Elevator steel wire rope fault diagnosis and fatigue prediction method and system

Abstract

The embodiment of the invention relates to the technical field of information, in particular to an elevator steel wire rope fault diagnosis and fatigue prediction method and system. The method comprises the following steps: collecting operation data when a single elevator normally operates, wherein the operation data comprises sound data and vibration data; according to the operation data during normal operation of the single ladder, establishing a fatigue baseline of the single ladder; collecting operation data when the elevator runs under the fault, and associating the operation data with a corresponding fault type; establishing a fault diagnosis model according to the operation data of the associated fault; acquiring operation data of a single ladder, and comparing the operation data with the fault diagnosis model to obtain a single ladder fault diagnosis result; and comparing the operation data with the single-ladder fatigue baseline to obtain a single-ladder fatigue prediction result.

Description

Technical Field

[0001] Multiple embodiments of this specification relate to the field of information technology, and specifically to a method and system for diagnosing elevator wire rope faults and predicting fatigue. Background Art

[0002] The elevator wire rope is used as an elevator suspension device, connecting the counterweight and the elevator car, and is suspended on the traction wheel and the anti-ropes wheel. Its normal operation depends on the friction between the traction rope groove and the wire rope to ensure the up and down movement of the elevator car and the counterweight. As the core load-bearing component of the traction elevator, the elevator wire rope faces the risk of being easily damaged after frequent use. Faults such as uneven wire rope tension, wear, and breakage may cause the elevator to run unsteadily, affecting riding comfort and normal use, or even cause the car to slip and hit the top, causing a serious safety accident. Therefore, in elevator maintenance, the monitoring and analysis of elevator wire ropes is particularly important. The existing monitoring technologies are mainly based on video detection, electromagnetic data, and tension monitoring and analysis, and a few use vibration analysis methods. On-site detection methods such as electromagnetic and tension require high on-site installation and detection conditions, and it is difficult to conduct continuous monitoring during the daily use and operation of the elevator. For example, the patent "Elevator Wire Rope Fault Detection Method Based on the Internet of Things" discloses a method of collecting acceleration data during the operation of the elevator car, uploading it to the cloud, and using the model in the online database to determine the wire rope fault. The technical problems it has are: 1) It is mainly based on pattern judgment (wavelet analysis) and big data analysis, without considering the unique vibration characteristics of elevators of the same structure caused by different structural models and installation reasons, and the model generalization ability is weak; 2) The collected acceleration data is not expanded according to multiple axes, the collection frequency is low, and it is impossible to obtain real and complete data characteristics. The patent application "A method for detecting elevator traction, wire rope slippage and defects" discloses a machine vision and its detection system through an AI algorithm camera, which collects images and videos of wire rope movement during elevator braking, and performs pattern recognition on images of defective parts of elevator wire ropes. The technical problems it has are: 1) It is impossible to detect wire rope faults without surface damage, such as uneven tension, inconsistent spring compression at the fixed end of the wire rope head, etc.; 2) It only has the ability to detect, and the fault standard is determined manually, and it is impossible to predict faults. Summary of the invention

[0003] Multiple embodiments of this specification describe a method and system for diagnosing elevator wire rope faults and predicting fatigue.

[0004] In a first aspect, the embodiments of this specification provide a method for diagnosing elevator wire rope faults and predicting fatigue, comprising the steps of: Collecting operation data of a single elevator during normal operation, wherein the operation data includes sound data and vibration data; Establishing a fatigue baseline of a single elevator based on the operating data during normal operation of the single elevator; Collect the running data of the elevator when it is running under fault conditions and associate it with the corresponding fault type; Establishing a fault diagnosis model based on the operating data associated with the fault; Collecting operation data of a single elevator, comparing the operation data with the fault diagnosis model, and obtaining a single elevator fault diagnosis result; The operating data is compared with the single-ladder fatigue baseline to obtain a single-ladder fatigue prediction result.

[0005] In a second aspect, the embodiments of this specification provide an elevator wire rope fault diagnosis and fatigue prediction system, including: A collection module, for collecting operation data of a single elevator during normal operation, wherein the operation data includes sound data and vibration data; A baseline module, establishing a fatigue baseline of a single elevator according to the operating data of the single elevator during normal operation; The fault module collects the running data of the elevator under fault and associates it with the corresponding fault type; A model module, establishing a fault diagnosis model according to the operation data associated with the fault; A diagnosis module collects the operation data of a single elevator, compares the operation data with the fault diagnosis model, and obtains a single elevator fault diagnosis result; The prediction module compares the operating data with the single-ladder fatigue baseline to obtain a single-ladder fatigue prediction result.

[0006] In a third aspect, an embodiment of this specification provides an electronic device, including a processor and a memory; The processor is connected to the memory; The memory is used to store executable program code; The processor runs a program corresponding to the executable program code by reading the executable program code stored in the memory, so as to execute the method described in any one of the above aspects.

[0007] In a fourth aspect, an embodiment of the present specification provides a computer-readable storage medium having a computer program stored thereon, wherein the computer program, when executed by a processor, implements the method described in any of the above aspects.

[0008] In a fifth aspect, an embodiment of this specification provides a computer program product, including a computer program, which implements the method described in any of the above aspects when executed by a processor.

[0009] The beneficial effects brought by the technical solutions provided by some embodiments of this specification include at least: In multiple embodiments of the present specification, the provided fault diagnosis and fatigue prediction method diagnoses the fault of the wire rope and analyzes and predicts the fatigue according to the sound and vibration data collected at the maintenance site. It can not only obtain the fault condition of the wire rope, but also obtain the fatigue prediction result of the wire rope according to the comparison of the fatigue baseline of the single ladder, which is helpful to more accurately grasp the fatigue of the wire rope and ensure the normal operation of the single ladder. It can realize the identification of fault precursors and provide relevant information when the fault precursor occurs, which is helpful to carry out targeted maintenance and prevent the occurrence of faults, so that the diagnosis of faults no longer depends on the experience of maintenance personnel. By accumulating operation data and corresponding labels, the accumulation of big data is realized, and a general aging trend model of wire rope is established based on the accumulated data, which is helpful to predict the fatigue life of the wire rope of the newly installed single ladder, provide reference data for maintenance, and ensure the normal and stable operation of the single ladder.

