Generator bearing wear resistance detection method and device based on laser scanning

Abstract

The invention discloses a generator bearing wear resistance detection method and device based on laser scanning, and relates to the technical field of bearing wear resistance detection, and the method comprises the steps: obtaining a first optical signal; reading a predetermined light feature, and performing feature extraction analysis on the first light signal to obtain first light feature information; traversing the first light feature information in a light feature database to obtain a first traversing result, and determining a first bearing shape; performing three-dimensional reconstruction by combining the first point location and the first bearing morphology to obtain a target three-dimensional morphology; comparing the target three-dimensional morphology with a predetermined three-dimensional morphology to obtain a morphology comparison result; and evaluating and analyzing the morphology comparison result to obtain a target wear resistance index. The technical problems that a traditional generator bearing abrasion resistance detection method is low in efficiency and insufficient in precision, and precise measurement and evaluation of the tiny abrasion of the surface of the bearing are difficult to achieve are solved, and the effect of improving the precision and efficiency of measurement of the tiny abrasion of the surface of the bearing through three-dimensional shape comparison is achieved.

Description

Technical Field

[0001] This application relates to the technical field of bearing wear resistance testing, specifically to a method and apparatus for testing the wear resistance of generator bearings based on laser scanning. Background Technology

[0002] With the continuous development of the power industry and the ongoing upgrading of generator sets, the performance of generator bearings plays a crucial role in ensuring the safe and stable operation of generator sets. However, traditional methods for testing the wear resistance of generator bearings, such as physical measurements and chemical analysis, while capable of assessing bearing wear to some extent, are increasingly unable to meet the demands of the modern power industry for efficient and accurate testing due to their cumbersome operation, long testing cycles, and potential for secondary damage to the bearings. In particular, with the expansion of production scale and the increasing complexity of processes, the working environment and load conditions of generator bearings have become more demanding, posing greater challenges to the monitoring and management of bearing wear. Traditional testing methods have significant shortcomings in terms of control accuracy, response speed, and stability, making it difficult to provide timely and accurate data for the maintenance and repair of generator sets. Summary of the Invention

[0003] This application provides a laser scanning-based method and apparatus for detecting the wear resistance of generator bearings, which solves the technical problems of low efficiency and insufficient accuracy of traditional generator bearing wear resistance detection methods, making it difficult to accurately measure and evaluate the minute wear on the bearing surface. It achieves the effect of improving the accuracy and efficiency of measuring the minute wear on the bearing surface through the comparison of three-dimensional morphology.

[0004] This application provides a laser scanning-based method for detecting the wear resistance of generator bearings. The method includes: acquiring a first optical signal, which refers to the laser beam reflection and transmission signal collected by a detector at a first point on a target generator bearing; reading predetermined optical features and performing feature extraction and analysis on the first optical signal based on the predetermined optical features to obtain first optical feature information; traversing the first optical feature information in an optical feature database to obtain a first traversal result, and analyzing the first traversal result to determine the morphology of a first bearing; performing three-dimensional reconstruction of the target generator bearing by combining the first point and the first bearing morphology to obtain a target three-dimensional morphology; comparing the target three-dimensional morphology with the predetermined three-dimensional morphology of the predetermined generator bearing to obtain a morphology comparison result; and introducing a wear resistance evaluation function to evaluate and analyze the morphology comparison result to obtain a target wear resistance index of the target generator bearing.

[0005] This application also provides a laser scanning-based generator bearing wear resistance detection device, comprising: an optical signal acquisition module for acquiring a first optical signal, wherein the first optical signal refers to the laser beam reflection and transmission light signal collected by a detector at a first point of the target generator bearing; a feature extraction and analysis module for reading predetermined optical features and performing feature extraction and analysis on the first optical signal based on the predetermined optical features to obtain first optical feature information; a traversal result analysis module for traversing the first optical feature information in an optical feature database to obtain a first traversal result and analyzing the first traversal result to determine the morphology of a first bearing; a three-dimensional reconstruction module for performing three-dimensional reconstruction of the target generator bearing by combining the first point and the first bearing morphology to obtain a target three-dimensional morphology; a three-dimensional morphology comparison module for comparing the target three-dimensional morphology with the predetermined three-dimensional morphology of the predetermined generator bearing to obtain a morphology comparison result; and an evaluation and analysis module for introducing a wear resistance evaluation function to evaluate and analyze the morphology comparison result to obtain a target wear resistance index of the target generator bearing.

