Method and system for monitoring operation state of hoisting mechanism of crane

Abstract

The invention provides a crane hoisting mechanism operation state monitoring method and system, and the method can obtain the real-time working condition data and state data of a crane hoisting mechanism, integrates the collected vibration signals of the crane hoisting mechanism with the working condition data, and achieves the real-time monitoring of the operation state of the crane hoisting mechanism through professional analysis and big data analysis means. And determining the fault abnormal condition of the hoisting mechanism of the crane. The equipment state of the hoisting mechanism of the crane can be monitored in real time, the possibility of operation accidents of the hoisting mechanism of the crane is predicted according to the equipment state, and corresponding measures are taken in time, so that the production safety of the crane is improved, the operation accidents of the crane are avoided, and the working efficiency of the crane is improved.

Description

Technical Field

[0001] The present invention relates to the field of special equipment operation and maintenance, and in particular to a method and system for monitoring the operating status of a crane hoisting mechanism. Background Art

[0002] With the advancement of my country's infrastructure construction and the overall development of the heavy industry of the national economy, cranes have become important equipment for the rapid development of the national economy and have been widely used in many national economic sectors. With the development of cranes towards large-scale, high-speed and automated directions, the safety of cranes has received more and more attention. Among all kinds of special equipment accidents, crane accidents not only account for a high proportion, but are usually major and extremely serious accidents, which seriously threatens the safety of production safety and the safety of people's lives and property.

[0003] A large number of crane accidents occur during crane operation, such as collisions between cranes and surrounding obstacles, interference between multiple cranes, and injuries to construction workers during crane operation. The real-time equipment status of the crane is difficult to judge based on static experience, and often requires manual analysis and processing by safety management personnel, which reduces the safety of the crane during operation and the crane's work efficiency.

[0004] Therefore, it is necessary to monitor the equipment status of the crane in real time and predict possible accidents during crane operation in advance to avoid accidents during crane operation. Summary of the invention

[0005] In order to overcome the problem that it is difficult to monitor the equipment operating status of lifting equipment in real time, the present application provides a method and system for monitoring the operating status of a crane hoisting mechanism.

[0006] To achieve the above-mentioned purpose, some embodiments of the present application provide a method for monitoring the operating status of a crane hoisting mechanism, comprising the following steps: Step 1. Acquire real-time working condition data and status data of the crane hoisting mechanism, wherein the working condition data is information data related to the working condition of the crane hoisting mechanism; the status data is vibration monitoring data of the hoisting mechanism working status collected by a vibration acceleration sensor deployed on the hoisting mechanism;

[0007] Step 2. Combine the collected vibration monitoring data of the crane hoisting mechanism with the operating condition data, and use professional analysis and big data analysis methods to determine the abnormal fault of the crane hoisting mechanism.

[0008] Furthermore, the vibration acceleration sensor is arranged at the high-speed shaft end of the reducer, the low-speed shaft end of the reducer, the pinion bearing seat near the reduction box, the pinion bearing seat near the drum, the drum bearing seat near the reduction box and the drum bearing seat of the far reduction box of the crane hoisting mechanism.

[0009] Furthermore, the professional analysis method in step 2 includes obtaining the characteristic frequency of the hoisting mechanism gear fault, and the gear fault characteristic frequency f 齿轮 As follows:

[0010]

[0011] Among them, f r , f m and f c are the gear rotation frequency, gear meshing frequency and gear natural frequency when the hoisting mechanism encounters a fault, A and B are the fault condition data and fault vibration monitoring status data of a single hoisting mechanism when the hoisting mechanism of the crane encounters a fault, respectively. ele and f des are the motor operating frequencies corresponding to the A and B fault data encountered by the lifting mechanism during the ascending and descending processes, respectively; P is the number of pole pairs; k and m are the gear pair equivalent mass and gear average stiffness corresponding to the A and B fault data encountered by the lifting mechanism during the ascending and descending processes, respectively;

[0012] The professional analysis method also includes obtaining the characteristic frequency of the hoisting mechanism bearing fault, and the characteristic frequency of the bearing fault includes:

[0013] The comprehensive frequency f of the fault characteristics of impact single-sided raceway bearing 1轴承 :

[0014]

[0015] Comprehensive frequency f of fault characteristics of impact double-sided raceway bearing 2轴承 :

[0016]

[0017] Among them, f o-out , f o-in , f bc1 and f bc2 are the characteristic frequencies of the outer ring fault of the lifting mechanism, the inner ring fault of the bearing, the rolling element fault of the bearing when impacting the single-side raceway, and the rolling element fault of the bearing when impacting the double-side raceway. d is the diameter of the rolling element of the lifting mechanism bearing, D m is the pitch diameter of the lifting mechanism bearing, z is the number of rolling elements in the lifting mechanism bearing, and α is the contact angle of the lifting mechanism bearing.

