Towel rack transverse and vertical pipe assembly regulation and control system and method based on big data

The towel rack horizontal and vertical tube assembly control system, built using big data technology, utilizes KNN and SVM classification models to automatically adjust the assembly equipment, solving the problem of assembly failures caused by operators' limited knowledge and achieving a higher assembly success rate.

CN118918635APending Publication Date: 2024-11-08JIANGXI AVONFLOW HVAC TECH CO LTD
View PDF 0 Cites 0 Cited by

Patent Information

Application Number
CN202410912288.3
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2024-07-09
Publication Date
2024-11-08

AI Technical Summary

Technical Problem

In the current process of assembling the horizontal and vertical tubes of towel racks, the limited knowledge of the operators leads to incomplete and inaccurate control of the assembly equipment, which increases the possibility of assembly failure.

Method used

A big data-based towel rack horizontal and vertical tube assembly control system is adopted. Through parameter acquisition module, feature extraction module, verification module, KNN classification model and SVM classification model, a method for controlling the assembly of towel rack horizontal and vertical tubes is constructed to realize the automatic adjustment of the assembly equipment.

Benefits of technology

It improves the comprehensiveness and accuracy of assembly equipment control, reduces the risk of assembly failure, and ensures the success rate of assembling the horizontal and vertical tubes of the towel rack.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN118918635A_ABST
    Figure CN118918635A_ABST
Patent Text Reader

Abstract

The invention discloses a big data-based towel rack transverse and vertical pipe assembly regulation and control system and method, and relates to the field of assembly regulation and control systems.The method provides a data basis for subsequent construction of a final KNN classification model and a final SVM classification model by collecting multiple groups of non-operational parameter data and action image data during assembly of towel rack transverse and vertical pipes; the method comprises the following steps: constructing a final KNN classification model and a final SVM classification model; classifying the parameters of the to-be-adjusted assembly equipment, and judging whether the parameters need to be adjusted or not; the operational parameters and the non-operational parameters are adjusted respectively, and the obtained final towel rack transverse and vertical pipe assembling equipment can well complete assembling operation on the towel rack transverse and vertical pipes; the assembling equipment is optimized and adjusted from the two aspects of parameters of the assembling equipment and action parameters in the operation process, so that the assembling equipment is regulated and controlled more comprehensively and more accurately.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present invention belongs to the field of assembly control systems, and in particular, relates to a towel rack horizontal and vertical pipe assembly control system and method based on big data. Background Art

[0002] Currently, the assembly process of the horizontal and vertical tubes of the towel rack is usually regulated by the assembly equipment, and manual adjustments are required by operators. Since the operators' understanding of the assembly equipment is mostly limited and one-sided, they are unable to comprehensively and accurately regulate the assembly equipment, which greatly increases the possibility of final assembly failure. Summary of the invention

[0003] In view of the problems in the related technology, the present invention proposes a towel rack horizontal and vertical pipe assembly control system and method based on big data to overcome the above-mentioned technical problems existing in the existing related technology.

[0004] In order to solve the above technical problems, the present invention is achieved through the following technical solutions:

[0005] The present invention is a method for assembling and controlling horizontal and vertical pipes of a towel rack based on big data, comprising the following steps:

[0006] S1, the parameter collection module collects non-operational parameter data and action image data when assembling multiple groups of horizontal and vertical pipes of the towel rack;

[0007] The feature extraction module extracts features from the action image data to obtain a device operation image feature data matrix;

[0008] The verification module performs a legality verification on the non-operational parameter data;

[0009] S2, the first classification model construction module uses the verified non-operational parameter data to construct the final KNN classification model;

[0010] The second classification model building module uses the device operation image feature data matrix to build the final SVM classification model;

[0011] S3, the assembly equipment to be adjusted data acquisition module collects the parameters of the assembly equipment to be adjusted to obtain an image data set of the assembly equipment to be adjusted and a parameter data set of the assembly equipment to be adjusted;

[0012] The primary adjustment module iteratively adjusts the non-operational parameters of the assembly equipment to be adjusted according to the classification result of the assembly equipment parameter data set to be adjusted by the final KNN classification model, and obtains the first towel rack horizontal and vertical tube assembly equipment;

[0013] The secondary adjustment module iteratively adjusts the operational parameters of the assembly equipment to be adjusted according to the classification results of the features of the assembly equipment image data set to be adjusted by the final SVM classification model, and obtains the final towel rack horizontal and vertical tube assembly equipment;