[0010] Other features and advantages of the various embodiments of the present specification will be further disclosed in the following detailed description and drawings. BRIEF DESCRIPTION OF THE DRAWINGS

[0011] In order to more clearly illustrate the technical solutions in the embodiments of this specification, the drawings required for use in the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of this specification. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0012] Figure 1 A schematic diagram of the application architecture of the fault diagnosis and fatigue prediction method provided in the embodiments of this specification.

[0013] Figure 2 A schematic diagram of the fault diagnosis and fatigue prediction method provided in the embodiments of this specification.

[0014] Figure 3 A schematic diagram of a single-ladder fatigue baseline provided in an embodiment of this specification.

[0015] Figure 4 A schematic diagram of a method for obtaining a single-ladder fatigue prediction result provided in an embodiment of this specification.

[0016] Figure 5 A schematic diagram of a method for establishing a fault diagnosis model provided in an embodiment of this specification.

[0017] Figure 6 Schematic diagram of the fault diagnosis and fatigue prediction system provided in the embodiments of this specification.

[0018] Figure 7 A schematic diagram of an electronic device provided in an embodiment of this specification. DETAILED DESCRIPTION

[0019] The following is an explanation and description of the technical solutions of the embodiments of this specification in conjunction with the drawings of the embodiments of this specification, but the following embodiments are only preferred embodiments of this specification, not all. Based on the embodiments in the implementation mode, other embodiments obtained by those skilled in the art without creative work are all within the scope of protection of this specification.

[0020] The terms "first", "second", "third", etc. in the description and claims of this specification and the above-mentioned drawings are used to distinguish different objects, rather than to describe a specific order. In addition, the terms "including" and "having" and any variations thereof are intended to cover non-exclusive inclusions. For example, a process, method, system, product or device that includes a series of steps or units is not limited to the listed steps or units, but optionally includes steps or units that are not listed, or optionally includes other steps or units inherent to these processes, methods, products or devices.

[0021] In the following description, terms such as "inside", "outside", "up", "down", "left", "right", etc. that indicate directions or positional relationships are only used to facilitate the description of the embodiments and simplify the description, and do not indicate or imply that the device or element referred to must have a specific direction, be constructed and operated in a specific direction, and therefore should not be understood as a limitation of this specification.

[0022] The data involved in this application are all information and data authorized by the user or fully authorized by all parties, and the collection of relevant data complies with the relevant laws, regulations and standards of relevant countries and regions.

[0023] Before describing the technical solution in this specification, the application scenarios and related technologies of the technical solution are introduced.

[0024] As an important vertical transportation tool, elevators are widely used in various buildings, such as residential buildings, office buildings, hotels, hospitals, etc. The basic structure of an elevator includes a car, a door system, a traction system, a guide system, an electrical control system, etc. With the development of technology, modern elevators have not only achieved efficient and fast transportation functions, but also made significant progress in energy saving, safety, and intelligence. Elevator maintenance is an important part of ensuring the normal operation of elevators, extending their service life, and ensuring the safety of passengers. Elevator maintenance refers to a series of preventive maintenance work carried out to ensure the safe and reliable operation of elevators. Effective elevator maintenance can extend the service life of elevators, reduce the occurrence of failures, and ensure the personal safety of passengers. Maintenance work usually includes regular inspections of the status of various elevator components, such as guide rails, rollers, door machine systems, brakes, limit switches, etc., as well as testing of control systems and electrical circuits. In addition, it also includes operations such as lubrication, tightening, and adjustment to ensure that each component is in the best working condition.

[0025] The maintenance cycle is generally divided into weekly inspection, monthly inspection, quarterly inspection, semi-annual inspection and annual inspection. The specific frequency is based on the frequency of elevator use and the manufacturer's recommendation. Maintenance work is performed by professional elevator companies or third-party service agencies, who are usually equipped with professional technicians and advanced detection tools. Common elevator faults include but are not limited to elevator failure to start, vibration or abnormal noise during operation, door switch failure, overload alarm, etc. These problems may be caused by failures in the electrical control system, wear of mechanical parts, sensor failure, etc. For these problems, maintenance personnel will conduct detailed inspections based on the fault phenomenon and take corresponding repair measures. For example, for electrical control system failures, it may be necessary to check the circuit connection and replace damaged components; for wear of mechanical parts, it is necessary to replace or adjust the corresponding parts.

[0026] As an indispensable means of transportation in modern buildings, the safe and stable operation of elevators is of vital importance. Through scientific and reasonable maintenance plans and professional and meticulous maintenance work, the safety and reliability of elevators can be maximized and the user experience can be improved. However, the current elevator maintenance work only relies on manual fault diagnosis and lacks the ability to predict elevator faults and health, which makes the elevator still in a risky state. Therefore, it is necessary to continue to study elevator maintenance technology.

[0027] Please refer to the attached Figure 1 This specification records the schematic diagram of the application scenario of the technical solution. In the elevator maintenance task, this specification discloses the technical solution to install the collection module 11 in the designated parts and positions of the elevator to collect the relevant operation data of the wire rope 10 during the up and down process of the single elevator. Exemplarily, it is installed near the anti-ropes of the elevator, and the specific installation position is determined according to the disclosed data collection technology. Specifically, for the elevator with the anti-ropes at the top of the car, the collection module 11 is installed near the anti-ropes at the top of the car; for the elevator with the anti-ropes at the bottom of the car, the collection module 11 is installed at the bottom of the car. The collection module 11 is recommended to collect vibration data (6-axis vibration data) and sound data. Temperature and humidity data can also be collected. The sound data is collected by propagating to the collection module 11 through the air, so a shock-absorbing structure or shock-absorbing material is set between the device for collecting sound data and the elevator. The device for collecting vibration data needs to be as rigidly connected to the elevator as possible. The data collected by the collection module 11 is sent to the terminal device 20 at the maintenance site, and the terminal device 20 has a communication connection with the server 30.