[0018] Further, based on the characteristic frequency, signal processing is combined with the fault mechanism, and the analysis means include time domain analysis, frequency domain analysis, envelope analysis and wavelet analysis;

[0019] The time domain analysis is to directly analyze the time waveform of the vibration signal to identify abnormal amplitude and impact signal characteristics, and is used to detect vibration, impact and imbalance faults; the indicators involved include mean, variance and peak factor;

[0020] The frequency domain analysis uses Fourier transform FFT to convert the vibration signal from the time domain to the frequency domain, analyzes the amplitude and phase of different frequency components in the signal, and effectively identifies the fault characteristics of bearings, gears or other rotating parts through spectrum analysis. The faults include gear wear and bearing damage.

[0021] The envelope analysis is to diagnose bearing defects by extracting the envelope curve of the vibration signal and identifying the specific frequencies of the bearing rolling elements, inner ring and outer ring;

[0022] The wavelet analysis performs local analysis in the time domain and the frequency domain through wavelet transform, and is used to capture transient changes in non-stationary signals.

[0023] Furthermore, in step 2, big data analysis is used to predict vibration characteristics under specified working conditions, including using the extreme gradient boosting algorithm XGBoost model in the supervised learning field and a multi-output regression model outside the XGBoost model to predict the characteristic values ​​of sensor vibration monitoring data, wherein the characteristic values ​​reflect the vibration characteristics of different dimensions at different positions of the lifting mechanism.

[0024] Furthermore, the supervised learning model XGBoost selects an objective function as the termination target of model training. The objective function of XGBoost is as follows. The formula includes three parts. The first part is the loss function, which is the main part of the objective function; the second part is the number of leaves, which indicates the complexity of the model tree; and the last part is the regularization term, which is used to prevent the model from overfitting.

[0025]

[0026] in, is the output parameter predicted by the model of the i-th sample, yi is the actual output parameter of the i-th sample, m represents the number of samples, T represents the number of leaf nodes of the regression tree, and γ is a hyperparameter.

[0027] Furthermore, the XGBoost model data includes:

[0028] ① Independent variables and dependent variables

[0029] The research object is the vibration phenomenon of the crane hoisting mechanism, and the crane working condition data is selected as the independent variable of the model; the sampling frequency of the vibration monitoring data and the working condition data is inconsistent, and the characteristics of the vibration monitoring data are refined into characteristic values ​​according to the sampling frequency of the working condition data as the dependent variable of the model;

[0030] The step of extracting the features of the vibration monitoring data into feature values ​​according to the sampling frequency of the working condition data includes: extracting features of the vibration monitoring data within the time interval of each working condition, and selecting feature values ​​of the vibration monitoring data of the vibration acceleration sensor in the time domain. The specific steps are as follows:

[0031] S1. Establish a dynamic time window of corresponding working conditions based on crane working condition statistics and divide the vibration monitoring data;

[0032] S2. Extracting time domain features from the vibration monitoring data in each time window;

[0033] S3, the characteristic value in each time window is taken as the vibration characteristic of the working condition;

[0034] ②Training set and test set

[0035] After the model training is completed, the generalization error of the trained model is evaluated through experimental testing. A test set data is used to test the model's ability to discriminate new samples. The error on the test set is used as the generalization error of the model, and the test set data should not overlap with the training set data.

[0036] Furthermore, the XGBoost model evaluation includes: after the model training is completed, judging the quality of the model's prediction by the statistics of the difference between the test set and the real data;

[0037] The XGBoost model parameter adjustment includes: using the XGBoost package in Python to build a vibration eigenvalue prediction model: selecting a set of data, and adjusting the model parameters by cross-validation of the training set and the test set of the XGBoost model under different numbers of classification trees, and the result of the parameter adjustment enables the trained model to perform better on the test set.

[0038] Furthermore, the XGBoost model training and prediction: Big data analysis reflects the changes in the vibration state of the lifting mechanism by comparing the differences between the predicted data and the actual data. When the difference between the data predicted by the model and the actual value becomes larger, the vibration state of the lifting mechanism is likely to have changed.

[0039] On the other hand, the present application also provides a crane hoisting mechanism operating status monitoring system, which includes a memory for storing computer program instructions and a processor for executing program instructions, wherein when the computer program instructions are executed by the processor, the system is triggered to execute the above method. BRIEF DESCRIPTION OF THE DRAWINGS

[0040] Figure 1 It is a schematic diagram of the vibration monitoring position of the crane hoisting mechanism of the present invention;

[0041] Figure 2 It is a cross-validation diagram of the training set and the test set of the XGBoost model of the present invention under different numbers of classification trees;

[0042] Figure 3A It is a schematic diagram of vibration monitoring data analysis of 100t and 40t lifting loads at 100% lifting speed of the crane No. 2 lifting mechanism of the present invention;

[0043] Figure 3B It is a schematic diagram of vibration monitoring data analysis of 100% and -100% lifting speeds of the crane No. 2 lifting mechanism of the present invention when the crane is unloaded with a capacity of 10t;