[0014] When assembling the horizontal and vertical tubes of the towel rack automatically, assembly equipment is needed to assemble the horizontal and vertical tubes. Because the assembly equipment needs to adjust some parameters before and during the assembly of the horizontal and vertical tubes to ensure that the assembly is successful, based on the above reasons, this solution collects multiple sets of non-operational parameter data and action image data when assembling the horizontal and vertical tubes of the towel rack to find out the data connection between the non-operational parameter data and whether the assembly result is successful, as well as the data connection between the feature data of various actions contained in the action image data and whether the assembly result is successful. This provides a data basis for the subsequent construction of the final KNN classification model and the final SVM classification model. Feature extraction is performed on the action image data to obtain feature data of various actions, including the tilt of the robot arm. Angle, displacement length, etc., based on these feature data, the SVM classification model can be easily trained; by verifying the legality of non-operational parameter data, the quality and accuracy of the non-operational parameter data obtained are improved; by constructing the final KNN classification model and the final SVM classification model, an adjustment basis is provided for the subsequent primary and secondary adjustments of the parameters of the assembly equipment to be adjusted; by collecting the parameters of the assembly equipment to be adjusted, these parameters can be classified to determine whether these parameters need to be adjusted; by adjusting the operational parameters and non-operational parameters separately, the final towel rack horizontal and vertical tube assembly equipment obtained can better complete the assembly operation of the towel rack horizontal and vertical tubes, so that the purpose of regulating the towel rack assembly process is completed.

[0015] Preferably, the S1 comprises the following steps:

[0016] S11. Set the towel rack horizontal and vertical pipe parameter type, assembly equipment parameter type and non-operational assembly fault type to obtain the horizontal and vertical pipe parameter type set. Assembling device parameter type sets First set of fault types and assembly success type; a 1i 、a 2i and a 3i They respectively represent the set i-th horizontal and vertical pipe parameter type, the i-th assembly equipment parameter type, and the i-th assembly fault type; a1′, a′2, and a3′ respectively represent the total number of set horizontal and vertical pipe parameter types, the total number of set assembly equipment parameter types, and the total number of set assembly fault types;

[0017] According to the horizontal and vertical pipe parameter type set Assembling device parameter type sets First set of fault types The horizontal and vertical tube parameter data, assembly equipment parameter data and assembly failure data of the horizontal and vertical tubes of the towel racks at multiple historical moments are collected, and the horizontal and vertical tube parameter data matrix a1′, the assembly equipment parameter data matrix a′2 and the first assembly result data set are obtained respectively. a3′ i represents the non-operational assembly result data of the assembled towel rack horizontal and vertical tubes at the i-th historical moment, Indicates the total number of groups of data collected at the historical moment of assembling the horizontal and vertical pipes of the towel rack; a1′ and a′2 are as follows:

[0018]

[0019] Among them, a1′ ij1 and a1′ ij2 Respectively represent the parameter data of the jth type of the horizontal pipe and the jth type of the vertical pipe of the assembled horizontal and vertical pipes of the towel rack at the i-th historical moment; a′ 2ij Represents parameter data of the jth type of assembly equipment for assembling horizontal and vertical pipes of towel racks at the i-th historical moment;

[0020] The non-operational assembly failure type refers to a failure type that is not caused by improper operation;

[0021] The horizontal and vertical tube parameter type set includes the size parameters, shape parameters and material parameters of the horizontal and vertical tubes; the assembly equipment parameter type set includes assembly size, position accuracy, power supply voltage and power required by the equipment, length, width, height and weight of the equipment.

[0022] S12, setting the second assembly fault type set And the action set a 4i represents the set i-th operational assembly fault type, a′4 represents the total number of set operational assembly fault types; represents the i-th operation action set, and d represents the total number of operation actions set;

[0023] According to the second assembly fault type set Assembly success type and operation action set collection The image data of the equipment during the assembly operation of the horizontal and vertical pipes of the towel rack at the historical moment and the corresponding operational assembly fault type data are obtained to obtain the equipment operation image data matrix And the second assembly result dataset represents the operational assembly result data of the assembled horizontal and vertical pipes of the towel rack at the i-th historical moment; as follows,

[0024]

[0025] in, Image data representing the j-th operation action of the device during the assembly operation of assembling the horizontal and vertical tubes of the towel rack at the i-th historical moment;

[0026] S13. Set operation action set The operational feature type of each operation action in is obtained to obtain the operational feature type matrix d′; as follows,

[0027]

[0028] Among them, d j ' k Indicates the kth operational feature type of the jth operation action set, Indicates the total number of operational feature types set for the jth operation action;

[0029] S14, according to the operational feature type matrix d', the device operation image data matrix Extract features from each image in the image to obtain a device operation image feature data matrix;

[0030] S15, setting the validity verification rules corresponding to each horizontal and vertical pipe parameter type in the horizontal and vertical pipe parameter type set and each assembly equipment parameter type in the assembly equipment parameter type set to obtain a first rule set and a second rule set;

[0031] Using the first rule set and the second rule set to respectively perform a legality check on each horizontal and vertical tube parameter data in the horizontal and vertical tube parameter data matrix and each assembly equipment parameter data in the assembly equipment parameter data matrix, to obtain a verified horizontal and vertical tube parameter data matrix and a verified assembly equipment parameter data matrix;

[0032] By collecting the non-operational parameters of the assembly equipment through the first assembly fault type set and the assembly success type, the diversity and richness of the non-operational parameter data are improved, so that the final KNN classification model obtained by training and testing the initial KNN classification model through these data is more robust; then, the operational parameters of the assembly equipment are collected through the second assembly fault type set and the assembly success type, the diversity and richness of the operational parameter data are improved, so that the final SVM classification model obtained by training and testing the initial SVM classification model through these data is more robust; secondly, by decomposing the assembly equipment into actions, since each action has different feature types, separate analysis helps to improve the accuracy of subsequent model construction.