[0028] In the technical solution, the server 30 can be an independent physical server, or a server cluster or distributed system composed of multiple physical servers, or a cloud server that provides basic cloud computing services such as cloud services, cloud databases, cloud computing, cloud functions, cloud storage, network services, cloud communications, middleware services, domain name services, security services, content delivery networks (CDN), and big data and artificial intelligence platforms. The terminal device 20 can be a smart phone, a tablet computer, a laptop, a PDA, a personal computer, a smart speaker, a smart TV, a smart watch, a vehicle-mounted device, a wearable device, etc., but is not limited to this. The terminal device 20 and the server 30 can be directly or indirectly connected by wired or wireless communication, and this application does not limit this. The number of servers 30 and terminal devices 20 is also not limited. The fault diagnosis and fatigue prediction of the elevator wire rope 10 can be completed independently by the terminal device 20, or by the server 30, or by the terminal device 20 and the server 30 in cooperation, and this application does not make specific limitations.

[0029] Since this application involves some professional terms, these professional terms will be introduced below.

[0030] Wire rope bending times The number of bending times (or bending fatigue life) of a wire rope refers to the number of cycles that the wire rope 10 can withstand during repeated bending until it breaks or reaches a predetermined performance degradation index. The number of bending times of the wire rope 10 is affected by many factors, including but not limited to the design, manufacturing process, material hardness, diameter, rope core type, lubrication, and use conditions of the wire rope 10. Generally speaking, the more reasonable the design of the wire rope 10 and the better the material, the more times it can be bent. In order to evaluate the number of bending times of the wire rope 10, a bending fatigue test under laboratory conditions is usually carried out. During the test, the wire rope 10 will repeatedly pass through a bending wheel with a certain radius of curvature in a special device, and the number of cycles before breaking will be recorded. This data can be used to guide the selection and maintenance strategy of the wire rope 10 in practical applications. In actual use, in order to ensure safety and extend service life, a certain safety factor is usually considered during design, and regular inspection and maintenance are carried out according to the specific use environment. For example, in the elevator industry, the wire rope 10 is one of the key safety components, and its inspection and replacement have strict standards. The maintenance personnel will regularly check the wear condition, number of broken wires and other indicators of the wire rope 10, and perform maintenance or replacement according to the manufacturer's recommendation.

[0031] Running mileage The elevator moves up and down between different floors every day, and the total running distance accumulated is expressed in mileage. The running mileage can be used to evaluate the frequency of use, maintenance needs and life expectancy of the elevator. The running mileage of the elevator can be recorded by the counter installed in the elevator control system every time the elevator goes from one floor to another and then returns, and multiplied by the distance between floors. Modern elevator systems usually integrate intelligent monitoring systems that can automatically record the operation data of the elevator and upload it to the cloud for analysis.

[0032] Theoretical Baseline The theoretical baseline is the vibration and noise of the wire rope 10 during operation, obtained under laboratory conditions or theoretical simulation results. The theoretical baseline is derived from an idealized theoretical model or data under specific experimental conditions, and can provide a reference data. It includes the vibration and noise curves of each operating state under the elevator maintenance condition. The elevator maintenance condition is determined by the publicly available technology in the art.

[0033] Machine Learning Models Machine learning models refer to methods that use algorithms and statistical models to enable computer systems to learn from data to improve their performance or make predictions without being explicitly programmed. Machine learning is a subfield of artificial intelligence that allows computers to learn from experience and improve their capabilities without being explicitly programmed. Machine learning models can be divided into the following categories: supervised learning, unsupervised learning, semi-supervised learning, and reinforcement learning.

[0034] Clustering Algorithms Clustering algorithm is an unsupervised learning method that aims to divide the objects in the data set into several clusters (clusters) so that the objects within the cluster are similar to each other, while the objects between different clusters are quite different. Clustering algorithms are widely used in many fields such as data analysis, pattern recognition, market segmentation, social network analysis, etc. By measuring the similarity or distance between data points, the data set is divided into several groups, and the data points in each group share certain characteristics. The goal of clustering is to make the data points within the cluster as similar as possible, while the differences between clusters are as large as possible. Clustering algorithms include partitioning clustering such as K-means. K-means is one of the most commonly used clustering algorithms. It divides the data into a preset number of clusters through an iterative process. The center point (centroid) of each cluster represents the average attribute of the cluster. K-medoids is similar to K-means, but the centroid must be the actual data point in the cluster. Hierarchical clustering, such as agglomerative hierarchical clustering, merges the nearest clusters from the bottom up to form a tree structure. Divisive hierarchical clustering divides the data set into smaller and smaller clusters from the top down. Density-based clustering, such as DBSCAN, discovers clusters of arbitrary shapes by measuring the density of the neighborhood around a point and can identify noise points. OPTICS extends the idea of ​​DBSCAN and provides a hierarchical density clustering method. Grid-based clustering, such as STING, divides the space into cells and performs statistical analysis at the cell level. CLIQUE combines the advantages of grid division and hierarchical clustering and can efficiently process large-scale data sets. Model-based clustering, such as the EM algorithm, uses a mixed model (such as a Gaussian mixture model) for clustering and estimates the probability distribution by iteratively optimizing parameters.

[0035] Autoencoder Model The autoencoder model is an autoencoder. Autoencoder is an unsupervised learning algorithm that is mainly used for data dimensionality reduction, feature extraction and data reconstruction. It implements this process through two main parts: encoder and decoder. The autoencoder includes encoder, latent space and decoder. Among them, the encoder compresses the input data into a low-dimensional latent space representation. This process is usually implemented through several layers of neural networks, aiming to extract key features or information from the input data. Latent space is a low-dimensional representation of the encoder output, also known as code or bottleneck. The representation in this space is a compressed form of the input data and contains the key information needed to reconstruct the original data. The decoder decodes the low-dimensional latent space representation back to the dimension of the original data. The decoder also consists of several layers of neural networks, and its goal is to reconstruct the original input data as accurately as possible. The autoencoder is trained and used in two stages. In the training stage, the input data x is mapped to the latent space z through the encoder, expressed as z=f(x). Then, the latent space representation z is used to reconstruct the original data x' through the decoder, expressed as x'=g(z)=g(f(x)). The training goal is to minimize the reconstruction error, that is, the difference between x and x'. The commonly used loss function is the mean square error (MSE). The use stage is after the training is completed. The encoder can be used to map new data to a low-dimensional latent space for feature extraction or dimensionality reduction; the decoder can generate data from the latent space representation and be applied to tasks such as generating models.

