An operation and maintenance method for a ball screw automatic loading and unloading mechanism

Abstract

The present invention relates to the field of equipment operation and maintenance technology, and in particular to an operation and maintenance method for an automatic loading and unloading mechanism of a ball screw. Based on a wavelet transform algorithm, the laser point cloud data set of a target component is corrected and processed; a characteristic state model diagram of the target component is constructed according to the corrected laser point cloud data set; the degree of overlap between the characteristic state model diagram and the standard state model diagram is calculated; if the degree of overlap is not greater than a preset threshold, the target component is marked as a scrapped component, and the scrapping reason of the scrapped component is obtained; an evaluation index system is constructed based on a gray correlation analysis method, and a correlation analysis is performed on each assembly component in the automatic loading and unloading mechanism of a ball screw based on the evaluation index system combined with the scrapping reason of the scrapped component, and suspicious assembly components are listed as target components. This method can quickly and accurately diagnose faulty components, providing strong support for the maintenance and repair of the automatic loading and unloading mechanism of a ball screw.

Description

Technical Field

[0001] The invention relates to the technical field of equipment operation and maintenance, and in particular to an operation and maintenance method of a ball screw automatic loading and unloading mechanism. Background Art

[0002] The ball screw automatic loading and unloading mechanism plays a vital role in modern automated production lines. It not only improves production efficiency and reduces labor costs, but also enhances the flexibility and reliability of the production line. However, during the long-term operation of this type of equipment, due to frequent mechanical movement, environmental factors and uncertainty in human operation, it often faces a series of problems, including but not limited to wear, corrosion, overload, fault shutdown, etc. These problems directly affect the normal operation, production efficiency and product quality of the equipment. Therefore, it is particularly important to carry out effective operation and maintenance management of the ball screw automatic loading and unloading mechanism. However, its operation and maintenance methods still face some technical obstacles: First, the complexity of the ball screw automatic loading and unloading mechanism increases the difficulty of fault diagnosis and maintenance. The mechanism consists of multiple components, including ball screws, guide rails, motors, sensors, etc. The failure of any component may cause the failure of the entire system. Therefore, professional technicians are required to perform fault diagnosis and maintenance, which increases the cost and time of operation and maintenance. Secondly, the operation and maintenance of the ball screw automatic loading and unloading mechanism needs to rely on advanced monitoring and diagnostic technologies. At present, although there are some monitoring and diagnostic technologies that can be used for the operation and maintenance of ball screw automatic loading and unloading mechanisms, these technologies still have some limitations, such as low accuracy, insufficient reliability, and high cost. Therefore, it is necessary to further develop and improve these technologies to improve the operation and maintenance level of ball screw automatic loading and unloading mechanisms. Summary of the invention

[0003] The present invention overcomes the deficiencies of the prior art and provides an operation and maintenance method for a ball screw automatic loading and unloading mechanism.

[0004] To achieve the above-mentioned purpose, the technical solution adopted by the present invention is:

[0005] The first aspect of the present invention discloses an operation and maintenance method of a ball screw automatic loading and unloading mechanism, comprising the following steps:

[0006] Scan and detect the target components in the automatic loading and unloading mechanism of the ball screw by using a laser scanning device, collect the laser point cloud data fed back by the target components, and obtain the laser point cloud data set of the target components; wherein the target components include the ball screw, the bearing, the guide block and the guide rail;

[0007] The laser point cloud data set of the target component is corrected based on the wavelet transform algorithm to obtain a corrected laser point cloud data set; a characteristic state model diagram of the target component is constructed according to the corrected laser point cloud data set;

[0008] Obtaining a standard state model diagram of a target component, calculating the degree of overlap between the characteristic state model diagram and the standard state model diagram; and comparing the degree of overlap with a preset threshold;

[0009] If the overlap degree is not greater than a preset threshold, the target component is marked as a scrapped component, and the characteristic state model diagram of the target component is imported into the label knowledge base for matching to obtain the scrap reason of the scrapped component;

[0010] An evaluation index system is constructed based on the grey correlation analysis method, and based on the evaluation index system and the scrapping reasons of the scrapped parts, a correlation analysis is performed on each assembly part in the automatic loading and unloading mechanism of the ball screw to obtain suspicious assembly parts;

[0011] The suspicious assembly parts are listed as target parts, and the suspicious assembly parts listed as target parts are subjected to laser scanning detection and evaluation to evaluate whether the suspicious assembly parts are also scrapped parts.

[0012] Furthermore, the laser point cloud data set of the target component is corrected based on the wavelet transform algorithm to obtain a corrected laser point cloud data set, specifically:

[0013] Introducing a wavelet transform algorithm, and selecting a wavelet basis function as an objective function; wherein the wavelet basis function includes a Daubechies wavelet basis function, a Morlet wavelet basis function, and a Haar wavelet basis function;

[0014] The scale range of the analysis is set, and discrete wavelet transform is performed on each laser point cloud data in the laser point cloud data set based on the objective function to obtain a set of wavelet coefficients; wherein each wavelet coefficient corresponds to the decomposition of the original laser point cloud data at a specific scale;

[0015] The wavelet coefficient matrix of the laser point cloud data set is constructed according to the obtained wavelet coefficients; each row in the wavelet coefficient matrix represents a time point of the original laser point cloud data, each column represents a specific scale, and the size of the element value reflects the oscillation intensity of the original laser point cloud data at a specific scale;

[0016] Compare each element in the wavelet coefficient matrix with a preset coefficient value; if an element in the wavelet coefficient matrix is ​​less than the preset coefficient value, mark the original laser point cloud data corresponding to the element as noise data, and filter out the original laser point cloud data marked as noise data in the laser point cloud data set;

[0017] Calculate the information entropy between each element in the wavelet coefficient matrix, and compare the information entropy between each element with a preset entropy value; if the information entropy between two elements is less than the preset entropy value, mark the original laser point cloud data corresponding to the two elements as redundant data, and filter out one of the original laser point cloud data marked as redundant data in the laser point cloud data set;

[0018] Repeat the above steps until the noise data and redundant data in the laser point cloud dataset are screened out to obtain the corrected laser point cloud dataset.