[0006] This application proposes a laser scanning-based method and apparatus for detecting the wear resistance of generator bearings. The method involves acquiring a first optical signal, which refers to the reflected and transmitted light signal from a laser beam at a first point on the target generator bearing, collected by a detector. Predetermined optical features are read, and feature extraction and analysis are performed on the first optical signal based on these features to obtain first optical feature information. This first optical feature information is then traversed through an optical feature database to obtain a first traversal result, and the first traversal result is analyzed to determine the morphology of the first bearing. The target generator bearing is then reconstructed in three dimensions by combining the first point and the first bearing morphology to obtain a target three-dimensional morphology. The target three-dimensional morphology is compared with the predetermined three-dimensional morphology of the target generator bearing to obtain a morphology comparison result. A wear resistance evaluation function is introduced to evaluate and analyze the morphology comparison result to obtain a target wear resistance index for the target generator bearing. This method solves the technical problems of low efficiency and insufficient accuracy in traditional generator bearing wear resistance detection methods, making it difficult to accurately measure and evaluate minute wear on the bearing surface. It achieves the effect of improving the accuracy and efficiency of measuring minute wear on the bearing surface through comparison of three-dimensional morphologies. Attached Figure Description

[0007] To more clearly illustrate the technical solutions of the embodiments of this disclosure, the accompanying drawings of the embodiments of this disclosure will be briefly described below. Flowcharts are used in this application to illustrate the operations performed by the apparatus according to the embodiments of this application. It should be understood that the preceding or following operations are not necessarily performed precisely in sequence. Instead, various steps can be processed in reverse order or simultaneously as needed. Furthermore, other operations can be added to these processes, or one or more steps can be removed from these processes.

[0008] Figure 1 A schematic flowchart of a laser scanning-based generator bearing wear resistance detection method provided in this application embodiment; Figure 2 A schematic diagram of the structure of a laser scanning-based generator bearing wear resistance testing device provided in an embodiment of this application.

[0009] Figure labeling: 1. Optical signal acquisition module; 2. Feature extraction and analysis module; 3. Traversal result analysis module; 4. 3D reconstruction module; 5. 3D morphology comparison module; 6. Evaluation and analysis module. Detailed Implementation

[0010] The above description is only an overview of the technical solution of this application. In order to better understand the technical means of this application and to implement it in accordance with the contents of the specification, and to make the above and other objects, features and advantages of this application more obvious and understandable, the following are specific embodiments of this application.

[0011] To make the objectives, technical solutions, and advantages of this application clearer, the following detailed description of this application will be provided in conjunction with the accompanying drawings. The described embodiments should not be considered as limitations on this application. All other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this application.

[0012] In the following description, references to "some embodiments" describe a subset of all possible embodiments. However, it is understood that "some embodiments" can be the same or different subsets of all possible embodiments and can be combined with each other without conflict. The terms "first" and "second" are used merely to distinguish similar objects and do not represent a specific ordering of objects. The terms "comprising" and "having," and any variations thereof, are intended to cover non-exclusive inclusion. For example, a process, method, apparatus, product, or server that includes a series of steps or units is not necessarily limited to those steps or units explicitly listed, but may include other steps or modules not explicitly listed or inherent to these processes, methods, products, or apparatuses. Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this application belongs. The terminology used herein is for the purpose of describing embodiments of this application only.

[0013] This application provides a method for detecting the wear resistance of generator bearings based on laser scanning, such as... Figure 1 As shown, the method includes: Acquire the first optical signal, which refers to the laser beam reflection and transmission light signal at the first point of the target generator bearing collected by the detector.

[0014] In this embodiment, when detecting the wear resistance of a generator bearing, the system terminal uses a detector to acquire optical signal information from the bearing surface. This detector can capture the reflected and transmitted light signals of a laser beam at a first point on the target generator bearing, generating a first optical signal. Here, the first point refers to a random point on the target generator bearing, and the first optical signal refers to the reflected and transmitted light signals of the laser beam at that first point. These optical signals contain important information about the bearing surface morphology, material, and degree of wear. After collecting these optical signals, the detector transmits them to the system terminal for further processing and evaluation, thereby helping to understand the wear resistance of the generator bearing.

[0015] Furthermore, this application provides a detector for collecting laser scanning light signals, including: The detector collects the laser scanning light signal of the target generator bearing based on a predetermined laser scanning scheme to obtain the target light signal, wherein the target light signal includes the first light signal.

[0016] Preferably, when evaluating the wear resistance of generator bearings, the system terminal uses a detector to collect the optical signals generated during laser scanning. This process is based on a pre-set laser scanning scheme, ensuring that the laser beam scans the surface of the generator bearing according to a predetermined path and method. During signal collection, the detector captures and records all reflected and transmitted optical signals of the laser beam on the bearing surface and transmits them to the system terminal. The system terminal processes the received multiple optical signals to obtain a target optical signal containing the first optical signal. These optical signals reflect information such as the degree of wear and surface morphology at various points on the bearing surface.

[0017] Furthermore, this application provides a predetermined laser scanning scheme, including: The predetermined laser scanning scheme includes a predetermined scanning path and predetermined scanning parameters. A first scanning point corresponding to the first point is matched in the predetermined scanning path, and the optical signal corresponding to the first scanning point is recorded as the first optical signal. The predetermined scanning parameters include predetermined scanning speed parameters and predetermined resolution parameters.