[0036] This manual first provides a method for diagnosing elevator wire rope faults and predicting fatigue. Figure 2 , including the steps of: Step S101) collecting operation data of a single elevator during normal operation, wherein the operation data includes sound data and vibration data; Step S102) establishing a fatigue baseline of a single elevator according to the operating data of the single elevator during normal operation; Step S103) collecting the running data of the elevator when it is running under fault conditions and associating it with the corresponding fault type; Step S104) establishing a fault diagnosis model according to the operation data associated with the fault; Step S105) collecting the operation data of a single elevator, comparing the operation data with the fault diagnosis model, and obtaining a single elevator fault diagnosis result; Step S106) Compare the operating data with the single-elevator fatigue baseline to obtain a single-elevator fatigue prediction result.

[0037] Sound data refers to the sound that is generated by the air during the operation of a single elevator. Vibration data refers to the mechanical vibration data collected by a collection device that is tightly connected to the car. Vibration data can also be subdivided into three-axis vibration data or six-axis vibration data.

[0038] The single-echelon fatigue baseline corresponds to the sound data and vibration data. This correspondence includes the following situations: weighted fusion of sound data and vibration data to obtain fused data, and the single-echelon fatigue baseline represents the relationship baseline between the running mileage and the frequency domain composition of the fused data; three-axis vibration data or six-axis vibration data are fused to obtain vibration fusion data, and the single-echelon fatigue baseline includes the relationship baseline between the frequency domain composition of the running mileage and the sound data, and the relationship baseline between the frequency domain composition of the running mileage and the vibration fusion data; the vibration data is three-axis vibration data, and the single-echelon fatigue baseline includes the relationship baseline between the frequency domain composition of the running mileage and the sound data, and three relationship baselines that respectively represent the frequency domain composition of the running mileage and the three-axis vibration fusion data; the vibration data is six-axis vibration data, and the single-echelon fatigue baseline includes the relationship baseline between the frequency domain composition of the running mileage and the sound data, and six relationship baselines that respectively represent the frequency domain composition of the running mileage and the six-axis vibration fusion data.

[0039] The above is for the sound data and vibration data without distinguishing between the single elevator going up and down. When distinguishing between the single elevator going up and down, the number of corresponding relationship curves will double. Please refer to the case of not distinguishing between the single elevator going up and down for processing.

[0040] Please refer to the attached Figure 3 , is an example of a single-step fatigue baseline, which is a curve fitted by a series of points. Each point includes the running mileage, the corresponding sound data frequency domain component 12 and the vibration fusion data frequency domain component 13. On the other hand, the time-frequency domain features can also be used to participate in the characterization of the relationship baseline between the running mileage and the sound data, and the relationship between the running mileage and the vibration fusion data.

[0041] When the operating data is compared with the single-elevator fatigue baseline, the three-axis vibration data or the six-axis vibration data is fused to obtain vibration fusion data. The frequency domain composition of the sound data and the vibration fusion data is extracted. The frequency domain composition of the sound data is compared with the single-elevator fatigue baseline, and the point on the single-elevator fatigue baseline that is closest to the frequency domain composition of the sound data is found, and the similarity between the two is obtained. Then the frequency domain composition of the vibration fusion data is compared with the single-elevator fatigue baseline, and the point on the single-elevator fatigue baseline that is closest to the frequency domain composition of the vibration fusion data is found, and the similarity between the two is obtained. The two points correspond to a running mileage respectively, and the weighted mean of the two running mileages is calculated as the equivalent running mileage of the single elevator maintained this time. The equivalent running mileage is the quantity that characterizes the fatigue of the single elevator. When calculating the weighted mean, the weights can be determined according to the two similarities. For example, the running mileage corresponding to the most matching point obtained according to the sound data is 180 kilometers, with a similarity of 80%, and the running mileage corresponding to the most matching point obtained according to the vibration data is 200 kilometers, with a similarity of 75%. The weighted mean is 180*(80 / (80+75))+200*(75 / (80+75))=189.67. That is, the final equivalent running mileage is 189.67 kilometers. 189.67 kilometers represents the fatigue prediction result of a single ladder.

[0042] Similarly, when the vibration data is six-axis vibration data, and the fatigue baseline of the single elevator includes a baseline representing the relationship between the running mileage and the sound data and six baselines representing the relationship between the running mileage and the six-axis vibration fusion data, a total of seven comparisons will be required, and each comparison can obtain a corresponding running mileage and the corresponding similarity. The weighted mean of the running mileage can be calculated according to the similarity. For example, the actual running mileage of a single elevator is 150 kilometers, but the sound data and vibration data collected by it are compared with the fatigue baseline of the single elevator, and it is found that the sound data frequency domain composition 12 and the vibration fusion data frequency domain composition 13 corresponding to 180 kilometers are the closest. In other words, although the single elevator has only run 150 kilometers, its working condition is poor, there are deficiencies in assembly, frequent heavy load operation, poor lubrication, etc., resulting in its equivalent running mileage reaching 180 kilometers, that is, the single elevator is fatigued faster than expected. Through the technology recorded in this embodiment, the fatigue of the single elevator can be predicted more accurately. It is conducive to more accurately identifying the current situation of the single elevator, so as to formulate corresponding maintenance plans and ensure the normal operation of the single elevator.

[0043] On the other hand, in another embodiment, please refer to the attached Figure 4 , comparing the operating data with the single-ladder fatigue baseline to obtain a single-ladder fatigue prediction result includes: Step S201) obtaining the rate of change of the sound data and the vibration data with the fatigue degree according to the single-step fatigue degree baseline; Step S202) setting the weights of the sound data and the vibration data according to the change rate; Step S203) obtaining the part of the sound data and the vibration data that best matches the single-step fatigue baseline according to the weight; Step S204) obtaining a single-ladder fatigue prediction result according to the best matching part.