[0033] Preferably, S2 comprises the following steps:

[0034] S21, constructing an initial KNN classification model, setting the number of nearest neighbors considered by the initial KNN classification model when determining the category of a data point to b; setting a value range of b; and evenly dividing the value range of b to obtain a value set of the number of nearest neighbors;

[0035] S22, traversing the value set of the number of nearest neighbors, and sequentially substituting them into the initial KNN classification model; then using the verified horizontal and vertical tube parameter data matrix, the verified assembly equipment parameter data matrix and the first assembly result data set to train and test the initial KNN classification model with different b values, after the training and testing are completed, the optimal number of nearest neighbors and the final KNN classification model are obtained;

[0036] S23, constructing an initial SVM classification model; using the device operation image feature data matrix e and the second assembly result data set to train and test the initial SVM classification model, and after the training and testing are completed, obtaining the final SVM classification model;

[0037] The initial KNN classification model is trained and tested by using the verified horizontal and vertical tube parameter data matrix, the verified assembly equipment parameter data matrix, and the first assembly result data set, so that the KNN classification model has a good classification ability for the horizontal and vertical tube parameter data and the assembly equipment parameter data, providing an accurate classification model for the subsequent one-time adjustment; the initial SVM classification model is then trained and tested by using the equipment operation image feature data matrix and the second assembly result data set, so that the SVM classification model has a good classification ability for the image feature data, providing an accurate classification model for the subsequent secondary adjustment.

[0038] Preferably, the S22 comprises the following steps:

[0039] S221, normalizing the verified horizontal and vertical pipe parameter data matrix and the verified assembly equipment parameter data matrix to obtain a processed horizontal and vertical pipe parameter data matrix and a processed assembly equipment parameter data matrix;

[0040] S222, traverse the value set of the number of nearest neighbors, substitute the first value of the value set of the number of nearest neighbors into the initial KNN classification model, and obtain the first initial KNN classification model; set the first value of the value set of the number of nearest neighbors as the current optimal number of nearest neighbors; set the first training data ratio; divide the processed horizontal and vertical tube parameter data matrix, the processed assembly equipment parameter data matrix, and the first assembly result data set according to the first training data ratio to obtain the horizontal and vertical tube parameter training data matrix, the horizontal and vertical tube parameter test data matrix, the assembly equipment parameter training data matrix, the assembly equipment parameter test data matrix, the first assembly result training data set, and the first assembly result test data set;

[0041] S223, setting a first training error threshold; inputting the horizontal and vertical tube parameter training data matrix, the assembly equipment parameter training data matrix and the first assembly result training data set into the first initial KNN classification model for training operation; during the training process, when the training error is less than the first training error threshold, stopping the training to obtain the trained first initial KNN classification model;

[0042] S224, set the first test accuracy threshold c1 and the initial optimal nearest neighbor number numerical database; input the horizontal and vertical tube parameter test data matrix, the assembly equipment parameter test data matrix and the first assembly result test data set into the trained first initial KNN classification model for testing; after the test is completed, obtain the first test accuracy c1′; when c1′≥c1, store the first test accuracy c1′ in the optimal nearest neighbor number numerical database; and perform the next round of traversal, repeat S222, S223 and S234; when c1′<c1, perform the next round of traversal, repeat S222, S223 and S234;

[0043] S225, after the traversal is completed, a final optimal nearest neighbor number numerical database is obtained; the value with the largest value of the nearest neighbor number in the final optimal nearest neighbor number numerical database is selected as the optimal nearest neighbor number; the optimal nearest neighbor number is substituted into the initial KNN classification model to obtain the final KNN classification model;

[0044] By normalizing the verified horizontal and vertical pipe parameter data matrix and the verified assembly equipment parameter data matrix, the differences between the horizontal and vertical pipe parameter data and the assembly equipment parameter data are reduced; through multiple training and testing, the optimal number of nearest neighbors of the KNN classification model is obtained, so that the KNN classification model has a good classification effect for specific data such as horizontal and vertical pipe parameter data and assembly equipment parameter data.

[0045] Preferably, the S23 comprises the following steps:

[0046] S231, setting a second training data ratio; dividing the device operation image feature data matrix and the second assembly result data set according to the second training data ratio to obtain a device operation image feature training data matrix, a device operation image feature test data matrix, a second assembly result training data set, and a second assembly result test data set;

[0047] S232, setting a second training error threshold; inputting the device operation image feature training data matrix and the second assembly result training data set into the initial SVM classification model for training; during the training process, when the training error is less than the second training error threshold, stopping the training to obtain a trained SVM classification model;

[0048] S233, setting a first test accuracy threshold c2; inputting the device operation image feature test data matrix and the second assembly result test data set into the trained SVM classification model for testing; after the test is completed, obtaining a first test accuracy c′2; when c′2≥c2, using the trained SVM classification model as the final SVM classification model; otherwise, continuing the training operation of S232 until c′2≥c2, and obtaining the final SVM classification model;

[0049] By training and testing the initial SVM, the SVM classification model has a good classification effect on specific data such as device operation image feature data.