[0018] Optionally, the predetermined laser scanning scheme is designed in advance based on the geometric characteristics of the generator bearing, the expected detection accuracy, and the detection requirements. It includes a predetermined scanning path and predetermined scanning parameters. The predetermined scanning path is pre-planned to ensure that the laser beam scans each area of ​​the bearing surface point by point in a specific order, ensuring that every point on the bearing surface is accurately scanned. Simultaneously, due to the three-dimensional structure of the bearing, the scanning path considers different scanning depths to achieve a comprehensive scan of all three-dimensional coordinate points of the bearing. This design ensures the comprehensiveness and depth of the scan, providing a rich data foundation for subsequent wear resistance assessment. The predetermined scanning parameters refine the specific requirements of the scanning process, including predetermined scanning speed parameters and predetermined resolution parameters. The predetermined scanning speed parameter ensures the efficiency and stability of the scan, avoiding the impact of excessively fast or slow scanning speeds on the results. The predetermined resolution parameter determines the level of detail in the scan; high-resolution scanning can capture more detailed information, thus more accurately reflecting the wear degree and surface morphology of the bearing surface. During the scanning process, the system terminal randomly selects a point as the first point. When the scanning path matches this first point, the corresponding optical signal is specially marked as the first optical signal. This step ensures the accurate extraction of key information, providing strong support for the wear resistance assessment of generator bearings.

[0019] Read the predetermined optical features, and perform feature extraction and analysis on the first optical signal based on the predetermined optical features to obtain the first optical feature information.

[0020] In one embodiment, during the assessment of generator bearing wear resistance, the system terminal not only performs laser scanning to collect optical signals but also reads pre-defined optical features as a reference. These pre-defined optical features are based on the optical signal characteristics of the bearing surface under normal conditions; they represent the optical signal characteristics of the bearing under ideal conditions, including reflected light intensity and transmitted light intensity. Subsequently, the system terminal performs feature extraction analysis on the first optical signal based on these pre-defined optical features. By comparing the collected first optical signal with the pre-defined optical features, it identifies the feature items that differ between the two and summarizes these feature items to form first optical feature information. This information reflects the actual condition of the bearing at a first point. By comparing this information with the pre-defined optical features, the system terminal can more accurately assess the bearing wear resistance, providing a scientific basis for future maintenance and replacement decisions.

[0021] Furthermore, this application provides predetermined optical features, including: The predetermined optical characteristics include a predetermined reflected light signal and a predetermined transmitted light signal. The predetermined reflected light signal includes reflected light intensity, reflected light polarization, reflected light wavefront phase, reflected light wavelength, reflected light polarization dependence, reflected light time dependence, and emitted light spatial distribution. The predetermined transmitted light signal includes transmitted light intensity, transmitted light polarization, transmitted light wavefront phase, transmitted light wavelength, transmitted light polarization dependence, transmitted light time dependence, and transmitted light spatial distribution.

[0022] Preferably, the predetermined optical characteristics describe the expected behavior of the laser when it interacts with the target generator bearing. These characteristics include two aspects: predetermined reflected light signals and predetermined transmitted light signals. The predetermined reflected light signals include reflected light intensity, reflected light polarization, reflected light wavefront phase, reflected light wavelength, reflected light polarization dependence, reflected light time dependence, and emitted light spatial distribution. The predetermined transmitted light signals include transmitted light intensity, transmitted light polarization, transmitted light wavefront phase, transmitted light wavelength, transmitted light polarization dependence, transmitted light time dependence, and transmitted light spatial distribution. Specifically, reflected light intensity refers to the intensity of light reflected from the bearing surface, used to determine the degree of wear on the bearing surface. Reflected light wavelength is the color of the reflected light, used to identify whether there has been a material change or contamination on the bearing surface. Reflected light polarization describes the direction or polarization state of the electric field vector of the reflected light, which is related to the texture or stress state of the bearing surface. Reflected light wavefront phase is the phase change of the light wave during reflection, revealing the microstructure or deformation of the bearing surface. Reflected light polarization dependence refers to whether the polarization state of the reflected light depends on the polarization state of the incident light. The time dependence of reflected light describes the characteristics of reflected light changing over time, such as the changes in optical signals during the dynamic operation of a bearing. The spatial distribution of reflected light is the distribution pattern of light after reflection from the bearing surface, reflecting the geometry and roughness of the bearing surface. Transmitted light intensity is the intensity of light passing through the bearing surface, providing information about the internal material or structure of the bearing. Transmitted light wavelength is the color of the transmitted light, reflecting the characteristics of the internal material of the bearing. Transmitted light polarization describes the polarization state of the transmitted light, which is related to the internal structure or stress state of the bearing. The wavefront phase of transmitted light is the phase change of the light wave during transmission, revealing the internal structure or deformation of the bearing. The polarization dependence of transmitted light refers to whether the polarization state of the transmitted light depends on the polarization state of the incident light. The time dependence of transmitted light describes the characteristics of transmitted light changing over time, revealing changes in the internal material or structure of the bearing. The spatial distribution of transmitted light is the distribution pattern of light after transmission within the bearing, related to the internal structure or material of the bearing. In the process of evaluating the wear resistance of generator bearings, predetermined optical characteristics are used as a reference standard to help the system terminal analyze and evaluate the actually collected optical signals, thereby determining the actual condition of the bearing surface.