[0044] Figure 3C It is a schematic diagram of vibration monitoring data analysis of the crane No. 2 hoisting mechanism of the present invention at 70% and -35% hoisting speeds when the crane is unloaded with a load of 10t;

[0045] Figure 4A It is a schematic diagram of vibration monitoring data analysis of 90t and 20t lifting loads at 100% lifting speed of the crane 3# lifting mechanism of the present invention;

[0046] Figure 4B It is a schematic diagram of vibration monitoring data analysis of 90t and 20t lifting loads at -100% lifting speed of the crane 3# lifting mechanism of the present invention;

[0047] Figure 4C It is a schematic diagram of the vibration monitoring data analysis of the crane 3# hoisting mechanism of the present invention at -80% and -39% hoisting speeds under a 20t hoisting load condition. DETAILED DESCRIPTION

[0048] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative work are within the scope of protection of the present invention.

[0049] The embodiments of the present invention are described in detail below with reference to the accompanying drawings.

[0050] The crane hoisting mechanism operating status monitoring method and system provided in the embodiments of the present application are suitable for scenarios in which the operating status of the crane hoisting mechanism is monitored in real time and faults are prevented. Here, the operating status of the crane hoisting mechanism is the real-time equipment status of the crane.

[0051] There are corresponding risk areas when the crane is working. For example, the crane's lifting mechanism may cause the load to fall due to brake failure, the gantry may overturn due to structural instability and deformation, and the swivel mechanism may cause the boom rod system or the load to collide due to coupling failure. However, it is difficult for on-site safety managers to monitor the equipment status of the crane in real time. The equipment status of the crane is mainly judged based on static experience. It is impossible to evaluate the real-time operation risk of the crane in real time, and it is difficult to predict potential crane operation accidents. Remedies can only be carried out after the operation accident occurs, and personal injuries to staff cannot be avoided in time. The serious consequences are irreversible.

[0052] The operating status monitoring method of the crane hoisting mechanism provided in the embodiment of the present application can obtain the real-time operating data and status data of the crane hoisting mechanism, and based on the real-time operating data and status data of the crane hoisting mechanism, use professional analysis and big data analysis methods to fully combine professional mechanisms, industrial data, and algorithm models to find out the rules or abnormal characteristics in the data, discover the fault degradation trend and cause, and then evaluate the remaining life and reliability of the equipment, realize daily status evaluation and abnormal alarm and other functions, and predict the possibility of crane operation accidents according to the equipment status and take corresponding measures in time, thereby improving the safety of crane production, avoiding the occurrence of crane operation accidents, and improving the work efficiency of the crane.

[0053] Some embodiments of the present application provide a method for monitoring the operating status of a crane hoisting mechanism, comprising the following steps:

[0054] Step 1. Obtain the real-time working condition data and status data of the crane hoisting mechanism.

[0055] A crane refers to a multi-action lifting machine that can lift and carry heavy objects vertically and horizontally within a certain range. It is also called an overhead crane, an overhead crane, or a crane. The real-time working condition data of a crane's hoisting mechanism refers to the information data related to the working condition of the crane's hoisting mechanism; the status data of a crane's hoisting mechanism refers to the vibration monitoring data of the working condition of the hoisting mechanism obtained by the vibration sensor deployed on the hoisting mechanism. Combining the working condition data and vibration monitoring data of the crane's hoisting mechanism, through professional analysis and big data analysis methods, the abnormal fault of the crane's hoisting mechanism is determined, and the system warning is further triggered to remind relevant personnel to check and maintain the fault diagnosis site.

[0056] In some embodiments, the corresponding working condition data is obtained through the data collected by the PLC. The working condition data of the lifting mechanism generally describes the operating state of the lifting mechanism at a certain time point or in a short period of time. The working condition can change at any time. The working condition data of the crane lifting mechanism includes but is not limited to: lifting weight, lifting speed, lifting height, travel position, travel speed, etc.

[0057] In some embodiments, the status data refers to the vibration status data of the crane hoisting mechanism during the lifting process. By installing a vibration sensor on the bearing seat at the key transmission position of the hoisting mechanism, the sensor collects real-time vibration sensor data according to a preset collection frequency and provides it to the corresponding background processing equipment.

[0058] By analyzing the vibration monitoring data of the lifting mechanism under different working conditions, the data features related to the characteristic frequency of the lifting mechanism fault are found in the data, and appropriate characteristic values ​​are selected to describe them. The reason why the crane will produce violent vibration during lifting action is inferred, and then the lifting mechanism fault diagnosis is carried out based on the fault characteristics.

[0059] Exemplarily, sensors for collecting different types of data are respectively arranged at different parts of the crane, for example Figure 1 As shown in the figure, the sensor for collecting the vibration acceleration of the hoisting mechanism of the crane can be set on the housing of the reducer high-speed shaft end, the reducer low-speed shaft end, the pinion bearing seat (near the reduction box), the pinion bearing seat (near the drum), the drum bearing seat (near the reduction box) and the drum bearing seat (far from the reduction box) of the hoisting mechanism, with a total of six measuring points. Two piezoelectric acceleration sensors are set at each measuring point to monitor the vibration signals in the horizontal and vertical directions respectively. The specific installation positions are shown in Table 1 below.