[0050] Preferably, S3 comprises the following steps:

[0051] S31, setting the assembly equipment of the horizontal and vertical pipes of the towel rack to be adjusted and the horizontal and vertical pipes to be assembled; collecting data on the assembly equipment of the horizontal and vertical pipes of the towel rack to be adjusted and the horizontal and vertical pipes to be assembled according to the horizontal and vertical pipe parameter type set and the assembly equipment parameter type set, and performing legality verification and normalization processing on the collected data in the manner of S15 and S221 to obtain the parameter data set of the assembly equipment to be adjusted and the parameter data set of the horizontal and vertical pipes to be assembled;

[0052] S32, collecting image data of each operation action when the horizontal and vertical tube assembly device of the towel rack to be adjusted is assembled according to the operation action set, to obtain an image data set of the assembly device to be adjusted;

[0053] S33, adjusting the towel rack horizontal and vertical tube assembly device to be adjusted once according to the assembly device parameter data set to be adjusted and the horizontal and vertical tube parameter data set to be assembled, to obtain a first towel rack horizontal and vertical tube assembly device;

[0054] S34, performing secondary adjustment on the first towel rack horizontal and vertical tube assembly device according to the assembly device to be adjusted image data set to obtain a final towel rack horizontal and vertical tube assembly device;

[0055] By collecting non-operational parameter data and action image data of the towel rack horizontal and vertical tube assembly equipment to be adjusted, the non-operational parameter data of the towel rack horizontal and vertical tube assembly equipment to be adjusted is verified for legality and normalized, so that the collected data quality is higher and the subsequent parameter adjustment of the towel rack horizontal and vertical tube assembly equipment to be adjusted is more accurate; by adjusting the non-operational parameter data and action image data of the assembly equipment separately, the adjustment is more targeted and the adjustment efficiency is higher.

[0056] Preferably, the S33 comprises the following steps:

[0057] S331, setting a first repetition test number g1; repeating g1 times to input the parameter data set of the assembly equipment to be adjusted and the parameter data set of the horizontal and vertical pipes to be assembled into the final KNN classification model to obtain a first non-operational result data set of the assembly equipment to be adjusted;

[0058] S332, setting a first success probability threshold g1′; calculating the probability of occurrence of the result data of the success type in the non-operational result data set of the first assembly device to be adjusted, and obtaining the first success probability g′2; when g′2≥g1′, there is no need to adjust the towel rack horizontal and vertical tube assembly device to be adjusted once, and the towel rack horizontal and vertical tube assembly device to be adjusted is used as the first towel rack horizontal and vertical tube assembly device; otherwise, adjusting the parameter data of the towel rack horizontal and vertical tube assembly device to be adjusted once; after the adjustment is completed once, the first towel rack horizontal and vertical tube assembly device is obtained; specifically comprising the following steps:

[0059] S3321, construct a first sparrow population; set the maximum number of iterations of the first sparrow population to And the current number of iterations is They are respectively recorded as the first maximum number of iterations and the first current number of iterations;

[0060] S3322, setting the initial position of each sparrow in the first sparrow population according to the assembly equipment parameter data set to be adjusted, to obtain a first initial position matrix;

[0061] S3323, constructing a fitness function of the first sparrow population according to the first success probability g′2 and the first success probability threshold g1′;

[0062] S3324, start iterative operation; in each round of iteration, calculate the fitness of each sparrow according to the fitness function g~1 of the first sparrow population, obtain the best fitness and the corresponding best individual position; update the position of each sparrow according to the best fitness and the corresponding best individual position; when When , the iteration is stopped to obtain a first final position matrix; the fitness value of each position in the first final position matrix is ​​calculated to obtain a first final fitness value set; the final position corresponding to the largest fitness value in the first final fitness value set is selected as the first optimal position;

[0063] S3325, repeat g1 times to input the first optimal position and the parameter data set of the horizontal and vertical tubes to be assembled into the final KNN classification model to obtain the second non-operational result data set of the assembly equipment to be adjusted; calculate the probability of the occurrence of the result data of the success type in the non-operational result data set of the second assembly equipment to be adjusted to obtain the second success probability g3′; when g3′≥g1′, substitute each data in the first optimal position into the towel rack horizontal and vertical tube assembly equipment to be adjusted to obtain the first towel rack horizontal and vertical tube assembly equipment; otherwise, continue the iterative operation of S3314 until g3′≥g1′, and obtain the first towel rack horizontal and vertical tube assembly equipment;

[0064] The sparrow optimization algorithm has the characteristics of fast convergence and will not fall into local optimality. Therefore, the sparrow optimization algorithm is used to simultaneously adjust multiple non-operational data of the assembly equipment, and the final KNN classification model is used to perform real-time classification on the adjusted data in each iteration. The probability of successful occurrence is used as the fitness function based on the results of multiple real-time classifications; this makes it possible to more efficiently adjust to a set of optimal non-operational data, so that the assembly equipment can ensure that the assembly results are successful.