[0023] The first optical feature information is traversed in the optical feature database to obtain the first traversal result, and the first traversal result is analyzed to determine the morphology of the first bearing.

[0024] In one embodiment, during the evaluation of generator bearing wear resistance, after the system terminal obtains the first optical feature information, this information is compared and matched with a pre-established optical feature database. The optical feature database stores a large number of bearings with different wear resistances and their corresponding optical feature information. The system terminal iterates through the bearing optical feature information in the database, calculating the similarity with the first optical feature information in each iteration. When bearing optical feature information that meets the similarity limit is found, the system terminal acquires the bearing corresponding to the bearing optical feature information and uses the morphology of this bearing as the first bearing morphology, providing an important basis for subsequent wear resistance evaluation.

[0025] Furthermore, this application provides methods for determining the morphology of the first bearing, including: Extract the first data from the optical feature database, where the first data refers to the optical feature information of the first bearing corresponding to the first bearing; read the predetermined label scheme, and perform label processing on the first optical feature information and the first bearing optical feature information in sequence based on the predetermined label scheme to obtain the first vector and the first bearing vector respectively.

[0026] Preferably, during the evaluation of generator bearing wear resistance, the system terminal randomly extracts a data point from the optical feature database as the first data. This data is the optical feature information of the first bearing corresponding to the first bearing. The first bearing is a bearing randomly extracted from the optical feature database. The optical feature information of the first bearing is the optical feature data corresponding to this bearing. These data record the optical characteristics of the bearing. Subsequently, the system terminal reads a predetermined tagging scheme. This tagging scheme is a set of rules that defines how to convert complex optical feature information into a comparable and easily processed vector form. The tagging scheme includes steps such as feature selection, feature encoding, and feature scaling. Specifically, the system terminal processes the first optical feature information based on the rules set in the predetermined tagging scheme. First, it selects the feature parameters most relevant to the bearing wear degree from the first optical feature information. Then, it converts the selected features into numerical form. For example, for reflected light intensity, the system terminal reads the signal value and converts it into a floating-point format. For polarization characteristics, the system terminal extracts parameters such as polarization angle and degree of polarization and converts them into numerical form. Subsequently, to ensure the numerical comparability of different features, the system terminal performs scaling on these features using a min-max normalization method, scaling each feature to ensure its value is between 0 and 1. After these processing steps, the system terminal obtains a vector containing multiple numerical features; this vector is the numerical representation of the first optical feature information, i.e., the first vector. Following a similar process to the first optical feature information processing, the system terminal also processes the first bearing optical feature information according to the rules of a predetermined labeling scheme. This process also includes feature selection, feature encoding, and feature scaling. The processed first bearing optical feature information is also converted into a numerical vector, i.e., the first bearing vector. Both vectors represent their respective optical feature information in numerical form and can be compared on the same dimension, facilitating subsequent comparative analysis and bearing status determination.

[0027] The first similarity between the first vector and the first bearing vector is obtained using the Tanimoto similarity coefficient algorithm; when the first similarity meets the similarity limit, the first shape of the first bearing is taken as the first bearing shape.

[0028] Preferably, after the system terminal obtains the first vector and the first bearing vector, it uses the Tanimoto similarity coefficient algorithm to calculate the similarity between the two vectors. The Tanimoto similarity coefficient algorithm is a similarity calculation method used in cheminformatics and bioinformatics, applicable to binary or real-number vectors. For real-number vectors, the Tanimoto similarity coefficient can be viewed as a variant of the cosine of the angle between vectors, but it takes into account the length of the vectors, i.e., the norm. Specifically, the system terminal multiplies the corresponding features between the first vector and the first bearing vector, then adds all the products to obtain the dot product. Subsequently, the system terminal squares each feature in the first vector, adds the square root, and obtains the first vector norm. Then, the same steps are used to calculate the norm of the first bearing vector. Finally, the system terminal calculates the ratio of the dot product to the sum of the squares of the two vector norms minus the dot product to obtain the first similarity. This similarity value reflects the numerical closeness of the two vectors, i.e., the similarity of the two bearings in terms of optical feature information. After obtaining the first similarity score, the system terminal compares this score with a similarity limit to determine if it is greater than or equal to the limit. This similarity limit is determined based on the needs of practical applications and historical experience. When the calculated first similarity score is greater than or equal to the limit, the system terminal determines that the bearing represented by the first vector is similar to the first bearing in terms of optical feature information, and therefore uses the first shape of the first bearing as the first bearing shape. If the score is less than the limit, the system terminal selects a new data point from the optical feature database as the second data point, and repeats the above process until a bearing that meets the similarity limit is found.