[0060] Table 1: Installation position table of crane hoisting mechanism sensors

[0061]

[0062] In addition, in some embodiments, the types of sensors may include but are not limited to: acceleration sensors and strain sensors, etc. The strain sensor is used to collect structural strain data of the bottom plate under the crane main beam.

[0063] The data collected by the programmable logic controller PLC is collected by connecting the PLC to different parts of the crane and performing corresponding data processing. For example, the PLC is used to obtain the lifting weight, lifting height and lowering height at the lifting position of the crane, and the current, frequency, voltage, motor speed, etc. during operation are obtained at the trolley position of the crane. The embodiments of the present application do not make specific restrictions on this.

[0064] Step 2. Integrate the collected vibration monitoring data of the crane hoisting mechanism with the operating condition data, and use professional analysis and big data analysis methods to determine the abnormal fault of the crane hoisting mechanism.

[0065] Exemplarily, the vibration monitoring of the crane hoisting mechanism utilizes vibration sensors installed on the reducer and the bearing housing to collect vibration signals of the transmission process at a frequency of 1000 Hz.

[0066] Among them, the professional analysis based on the collected data of the crane hoisting mechanism and working condition data is as follows:

[0067] 1) Characteristic frequency of hoisting mechanism gear failure

[0068] The gear fault characteristic frequency f 齿轮 As follows:

[0069]

[0070] Among them, f r , f m and f c are the gear rotation frequency, gear meshing frequency and gear natural frequency when the hoisting mechanism encounters a fault, A and B are the fault condition data and fault vibration monitoring status data of a single hoisting mechanism when the hoisting mechanism of the crane encounters a fault, respectively. ele and f des are the motor operating frequencies corresponding to the A and B fault data encountered by the lifting mechanism during the ascending and descending processes, respectively; P is the number of pole pairs; k and m are the gear pair equivalent mass and gear average stiffness corresponding to the A and B fault data encountered by the lifting mechanism during the ascending and descending processes, respectively;

[0071] 2) Characteristic frequency of hoisting mechanism bearing failure

[0072] The comprehensive frequency f of the fault characteristics of impact single-sided raceway bearing 1轴承 :

[0073]

[0074] Comprehensive frequency f of fault characteristics of impact double-sided raceway bearing 2轴承 :

[0075]

[0076] Among them, f o-out , f o-in , f bc1 and f bc2are the characteristic frequencies of the outer ring fault of the lifting mechanism, the inner ring fault of the bearing, the rolling element fault of the bearing when impacting the single-side raceway, and the rolling element fault of the bearing when impacting the double-side raceway. d is the diameter of the rolling element of the lifting mechanism bearing, D m is the pitch diameter of the lifting mechanism bearing, z is the number of rolling elements in the lifting mechanism bearing, and α is the contact angle of the lifting mechanism bearing.

[0077] The professional analysis process mainly focuses on signal processing and fault mechanism analysis, including the following analysis methods:

[0078] 1. Time domain analysis: By directly analyzing the time waveform of the vibration signal, the abnormal amplitude, impact and other signal characteristics are identified to detect faults such as vibration, impact and imbalance. Indicators include mean, variance, peak factor, etc.

[0079] 2. Frequency domain analysis: Use Fourier transform (FFT) to convert the vibration signal from the time domain to the frequency domain, and analyze the amplitude and phase of different frequency components in the signal. Spectral analysis can effectively identify the fault characteristics of bearings, gears or other rotating parts, such as gear wear, bearing damage, etc.

[0080] 3. Envelope analysis: It is effective in detecting bearing faults. By extracting the envelope curve of the vibration signal, the specific frequencies of the bearing rolling elements, inner ring, and outer ring are identified to help diagnose bearing defects.

[0081] 4. Wavelet analysis: Wavelet transform performs local analysis in the time and frequency domains and is suitable for capturing transient changes in non-stationary signals, such as shock or fault characteristics.

[0082] By comparing the mechanism model with the collected vibration signals through the above-mentioned professional analysis methods, the key fault points in the crane hoisting mechanism can be identified.

[0083] On the other hand, big data analysis is performed on the data and working condition data of the crane hoisting mechanism. The purpose of big data analysis is to predict the vibration characteristics under specified working conditions, which belongs to the category of regression. The model extreme gradient boosting algorithm XGBoost (eXtreme Gradient Boosting) in the supervised learning field is selected. In addition, because the eigenvalues ​​of the sensor vibration monitoring data reflect the vibration characteristics of different dimensions at different positions of the hoisting mechanism, there is no way to use only one XGBoost model to predict all eigenvalues. Therefore, it is necessary to put a multi-output regression model (MultiOutputRegressor) outside the XGBoost training model algorithm to achieve a set of prediction models for different eigenvalues.

