (1) The initial setting of the relevant parameters of the model, the number of trees in the forest n_estimators (i.e. B value) is 500, the maximum number of features considered at each split max_feature is 1, the maximum depth of each decision tree max_depth is None (let the tree grow naturally), the minimum number of samples required to split min_samples_split is 5, and min_samoles_leaf is 2.

[0063] (2) In the data back sampling, according to B is 500 and formula (20), 500 times of sampling are carried out, and it is ensured that about 36.8% of the out-of-bag (OOB) samples do not participate in the tree training.

[0064] (3) In the generation of single decision regression tree, at each node, only the best split is selected from the randomly selected 1 (max_feature=1) candidate features, and the best split is found according to formula (21) Minimize the square error until the tree depth, node sample number, leaf node sample number meet the initial setting value, stop splitting.

[0065] (4) Training and out-of-bag validation, in the prediction training, the out-of-bag sample (OOB) is used for monitoring, and the OOB-R 2 Value calculation is carried out through formula (23), if the value is close to 1, there is no need to increase the tree.

[0066] (5) Hyperparameter grid fine tuning and cross-validation, the grid range is set as: n_estimators∈{300, 500, 800}, max_features∈{1, 2}, max_depth∈{None, 15, 25}, min_samples_split∈{2, 5, 10}. Through formula (24) to calculate the negative root mean square error NRSME (NRSME=-RSME) of the prediction result, finally get the optimal combination of n_estimators, max_feature, max_depth, min_samples_split, min_samoles_leaf.

[0067] (6) Final model training and evaluation, using all the training set, retraining under the obtained optimal parameters to get the final fatigue life prediction model based on random forest regression algorithm.

[0068] (6) Evaluation of the prediction model test In the model test stage, according to the relevant data of the test set, and according to formula (23), (24), the obtained prediction model is evaluated to judge the effectiveness of the constructed model.

[0069] (VII) Prediction of fatigue life of traction machine main shaft The displacement of each test point of the monitored hoisting machine main shaft is obtained by a laser displacement sensor, and the bending deformation of the main shaft under the actual state is calculated by formulas (16)-(19). The obtained deformation and main shaft load conditions are input into the constructed fatigue life prediction model. The output of the prediction model is the prediction of the fatigue life of the high-parameter elevator traction machine main shaft.

[0070] Based on the high-parameter elevator traction machine main shaft fatigue life prediction method according to the embodiment of the present invention, the present invention also provides a high-parameter elevator traction machine main shaft fatigue life prediction system for implementing the method according to any embodiment of the present invention. In an exemplary embodiment, the system includes: a digital modeling module 501 for constructing a digital model of the main shaft based on the structural parameters and operating parameters of the hoisting machine main shaft; a sensor detection module 502, including a laser displacement sensor installed on the upper end cover of the traction machine, for detecting the actual bending deformation of the main shaft under at least one load state; a digital model verification module 503, for comparing the actual bending deformation with the simulated bending deformation obtained by simulating the main shaft digital model to verify and correct the main shaft digital model; a data analysis module 504, for using the corrected main shaft digital model to simulate and obtain fatigue life data of the traction machine main shaft under different bending deformations; a prediction model training module 505, for using the fatigue life data obtained by simulation to train a fatigue life prediction model constructed based on a random forest regression algorithm; and a real-time prediction module 506, for inputting the actual bending deformation and the current load state obtained in real time by the sensor detection module 502 into the trained fatigue life prediction model, so that the fatigue life prediction model outputs a predicted value for the remaining fatigue life of the traction machine main shaft.

[0071] It should be understood that the steps, processes, operations, etc. described in the above embodiments can be implemented by computer software programs, therefore, the embodiments of the present application also relate to a computer program product comprising a computer program which, when executed by a processor, implements the steps of the method or the processes, operations of the system described in any one of the embodiments of the present disclosure. In some embodiments, the program code of the computer program for performing the embodiments of the present application can be written in any combination of one or more programming languages, for example, the computer program can be written using high-level procedural and / or object-oriented programming languages, and / or assembly / machine languages. The programming languages include, but are not limited to, such as Java, C++, python, "C" language or similar programming languages. The program code can be executed entirely on a personal computing device, partially on a personal computing device, partially on a remote computing device, or entirely on a remote computing device or server. In the case involving a remote computing device, the remote computing device can be connected to the personal computing device through any kind of network, including a local area network (LAN) or a wide area network (WAN), or can be connected to an external computing device (for example, connected to the Internet through an Internet service provider).