[0029] By combining the first point location with the first bearing morphology, the target generator bearing is reconstructed in three dimensions to obtain the target three-dimensional morphology.

[0030] In one embodiment, after obtaining the first bearing morphology, the system terminal imports the 3D model of the first bearing morphology into a 3D modeling component as the initial 3D morphology. Subsequently, positions corresponding to the first point are matched on the initial 3D morphology. This is done based on bearing design knowledge and the specific information of the first point. Once the first point is aligned, the system terminal fine-tunes the initial 3D morphology based on the detailed information of the first point, for example, modifying local fillets. After fine-tuning, the system terminal outputs the initial 3D morphology to generate the target 3D morphology. This target 3D morphology is of great significance for understanding the current state of the bearing and for conducting wear resistance testing.

[0031] The morphology comparison results are obtained by comparing the three-dimensional morphology of the target with the predetermined three-dimensional morphology of the predetermined generator bearing.

[0032] In one embodiment, after the target 3D topography is reconstructed, the system terminal compares the target 3D topography with the predetermined 3D topography of the predetermined generator bearing to check the similarity and differences between the two 3D models. This process is performed by precisely analyzing the geometry, dimensions, and possible deviations of the two models. First, the system terminal aligns the target 3D topography and the predetermined 3D topography to ensure that their positions and orientations in space are consistent, enabling accurate comparison. Then, the similarity and differences between the two models are analyzed, and the specific values ​​of the deviations are recorded, calculated by distributing the 3D coordinates of the predetermined 3D topography at the deviation position to the 3D coordinates of the target 3D topography. Afterward, the system terminal summarizes the deviation distances at each deviation position to form a topography comparison result. This result will be used for the subsequent generation of the target wear resistance index.

[0033] A wear resistance evaluation function is introduced to evaluate and analyze the morphology comparison results, and the target wear resistance index of the target generator bearing is obtained.

[0034] In one embodiment, when evaluating the wear resistance of a target generator bearing, the system terminal introduces a wear resistance evaluation function to analyze the results in conjunction with morphological comparison. First, the system terminal clarifies the key factors considered in the wear resistance evaluation function. These factors include wear index parameters and wear index weights. Then, the system terminal analyzes the morphological comparison results to generate n wear index parameters, which are the calculated deviation distances. Next, based on historical experience, the system terminal assigns a corresponding weight to each deviation distance. The larger the deviation distance, the greater the weight. Then, these wear index parameters and their corresponding weights are input into the wear resistance evaluation function for calculation, generating a target wear resistance index. This index is a quantitative value used to represent the wear resistance performance of the target generator bearing relative to a predetermined generator bearing. A higher index indicates that the target generator bearing performs well in terms of wear resistance and has good durability. Conversely, a lower index indicates poor durability.

[0035] Furthermore, this application provides the wear resistance evaluation function, including: The expression for the wear resistance evaluation function is as follows: ; in, This refers to the target generator bearing. The target wear resistance index, This refers to the target generator bearing. The target wear level, This refers to the first characterization of the target wear degree. The parameters of each wear index, It refers to the first The weights of each wear index, the target wear degree being determined by... A comprehensive characterization of wear indicators is performed, and .

[0036] Preferably, the wear resistance evaluation function is an important tool for quantifying the wear resistance performance of generator bearings. The function expression is as follows: ; in, Refers to the target generator bearing The target abrasion resistance index represents the abrasion resistance performance to be quantified. Refers to the target generator bearing The target wear degree is such that the greater the current target wear degree, the smaller the corresponding target wear resistance index. This refers to the first degree of target wear. The parameter of the wear index, i.e., the first wear index in the morphology comparison result. One deviation distance, It refers to the first The weight of each wear index reflects its importance in the overall wear resistance performance. It is the total number of wear indicators, and the target wear degree is determined by... A set of wear indicators is used to comprehensively characterize the bearing, where n is not equal to 0. The wear resistance evaluation function f(x) is used to calculate the target wear resistance index of the target generator bearing x. This function calculates the product of each wear indicator and its corresponding weight, and then adds these products to obtain the target wear degree of the target generator bearing. Subsequently, the difference between 1 and the target wear degree is calculated to obtain the target wear resistance index.

[0037] Furthermore, this application provides methods for adjusting the target wear level, including: Obtain the target lubricating oil of the target generator bearing; perform spectral analysis on the target lubricating oil to obtain the target spectral signal; read the predetermined spectral features, and perform multi-dimensional feature acquisition on the target spectral signal based on the predetermined spectral features to obtain the target signal features.