[0084] 1) XGBoost Model

[0085] XGBoost is one of the boosting algorithms. The idea of ​​the boosting algorithm is to integrate many weak classifiers together to form a strong classifier. XGBoost is a boosted tree model that integrates many tree models together to form a strong classifier.

[0086] The supervised learning model XGBoost needs to determine an objective function as the termination target of model training. The objective function of XGBoost is shown in the following formula and consists of three parts. The first part is the loss function, which is the main part of the objective function; the second part is the number of leaves, which indicates the complexity of the model tree; and the last part is the regularization term, which is used to prevent the model from overfitting.

[0087]

[0088] in, is the output parameter predicted by the model for the i-th sample, y i is the actual output parameter of the i-th sample, m represents the number of samples, T represents the number of leaf nodes of the regression tree, and γ is a hyperparameter.

[0089] 2) Model data

[0090] ① Independent variables and dependent variables

[0091] The research object is the vibration phenomenon of the lifting mechanism, so the crane operating condition data is the independent variable of the model; the sampling frequencies of the vibration monitoring data and the operating condition data are inconsistent, and it is impossible to directly train and predict the vibration monitoring data with the operating condition data one-to-many, so it is necessary to extract the characteristics of the vibration monitoring data into characteristic values ​​according to the sampling frequency of the operating condition data as the dependent variable of the model.

[0092] Among them, the characteristics of the vibration monitoring data are refined into characteristic values ​​according to the sampling frequency of the operating condition data, including: feature extraction of the vibration monitoring data within the time interval of each operating condition, and selection of characteristic values ​​of the sensor vibration monitoring data in the time domain, such as root mean square, average amplitude, variance, root mean square amplitude, peak value, peak-to-peak value, skewness, kurtosis, etc.

[0093] The specific steps of the refining process are as follows:

[0094] S1. Establish a dynamic time window of corresponding working conditions based on crane working condition statistics and divide the vibration monitoring data;

[0095] S2. Extracting time domain features from the vibration monitoring data in each time window;

[0096] S3. The characteristic value in each time window will be used as the vibration characteristic of the working condition.

[0097] ② After the training of the model is completed, the generalization error of the trained model needs to be evaluated through experimental testing. Therefore, a test set data is needed to test the model's ability to discriminate new samples, and the error on the test set is used as the "generalization error" of the model. In order to reflect the authenticity of the model's robustness, the test set data should not overlap with the training set data.

[0098] 3) Model evaluation

[0099] Model evaluation is the only way to judge the quality of the model after model training is completed. That is, some statistics that can reflect the difference between the test set and the real data are needed to judge the quality of the model's prediction. The R-square (adjusted mean square error) of the difference is selected to reflect the generalization error of the model.

[0100] 4) Model parameter adjustment

[0101] Directly use the XGBoost package in Python to build the vibration eigenvalue prediction model.

[0102] Select a set of data and adjust the model parameters through cross-validation. Take n_estimators (number of weak classifiers, number of classification trees) as an example. Figure 2 This is a cross-validation diagram of the model's training set and test set with different numbers of classification trees. Our parameter adjustment goal is to make the two lines in the figure below as close as possible so that the trained model can perform better on the test set.

[0103] 5) Model training and prediction

[0104] The main purpose of big data analysis is to reflect the changes in the vibration state of the lifting mechanism by comparing the difference between the predicted data and the actual data. In short, when the difference between the data predicted by the model and the actual value becomes larger, the vibration state of the lifting mechanism is likely to have changed.

[0105] Example 1

[0106] Part of the data from May 2024 was selected as the training set, and a group of models with better performance were trained as the prediction model for the vibration eigenvalues ​​of the lifting mechanism. Then, a few days of data were selected in each month as the test set, the eigenvalues ​​were predicted, and the difference between the predicted values ​​and the true values ​​was scored to judge the accuracy of the model prediction.

[0107] Since the vibrations exhibited by the hoisting mechanism of the crane at different hoisting speeds are different, three hoisting speeds that are more common in the working condition data are selected for comparison. For example, if the rated speed of the motor of the hoisting mechanism is 745rpm and the reduction ratio is 72.881, the characteristic frequencies corresponding to the speeds of 100%, 70%, and 35% are shown in Table 2 below, where the characteristic frequencies may include the motor output frequency and the reduction box output frequency, in Hz.