[0065] Preferably, the S34 comprises the following steps:

[0066] S341, performing feature extraction on each image data in the image data set of the assembly equipment to be adjusted in the manner of S14, to obtain a feature data matrix of the operation image of the assembly equipment to be adjusted;

[0067] S342, setting a second repetition test number g2; repeating g2 times to input the operation image feature data matrix of the assembly equipment to be adjusted into the final SVM classification model to obtain the first assembly equipment to be adjusted operability result data set;

[0068] S343, setting a second success probability threshold g′4; calculating the probability of occurrence of the result data of the success type in the operational result data set of the first assembly device to be adjusted, and obtaining a third success probability g5′; when g5′≥g′4, there is no need to perform a secondary adjustment on the first towel rack horizontal and vertical tube assembly device, and obtain the final towel rack horizontal and vertical tube assembly device; otherwise, perform a secondary adjustment on the first towel rack horizontal and vertical tube assembly device, and after the secondary adjustment is completed, obtain the final towel rack horizontal and vertical tube assembly device;

[0069] By adopting the final SVM model to classify the collected image data of the operating actions of the assembly equipment to be adjusted, it is possible to determine whether the various feature data of the current operating actions can make the final assembly result successful; if not, the action feature data is adjusted to further ensure that the assembly equipment can be assembled successfully.

[0070] Preferably, in S343, performing a secondary adjustment on the first towel rack horizontal and vertical tube assembly device, and after the secondary adjustment is completed, obtaining the final towel rack horizontal and vertical tube assembly device includes the following steps:

[0071] S3431, construct a second sparrow population; set the maximum number of iterations of the second sparrow population to And the current number of iterations is Denoted as the second maximum number of iterations and the second current number of iterations respectively;

[0072] S3432, setting the initial position of each sparrow in the second sparrow population according to the assembly equipment operation image feature data matrix h′ to obtain a second initial position matrix set;

[0073] S3433, constructing a fitness function of the second sparrow population according to the second success probability threshold g′4 and the third success probability g5′;

[0074] S3434, start iterative operation; in each round of iteration, calculate the fitness of each sparrow according to the fitness function of the second sparrow population, obtain the best fitness and the corresponding best individual position; update the position of each sparrow according to the best fitness and the corresponding best individual position; when When , the iteration is stopped to obtain a second final position matrix set; the fitness value of each position in the second final position matrix set is calculated to obtain a second final fitness value set; the final position corresponding to the maximum fitness value in the second final fitness value set is selected as the second optimal position;

[0075] S3435, repeat g2 times to input the second optimal position into the final SVM classification model to obtain the second assembly equipment to be adjusted operational result data set; calculate the probability of the occurrence of the result data of the success type in the second assembly equipment to be adjusted operational result data set to obtain the fourth success probability g′6; when g′6≥g′4, substitute each data in the second optimal position into the first towel rack horizontal and vertical tube assembly equipment to obtain the final towel rack horizontal and vertical tube assembly equipment; otherwise, continue the iterative operation of S3434 until g′6≥g′4 to obtain the final towel rack horizontal and vertical tube assembly equipment;

[0076] The sparrow optimization algorithm is used to simultaneously adjust the feature data of multiple operating actions of the assembly equipment, and the adjusted data in each iteration is classified in real time through the final SVM classification model. The probability of successful occurrence is used as the fitness function according to the results of multiple real-time classifications; this makes it possible to more efficiently adjust to a set of optimal operational data, so that the assembly equipment can further ensure that the assembly result can be successful.

[0077] A towel rack horizontal and vertical pipe assembly control system based on big data includes a parameter acquisition module, a feature extraction module, a verification module, a first classification model construction module, a second classification model construction module, a to-be-adjusted assembly equipment data acquisition module, a primary adjustment module and a secondary adjustment module.

[0078] The present invention has the following beneficial effects:

[0079] 1. In the present invention, by collecting multiple groups of non-operational parameter data and action image data when assembling the horizontal and vertical tubes of the towel rack, a data basis is provided for the subsequent construction of the final KNN classification model and the final SVM classification model; by constructing the final KNN classification model and the final SVM classification model; the parameters of the assembly equipment to be adjusted are classified to determine whether these parameters need to be adjusted; by adjusting the operational parameters and non-operational parameters respectively, the final towel rack horizontal and vertical tube assembly equipment obtained can better complete the assembly operation of the towel rack horizontal and vertical tubes; the assembly equipment is optimized and adjusted from two aspects of the assembly equipment's own parameters and the action parameters during the operation process, so that the regulation of the assembly equipment is more comprehensive and more accurate.