[0019] Furthermore, the degree of overlap between the characteristic state model diagram and the standard state model diagram is calculated, specifically:

[0020] Constructing a three-dimensional grid space and equally dividing the three-dimensional grid space into a plurality of sub-grids of unit volume; inputting the characteristic state model diagram and the standard state model diagram into the three-dimensional grid space;

[0021] Aligning the characteristic state model diagram with the standard state model diagram according to the reference features of the characteristic state model diagram and the standard state model diagram;

[0022] Constructing a counting table, traversing each subgrid of the three-dimensional grid space, if a subgrid has both a characteristic state model diagram and a standard state model diagram, then filling 1 in the initial blank position of the counting table;

[0023] If only characteristic state model diagrams or only standard state model diagrams exist in a subgrid, fill in -1 in the next blank position of the counting table;

[0024] If neither the characteristic state model diagram nor the standard state model diagram exists in a subgrid, fill in 0 in the next blank position of the counting table;

[0025] After the traversal is completed, the sum of all the counts of 1 in the counting table is counted to obtain the first volume value; and the sum of all the counts of -1 in the counting table is counted, and the sum of all the counts of -1 in the counting table is processed as an absolute value to obtain the second volume value;

[0026] The first volume value and the second volume value are ratio-processed to obtain a volume ratio, and the volume ratio is weighted to obtain the degree of overlap between the characteristic state model diagram and the standard state model diagram.

[0027] Furthermore, the characteristic state model diagram of the target component is imported into the label knowledge base for matching, and the scrapping reason of the scrapped component is obtained, which is specifically:

[0028] Retrieving scrap feature model diagrams corresponding to various types of scrapped products of target components based on a big data network; Building a knowledge base, dividing the knowledge base into a plurality of sub-storage spaces, and assigning corresponding scrap type labels to each sub-storage space;

[0029] The scrap feature model graphs corresponding to each type of scrapped product are stored in the sub-storage space assigned with the corresponding scrap type label to obtain a label knowledge base;

[0030] Acquire a characteristic state model diagram of a target component, calculate similarity values ​​between the characteristic state model diagram and the scrap characteristic model diagrams in each sub-storage space in a label knowledge base based on a cosine similarity algorithm, and obtain similarity values ​​between the characteristic state model diagram and each scrap characteristic model diagram;

[0031] Acquire a scrap feature model diagram having the largest similarity value to the feature state model diagram, and extract a scrap type label of a sub-storage space to which the scrap feature model diagram having the largest similarity value to the feature state model diagram belongs;

[0032] The scrapping reason of the scrapped part is determined according to the extracted scrapping type label.

[0033] Furthermore, an evaluation index system is constructed based on the grey correlation analysis method, and based on the evaluation index system and the scrapping reasons of the scrapped parts, a correlation analysis is performed on each assembly part in the automatic loading and unloading mechanism of the ball screw, and the suspicious assembly parts are obtained by analysis, specifically:

[0034] Acquire assembly engineering drawing information of the ball screw automatic loading and unloading mechanism, acquire assembly relationships between various assembly components in the ball screw automatic loading and unloading mechanism according to the assembly engineering drawing information, and acquire functional characteristic information of various assembly components;

[0035] Determine the assembly relationship index between the assembly components according to the assembly relationship between the assembly components; determine the functional characteristic index of each assembly component according to the functional characteristic information of each assembly component;

[0036] Based on the big data network, all possible scrapping reasons of each assembly component are obtained, and a unique code is assigned to each scrapping reason to obtain the scrapping reason code of each assembly component;

[0037] Using the assembly relationship index as a comparison sequence, the functional characteristic index as a benchmark sequence, and the scrapping reason code of each assembly component as a reference sequence;

[0038] Calculating the grey correlation weight values ​​between the comparison sequence, the benchmark sequence and the reference sequence based on the grey correlation analysis method; constructing an evaluation index system according to the grey correlation weight values ​​between the comparison sequence, the benchmark sequence and the reference sequence;

[0039] Obtaining the scrapping reasons of the scrapped parts, importing the scrapping reasons of the scrapped parts into the evaluation index system for evaluation, and obtaining a plurality of grey correlation degrees;

[0040] Among the obtained several grey correlation degrees, assembly components whose grey correlation degree is greater than a preset correlation degree threshold are screened out, and the screened assembly components are defined as suspicious assembly components.

[0041] Furthermore, the scrap type labels include scrap due to wear, scrap due to deformation, scrap due to fatigue, scrap due to cracking, and scrap due to corrosion.

[0042] Furthermore, the assembly relationship index is used to reflect the connection mode, relative position and movement relationship between the various assembly components, including direct connection, indirect connection, fixed connection and sliding connection; the functional characteristic index is used to reflect the functional role, working principle and performance parameters of each assembly component, including support, guidance, transmission, sealing and lubrication.

[0043] Furthermore, all possible reasons for failure of the assembly components include wear, fatigue, fracture, corrosion and deformation.

[0044] The second aspect of the present invention discloses an operation and maintenance system for a ball screw automatic loading and unloading mechanism, the operation and maintenance system comprising a memory and a processor, the memory storing an operation and maintenance method program for the ball screw automatic loading and unloading mechanism, when the operation and maintenance method program for the ball screw automatic loading and unloading mechanism is executed by the processor, any one of the operation and maintenance method steps for the ball screw automatic loading and unloading mechanism is implemented.