[0108] Table 2: Characteristic frequencies corresponding to 100%, 70%, and 35% lifting speeds

[0109] Lifting speed (%) Motor output frequency (Hz) Gearbox output frequency (Hz) 100 1 242 0.170 70 8.69 0.119 35 434 0.060

[0110] 1) Analysis of vibration monitoring data of 2# hoisting mechanism:

[0111] Vibration monitoring: The vibration monitoring data of the 2# lifting mechanism under different working conditions were selected for time domain and frequency domain analysis. The analysis diagram is as follows: Figure 3A , 3B As shown in 3C, by comparison, the lifting vibration of the 2# lifting mechanism is relatively large under all working conditions, and the vibration in the vertical direction is generally larger than that in the horizontal direction. The vibration at the high-speed shaft end of the reduction gearbox (near the brake) is relatively minimal; the vertical vibration at the low-speed shaft end of the reduction gearbox is the largest; and the vibration component on the bearing seat at the driven end of the drum is the most complex. The vibration of the 2# lifting mechanism during the "descending" action is more serious than the vibration of the "ascending" action, and the impact vibration is more obvious; the vibration waveform during the "ascending" action shows a certain asymmetry, which is mainly manifested in the reduction gearbox. In addition, the vibration of the 2# lifting mechanism has shown a significant deterioration trend since monitoring, indicating that the existing fault is becoming increasingly serious and should be repaired as soon as possible.

[0112] ② Vibration source location analysis

[0113] Analysis of the vibration amplitude revealed that the time domain waveform of the 2# hoisting mechanism showed a certain asymmetry under the "rising" condition. The vibration monitoring data on the gearbox was particularly obvious. It was speculated that there was a loose friction fault in the gearbox, and the prominent peak period was consistent with the output frequency of the gearbox. Under the "descending" condition, the asymmetric vibration characteristics disappeared. It was speculated that the friction problem should be caused by the output rotation of the gearbox, and the constraint of the base at the low-speed shaft end of the gearbox was weakened. In addition, there was obvious impact vibration in the vibration waveform. The period of the impact vibration was about the output frequency of the gearbox, and it was more obvious at both ends of the drum. It was speculated that it was caused by the deterioration of the alignment between the drum and the low-speed shaft of the gearbox.

[0114] In addition, among the vibration frequencies of the sensors on the reduction gearbox, the frequencies with larger amplitudes are ≈17 times and 34 times the output frequency of the motor, and are surrounded by sideband clusters spaced at the motor output frequency. The overall amplitude is shown as higher at the high-speed shaft end than at the low-speed shaft end, and higher at the motor end than at the brake end. Therefore, it is inferred that the vibration source is the primary meshing frequency in the reduction gearbox, and when the lifting speed is 100%, the vibration amplitude of this frequency is particularly prominent, which is inferred to have caused resonance. In addition, the frequency ≈8 times the output frequency of the motor and its multiples are also relatively obvious, and mostly appear on the high-speed shaft of the reduction gearbox. It is suspected to be the secondary meshing frequency in the reduction gearbox. It is recommended to check the reduction gearbox for faults.

[0115] 2) Analysis of vibration monitoring data of 3# hoisting mechanism:

[0116] ① Time-frequency domain feature analysis

[0117] The vibration monitoring data of the 3# lifting mechanism under different working conditions were selected for time domain and frequency domain analysis. The analysis diagram is shown in Figure 4A , 4B and 4C.

[0118] By comparison, the vibration of the 3# lifting mechanism is good, and the vibration is greatly affected by the lifting load. The vibration at the high-speed shaft end of the gearbox (near the motor) is relatively large; the vertical vibration of the low-speed shaft end of the gearbox is the largest; the vibration components on the bearing seat at the driven end of the drum are the most complex.

[0119] ② Vibration source location analysis

[0120] Analysis of the vibration amplitude revealed that when the 3# lifting mechanism was lifting under load, the frequencies of the sensor vibration on the reduction gearbox had larger amplitudes of ≈17 times and 34 times the motor output frequency, and were surrounded by side frequency clusters at intervals of the motor output frequency. It was inferred that the vibration source was the primary meshing frequency in the reduction gearbox. In addition, a frequency of ≈8 times the motor output frequency was also relatively obvious, and mostly appeared on the high-speed shaft of the reduction gearbox, suspected to be the secondary meshing frequency in the reduction gearbox. Multiple frequency vibrations occurred in the vibration, but the overall vibration amplitude was not large, and it can be inferred that there is no obvious fault manifestation for the time being.

[0121] Some embodiments of the present application also provide a crane hoisting mechanism operating status monitoring system, the system comprising a memory for storing computer program instructions and a processor for executing the program instructions, wherein when the computer program instructions are executed by the processor, the system is triggered to execute the aforementioned crane hoisting mechanism operating status monitoring method.

[0122] In summary, the solution provided by the present application can obtain the real-time working condition data and status data of the crane hoisting mechanism, and integrate the collected vibration signal of the crane hoisting mechanism with the working condition data, and use professional analysis and big data analysis methods to determine the abnormal fault of the crane hoisting mechanism. The present application can monitor the equipment status of the crane hoisting mechanism in real time, and predict the possibility of an operation accident of the crane hoisting mechanism according to the equipment status and take corresponding measures in time, thereby improving the safety of the crane during production, avoiding the occurrence of crane operation accidents, and improving the working efficiency of the crane.