[0044] Exemplarily, when the mileage changes from 180 kilometers to 200 kilometers, the change in sound data is relatively small, while the change in vibration data is relatively large. When making a specific comparison, the percentage change method can be used. Historically, the average change rate of the amplitude of the frequency component of the sound data and the average change rate of the amplitude of the frequency component of the vibration data are calculated. Weights are obtained based on the two average change rates. When the average change rate is large, the weight is also relatively large. That is, if the average change rate of the sound data is larger, the weight of the sound data is relatively larger. By generating weights based on the change rate of the sound data and the vibration data with fatigue, the accuracy of the single-ladder fatigue prediction results can be improved.

[0045] On the other hand, in another embodiment, the number of cycles can be used instead of the running mileage to characterize the fatigue of a single elevator. The number of cycles refers to the number of times a single elevator completes a running cycle. For some single elevators, conversion is required. For example, a single elevator runs from 1 to 10 floors. When it runs from the 1st floor to the 10th floor, and then from the 10th floor to the 1st floor, a running cycle is completed. When it runs from the 1st floor to the 5th floor, and then from the 5th floor to the 1st floor, 0.5 running cycles are completed. When it runs from the 1st floor to the 3rd floor, and then runs to the 2nd floor, for floors 2 to 3, one up and one down are completed, which is counted as 0.1 running cycles, and when floors 1 to 2 complete one up, it waits until floors 1 to 2 complete one down before counting 0.1 running cycles into the total running cycles. Since the single elevator is a reciprocating cycle, and the up and down must appear alternately between the two floors, the pairing can be completed quickly. That is, there will not be two consecutive ups or two consecutive downs between the two floors. Representing the life of a single elevator by its operating mileage can eliminate the impact of different operating distances of the single elevator.

[0046] On the other hand, the single-ladder fatigue degree can also be characterized by the technology already disclosed in the art. For example, the fatigue damage model in the frequency domain is used. Exemplarily, the time domain signal (i.e., the upward sound curve, the upward vibration curve, the downward sound curve, the downward vibration curve) is converted into a frequency domain signal by Fourier transform to obtain the stress power spectrum density (i.e., PSD graph). The PSD graph describes the mean square value of each sine wave for its frequency, and is the basis of frequency domain analysis. The fatigue damage model can use the Dirlik method, the Bendat method, etc. to extract fatigue damage information from the PSD graph. These methods are usually based on statistics and probability theory, and can take into account the randomness and dynamic characteristics of the load. The time domain signal needs to be preprocessed before being converted into a frequency signal. Exemplarily, the preprocessing is filtering, denoising, etc. to improve the signal quality.

[0047] The fault diagnosis model establishes a correspondence between the fault type and the sound data and vibration data. This correspondence has the following situations: the sound data and vibration data are weightedly fused to obtain fused data, and the frequency domain composition of the fused data is extracted. The fault diagnosis model records the correspondence between the fault type and the frequency domain composition of the fused data; the three-axis vibration data or six-axis vibration data are fused to obtain vibration fusion data, and the fault diagnosis model includes the correspondence between the frequency domain composition of the fault type and the sound data and the correspondence between the fault type and the frequency domain composition of the vibration fusion data; the vibration data is three-axis vibration data, and the fault diagnosis model includes the correspondence between the frequency domain composition of the fault type and the sound data and three correspondences between the fault type and the frequency domain composition of the three-axis vibration fusion data; the vibration data is six-axis vibration data, and the fault diagnosis model includes the correspondence between the frequency domain composition of the fault type and the sound data and six correspondences between the fault type and the frequency domain composition of the six-axis vibration fusion data.

[0048] Collect the operation data of a single elevator. For example, collect the data during planned maintenance, troubleshooting, temporary inspection, etc., and fuse the three-axis vibration data or the six-axis vibration data to obtain vibration fusion data. Extract the frequency domain composition of the sound data and the frequency domain composition of the vibration fusion data respectively. Compare the frequency domain composition of the sound data with the fault diagnosis model to obtain the fault diagnosis type obtained based on the sound data. Compare the frequency domain composition of the vibration fusion data with the fault diagnosis model to obtain the fault diagnosis type obtained based on the vibration fusion data. The two fault diagnosis types constitute the single elevator fault diagnosis result.

[0049] On the other hand, in another embodiment, according to the operation data of the single elevator during normal operation, the method for establishing a fatigue baseline of the single elevator includes: marking the running mileage of the running data, and extracting the time-frequency domain features of the running data; Establishing and using the time-frequency domain features to train an autoencoder model, and obtaining a feature expression of the operating data according to the autoencoder model; According to the characteristic expression and the corresponding running mileage, a single-level fatigue baseline is obtained.

[0050] That is, the single-level fatigue baseline is established using the time-frequency domain features of sound data and vibration data, and the autoencoder model is used for specific implementation.

[0051] The operation data is collected multiple times, and multiple operation data with operation mileage marked are obtained.

[0052] On the other hand, please see the attached Figure 5 , according to the operating data of the associated fault, the method for establishing a fault diagnosis model comprises: Step S301) extracting time-frequency domain features of the operation data; Step S302) clustering the time-frequency domain features using a clustering algorithm and filtering out time-frequency domain features with a high degree of discreteness; Step S303) obtaining sample data according to the fault type associated with the clustered time-frequency domain features; Step S304) Establishing and using the sample data to train a machine learning model; Step S305) Obtain a fault diagnosis model based on the trained machine learning model.