[0038] Optionally, the system terminal acquires the target lubricating oil used in the target generator bearing. Lubricating oil plays a crucial role in the operation of generator bearings, not only reducing friction and wear but also cooling the bearings and preventing corrosion. Therefore, understanding the performance of the lubricating oil is essential for evaluating the wear resistance and other properties of the generator bearings. Subsequently, spectral analysis is performed on the target lubricating oil. Spectral analysis is a method for studying the composition and properties of a substance by measuring the interaction between the substance and light. Through spectral analysis, the spectral signal of the target lubricating oil can be obtained, which contains information about the various components in the lubricating oil. To analyze the spectral signal of the target lubricating oil more accurately, the system terminal reads predetermined spectral features. Predetermined spectral features are a set of known spectral data features, including frequency domain features, i.e., the signal's performance in the frequency domain, such as frequency distribution, frequency peaks, etc., as well as other features such as amplitude and phase. These features are derived based on historical experience and professional knowledge. Based on these predetermined spectral features, the system terminal performs multi-dimensional feature acquisition on the target spectral signal. Multi-dimensional feature acquisition refers to extracting information from the signal from multiple perspectives to comprehensively describe the characteristics of the signal. In this process, the system terminal uses predetermined spectral characteristics as a reference to extract multiple related feature values ​​from the target spectral signal. These feature values ​​help the system terminal to gain a more comprehensive understanding of the composition, properties, and potential problems of the target lubricating oil. Through the above process, target signal characteristics are obtained, which will serve as the basis for subsequent analysis to evaluate the wear resistance of the target generator bearing.

[0039] The wear assessment model is invoked to evaluate and analyze the characteristics of the target signal to obtain a predicted target wear degree. The wear assessment model is an intelligent model obtained by machine learning based on the principle of neural network to perform machine learning on historical lubricating oil detection data. The target wear degree is then adjusted based on the predicted target wear degree.

[0040] Optionally, to more accurately assess the wear condition of the target generator bearing, the system terminal invokes a wear assessment model. This model is built upon neural network principles. Specifically, the system terminal collects historical lubricating oil testing data and divides this data into training, validation, and test sets. Then, based on the complexity of the problem and the characteristics of the data, the number of layers, nodes, activation functions, and other parameters of the neural network are determined. Next, the neural network is trained using the training set data, continuously adjusting the connection weights of neurons through backpropagation to minimize the error between predicted and true values. During training, the performance of the neural network is monitored using validation set data to prevent overfitting. Training stops when the neural network's performance on the validation set reaches a preset standard. Then, the trained neural network is evaluated using test set data, calculating its accuracy to measure its performance. If the neural network performance does not meet the requirements, the system terminal optimizes the neural network, such as adjusting the network structure or changing the learning rate. Conversely, the system terminal outputs the trained neural network to generate the wear assessment model. This model learns from historical lubricating oil testing data to understand the complex relationship between lubricating oil spectral signals and generator bearing wear. After the wear assessment model is constructed, the system terminal inputs the target signal features extracted from the target lubricating oil spectral signal into the model. The model uses these features as input, performs calculations and analyses through a complex internal neural network structure, and outputs a predicted target wear degree. This predicted target wear degree is derived from the current lubricating oil condition and the learning results of historical data, helping the system terminal understand the possible wear condition of the target generator bearing. Finally, the system terminal calculates the difference between the predicted target wear degree and the target wear degree, and takes the absolute value. If the calculated result exceeds the wear deviation threshold, the system terminal replaces the target wear degree with the predicted target wear degree. Otherwise, the predicted target wear degree and the target wear degree are averaged. By combining the predicted target wear degree and the actual calculated target wear degree, the system terminal can obtain a more accurate and comprehensive wear degree, providing strong support for the accurate calculation of the target wear resistance index.

[0041] In the above text, refer to Figure 1 A laser scanning-based method for detecting the wear resistance of generator bearings according to embodiments of the present invention is described in detail. Next, reference will be made to... Figure 2 A laser scanning-based generator bearing wear resistance testing device according to an embodiment of the present invention is described.

[0042] The laser scanning-based generator bearing wear resistance testing device according to embodiments of the present invention addresses the technical problems of low efficiency and insufficient accuracy in traditional generator bearing wear resistance testing methods, making it difficult to accurately measure and evaluate minute wear on the bearing surface. It achieves the effect of improving the accuracy and efficiency of measuring minute wear on the bearing surface through three-dimensional morphology comparison. The laser scanning-based generator bearing wear resistance testing device includes: an optical signal acquisition module 1, a feature extraction and analysis module 2, a traversal result analysis module 3, a three-dimensional reconstruction module 4, a three-dimensional morphology comparison module 5, and an evaluation and analysis module 6.