[0072] It should also be understood that the computer program product can be stored on a computer readable storage medium, such as including: hard disk, floppy disk, magnetic tape, optical disc, solid state disk, flash memory, etc.

[0073] Further, it can be understood that the method of the present application can be executed by a computer device. Therefore, the embodiments of the present application also relate to a computer device comprising: a memory having a computer program stored thereon; a processor which, when executing the computer program, implements the steps (processes, operations, etc.) of the method described in any one of the embodiments of the present disclosure.

[0074] Figure 6 An exemplary structure of a computer device is shown. As shown in Figure 6 The computer device 100 can at least include a processor 101 and a memory 102, wherein the memory 102 can be or include a storage medium having a computer program (or computer readable instructions) stored thereon, and the processor 101 executes the computer program to perform part or all of the operations in the method described in any one of the embodiments of the present disclosure.

[0075] In some embodiments, the computer device 100 can further include a data storage device 103, a display 104, a speaker 105, and a communication module 106. Among them, the processor 101, the memory 102, and the data storage device 103 communicate with each other through a bus, and interact with peripheral devices such as the display 104 and the communication module 106 through the bus and the I / O module 107.

[0076] Specifically, the data storage device 103 can store application programs and various configuration files, data thereof. The memory 102 stores computer programs. The processor 101 executes the computer programs to perform various processes, operations or steps of the configuration center-based interface simulation method. The display 104 can be used to display various data, images, etc.

[0077] In optional embodiments, the communication module 106 can be omitted in the case where it is not necessary to communicate with the other systems, devices or equipment. In some embodiments, the computer device 100 can further include a speaker 105 for voice output.

[0078] In some embodiments, the processor 101 can include any suitable semiconductor-based electronic processing unit, chip, microchip or integrated circuit (IC). The memory 102 is a programmable memory, which can include any suitable electronic storage device configured to store instructions and can be reprogrammed. For example, the programmable memory can include an erasable programmable read-only memory (EPROM) device. The data storage device 103 can be a persistent storage device, which can include any suitable electronic memory configured to retain stored information when power is cycled. For example, the data storage device 103 can include a hard disk drive, a solid state drive (SSD), a flash drive, a hybrid drive, etc., or any combination of these.

[0079] Therefore, the processor 101 can control the output of information on the I / O module 107 by storing information to the memory and / or executing programs / instructions stored in the memory. For example, several aspects of the methods described herein can be performed by the processor 101 according to programs / instructions stored in the memory (e.g., the memory 102 and / or the data storage device 103).

[0080] Also, the processor 101 can be in electronic communication with the I / O module 107 and / or the communication module 106 to receive or transmit relevant instructions and information. The I / O module 107 can include any suitable mechanical or virtual user interface configured to enable a user to interact with the computer device 100 or to perform one or more functions of the computer device 100 itself, such as a graphical user interface (GUI) on a screen or other display. In some examples, the user interface can include a voice interface capable of voice recognition by which an operator can provide voice commands to the processor.

[0081] The communication module 106 can include any suitable devices and / or structures configured to facilitate the exchange of information between the computer device 100 and external electronic devices, such as sensor detection modules. The communication module 106 can include devices configured to send and / or receive wireless or wired information with other devices. For example, the communication module 106 can include an antenna, a transceiver, a connector for wired reception and / or transmission of data, a data exchange device, and the like, or any combination of these. The communication module 106 can also include ancillary equipment, such as filter circuits, encryption / decryption circuits, and / or integrated circuit (IC) chips for processing signals (e.g., Bluetooth® chips). In some embodiments, the communication module 106 can include a WiFi device configured to connect to a local wireless network.

[0082] In some embodiments, the computer device 100 can include a smartphone, a wearable computer, a portable / mobile electronic device, a tablet computer, a smartwatch, a personal digital assistant (PDA), a personal computer (PC), a desktop computer, a notebook computer, a server, and the like. The computer device 100 can include or be installed with one or more application programs (APPs), one of which is configured to perform the dynamic data service system and method described herein.