[0045] The third aspect of the present invention discloses a computer-readable storage medium, which includes an operation and maintenance method program for an automatic loading and unloading mechanism of a ball screw. When the operation and maintenance method program for the automatic loading and unloading mechanism of a ball screw is executed by a processor, any one of the operation and maintenance method steps for the automatic loading and unloading mechanism of a ball screw is implemented.

[0046] The present invention solves the technical defects existing in the background technology, and has the following beneficial effects: based on the wavelet transform algorithm, the laser point cloud data set of the target component is corrected and processed; the characteristic state model diagram of the target component is constructed according to the corrected laser point cloud data set; the overlap degree between the characteristic state model diagram and the standard state model diagram is calculated; if the overlap degree is not greater than the preset threshold, the target component is marked as a scrapped component, and the characteristic state model diagram of the target component is imported into the label knowledge base for matching to obtain the scrapping reason of the scrapped component; an evaluation index system is constructed based on the gray correlation analysis method, and the evaluation index system is combined with the scrapping reason of the scrapped component to perform correlation analysis on each assembly component in the automatic loading and unloading mechanism of the ball screw; suspicious assembly components are listed as target components, and the suspicious assembly components listed as target components are subjected to laser scanning detection and evaluation. The present method can quickly and accurately identify scrapped components and analyze the scrapping reasons, providing strong support for the maintenance and repair of the automatic loading and unloading mechanism of the ball screw. BRIEF DESCRIPTION OF THE DRAWINGS

[0047] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, drawings of other embodiments can be obtained based on these drawings without paying creative work.

[0048] Figure 1 It is a flow chart of the overall method of operation and maintenance method of ball screw automatic loading and unloading mechanism;

[0049] Figure 2 It is a partial method flow chart of an operation and maintenance method of a ball screw automatic loading and unloading mechanism;

[0050] Figure 3 The system block diagram of an operation and maintenance system for a ball screw automatic loading and unloading mechanism. DETAILED DESCRIPTION

[0051] In order to more clearly understand the above-mentioned purpose, features and advantages of the present invention, the present invention is further described in detail below in conjunction with the accompanying drawings and specific embodiments. It should be noted that the embodiments of the present application and the features in the embodiments can be combined with each other without conflict.

[0052] In the following description, many specific details are set forth to facilitate a full understanding of the present invention. However, the present invention may also be implemented in other ways different from those described herein. Therefore, the protection scope of the present invention is not limited to the specific embodiments disclosed below.

[0053] like Figure 1As shown, the first aspect of the present invention discloses an operation and maintenance method of a ball screw automatic loading and unloading mechanism, comprising the following steps:

[0054] S102: Scan and detect the target component in the automatic loading and unloading mechanism of the ball screw by using a laser scanning device, collect laser point cloud data fed back by the target component, and obtain a laser point cloud data set of the target component; wherein the target component includes a ball screw, a bearing, a guide block, and a guide rail;

[0055] S104: Correcting the laser point cloud data set of the target component based on a wavelet transform algorithm to obtain a corrected laser point cloud data set; constructing a characteristic state model diagram of the target component according to the corrected laser point cloud data set;

[0056] S106: Acquire a standard state model diagram of the target component, calculate the overlap degree between the characteristic state model diagram and the standard state model diagram; and compare the overlap degree with a preset threshold;

[0057] S108: If the overlap degree is not greater than a preset threshold, the target component is marked as a scrapped component, and the characteristic state model diagram of the target component is imported into a label knowledge base for matching to obtain the scrap reason of the scrapped component;

[0058] S110: constructing an evaluation index system based on a grey correlation analysis method, and performing correlation analysis on each assembly component in the ball screw automatic loading and unloading mechanism based on the evaluation index system and the scrapping reasons of the scrapped components, and obtaining suspicious assembly components through analysis;

[0059] S112: List the suspicious assembly components as target components, and perform laser scanning detection and evaluation on the suspicious assembly components listed as target components to evaluate whether the suspicious assembly components are also scrapped components.

[0060] It should be noted that the laser point cloud data set of the target component is collected by laser scanning equipment, and then the data is corrected and processed based on the wavelet transform algorithm to construct the characteristic state model diagram of the target component. Then, the characteristic state model diagram is compared with the standard state model diagram. If the degree of overlap is not greater than the preset threshold, the target component is marked as a scrapped component, and the scrapping reason is obtained by matching the label knowledge base. Then, an evaluation index system is constructed based on the gray correlation analysis method, and the association analysis of each assembly component is performed in combination with the scrapping reason to obtain the suspicious assembly component. Finally, the suspicious assembly component is laser scanned and detected to determine whether it is a scrapped component. This method can realize accurate scanning, detection and analysis of the target components of the automatic loading and unloading mechanism of the ball screw. Through technical means such as wavelet transform algorithm and gray correlation analysis method, the accuracy and reliability of the data are improved, and the scrapped components can be quickly and accurately identified, and the scrapping reason is analyzed, which provides strong support for the maintenance and repair of the automatic loading and unloading mechanism of the ball screw.