[0080] 2. In the present invention, through multiple training and testing, the KNN classification model has a good classification effect on non-operational data, and then through training and testing the initial SVM, the SVM classification model has a good classification effect on the device operation image feature data.

[0081] 3. In the present invention, the multiple non-operational data and operations of the assembly equipment are adjusted simultaneously by adopting the sparrow optimization algorithm, and the adjusted data in each round of iteration is classified in real time by the final KNN classification model and the final SVM classification model, and the probability of successful occurrence is used as the fitness function according to the results of multiple real-time classifications; so that a set of optimal non-operational data and operational data can be adjusted more efficiently, so that the assembly equipment can ensure that the assembly result can be successful.

[0082] Of course, any product implementing the present invention does not necessarily need to achieve all of the advantages described above at the same time. BRIEF DESCRIPTION OF THE DRAWINGS

[0083] In order to more clearly illustrate the technical solutions of the embodiments of the invention, the drawings required for describing the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the invention. For ordinary technicians in this field, they can also obtain drawings based on these drawings without paying creative work.

[0084] Figure 1 The present invention is a schematic diagram of a process of regulating a towel rack horizontal and vertical tube assembly device by a towel rack horizontal and vertical tube assembly control system based on big data. DETAILED DESCRIPTION

[0085] The following will be combined with the drawings in the embodiments of the invention to clearly and completely describe the technical solutions in the embodiments of the invention. Obviously, the described embodiments are only part of the embodiments of the invention, not all of the embodiments. Based on the embodiments in the invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the invention.

[0086] In the description of the present invention, it is necessary to understand that the terms "opening", "upper", "lower", "top", "middle", "inside" and the like indicating orientation or positional relationship are only for the convenience of describing the invention and simplifying the description, and do not indicate or imply that the components or elements referred to must have a specific orientation, be constructed and operate in a specific orientation, and therefore cannot be understood as limiting the invention.

[0087] Embodiment 1

[0088] This embodiment is a method for assembling and controlling horizontal and vertical pipes of a towel rack based on big data, comprising the following steps:

[0089] S1, the parameter collection module collects non-operational parameter data and action image data when assembling multiple groups of horizontal and vertical pipes of the towel rack;

[0090] The feature extraction module extracts features from the action image data to obtain a device operation image feature data matrix;

[0091] The verification module performs a legality verification on the non-operational parameter data;

[0092] The S1 comprises the following steps:

[0093] S11. Set the towel rack horizontal and vertical pipe parameter type, assembly equipment parameter type and non-operational assembly fault type to obtain the horizontal and vertical pipe parameter type set. Assembling device parameter type sets First set of fault types and assembly success type; a 1i 、a 2i and a 3i They respectively represent the set i-th horizontal and vertical pipe parameter type, the i-th assembly equipment parameter type, and the i-th assembly fault type; a1′, a′2, and a3′ respectively represent the total number of set horizontal and vertical pipe parameter types, the total number of set assembly equipment parameter types, and the total number of set assembly fault types;

[0094] According to the horizontal and vertical pipe parameter type set Assembling device parameter type sets First set of fault types The horizontal and vertical tube parameter data, assembly equipment parameter data and assembly failure data of the horizontal and vertical tubes of the towel racks at multiple historical moments are collected, and the horizontal and vertical tube parameter data matrix a1′, the assembly equipment parameter data matrix a′2 and the first assembly result data set are obtained respectively. a3′ i represents the non-operational assembly result data of the assembled towel rack horizontal and vertical tubes at the i-th historical moment, Indicates the total number of groups of data collected at the historical moment of assembling the horizontal and vertical pipes of the towel rack; a1′ and a′2 are as follows:

[0095]

[0096] Among them, a1′ ij1 and a1′ ij2 Respectively represent the parameter data of the jth type of the horizontal pipe and the jth type of the vertical pipe of the assembled horizontal and vertical pipes of the towel rack at the i-th historical moment; a′ 2ij Represents parameter data of the jth type of assembly equipment for assembling horizontal and vertical pipes of towel racks at the i-th historical moment;

[0097] S12, setting the second assembly fault type set And the action set a 4i represents the set i-th operational assembly fault type, a′4 represents the total number of set operational assembly fault types; represents the i-th operation action set, and d represents the total number of operation actions set;

[0098] According to the second assembly fault type set Assembly success type and operation action set collection The image data of the equipment during the assembly operation of the horizontal and vertical pipes of the towel rack at the historical moment and the corresponding operational assembly fault type data are obtained to obtain the equipment operation image data matrix And the second assembly result dataset represents the operational assembly result data of the assembled horizontal and vertical pipes of the towel rack at the i-th historical moment; as follows,

[0099]

[0100] in, Image data representing the j-th operation action of the device during the assembly operation of assembling the horizontal and vertical tubes of the towel rack at the i-th historical moment;

[0101] S13. Set operation action set The operational feature type of each operation action in is obtained to obtain the operational feature type matrix d′; as follows,

[0102]