[0083] Although not shown, it should be understood that the computer device 100 also includes a power supply component, which can include any suitable devices and / or structures configured to provide an electrical interface between the computer device 100 and a power source. The power source can include any suitable source of electrical energy, such as a battery, a plug, a capacitor, a fuel cell, and the like, or any combination of these. Additionally or alternatively, the power source can be included in the power supply component. For example, a battery or battery pack can be included within the computer device 100. In some embodiments, the battery can be rechargeable, such as by an interface provided by a cable or the power supply component. In some embodiments, the power supply component can share features or duplicate features of the communication module 106. For example, a USB or micro-USB cable connector can be included in the computer device 100, such that power supply or data communication is performed by the same component.

[0084] Those skilled in the art will understand that the above- described embodiments of the present application are merely illustrative of the present application and that many modifications, changes, substitutions, and the like can be suggested to one skilled in the art and can be made without departing from the spirit and scope of the present disclosure.

Claims

1. A high-parameter elevator traction machine main shaft fatigue life prediction method, characterized in that: include: Based on the structural parameters and operating parameters of the traction machine main shaft, a main shaft digital model is constructed; Using a laser displacement sensor installed on the upper end cover of the traction machine, the actual bending deformation of the main shaft under at least one load state is obtained, and the actual bending deformation is compared with the simulated bending deformation obtained by simulating the digital model of the main shaft to verify and correct the digital model of the main shaft; Using the modified digital model of the main shaft, simulation is performed to obtain fatigue life data of the traction machine main shaft under different bending deformations; Using the fatigue life data obtained by simulation to train a fatigue life prediction model constructed based on a random forest regression algorithm; The actual bending deformation and current load state of the traction machine main shaft are obtained in real time through the laser displacement sensor; The actual bending deformation amount and the current load state obtained in real time are input into the trained fatigue life prediction model, and the fatigue life prediction model outputs a predicted value of the remaining fatigue life of the traction machine main shaft.

2. The high-parameter elevator traction machine main shaft fatigue life prediction method according to claim 1 is characterized in that: The main shaft of the traction machine is distinguished to be in the elastic deformation stage or the plastic deformation stage and a digital model of the main shaft is constructed, wherein: For the elastic deformation stage, the spindle fatigue life is expressed by the Basquin equation as follows: N f Indicates the fatigue life of the spindle, is the material fatigue strength coefficient, b is the Basquin index, L is the spindle span, d is the spindle diameter, δ is the bending deformation of the spindle, and E is the elastic modulus; For the plastic deformation stage, the spindle fatigue life is expressed by the Manson-Coffin equation as follows: N f represents the fatigue life of the spindle, δ p Indicates the permanent bending deformation of the main shaft, is the fatigue ductility coefficient, c is the fatigue ductility index, L is the spindle span, and d is the spindle diameter.

3. The high-parameter elevator traction machine main shaft fatigue life prediction method according to claim 1 is characterized in that: The laser displacement sensor comprises: a first laser displacement sensor and a second laser displacement sensor respectively installed at the upper end cover of the traction machine near the two bearing fulcrums of the main shaft, and a third laser displacement sensor installed at the center of the upper end cover of the traction machine; Among them, the three-point test method is used to obtain the actual bending deformation of the main shaft, including: Using the first laser displacement sensor, the second laser displacement sensor, and the third laser displacement sensor, respectively measure the initial displacement value when the elevator is in an unloaded state during initial use and the displacement values ​​detected subsequently; The bending deformation of the main shaft is calculated by the following formula: δ = Δb-0.5× (Δa+Δc) Among them, δ is the bending deformation of the main shaft, which is used to represent the bending deformation of the main shaft, Δa represents the difference between the displacement value currently detected by the first laser displacement sensor and the corresponding initial displacement value, Δb represents the difference between the displacement value currently detected by the second laser displacement sensor and the corresponding initial displacement value, and Δc represents the difference between the displacement value currently detected by the third laser displacement sensor and the corresponding initial displacement value.