[0043] Optical signal acquisition module 1: The optical signal acquisition module 1 is used to acquire a first optical signal, which refers to the laser beam reflection and transmission light signal at the first point of the target generator bearing collected by the detector. Feature extraction and analysis module 2: The feature extraction and analysis module 2 is used to read predetermined optical features and perform feature extraction and analysis on the first optical signal based on the predetermined optical features to obtain first optical feature information.

[0044] Traversal result analysis module 3: The traversal result analysis module 3 is used to traverse the first optical feature information in the optical feature database to obtain the first traversal result, and analyze the first traversal result to determine the morphology of the first bearing.

[0045] 3D Reconstruction Module 4: The 3D reconstruction module 4 is used to perform 3D reconstruction of the target generator bearing by combining the first point and the first bearing shape to obtain the target 3D shape.

[0046] Three-dimensional morphology comparison module 5: The three-dimensional morphology comparison module 5 is used to compare the target three-dimensional morphology with the predetermined three-dimensional morphology of the predetermined generator bearing to obtain the morphology comparison result.

[0047] Evaluation and Analysis Module 6: The evaluation and analysis module 6 is used to introduce a wear resistance evaluation function to evaluate and analyze the morphology comparison results, and obtain the target wear resistance index of the target generator bearing.

[0048] Furthermore, the optical signal acquisition module 1 also includes: The detector collects the laser scanning light signal of the target generator bearing based on a predetermined laser scanning scheme to obtain the target light signal, wherein the target light signal includes the first light signal.

[0049] Furthermore, the optical signal acquisition module 1 also includes: The predetermined laser scanning scheme includes a predetermined scanning path and predetermined scanning parameters. A first scanning point corresponding to the first point is matched in the predetermined scanning path, and the optical signal corresponding to the first scanning point is recorded as the first optical signal. The predetermined scanning parameters include predetermined scanning speed parameters and predetermined resolution parameters.

[0050] Furthermore, the feature extraction and analysis module 2 also includes: The predetermined optical characteristics include a predetermined reflected light signal and a predetermined transmitted light signal. The predetermined reflected light signal includes reflected light intensity, reflected light polarization, reflected light wavefront phase, reflected light wavelength, reflected light polarization dependence, reflected light time dependence, and emitted light spatial distribution. The predetermined transmitted light signal includes transmitted light intensity, transmitted light polarization, transmitted light wavefront phase, transmitted light wavelength, transmitted light polarization dependence, transmitted light time dependence, and transmitted light spatial distribution.

[0051] Furthermore, the traversal result analysis module 3 also includes: Extract the first data from the optical feature database, where the first data refers to the optical feature information of the first bearing corresponding to the first bearing; read the predetermined labeling scheme and perform labeling processing on the first optical feature information and the first bearing optical feature information in sequence based on the predetermined labeling scheme to obtain the first vector and the first bearing vector respectively; use the Tanimoto similarity coefficient algorithm to obtain the first similarity between the first vector and the first bearing vector; when the first similarity meets the similarity limit, take the first morphology of the first bearing as the first bearing morphology.

[0052] Furthermore, the evaluation and analysis module 6 also includes: The expression for the wear resistance evaluation function is as follows: ; in, This refers to the target generator bearing. The target wear resistance index, This refers to the target generator bearing. The target wear level, This refers to the first characterization of the target wear degree. The parameters of each wear index, It refers to the first The weights of each wear index, the target wear degree being determined by... A comprehensive characterization of wear indicators is performed, and .

[0053] Furthermore, the evaluation and analysis module 6 also includes: The process involves: acquiring the target lubricating oil for the target generator bearing; performing spectral analysis on the target lubricating oil to obtain a target spectral signal; reading predetermined spectral features and acquiring multi-dimensional features of the target spectral signal based on the predetermined spectral features to obtain target signal features; calling a wear assessment model to evaluate and analyze the target signal features to obtain a predicted target wear degree, wherein the wear assessment model is an intelligent model obtained by machine learning based on neural network principles and historical lubricating oil detection data; and adjusting the target wear degree based on the predicted target wear degree.

[0054] The laser scanning-based generator bearing wear resistance testing device provided in this embodiment of the invention can execute the laser scanning-based generator bearing wear resistance testing method provided in any embodiment of the invention, and has the corresponding functional modules and beneficial effects of the method.

[0055] Although this application makes various references to certain modules in the apparatus according to the embodiments of this application, any number of different modules can be used and run on user terminals and / or servers. The various units and modules included are only divided according to functional logic, but are not limited to the above division, as long as the corresponding functions can be achieved; in addition, the specific names of each functional unit are only for easy distinction between each other and are not intended to limit the scope of protection of this invention.