[0061] Furthermore, if Figure 2 As shown, the laser point cloud data set of the target component is corrected based on the wavelet transform algorithm to obtain the corrected laser point cloud data set, which is specifically:

[0062] S202: Introduce a wavelet transform algorithm, and select a wavelet basis function as an objective function; wherein the wavelet basis function includes a Daubechies wavelet basis function, a Morlet wavelet basis function, and a Haar wavelet basis function;

[0063] S204: setting the scale range of the analysis, and performing discrete wavelet transform processing on each laser point cloud data in the laser point cloud data set based on the objective function to obtain a set of wavelet coefficients; wherein each wavelet coefficient corresponds to the decomposition of the original laser point cloud data at a specific scale;

[0064] S206: constructing a wavelet coefficient matrix of the laser point cloud data set according to the obtained wavelet coefficients; wherein each row in the wavelet coefficient matrix represents a time point of the original laser point cloud data, each column represents a specific scale, and the size of the element value reflects the oscillation intensity of the original laser point cloud data at a specific scale;

[0065] S208: Compare each element in the wavelet coefficient matrix with a preset coefficient value; if an element in the wavelet coefficient matrix is ​​less than the preset coefficient value, mark the original laser point cloud data corresponding to the element as noise data, and filter out the original laser point cloud data marked as noise data in the laser point cloud data set;

[0066] S210: calculating the information entropy between each element in the wavelet coefficient matrix, and comparing the information entropy between each element with a preset entropy value; if the information entropy between two elements is less than the preset entropy value, marking the original laser point cloud data corresponding to the two elements as redundant data, and filtering out one of the original laser point cloud data marked as redundant data in the laser point cloud data set;

[0067] S212: Repeat the above steps until the noise data and redundant data in the laser point cloud dataset are screened out to obtain a corrected laser point cloud dataset.

[0068] It should be noted that the laser point cloud data is processed by discrete wavelet transform by relevant technicians by setting or selecting the corresponding wavelet basis function as the objective function and setting the scale range of the analysis, and a series of wavelet coefficients are obtained. These coefficients reflect the decomposition of the original laser point cloud data at different scales and the oscillation intensity. Next, a wavelet coefficient matrix is ​​established, each row corresponds to the data at a time point, each column represents a specific scale, and the size of the matrix element represents the oscillation intensity of the original data at a specific scale. Next, the elements in the wavelet coefficient matrix are compared. If an element is less than the preset coefficient value, the original laser point cloud data corresponding to the element is considered to be noise data and is removed from the original data set. Subsequently, the information entropy between the elements in the matrix is ​​calculated. If the information entropy between the two elements is less than the preset entropy value, the original laser point cloud data corresponding to the two elements is considered to be redundant data, and one of them is removed from the data set. This process is repeated until all noise data and redundant data are removed, and finally a corrected laser point cloud data set is obtained. By comparing the wavelet coefficients with the preset values, the noise part in the laser point cloud data is effectively identified and removed, and the data quality is improved. By calculating the information entropy between elements and comparing them, redundant data can be identified and eliminated, reducing the amount of data while maintaining valid information in the data. After the above steps, the corrected laser point cloud data set is finally obtained, which is purer and more efficient, providing higher quality data support for subsequent analysis, modeling and other tasks. The entire process is automated, reducing manual intervention and improving the efficiency and consistency of data processing. In summary, this method effectively processes the noise and redundant information in the laser point cloud data through the wavelet transform algorithm, significantly improves the quality and availability of the data, and provides a solid foundation for subsequent analysis.

[0069] Furthermore, the degree of overlap between the characteristic state model diagram and the standard state model diagram is calculated, specifically:

[0070] Constructing a three-dimensional grid space and equally dividing the three-dimensional grid space into a plurality of sub-grids of unit volume; inputting the characteristic state model diagram and the standard state model diagram into the three-dimensional grid space;

[0071] Aligning the characteristic state model diagram with the standard state model diagram according to the reference features of the characteristic state model diagram and the standard state model diagram;

[0072] Constructing a counting table, traversing each subgrid of the three-dimensional grid space, if a subgrid has both a characteristic state model diagram and a standard state model diagram, then filling 1 in the initial blank position of the counting table;

[0073] If only characteristic state model diagrams or only standard state model diagrams exist in a subgrid, fill in -1 in the next blank position of the counting table;

[0074] If neither the characteristic state model diagram nor the standard state model diagram exists in a subgrid, fill in 0 in the next blank position of the counting table;

[0075] After the traversal is completed, the sum of all the counts of 1 in the counting table is counted to obtain the first volume value; and the sum of all the counts of -1 in the counting table is counted, and the sum of all the counts of -1 in the counting table is processed as an absolute value to obtain the second volume value;

[0076] The first volume value and the second volume value are ratio-processed to obtain a volume ratio, and the volume ratio is weighted to obtain the degree of overlap between the characteristic state model diagram and the standard state model diagram.

[0077] The reference features include assembly reference plane, positioning reference plane, axis and other features.

[0078] It should be noted that the three-dimensional grid space is constructed using industrial drawing software such as CAD and divided into multiple sub-grids of unit volume. Then, the characteristic state model diagram and the standard state model diagram are respectively input into this grid space. Next, by comparing the benchmark features of the two, the registration operation is performed to ensure that their spatial correspondence is correct. After registration, the entire three-dimensional grid space is traversed to record the coexistence state of the two models in each sub-grid: if both exist, the current position of the counting table is filled with 1; if only one exists, it is filled with -1; if neither exists, it is filled with 0. After the traversal is completed, the total number of 1 and -1 in the counting table is counted to obtain two volume values ​​respectively. Then, the ratio of the two volume values ​​is calculated, and the results are weighted to obtain the degree of overlap between the characteristic state model diagram and the standard state model diagram. Through three-dimensional gridding, accurate spatial matching and quantitative analysis between models are achieved, and the similarities or differences in spatial distribution of the two models can be carefully compared. The constructed counting table provides an intuitive way to quickly identify overlapping and non-overlapping areas between models, making it easier to understand the spatial distribution of the two. By calculating the volume ratio and performing weighted processing, a numerical overlap degree is obtained, making the comparison results more operational and interpretable. This method is suitable for the rapid analysis of large amounts of data, especially in scenarios where complex three-dimensional model comparisons need to be processed, and can significantly improve analysis efficiency.