[0053] When a large amount of operating data at the time of a fault is obtained, the operating data of the same fault is also different. By extracting time-frequency domain features from the operating data, clustering is performed, and the time-frequency domain features with high discreteness are screened out. The time-frequency domain features of the clustered operating data correspond to a fault type. Take the time-frequency domain features of the cluster center, associate a fault type, and you can get a sample data. Use the sample data to train the machine learning model. The time-frequency domain features of the operating data include the time-frequency domain features of the sound data and the time-frequency domain features of the vibration data. The vibration data can be vibration fusion data, three-axis vibration data, or six-axis vibration data. Sound data and vibration data can be further distinguished between uplink and downlink. Time-frequency domain features include frequency domain composition and its amplitude. Time-frequency domain features refer to features that take into account both time domain and frequency domain characteristics. Specifically, they are obtained through wavelet Fourier transform and short-time Fourier transform.

[0054] On the other hand, in another embodiment, the method further comprises the steps of: Collecting single elevator data, including the single elevator service year, average daily running mileage, running mileage, and wire rope bending times; Associating the single ladder data as a label to the corresponding operation data; A universal fatigue prediction model for elevators of corresponding structural types is established according to the operating data associated with the single elevator data of the elevator structural type.

[0055] Since the installation and operation conditions of each single elevator are different, the fatigue of the single elevator cannot be accurately predicted based only on the structure and model of the single elevator. After adding the year of use of the single elevator, the average daily mileage, the operating mileage, and the number of wire rope bending times, combined with the structure of the single elevator, a universal fatigue prediction model for elevators with the same single elevator structure can be established through the single elevators that have collected sound data and vibration data. The newly installed single elevators or those that are connected to this method for fatigue prediction for the first time can obtain an initial fatigue prediction result through a universal fatigue prediction model. It should be noted that a universal fatigue prediction model for elevators is established for each single elevator structure. The year of use of the single elevator, the average daily mileage, the operating mileage, and the number of wire rope bending times are inputs to the universal fatigue prediction model for elevators. The input of the universal fatigue prediction model for elevators also includes sound data and vibration data.

[0056] On the other hand, in another embodiment, the method of training a machine learning model using the sample data includes: Extracting the index features of the time-frequency domain features according to the preset indexes; The use year, average daily mileage, mileage, and number of wire rope bending times of the single elevator are included in the indicator characteristics; Taking the indicator feature as input and the fault type as a label value, training the machine learning model; After the preset training termination condition is reached, a fault diagnosis model is obtained according to the machine learning model.

[0057] Different from using the frequency domain composition and amplitude of time-frequency domain features to generate sample data, the machine learning model is trained. In this embodiment, sample data is generated by extracting preset indicators. It has faster processing speed and higher accuracy. The preset indicators include peak, root mean square, average, kurtosis, skewness, standard deviation, zero crossing rate, frequency and energy. The machine learning model includes a deep neural network or a recurrent neural network. The method for training the machine learning model includes: using ensemble learning to train the machine learning model, or the machine learning model is a pre-trained model, and the machine learning model is trained using transfer learning.

[0058] By collecting a large amount of operation data marked with relevant information, it is possible to obtain operation data associated with operation mileage and associated single-elevator operation data labels. The single elevators are classified according to the single-elevator operation data labels. The single-elevator classification can be classified according to the fact that all the single-elevator operation data are the same or fall into the same numerical range, or it can be classified according to the fact that some of the single-elevator operation data are the same or fall into the same numerical range. For example, single elevators with a service life of 5 years, an average daily operation number of 2,000 times, an operation mileage of 20,000 to 30,000 kilometers, and 10 bending times of the wire rope are between 10,000 and 100,000 times are classified into one category. All the operation data under the category are labeled with their corresponding operation mileage to constitute the sample data.

[0059] On the other hand, in another embodiment, the weight of each indicator feature is set according to the fault type, and the product of the indicator feature and the weight is used as input when training the machine learning model. Different fault types will cause significant changes in different indicators. By amplifying the effect of the corresponding indicator through weight selection, the accuracy of the fault diagnosis model can be further improved.

[0060] On the other hand, in another embodiment, when a single elevator fails for the first time, the fault type corresponding to the previous indicator feature within a preset time period is updated to a fault precursor, and the type of the fault precursor is consistent with the type of the fault; the fault diagnosis model is retrained based on the updated indicator feature. Thus, a fault reminder can be issued when a single elevator has a fault sign.

[0061] On the other hand, in another embodiment, according to the operation data of the single elevator during normal operation, the method for establishing a fatigue baseline of the single elevator includes: Establish a single-ladder theoretical model and obtain single-ladder theoretical operation data; Obtaining a difference between the operating data during normal operation of the single elevator and the theoretical operating data of the single elevator as differential operating data; marking the running mileage of the difference running data, and extracting the time-frequency domain features of the difference running data; Establishing and using the time-frequency domain features to train an autoencoder model, and obtaining a feature expression of the difference operation data according to the autoencoder model; According to the characteristic expression and the corresponding running mileage, a single-level fatigue baseline is obtained.

[0062] In this embodiment, the fatigue baseline of a single elevator no longer depends on all the collected operating data, but only on the part that is different from the theoretical operating data. The operating data of the single elevator that is the same as the theoretical operating data will be excluded. Therefore, more targeted operating data can be used to establish the fatigue baseline of the single elevator, which helps to improve the accuracy of the single elevator fatigue prediction. The establishment of a single elevator theoretical model is achieved through computer simulation technology. On the other hand, the operating data of a single elevator in normal operation can also be collected under laboratory conditions, and after de-noising and other processing, it can be used as the theoretical operating data of the single elevator. For a single elevator corresponding to the theoretical baseline, the vibration amplitude in the horizontal direction should be infinitely close to 0, and the vibration in the vertical direction should be close to the vibration determined by the structural characteristics of the elevator. Such as the inherent operating vibration characteristics caused by the traction machine speed, traction wheel diameter, traction ratio, balance coefficient and anti-ropes position under the operation of the traction machine.

[0063] Due to differences in installation, usage frequency, and maintenance and repair conditions, there are certain differences between the single ladder wire rope 10 and the theoretical operating data, that is, the current optimal operating condition of the single ladder is different from the theoretical operating data. Therefore, it is necessary to create a single ladder fatigue baseline for the wire rope 10 of the single ladder based on the difference, which will be more targeted.