[0056] The specific embodiments described above do not constitute a limitation on the scope of protection of this application. Those skilled in the art should understand that various modifications, combinations, and substitutions can be made according to design requirements and other factors. Any modifications, equivalent substitutions, and improvements made within the spirit and principles of this application should be included within the scope of protection of this application.

Claims

1. A method for detecting the wear resistance of a generator bearing based on laser scanning, characterized in that, The method comprises the following steps: acquiring a first optical signal, the first optical signal being a laser beam reflection and transmission optical signal collected by a detector at a first point of a target generator bearing; reading a predetermined optical feature and performing feature extraction analysis on the first optical signal based on the predetermined optical feature to obtain first optical feature information; traversing the first optical feature information in an optical feature database to obtain a first traversal result and analyzing the first traversal result to determine a first bearing topography; performing three-dimensional reconstruction on the target generator bearing in combination with the first point and the first bearing topography to obtain a target three-dimensional topography; comparing the target three-dimensional topography with a predetermined three-dimensional topography of a predetermined generator bearing to obtain a topography comparison result; introducing a wear resistance evaluation function to evaluate and analyze the topography comparison result to obtain a target wear resistance index of the target generator bearing.

2. The method of claim 1, wherein, The detector collects laser scanning optical signals of the target generator bearing based on a predetermined laser scanning scheme to obtain target optical signals, wherein the target optical signals include the first optical signal.

3. The method of claim 2, wherein, The predetermined laser scanning scheme includes a predetermined scanning path and predetermined scanning parameters, a first scanning point corresponding to the first point is matched in the predetermined scanning path, and an optical signal corresponding to the first scanning point is recorded as the first optical signal, wherein the predetermined scanning parameters include predetermined scanning speed parameters and predetermined resolution parameters.

4. The method of claim 1, wherein, The predetermined optical feature includes a predetermined reflection optical signal and a predetermined transmission optical signal, the predetermined reflection optical signal includes reflection light intensity, reflection light polarization, reflection light wavefront phase, reflection light wavelength, reflection light polarization dependence, reflection light time dependence, and emission light spatial distribution, and the predetermined transmission optical signal includes transmission light intensity, transmission light polarization, transmission light wavefront phase, transmission light wavelength, transmission light polarization dependence, transmission light time dependence, and transmission light spatial distribution.

5. The method of claim 1, wherein, The method comprises the following steps: extracting first data in the optical feature database, the first data being first bearing optical feature information corresponding to a first bearing; reading a predetermined label scheme and sequentially performing label processing on the first optical feature information and the first bearing optical feature information based on the predetermined label scheme to obtain a first vector and a first bearing vector, respectively; obtaining a first similarity of the first vector and the first bearing vector by using a Tanimoto similarity coefficient algorithm; when the first similarity satisfies a similarity limit value, a first topography of the first bearing is taken as the first bearing topography.

6. The method of claim 1, wherein, The expression of the wear resistance evaluation function is as follows: ； wherein refers to the target wear resistance index of the target generator bearing , refers to the target wear degree of the target generator bearing , refers to a parameter of a first wear indicator characterizing the target wear degree, refers to a weight of a first wear indicator, the target wear degree being synthetically characterized by wear indicators, and .​​ 7. The method of claim 6, wherein, The method further comprises the following steps: acquiring target lubricating oil of the target generator bearing; performing spectral analysis on the target lubricating oil to obtain a target spectral signal; reading a predetermined spectral feature and performing multi-dimensional feature collection on the target spectral signal based on the predetermined spectral feature to obtain target signal features; calling a wear evaluation model to evaluate and analyze the target signal features to obtain a predicted target wear degree, wherein the wear evaluation model is an intelligent model obtained by machine learning on historical lubricating oil detection data based on a neural network principle; adjust the target wear degree based on the predicted target wear degree.

8. A laser scanning based generator bearing wear resistance detection device, characterized in that, The device is used to implement the laser scanning based generator bearing wear resistance detection method of any one of claims 1-7, comprising: an optical signal acquisition module: acquiring a first optical signal, the first optical signal being a laser beam reflection and transmission optical signal collected by a detector at a first point of a target generator bearing; a feature extraction analysis module: reading a predetermined optical feature and performing feature extraction analysis on the first optical signal based on the predetermined optical feature to obtain first optical feature information; a traversal result analysis module: traversing the first optical feature information in an optical feature database to obtain a first traversal result and analyzing the first traversal result to determine a first bearing topography; a three-dimensional reconstruction module: combining the first point and the first bearing topography to perform three-dimensional reconstruction on the target generator bearing to obtain a target three-dimensional topography; a three-dimensional topography comparison module: comparing the target three-dimensional topography with a predetermined three-dimensional topography of a predetermined generator bearing to obtain a topography comparison result; an evaluation analysis module: introducing a wear resistance evaluation function to evaluate and analyze the topography comparison result to obtain a target wear resistance index of the target generator bearing.

Images