[0079] Furthermore, the characteristic state model diagram of the target component is imported into the label knowledge base for matching, and the scrapping reason of the scrapped component is obtained, which is specifically:

[0080] Retrieving scrap feature model diagrams corresponding to various types of scrapped products of target components based on a big data network; Building a knowledge base, dividing the knowledge base into a plurality of sub-storage spaces, and assigning corresponding scrap type labels to each sub-storage space;

[0081] The scrap feature model graphs corresponding to each type of scrapped product are stored in the sub-storage space assigned with the corresponding scrap type label to obtain a label knowledge base;

[0082] Acquire a characteristic state model diagram of a target component, calculate similarity values ​​between the characteristic state model diagram and the scrap characteristic model diagrams in each sub-storage space in a label knowledge base based on a cosine similarity algorithm, and obtain similarity values ​​between the characteristic state model diagram and each scrap characteristic model diagram;

[0083] Acquire a scrap feature model diagram having the largest similarity value to the feature state model diagram, and extract a scrap type label of a sub-storage space to which the scrap feature model diagram having the largest similarity value to the feature state model diagram belongs;

[0084] The scrapping reason of the scrapped part is determined according to the extracted scrapping type label.

[0085] It should be noted that the scrap feature model diagrams of various scrapped products of the target component are retrieved through the big data network. Then, a knowledge base is constructed and divided into several sub-storage spaces, and a corresponding scrap type label is assigned to each sub-storage space. Next, the scrap feature model diagrams of each type of scrapped product are stored in the corresponding sub-storage space to form a label knowledge base. After that, the characteristic state model diagram of the target component is obtained, and the cosine similarity algorithm is used to calculate its similarity value with each scrap feature model diagram in the label knowledge base. Finally, the scrap feature model diagram with the largest similarity value with the characteristic state model diagram is found, and the scrap type label of the sub-storage space to which it belongs is extracted, so as to determine the scrap reason of the scrapped component. In summary, this method can quickly and accurately determine the scrap reason of the scrapped component, and can effectively improve the operation and maintenance efficiency.

[0086] Furthermore, an evaluation index system is constructed based on the grey correlation analysis method, and based on the evaluation index system and the scrapping reasons of the scrapped parts, a correlation analysis is performed on each assembly part in the automatic loading and unloading mechanism of the ball screw, and the suspicious assembly parts are obtained by analysis, specifically:

[0087] Acquire assembly engineering drawing information of the ball screw automatic loading and unloading mechanism, acquire assembly relationships between various assembly components in the ball screw automatic loading and unloading mechanism according to the assembly engineering drawing information, and acquire functional characteristic information of various assembly components;

[0088] Determine the assembly relationship index between the assembly components according to the assembly relationship between the assembly components; determine the functional characteristic index of each assembly component according to the functional characteristic information of each assembly component;

[0089] Based on the big data network, all possible scrapping reasons of each assembly component are obtained, and a unique code is assigned to each scrapping reason to obtain the scrapping reason code of each assembly component;

[0090] Using the assembly relationship index as a comparison sequence, the functional characteristic index as a benchmark sequence, and the scrapping reason code of each assembly component as a reference sequence;

[0091] Calculating the grey correlation weight values ​​between the comparison sequence, the benchmark sequence and the reference sequence based on the grey correlation analysis method; constructing an evaluation index system according to the grey correlation weight values ​​between the comparison sequence, the benchmark sequence and the reference sequence;

[0092] Obtaining the scrapping reasons of the scrapped parts, importing the scrapping reasons of the scrapped parts into the evaluation index system for evaluation, and obtaining a plurality of grey correlation degrees;

[0093] Among the obtained several grey correlation degrees, assembly components whose grey correlation degree is greater than a preset correlation degree threshold are screened out, and the screened assembly components are defined as suspicious assembly components.

[0094] Among them, the assembly components in the ball screw automatic loading and unloading mechanism are all the components of the ball screw automatic loading and unloading mechanism; the functional characteristic information of each assembly component refers to the relevant information of the specific functions and characteristics of each component that constitutes a whole. This information describes the role of each component in the whole, its working principle, performance characteristics and its relationship with other components.

[0095] It should be noted that to assign a unique code to each scrapping reason, the following steps can be taken:

[0096] (1) Determine the classification system for scrapping reasons: First, it is necessary to determine the classification method for scrapping reasons, such as classification according to component failure, wear, damage, aging, etc.

[0097] (2) Assign a code range to each category: According to the classification system, assign a code range to each category, for example, 001-099 indicates component failure, 100-199 indicates wear, 200-299 indicates damage, 300-399 indicates aging, etc.

[0098] (3) Assign a unique code to each specific scrapping reason: Within each classification range, assign a unique code to each specific scrapping reason, for example, 001 indicates component A failure, 002 indicates component B failure, 101 indicates component C wear, 201 indicates component D damage, etc.

[0099] (4) Establish a scrapping reason coding table: All scrapping reason codes and their corresponding description information are established in a coding table for easy query and use.

[0100] (5) Code the scrapping reason for each assembly component: According to the scrapping reason coding table, assign a corresponding code to the scrapping reason of each assembly component.

[0101] Through the above steps, a unique code can be assigned to each scrapping reason, and the scrapping reason code of each assembly component can be obtained. This can easily count and analyze the scrapping reasons, which is helpful to improve the maintenance and management level of the ball screw automatic loading and unloading mechanism.