[0123] It should be noted that the present application can be implemented in software and / or a combination of software and hardware, for example, can be implemented using an application specific integrated circuit (ASIC), a general purpose computer or any other similar hardware device. In one embodiment, the software program of the present application can be executed by a processor to implement the steps or functions described above. Similarly, the software program of the present application (including relevant data structures) can be stored in a computer-readable recording medium, for example, a RAM memory, a magnetic or optical drive or a floppy disk and similar devices. In addition, some steps or functions of the present application can be implemented using hardware, for example, as a circuit that cooperates with a processor to perform each step or function.

[0124] In a typical configuration of the present application, the terminal and the network device each include one or more processors (CPU), input / output interface, network interface and memory.

[0125] The memory may include non-permanent storage in a computer-readable medium, random access memory (RAM) and / or non-volatile memory in the form of read-only memory (ROM) or flash RAM. The memory is an example of a computer-readable medium.

[0126] Computer readable media include permanent and non-permanent, removable and non-removable media that can be implemented by any method or technology to store information. Information can be computer readable instructions, data structures, program modules or other data. Examples of computer storage media include, but are not limited to, phase change memory (PRAM), static random access memory (SRAM), dynamic random access memory (DRAM), other types of random access memory (RAM), read-only memory (ROM), electrically erasable programmable read-only memory (EEPROM), flash memory or other memory technology, read-only compact disk (CD-ROM), digital versatile disk (DVD) or other optical storage, magnetic cassettes, magnetic tape disk storage or other magnetic storage devices or any other non-transmission media that can be used to store information that can be accessed by a computing device. As defined herein, computer readable media does not include non-transitory media such as modulated data signals and carrier waves.

[0127] In addition, a part of the present application may be applied as a computer program product, such as a computer program instruction, which, when executed by a computer, can call or provide the method and / or technical solution according to the present application through the operation of the computer. The program instruction for calling the method of the present application may be stored in a fixed or removable recording medium, and / or transmitted through a data stream in a broadcast or other signal-bearing medium, and / or stored in a working memory of a computer device that runs according to the program instruction. Here, according to an embodiment of the present application, a device is included, the device including a memory for storing computer program instructions and a processor for executing program instructions, wherein, when the computer program instruction is executed by the processor, the device is triggered to run the method and / or technical solution based on the aforementioned multiple embodiments according to the present application.

[0128] It is obvious to those skilled in the art that the present application is not limited to the details of the above exemplary embodiments, and that the present application can be implemented in other specific forms without departing from the spirit or basic features of the present application. Therefore, from any point of view, the embodiments should be regarded as exemplary and non-restrictive, and the scope of the present application is defined by the attached claims rather than the above description, and it is intended that all changes falling within the meaning and scope of the equivalent elements of the claims are included in the present application. Any figure mark in the claims should not be regarded as limiting the claims involved. In addition, it is obvious that the word "comprising" does not exclude other units or steps, and the singular does not exclude the plural. Multiple units or devices stated in the device claim can also be implemented by one unit or device through software or hardware.

Claims

1. A method for monitoring the operating status of a crane hoisting mechanism, characterized in that: The steps include: Step 1. Acquire real-time working condition data and status data of the crane hoisting mechanism, wherein the working condition data is information data related to the working condition of the crane hoisting mechanism; the status data is vibration monitoring data of the hoisting mechanism working condition collected by a vibration acceleration sensor deployed on the hoisting mechanism; Step 2. Combine the collected vibration monitoring data of the crane hoisting mechanism with the operating condition data, and use professional analysis and big data analysis methods to determine the abnormal fault of the crane hoisting mechanism.

2. The method according to claim 1, characterized in that The vibration acceleration sensor is arranged at the high-speed shaft end of the reducer, the low-speed shaft end of the reducer, the pinion bearing seat near the reduction box, the pinion bearing seat near the drum, the drum bearing seat near the reduction box and the drum bearing seat of the far reduction box of the crane hoisting mechanism.

3. The method according to claim 1, characterized in that The professional analysis method in step 2 includes obtaining the characteristic frequency of the hoisting mechanism gear failure, and the gear failure characteristic frequency f 齿轮 As follows: ； Among them, f r , f m and f c are the gear rotation frequency, gear meshing frequency and gear natural frequency when the hoisting mechanism encounters a fault, A and B are the fault condition data and fault vibration monitoring status data of a single hoisting mechanism when the hoisting mechanism of the crane encounters a fault, respectively. ele and f des are the motor operating frequencies corresponding to the A and B fault data encountered by the lifting mechanism during the ascending and descending processes, respectively; P is the number of pole pairs; k and m are the gear pair equivalent mass and gear average stiffness corresponding to the A and B fault data encountered by the lifting mechanism during the ascending and descending processes, respectively; The professional analysis method also includes obtaining the characteristic frequency of the hoisting mechanism bearing fault, and the characteristic frequency of the bearing fault includes: The comprehensive frequency f of the fault characteristics of impact single-sided raceway bearing 1轴承 : ； Comprehensive frequency f of fault characteristics of impact double-sided raceway bearing 2轴承 : ； Among them, f o-out , f o-in , f bc1 and f bc2 are the characteristic frequencies of the outer ring fault of the lifting mechanism, the inner ring fault of the bearing, the rolling element fault of the bearing when impacting the single-side raceway, and the rolling element fault of the bearing when impacting the double-side raceway. d is the diameter of the rolling element of the lifting mechanism bearing, D m is the pitch diameter of the lifting mechanism bearing, z is the number of rolling elements in the lifting mechanism bearing, and α is the contact angle of the lifting mechanism bearing.