4. The high-parameter elevator traction machine main shaft fatigue life prediction method according to claim 1, characterized in that: The simulation of obtaining fatigue life data of the traction machine main shaft under different bending deformations using the modified main shaft digital model includes: The different bending deformation amounts are taken in a predetermined range with preset increments, and fatigue life data covering the elastic deformation and plastic deformation stages are obtained through the spindle digital model.

5. The high-parameter elevator traction machine main shaft fatigue life prediction method according to claim 4 is characterized in that: Training a fatigue life prediction model based on a random forest regression algorithm using the fatigue life data obtained by simulation includes: The fatigue life data obtained by simulation is used to analyze the fatigue life of each bending deformation of the traction machine main shaft under the elevator no-load and full-load states, and the fatigue life values ​​obtained by analysis are corresponded to each bending deformation to form a data set; The fatigue life prediction model is trained using the data set, with bending deformation and load state as input features and fatigue life as output.

6. The high-parameter elevator traction machine main shaft fatigue life prediction method according to claim 5, characterized in that: The fatigue life values ​​obtained by the analysis correspond to the respective bending deformations, and the data set includes: The fatigue life values ​​are classified into early, middle and late stages, and are set as labels together with corresponding bending deformation amounts and load states to form a data set.

7. The high-parameter elevator traction machine main shaft fatigue life prediction method according to claim 5, characterized in that: Also includes: The dataset is divided into a training set and a test set according to a predetermined ratio, and stratified sampling is used to maintain data distribution consistency.

8. The high-parameter elevator traction machine main shaft fatigue life prediction method according to claim 7, characterized in that: Using the data set to train the fatigue life prediction model includes: Extracting multiple training subsets from the training set with replacement using a bootstrap method, and generating each decision tree based on one training subset; When splitting each node of a single decision tree, a predetermined number of candidate features are randomly selected from all input features, and the best split is selected from the predetermined number of candidate features to minimize the squared error; Based on the data of the test set, the fatigue life prediction model is evaluated by the out-of-bag determination coefficient and the root mean square error to determine the effectiveness of the prediction model.

9. The high-parameter elevator traction machine main shaft fatigue life prediction method according to claim 8, characterized in that: The method further includes: using a grid search method to optimize hyperparameters during the training process, where the optimized hyperparameters include at least: the number of decision trees, the maximum number of features, the maximum depth of the tree, the minimum number of samples required for node splitting, and the minimum number of samples required for leaf nodes.

10. A high-parameter elevator traction machine main shaft fatigue life prediction system, used to implement the high-parameter elevator traction machine main shaft fatigue life prediction method according to any one of claims 1 to 9, characterized in that: include: A digital modeling module is used to build a digital model of the main shaft based on the structural parameters and operating parameters of the traction machine main shaft; a sensor detection module, comprising a laser displacement sensor mounted on an upper end cover of the traction machine, for detecting an actual bending deformation of the main shaft under at least one load state; a digital model verification module, configured to compare the actual bending deformation with a simulated bending deformation obtained by simulating the spindle digital model, so as to verify and correct the spindle digital model; a data analysis module, configured to utilize the modified digital model of the main shaft to simulate and obtain fatigue life data of the traction machine main shaft under different bending deformations; A prediction model training module, configured to train a fatigue life prediction model constructed based on a random forest regression algorithm using the fatigue life data obtained through simulation; The real-time prediction module is used to input the actual bending deformation and current load state obtained in real time by the sensor detection module into the trained fatigue life prediction model, and the fatigue life prediction model outputs a predicted value of the remaining fatigue life of the traction machine main shaft.

11. A computer program product comprising a computer program, characterized in that When the computer program is executed by a processor, the steps of the high-parameter elevator traction machine main shaft fatigue life prediction method according to any one of claims 1 to 9 are implemented.

12. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the steps of the high-parameter elevator traction machine main shaft fatigue life prediction method according to any one of claims 1 to 9 are implemented.

13. A computer device, characterized in that: include: a memory having a computer program stored thereon; A processor that implements the steps of the high-parameter elevator traction machine main shaft fatigue life prediction method as described in any one of claims 1 to 9 when executing the computer program.

Images