[0102] It should be noted that by obtaining the assembly engineering drawing information of the ball screw automatic loading and unloading mechanism, the assembly relationship between each assembly component and the functional characteristic information of each component are parsed and obtained. Then, based on this information, the system further determines the assembly relationship index and functional characteristic index between each assembly component. In order to fully cover the possible causes of failure, the system uses the big data network to obtain all potential scrapping reasons of each assembly component, and assigns a unique code to each scrapping reason, forming a scrapping reason coding database for each assembly component. Through such a coding system, the system can accurately classify and track faults. The assembly relationship index is used as a comparison sequence, the functional characteristic index is used as a benchmark sequence, and the scrapping reason code is used as a reference sequence. The gray correlation analysis method is used to calculate the gray correlation weight value between the three sequences. The gray correlation analysis method is a data analysis method that effectively handles uncertainty and fuzziness. By calculating the weight value, the system constructs an evaluation index system for evaluating the status of each assembly component. Once the specific cause of the scrapped component is obtained, the scrapping reason of the component will be imported into the evaluation index system for evaluation, generating a series of gray correlation degrees. By analyzing these correlations, the system can screen out assembly parts with high correlation with the scrapping cause, define them as suspicious assembly parts, and indicate parts that may need special attention or repair. By obtaining complete assembly engineering drawing information and big data network resources, the system can comprehensively collect and analyze various possible fault causes and improve the accuracy of fault diagnosis. The evaluation index system is constructed using the gray correlation analysis method, which realizes the intelligent correlation between the fault cause and the component status and improves the intelligent level of fault prediction. By screening out suspicious assembly parts with high correlation, the system can accurately locate the problem, which helps to formulate more effective maintenance strategies and reduce unnecessary costs and time waste. Based on the results of predictive analysis, the system can warn of potential failure risks in advance, promote preventive maintenance, extend the service life of equipment, and improve production efficiency and safety. In short, by integrating assembly relationships, functional characteristics and scrapping cause information, and combining the evaluation index system constructed by big data and gray correlation analysis, this system can not only effectively diagnose problems of existing equipment, but also predict potential failures, helping enterprises make more informed decisions in maintenance strategies, resource allocation and production management, thereby improving overall operational efficiency and equipment reliability.

[0103] Furthermore, the scrap type labels include scrap due to wear, scrap due to deformation, scrap due to fatigue, scrap due to cracking, and scrap due to corrosion.

[0104] Furthermore, the assembly relationship index is used to reflect the connection mode, relative position and movement relationship between the various assembly components, including direct connection, indirect connection, fixed connection and sliding connection; the functional characteristic index is used to reflect the functional role, working principle and performance parameters of each assembly component, including support, guidance, transmission, sealing and lubrication.

[0105] Furthermore, all possible reasons for failure of the assembly components include wear, fatigue, fracture, corrosion and deformation.

[0106] In addition, this operation and maintenance method also includes the following steps:

[0107] If the overlap degree is greater than a preset threshold, the target component is scanned and detected by an ultrasonic scanning device to obtain crack acoustic wave data fed back by the target component, thereby obtaining a crack acoustic wave data set of the target component;

[0108] Based on the wavelet transform algorithm, the crack acoustic wave data set of the target component is corrected to obtain a corrected crack acoustic wave data set; crack characteristic data in the target component is obtained according to the corrected crack acoustic wave data set; the crack characteristic data includes crack type, crack position, crack length, crack width and crack height;

[0109] Calculating a crack concentration value in the target component according to the crack characteristic data in the target component; if the crack concentration value is not greater than a preset concentration value, calibrating the target component as qualified;

[0110] If the crack concentration value is greater than a preset concentration value, the target component is marked as a scrapped component, and the scrapping reason of the scrapped component is marked as crack scrapping;

[0111] If the scrapping reason of the scrapped component is crack scrapping, the operation log information of the ball screw automatic loading and unloading mechanism is obtained, and the historical working temperature of each preset position node within the regional range of the scrapped component during the historical working process is obtained according to the operation log information;

[0112] Constructing a temperature distribution diagram of the scrapped component during the historical working process according to the historical working temperature of each preset position node; calculating a hash value between the temperature distribution diagram of the scrapped component during the historical working process and a preset temperature distribution diagram based on a hash algorithm;

[0113] If the hash value is not greater than a preset hash value threshold, temperature abnormality warning information is generated.

[0114] It should be noted that when the overlap degree of the target component exceeds the preset threshold, it means that the external dimensions of the target component are normal. At this time, the ultrasonic scanning equipment will be started to further detect the internal situation of the component, collect crack acoustic wave data, and form a crack acoustic wave data set. These data are corrected and processed by the wavelet transform algorithm to improve the accuracy and stability of the data, and then the characteristic data of the cracks, including crack type, position, length, width and height, are extracted. The analysis results of these characteristic data are used to calculate the crack concentration value as a basis for judging whether the component is qualified. If the crack concentration value is lower than the preset threshold, the component is considered qualified; otherwise, if it exceeds the threshold, it is judged as a scrapped component and the reason for its scrapping is marked as a crack. For the components that are scrapped due to cracks, the system will further obtain its temperature distribution information during the historical working process. By analyzing the historical working temperature of each preset position node, a temperature distribution map is constructed. Then, the hash algorithm is used to calculate the hash value between the temperature distribution map of the current component and the preset temperature distribution map. If this hash value is lower than the preset threshold, it means that the temperature distribution of the current component is abnormal. At this time, the cooling system in the mechanism (such as the spray liquid cooling system) may also have a fault. The system will generate temperature abnormality warning information to further troubleshoot the relevant cooling system. In summary, through the comprehensive use of a variety of advanced data analysis methods, efficient and accurate management and early warning of crack problems in the ball screw automatic loading and unloading mechanism are achieved, which significantly improves the maintenance efficiency and operation safety of the equipment.

[0115] In addition, this operation and maintenance method also includes the following steps:

[0116] Obtain electrical parameter data of each electrical module in the ball screw automatic loading and unloading mechanism at several preset time nodes, and perform feature extraction on the electrical parameter data of each electrical module at several preset time nodes based on the principal component analysis method to obtain feature vector information of the electrical parameter data at the preset time nodes;

[0117] A Markov chain is introduced, and a state transition calculation is performed on the characteristic vector information of the electrical parameter data at each preset time node based on the Markov chain to obtain the state transition probability of the electrical module at several preset time nodes;

[0118] Constructing a state transition probability matrix according to the state transition probabilities of the electrical module at a number of preset time nodes; performing normalization processing on the state transition probability matrix constructed based on the state transition probabilities of the electrical module at a number of preset time nodes to obtain the overall state transition probability of the electrical module within a preset time period;

[0119] If the overall state transition probability of the electrical module within a preset time period is greater than a preset probability value, the electrical module is marked as a faulty electrical model.