4. The method according to claim 3, characterized in that Based on the characteristic frequency, signal processing is combined with the fault mechanism, and the analysis means include time domain analysis, frequency domain analysis, envelope analysis and wavelet analysis; The time domain analysis is to directly analyze the time waveform of the vibration signal to identify abnormal amplitude and impact signal characteristics, and is used to detect vibration, impact and imbalance faults; the indicators involved include mean, variance and peak factor; The frequency domain analysis uses Fourier transform FFT to convert the vibration signal from the time domain to the frequency domain, analyzes the amplitude and phase of different frequency components in the signal, and effectively identifies the fault characteristics of bearings, gears or other rotating parts through spectrum analysis. The faults include gear wear and bearing damage. The envelope analysis is to diagnose bearing defects by extracting the envelope curve of the vibration signal and identifying the specific frequencies of the bearing rolling elements, inner ring and outer ring; The wavelet analysis performs local analysis in the time domain and the frequency domain through wavelet transform, and is used to capture transient changes in non-stationary signals.

5. The method according to claim 4, characterized in that In step 2, big data analysis is used to predict the vibration characteristics under specified working conditions, including using the extreme gradient boosting algorithm XGBoost model in the supervised learning field and a multi-output regression model outside the XGBoost model to predict the characteristic values ​​of the sensor vibration monitoring data, wherein the characteristic values ​​reflect the vibration characteristics of different dimensions at different positions of the lifting mechanism.

6. The method according to claim 5, characterized in that The supervised learning model XGBoost selects an objective function as the termination target of model training. The objective function of XGBoost is as follows. The formula includes three parts. The first part is the loss function, which is the main part of the objective function; the second part is the number of leaves, which indicates the complexity of the model tree; the last part is the regularization term, which is used to prevent the model from overfitting; ； in, is the output parameter predicted by the model of the i-th sample, yi is the actual output parameter of the i-th sample, m represents the number of samples, T represents the number of leaf nodes of the regression tree, and γ is a hyperparameter.

7. The method according to claim 6, characterized in that The XGBoost model data includes: ① Independent variables and dependent variables The research object is the vibration phenomenon of the crane hoisting mechanism, and the crane working condition data is selected as the independent variable of the model; the sampling frequency of the vibration monitoring data and the working condition data is inconsistent, and the characteristics of the vibration monitoring data are refined into characteristic values ​​according to the sampling frequency of the working condition data as the dependent variable of the model; The step of extracting the features of the vibration monitoring data into feature values ​​according to the sampling frequency of the working condition data includes: extracting features of the vibration monitoring data within the time interval of each working condition, and selecting feature values ​​of the vibration monitoring data of the vibration acceleration sensor in the time domain. The specific steps are as follows: S1. Establish a dynamic time window of corresponding working conditions based on crane working condition statistics and divide the vibration monitoring data; S2. Extracting time domain features from the vibration monitoring data in each time window; S3, the characteristic value in each time window is taken as the vibration characteristic of the working condition; ②Training set and test set After the model training is completed, the generalization error of the trained model is evaluated through experimental testing. A test set data is used to test the model's ability to discriminate new samples. The error on the test set is used as the generalization error of the model, and the test set data should not overlap with the training set data.

8. The method according to claim 7, characterized in that The XGBoost model evaluation includes: After the model training is completed, the statistical difference between the test set and the real data is used to judge the quality of the model's prediction; The XGBoost model parameter adjustment includes: using the XGBoost package in Python to build a vibration eigenvalue prediction model: selecting a set of data, and adjusting the model parameters by cross-validation of the training set and the test set of the XGBoost model under different numbers of classification trees, and the result of the parameter adjustment enables the trained model to perform better on the test set.

9. The method according to claim 8, characterized in that The XGBoost model training and prediction: Big data analysis reflects the changes in the vibration state of the lifting mechanism by comparing the differences between the predicted data and the actual data. When the difference between the data predicted by the model and the actual value becomes larger, the vibration state of the lifting mechanism is likely to have changed.

10. A crane hoisting mechanism operating status monitoring system, the system comprising a memory for storing computer program instructions and a processor for executing program instructions, wherein: When the computer program instructions are executed by the processor, the system is triggered to execute the method according to any one of claims 1 to 9.

Images