[0064] On the other hand, this manual provides an elevator wire rope fault diagnosis and fatigue prediction system, please refer to the attached Figure 6 ,include: A collection module 100 collects operation data of a single elevator during normal operation, wherein the operation data includes sound data and vibration data; A baseline module 200, for establishing a fatigue baseline of a single elevator according to the operation data of the single elevator during normal operation; Fault module 300, collects the running data of the elevator when it is running under fault, and associates it with the corresponding fault type; Model module 400, establishing a fault diagnosis model according to the operation data associated with the fault; The diagnosis module 500 collects the operation data of a single elevator, compares the operation data with the fault diagnosis model, and obtains the single elevator fault diagnosis result; The prediction module 600 compares the operating data with the single-elevator fatigue baseline to obtain a single-elevator fatigue prediction result.

[0065] See also Figure 7 A schematic diagram of the structure of an electronic device provided in an embodiment of this specification is shown.

[0066] like Figure 7As shown, the electronic device 1100 may include: at least one processor 1101, at least one network interface 1104, a user interface 1103, a memory 1105 and at least one communication bus 1102. Among them, the communication bus 1102 can be used to realize the connection and communication of the above-mentioned components. Among them, the user interface 1103 may include a button, and the optional user interface may also include a standard wired interface and a wireless interface. Among them, the network interface 1104 may include but is not limited to a Bluetooth module, an NFC module, a Wi-Fi module, etc. Among them, the processor 1101 may include one or more processing cores. The processor 1101 uses various interfaces and lines to connect various parts in the entire electronic device 1100, and executes various functions and processes data of the routing device 1100 by running or executing instructions, programs, code sets or instruction sets stored in the memory 1105, and calling data stored in the memory 1105. Optionally, the processor 1101 can be implemented in at least one hardware form of DSP, FPGA, and PLA. The processor 1101 may integrate one or a combination of CPU, GPU and modem, etc. Among them, the CPU mainly processes the operating system, user interface and application programs, etc.; the GPU is responsible for rendering and drawing the content to be displayed on the display screen; the modem is used to process wireless communication.

[0067] It is understandable that the above-mentioned modem may not be integrated into the processor 1101, but may be implemented by a separate chip.

[0068] Among them, the memory 1105 may include RAM or ROM. Optionally, the memory 1105 includes a non-transitory computer-readable medium. The memory 1105 can be used to store instructions, programs, codes, code sets or instruction sets. The memory 1105 may include a program storage area and a data storage area, wherein the program storage area may store instructions for implementing an operating system, instructions for at least one function (such as a touch function, a sound playback function, an image playback function, etc.), instructions for implementing the above-mentioned various method embodiments, etc.; the data storage area may store data involved in the above-mentioned various method embodiments, etc. The memory 1105 may also be at least one storage device located away from the aforementioned processor 1101. The memory 1105 as a computer storage medium may include an operating system, a network communication module, a user interface module and an application. The processor 1101 may be used to call the application stored in the memory 1105 and execute the methods in the above-mentioned multiple embodiments.

[0069] The embodiments of this specification also provide a computer-readable storage medium, which stores instructions, and when the instructions are executed on a computer or a processor, the computer or the processor executes the multiple steps in the above embodiments. If the components of the above electronic device are implemented in the form of software functional units and sold or used as independent products, they can be stored in the computer-readable storage medium.

[0070] The embodiments of this specification also provide a computer program product, including a computer program, which implements multiple steps in the above embodiments when executed by a processor.

[0071] In the absence of conflict, the technical features in this embodiment and implementation scheme can be combined arbitrarily.

[0072] In the above embodiments, it can be implemented in whole or in part by software, hardware, firmware or any combination thereof. When implemented by software, it can be implemented in whole or in part in the form of a computer program product. The computer program product includes a plurality of computer instructions. When the computer program instructions are loaded and executed on a computer, the process or function described in the embodiment of this specification is generated in whole or in part. The computer may be a general-purpose computer, a special-purpose computer, a computer network, or other programmable device. The computer instructions may be stored in a computer-readable storage medium or transmitted through the computer-readable storage medium. The computer instructions may be transmitted from a website site, a computer, a server or a data center to another website site, a computer, a server or a data center by wired (e.g., coaxial cable, optical fiber, digital subscriber line (DSL)) or wireless (e.g., infrared, wireless, microwave, etc.) means. The computer-readable storage medium may be any available medium that a computer can access or a data storage device such as a server or a data center that includes multiple available media integrated. The available medium may be a magnetic medium (eg, a floppy disk, a hard disk, a magnetic tape), an optical medium (eg, a digital versatile disc (DVD)), or a semiconductor medium (eg, a solid state drive (SSD)).

[0073] When implemented by hardware or firmware, the aforementioned method flow is programmed into the hardware circuit to obtain the corresponding hardware circuit structure and realize the corresponding function. For example, a programmable logic device (PLD) (such as a field programmable gate array (FPGA)) is such an integrated circuit, and its logic function is determined by the user programming the device. The designer programs by himself to "integrate" a digital system on a PLD, without asking a chip manufacturer to design and make a dedicated integrated circuit chip. Moreover, nowadays, instead of manually making integrated circuit chips, this programming is mostly implemented by "logic compiler" software, which is similar to the software compiler used when writing program development, and the original code before compilation must also be written in a specific programming language, which is called hardware description language (HDL), and HDL is not just one, but many. Those skilled in the art should also be aware that it is only necessary to program the method flow slightly in the above-mentioned hardware description languages ​​and program it into the integrated circuit to easily obtain the hardware circuit that implements the logic method flow.

[0074] The embodiments described above are merely preferred embodiments of this specification and are not intended to limit the scope of this specification. Without departing from the design spirit of this specification, various modifications and improvements made to the technical solutions of this specification by ordinary technicians in this field should fall within the scope of protection determined by the claims of this specification.