[0120] The electrical modules include but are not limited to servo drive modules, power modules, PLC modules, relay modules, and inverter modules. Electrical parameters include current, voltage, resistance, power, frequency, phase, etc.

[0121] It should be noted that by obtaining the electrical parameter data of the electrical module at several preset time nodes, the principal component analysis method is used for feature extraction, and the complex data is converted into feature vector information, which simplifies the data structure and improves the data processing efficiency and analysis accuracy. Next, the Markov chain theory is introduced to construct the state transition probability matrix by calculating the state transition probability of the feature vector information of the electrical parameter data between different time points. This step is essentially to capture the regularity of the state change of the electrical module through time series analysis, which provides a basis for fault prediction. By further normalizing the state transition probability matrix, the overall state transition probability of the electrical module within the preset time period is obtained, so as to more accurately quantify the state change trend of the electrical module. Finally, a preset probability value is set as the judgment standard. When the overall state transition probability of the electrical module within the preset time period exceeds this threshold, the electrical module is marked as a faulty electrical model. This judgment mechanism can timely identify the potential faults of the electrical module, realize early warning of faults, and avoid production interruptions and economic losses caused by electrical faults. In summary, this method provides an effective fault diagnosis and prediction method for the electrical module of the ball screw automatic loading and unloading mechanism, which helps to improve the reliability and production efficiency of the equipment.

[0122] like Figure 3 As shown, the second aspect of the present invention discloses an operation and maintenance system for a ball screw automatic loading and unloading mechanism, the operation and maintenance system includes a memory 20 and a processor 60, the memory 20 stores an operation and maintenance method program for the ball screw automatic loading and unloading mechanism, when the operation and maintenance method program for the ball screw automatic loading and unloading mechanism is executed by the processor 60, any one of the operation and maintenance method steps for the ball screw automatic loading and unloading mechanism is implemented.

[0123] The third aspect of the present invention discloses a computer-readable storage medium, which includes an operation and maintenance method program for an automatic loading and unloading mechanism of a ball screw. When the operation and maintenance method program for the automatic loading and unloading mechanism of a ball screw is executed by a processor, any one of the operation and maintenance method steps for the automatic loading and unloading mechanism of a ball screw is implemented.

[0124] The above are only specific embodiments of the present invention, but the protection scope of the present invention is not limited thereto. Any technician familiar with the technical field can easily think of changes or substitutions within the technical scope disclosed by the present invention, which should be covered by the protection scope of the present invention.