[0103] Among them, d j ' k Indicates the kth operational feature type of the jth operation action set, Indicates the total number of operational feature types set for the jth operation action;

[0104] S14, according to the operational feature type matrix d', the device operation image data matrix Extract features from each image in the image to obtain the device operation image feature data matrix e; as follows,

[0105]

[0106] Among them, e ij A feature dataset representing an image of the jth operation action of the device during the assembly operation of assembling the horizontal and vertical tubes of the towel rack at the i-th historical moment; e ijk Indicates e ij Feature data of the kth operational feature type in ;

[0107] S15. Set the horizontal and vertical pipe parameter type set Each horizontal and vertical pipe parameter type and assembly equipment parameter type set The first rule set is obtained by checking the legality of each assembly device parameter type in and the second rule set and They respectively represent the legality verification rules corresponding to the i-th horizontal and vertical pipe parameter types and the legality verification rules corresponding to the i-th assembly equipment parameter types;

[0108] Using the first rule set and the second rule set The validity of each horizontal and vertical pipe parameter data in the horizontal and vertical pipe parameter data matrix a1′ and each assembly equipment parameter data in the assembly equipment parameter data matrix a′2 is checked respectively to obtain the checked horizontal and vertical pipe parameter data matrix a′4 and the checked assembly equipment parameter data matrix a5′; respectively as follows,

[0109]

[0110] Among them, a′ 4ij1 and a′ 4ij2 a5′ respectively represents the parameter data of the jth type of the horizontal pipe and the parameter data of the jth type of the vertical pipe of the assembled horizontal and vertical pipes of the towel rack at the i-th historical moment after the validity check; ij Represents parameter data of the jth type of assembly equipment for assembling horizontal and vertical pipes of towel racks at the i-th historical moment after legality verification;

[0111] S2, the first classification model construction module uses the verified non-operational parameter data to construct the final KNN classification model;

[0112] The second classification model building module uses the device operation image feature data matrix to build the final SVM classification model;

[0113] The S2 comprises the following steps:

[0114] S21, construct an initial KNN classification model, set the number of nearest neighbors considered by the initial KNN classification model when determining the category of a data point to b; set the value range of b to [b min ,b max ], b min and b max They represent the lower and upper limits of b respectively;

[0115] The value range of b is evenly divided to obtain the value set of the number of nearest neighbors b i It represents the value of the ith b divided equally from the value range of b. Indicates the total number of values ​​of b equally divided from the value range of b;

[0116] S22, taking a value set of the number of nearest neighbors Traverse and substitute the initial KNN classification model in turn; then use the verified horizontal and vertical pipe parameter data matrix a′4, the verified assembly equipment parameter data matrix a5′ and the first assembly result data set The initial KNN classification model with different b values ​​is trained and tested. After the training and testing are completed, the optimal number of nearest neighbors is obtained. And the final KNN classification model;

[0117] The S22 comprises the following steps:

[0118] S221, normalize the verified horizontal and vertical pipe parameter data matrix a′4 and the verified assembly equipment parameter data matrix a5′ to obtain the processed horizontal and vertical pipe parameter data matrix a′6 and the processed assembly equipment parameter data matrix a′7; respectively as follows,

[0119]

[0120] Among them, a′ 6ij1 and a′ 6ij2 Respectively represent a′ 4ij1 and a′ 4ij2 The parameter data of the jth type of the horizontal tube and the jth type of the vertical tube of the assembled towel rack horizontal and vertical tubes at the i-th historical moment obtained after normalization processing; a′ 7ij Indicates a5′ ij The parameter data of the jth type of the assembly equipment for assembling the horizontal and vertical tubes of the towel rack at the i-th historical moment are obtained after normalization processing; the calculation formulas are as follows:

[0121]

[0122] S222: taking a value set of the number of nearest neighbors Traverse and substitute b1′ of the nearest neighbor number value set into the initial KNN classification model to obtain the first initial KNN classification model; set b1′ to the current optimal nearest neighbor number;

[0123] Set a first training data ratio; according to the first training data ratio, process the horizontal and vertical tube parameter data matrix a′6, the processed assembly equipment parameter data matrix a′7 and the first assembly result data set Perform data division to obtain a horizontal and vertical tube parameter training data matrix, a horizontal and vertical tube parameter test data matrix, an assembly equipment parameter training data matrix, an assembly equipment parameter test data matrix, a first assembly result training data set, and a first assembly result test data set;

[0124] S223, setting a first training error threshold; inputting the horizontal and vertical tube parameter training data matrix, the assembly equipment parameter training data matrix and the first assembly result training data set into the first initial KNN classification model for training operation; during the training process, when the training error is less than the first training error threshold, stopping the training to obtain the trained first initial KNN classification model;

[0125] S224, setting a first test accuracy threshold c1 and an initial optimal nearest neighbor number numerical database; inputting the horizontal and vertical tube parameter test data matrix, the assembly equipment parameter test data matrix and the first assembly result test data set into the trained first initial KNN classification model for testing; after the test is completed, obtaining a first test accuracy c1′;