Claims

1. A method for diagnosing elevator wire rope faults and predicting fatigue, characterized in that: Includes steps: Collecting operation data of a single elevator during normal operation, wherein the operation data includes sound data and vibration data; Establishing a fatigue baseline of a single elevator based on the operating data during normal operation of the single elevator; Collect the running data of the elevator when it is running under fault conditions and associate it with the corresponding fault type; Establishing a fault diagnosis model based on the operating data associated with the fault; Collecting operation data of a single elevator, comparing the operation data with the fault diagnosis model, and obtaining a single elevator fault diagnosis result; The operating data is compared with the single-ladder fatigue baseline to obtain a single-ladder fatigue prediction result.

2. The elevator wire rope fault diagnosis and fatigue prediction method according to claim 1, characterized in that: The method further comprises the steps of: Collecting single elevator data, including the single elevator service year, average daily running mileage, running mileage, and wire rope bending times; Associating the single ladder data as a label to the corresponding operation data; A universal fatigue prediction model for elevators of corresponding structural types is established according to the operating data associated with the single elevator data of the elevator structural type.

3. A method for diagnosing elevator wire rope faults and predicting fatigue according to claim 1 or 2, characterized in that: According to the operating data of the single elevator during normal operation, the method for establishing the fatigue baseline of the single elevator includes: marking the running mileage of the running data, and extracting the time-frequency domain features of the running data; Establishing and using the time-frequency domain features to train an autoencoder model, and obtaining a feature expression of the operating data according to the autoencoder model; According to the characteristic expression and the corresponding running mileage, a single-level fatigue baseline is obtained.

4. A method for diagnosing elevator wire rope faults and predicting fatigue according to claim 1 or 2, characterized in that: According to the operating data of the single elevator during normal operation, the method for establishing the fatigue baseline of the single elevator includes: Establish a single-ladder theoretical model and obtain single-ladder theoretical operation data; Obtaining a difference between the operating data during normal operation of the single elevator and the theoretical operating data of the single elevator as differential operating data; marking the running mileage of the difference running data, and extracting the time-frequency domain features of the difference running data; Establishing and using the time-frequency domain features to train an autoencoder model, and obtaining a feature expression of the difference operation data according to the autoencoder model; According to the characteristic expression and the corresponding running mileage, a single-level fatigue baseline is obtained.

5. The elevator wire rope fault diagnosis and fatigue prediction method according to claim 1 or 2, characterized in that: The method of establishing a fault diagnosis model according to the operating data of the associated fault includes: Extracting time-frequency domain features of the operating data; Clustering the time-frequency domain features using a clustering algorithm and filtering out time-frequency domain features with a high degree of discreteness; Obtain sample data based on the fault types associated with the clustered time-frequency domain features; Establish and use the sample data to train a machine learning model; A fault diagnosis model is obtained based on the trained machine learning model.

6. The elevator wire rope fault diagnosis and fatigue prediction method according to claim 5, characterized in that: The method of using the sample data to train a machine learning model includes: Extracting the index features of the time-frequency domain features according to the preset indexes; The use year, average daily mileage, mileage, and number of wire rope bending times of the single elevator are included in the indicator characteristics; Taking the indicator feature as input and the fault type as a label value, training the machine learning model; After the preset training termination condition is reached, a fault diagnosis model is obtained according to the machine learning model.

7. The elevator wire rope fault diagnosis and fatigue prediction method according to claim 6, characterized in that: The preset indicators include peak value, root mean square, average value, kurtosis, skewness, standard deviation, zero crossing rate, frequency and energy; The machine learning model includes a deep neural network or a recurrent neural network.

8. The elevator wire rope fault diagnosis and fatigue prediction method according to claim 6, characterized in that: According to the fault type, a weight of each indicator feature is set, and when training the machine learning model, the product of the indicator feature and the weight is used as input.

9. The elevator wire rope fault diagnosis and fatigue prediction method according to claim 6, characterized in that: When a single elevator fails for the first time, the fault type corresponding to the previous indicator feature within a preset time period is updated to a fault precursor, and the type of the fault precursor is consistent with the type of the fault; The fault diagnosis model is retrained according to the updated indicator features.

10. The elevator wire rope fault diagnosis and fatigue prediction method according to claim 6, characterized in that: The method of training the machine learning model includes: The machine learning model is trained using ensemble learning, or the machine learning model is a pre-trained model and the machine learning model is trained using transfer learning.

11. A method for diagnosing elevator wire rope faults and predicting fatigue according to claim 1 or 2, characterized in that: The operation data includes an upward sound curve, an upward vibration curve, a downward sound curve, and a downward vibration curve.

12. The elevator wire rope fault diagnosis and fatigue prediction method according to claim 1 or 2, characterized in that: The method of comparing the operating data with the single-ladder fatigue baseline to obtain a single-ladder fatigue prediction result includes: According to the single-step fatigue baseline, obtaining the rate of change of the sound data and the vibration data with the fatigue degree; setting weights of the sound data and the vibration data according to the change rate; Obtaining, according to the weight, the part of the sound data and the vibration data that best matches the single-step fatigue baseline; A single-ladder fatigue prediction result is obtained according to the most matching part.

13. An elevator wire rope fault diagnosis and fatigue prediction system, characterized in that: include: A collection module, for collecting operation data of a single elevator during normal operation, wherein the operation data includes sound data and vibration data; A baseline module, establishing a fatigue baseline of a single elevator according to the operating data of the single elevator during normal operation; The fault module collects the running data of the elevator under fault and associates it with the corresponding fault type; A model module, establishing a fault diagnosis model according to the operation data associated with the fault; A diagnosis module collects operation data of a single elevator during single elevator maintenance, compares the operation data with the fault diagnosis model, and obtains a single elevator fault diagnosis result; The prediction module compares the operating data with the single-ladder fatigue baseline to obtain a single-ladder fatigue prediction result.

14. An electronic device, characterized in that including a processor and a memory; The processor is connected to the memory; The memory is used to store executable program code; The processor runs a program corresponding to the executable program code by reading the executable program code stored in the memory, so as to execute the method according to any one of claims 1 to 12.

15. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the method according to any one of claims 1 to 12 is implemented.

16. Computer program product comprising a computer program, characterized in that When the computer program is executed by a processor, the method according to any one of claims 1 to 12 is implemented.

Images