Claims

1. An operation and maintenance method for a ball screw automatic loading and unloading mechanism, characterized in that: The following steps are involved: Scan and detect the target components in the automatic loading and unloading mechanism of the ball screw by using a laser scanning device, collect the laser point cloud data fed back by the target components, and obtain the laser point cloud data set of the target components; wherein the target components include the ball screw, the bearing, the guide block and the guide rail; The laser point cloud data set of the target component is corrected based on the wavelet transform algorithm to obtain a corrected laser point cloud data set; a characteristic state model diagram of the target component is constructed according to the corrected laser point cloud data set; Obtaining a standard state model diagram of a target component, calculating the degree of overlap between the characteristic state model diagram and the standard state model diagram; and comparing the degree of overlap with a preset threshold; If the overlap degree is not greater than a preset threshold, the target component is marked as a scrapped component, and the characteristic state model diagram of the target component is imported into the label knowledge base for matching to obtain the scrap reason of the scrapped component; An evaluation index system is constructed based on the grey correlation analysis method, and based on the evaluation index system and the scrapping reasons of the scrapped parts, a correlation analysis is performed on each assembly part in the automatic loading and unloading mechanism of the ball screw to obtain suspicious assembly parts; Listing the suspicious assembly parts as target parts, and performing laser scanning detection and evaluation on the suspicious assembly parts listed as target parts to evaluate whether the suspicious assembly parts are also scrapped parts; Among them, the laser point cloud data set of the target component is corrected based on the wavelet transform algorithm to obtain the corrected laser point cloud data set, which is specifically: Introducing a wavelet transform algorithm, and selecting a wavelet basis function as an objective function; wherein the wavelet basis function includes a Daubechies wavelet basis function, a Morlet wavelet basis function, and a Haar wavelet basis function; The scale range of the analysis is set, and discrete wavelet transform is performed on each laser point cloud data in the laser point cloud data set based on the objective function to obtain a set of wavelet coefficients; wherein each wavelet coefficient corresponds to the decomposition of the original laser point cloud data at a specific scale; The wavelet coefficient matrix of the laser point cloud data set is constructed according to the obtained wavelet coefficients; each row in the wavelet coefficient matrix represents a time point of the original laser point cloud data, each column represents a specific scale, and the size of the element value reflects the oscillation intensity of the original laser point cloud data at a specific scale; Compare each element in the wavelet coefficient matrix with a preset coefficient value; if an element in the wavelet coefficient matrix is ​​less than the preset coefficient value, mark the original laser point cloud data corresponding to the element as noise data, and filter out the original laser point cloud data marked as noise data in the laser point cloud data set; Calculate the information entropy between each element in the wavelet coefficient matrix, and compare the information entropy between each element with a preset entropy value; if the information entropy between two elements is less than the preset entropy value, mark the original laser point cloud data corresponding to the two elements as redundant data, and filter out one of the original laser point cloud data marked as redundant data in the laser point cloud data set; Repeat the above steps until the noise data and redundant data in the laser point cloud dataset are screened out to obtain a corrected laser point cloud dataset; The degree of overlap between the characteristic state model diagram and the standard state model diagram is calculated as follows: Constructing a three-dimensional grid space and equally dividing the three-dimensional grid space into a plurality of sub-grids of unit volume; inputting the characteristic state model diagram and the standard state model diagram into the three-dimensional grid space; Aligning the characteristic state model diagram with the standard state model diagram according to the reference features of the characteristic state model diagram and the standard state model diagram; Constructing a counting table, traversing each subgrid of the three-dimensional grid space, if a subgrid has both a characteristic state model diagram and a standard state model diagram, then filling 1 in the initial blank position of the counting table; If only characteristic state model diagrams or only standard state model diagrams exist in a subgrid, fill in -1 in the next blank position of the counting table; If neither the characteristic state model diagram nor the standard state model diagram exists in a subgrid, fill in 0 in the next blank position of the counting table; After the traversal is completed, the sum of all the counts of 1 in the counting table is counted to obtain the first volume value; and the sum of all the counts of -1 in the counting table is counted, and the sum of all the counts of -1 in the counting table is processed as an absolute value to obtain the second volume value; Performing ratio processing on the first volume value and the second volume value to obtain a volume ratio, and performing weighted processing on the volume ratio to obtain the degree of overlap between the characteristic state model diagram and the standard state model diagram; The characteristic state model diagram of the target component is imported into the label knowledge base for matching, and the scrapping reason of the scrapped component is obtained, which is specifically: Retrieving scrap feature model diagrams corresponding to various types of scrapped products of target components based on a big data network; Building a knowledge base, dividing the knowledge base into a plurality of sub-storage spaces, and assigning corresponding scrap type labels to each sub-storage space; The scrap feature model graphs corresponding to each type of scrapped product are respectively stored in the sub-storage space assigned with the corresponding scrap type label to obtain a label knowledge base; Acquire a characteristic state model diagram of a target component, calculate similarity values ​​between the characteristic state model diagram and the scrap characteristic model diagrams in each sub-storage space in a label knowledge base based on a cosine similarity algorithm, and obtain similarity values ​​between the characteristic state model diagram and each scrap characteristic model diagram; Acquire a scrap feature model diagram having the largest similarity value to the feature state model diagram, and extract a scrap type label of a sub-storage space to which the scrap feature model diagram having the largest similarity value to the feature state model diagram belongs; Determine the scrapping reason of the scrapped component according to the extracted scrapping type label; Among them, an evaluation index system is constructed based on the grey correlation analysis method, and based on the evaluation index system and the scrap reasons of the scrapped parts, a correlation analysis is performed on the assembly parts in the automatic loading and unloading mechanism of the ball screw, and the suspicious assembly parts are obtained by analysis, specifically: Acquire assembly engineering drawing information of the ball screw automatic loading and unloading mechanism, acquire assembly relationships between various assembly components in the ball screw automatic loading and unloading mechanism according to the assembly engineering drawing information, and acquire functional characteristic information of various assembly components; Determine the assembly relationship index between the assembly components according to the assembly relationship between the assembly components; determine the functional characteristic index of each assembly component according to the functional characteristic information of each assembly component; Based on the big data network, all possible scrapping reasons of each assembly component are obtained, and a unique code is assigned to each scrapping reason to obtain the scrapping reason code of each assembly component; Using the assembly relationship index as a comparison sequence, the functional characteristic index as a benchmark sequence, and the scrapping reason code of each assembly component as a reference sequence; Calculating the grey correlation weight values ​​between the comparison sequence, the benchmark sequence and the reference sequence based on the grey correlation analysis method; constructing an evaluation index system according to the grey correlation weight values ​​between the comparison sequence, the benchmark sequence and the reference sequence; Obtaining the scrapping reasons of the scrapped parts, importing the scrapping reasons of the scrapped parts into the evaluation index system for evaluation, and obtaining a plurality of grey correlation degrees; Among the obtained several grey correlation degrees, assembly components whose grey correlation degree is greater than a preset correlation degree threshold are screened out, and the screened assembly components are defined as suspicious assembly components.

2. The operation and maintenance method of the ball screw automatic loading and unloading mechanism according to claim 1 is characterized by: The scrap type labels include wear scrap, deformation scrap, fatigue scrap, crack scrap and corrosion scrap.

3. The operation and maintenance method of the ball screw automatic loading and unloading mechanism according to claim 1 is characterized in that: The assembly relationship index is used to reflect the connection mode, relative position and movement relationship between the assembly components, including direct connection, indirect connection, fixed connection and sliding connection; The functional characteristic indicators are used to reflect the functional role, working principle and performance parameters of each assembly component, including support, guidance, transmission, sealing and lubrication.

4. The operation and maintenance method of the ball screw automatic loading and unloading mechanism according to claim 1 is characterized by: All possible failure causes of the assembly parts include wear, fatigue, breakage, corrosion and deformation.

5. An operation and maintenance system for a ball screw automatic loading and unloading mechanism, characterized in that: The operation and maintenance system includes a memory and a processor, wherein the memory stores an operation and maintenance method program of the ball screw automatic loading and unloading mechanism. When the operation and maintenance method program of the ball screw automatic loading and unloading mechanism is executed by the processor, the operation and maintenance method steps of the ball screw automatic loading and unloading mechanism as described in any one of claims 1 to 4 are implemented.

6. A computer-readable storage medium, characterized in that: The computer-readable storage medium includes an operation and maintenance method program for an automatic loading and unloading mechanism of a ball screw. When the operation and maintenance method program for an automatic loading and unloading mechanism of a ball screw is executed by a processor, the operation and maintenance method steps for an automatic loading and unloading mechanism of a ball screw as described in any one of claims 1 to 4 are implemented.

Images