[0126] When c1′≥c1, the first test accuracy c1′ is stored in the optimal nearest neighbor number database; and the next round of traversal is performed, and S222, S223 and S234 are repeated;

[0127] When c1′<c1, the next round of traversal is performed, and S222, S223 and S234 are repeated;

[0128] S225. After the traversal is completed, a final optimal nearest neighbor number numerical database is obtained; the value with the largest value of the number of nearest neighbors in the final optimal nearest neighbor number numerical database is selected as the optimal nearest neighbor number. The optimal number of nearest neighbors Substitute it into the initial KNN classification model to obtain the final KNN classification model;

[0129] S23, constructing an initial SVM classification model; using the device to operate the image feature data matrix e and the second assembly result data set The initial SVM classification model is trained and tested. After the training and testing are completed, the final SVM classification model is obtained;

[0130] The S23 comprises the following steps:

[0131] S231, setting a second training data ratio; and assembling the device operation image feature data matrix e and the second assembly result data set according to the second training data ratio. Perform data division to obtain a device operation image feature training data matrix, a device operation image feature test data matrix, a second assembly result training data set, and a second assembly result test data set;

[0132] S232, setting a second training error threshold; inputting the device operation image feature training data matrix and the second assembly result training data set into the initial SVM classification model for training; during the training process, when the training error is less than the second training error threshold, stopping the training to obtain a trained SVM classification model;

[0133] S233, setting a first test accuracy threshold c2; inputting the device operation image feature test data matrix and the second assembly result test data set into the trained SVM classification model for testing; after the test is completed, obtaining a first test accuracy c′2;

[0134] When c′2≥c2, the trained SVM classification model is used as the final SVM classification model; otherwise, the training operation of S232 is continued until c′2≥c2, and the final SVM classification model is obtained;

[0135] S3, the assembly equipment to be adjusted data acquisition module collects the parameters of the assembly equipment to be adjusted to obtain an image data set of the assembly equipment to be adjusted and a parameter data set of the assembly equipment to be adjusted;

[0136] The primary adjustment module iteratively adjusts the non-operational parameters of the assembly equipment to be adjusted according to the classification result of the assembly equipment parameter data set to be adjusted by the final KNN classification model, and obtains the first towel rack horizontal and vertical tube assembly equipment;

[0137] The secondary adjustment module iteratively adjusts the operational parameters of the assembly equipment to be adjusted according to the classification results of the features of the assembly equipment image data set to be adjusted by the final SVM classification model, and obtains the final towel rack horizontal and vertical tube assembly equipment;

[0138] The S3 comprises the following steps:

[0139] S31, setting the assembly equipment of the horizontal and vertical pipes of the towel rack to be adjusted and the horizontal and vertical pipes to be assembled; according to the horizontal and vertical pipe parameter type set And assemble device parameter type sets Data collection is performed on the horizontal and vertical pipe assembly equipment of the towel rack to be adjusted and the horizontal and vertical pipes to be assembled, and the collected data is verified for legality and normalized using S15 and S221 to obtain the parameter data set of the assembly equipment to be adjusted. And the parameter data set of the horizontal and vertical pipes to be assembled e1′ i e′ represents the parameter data of the i-th type of the towel rack horizontal and vertical pipe assembly device to be adjusted;2i1 and e′ 2i2 Respectively represent parameter data of the i-th type of the horizontal pipe to be assembled and parameter data of the i-th type of the vertical pipe to be assembled;

[0140] S32, according to the operation action set Collect image data of each operation action of the assembly device for the horizontal and vertical pipes of the towel rack to be adjusted during assembly, and obtain an image data set of the assembly device to be adjusted Represents the collected image data of the i-th operation action of the assembly equipment to be adjusted;

[0141] S33, assembling the equipment parameter data set to be adjusted And the parameter data set of the horizontal and vertical pipes to be assembled The towel rack horizontal and vertical tube assembly device to be adjusted is adjusted once to obtain a first towel rack horizontal and vertical tube assembly device;

[0142] The S33 comprises the following steps:

[0143] S331, set the first repetition test number g1; repeat g1 times to set the parameter data set of the assembly equipment to be adjusted And the parameter data set of the horizontal and vertical pipes to be assembled

[0144] Input into the final KNN classification model to obtain the first non-operational result data set of the assembly equipment to be adjusted Indicates the i-th time the parameter data set of the assembly equipment to be adjusted is And the parameter data set of the horizontal and vertical pipes to be assembled Non-operational result data input into the final KNN classification model;

[0145] S332, setting a first success probability threshold g1′; calculating the non-operational result data set of the first assembly equipment to be adjusted The probability of the occurrence of the result data of the success type is obtained to obtain the first success probability g′2;

[0146] When g′2≥g1′, there is no need to adjust the horizontal and vertical tube assembly device of the towel rack to be adjusted, and the horizontal and vertical tube assembly device of the towel rack to be adjusted is used as the first horizontal and vertical tube assembly device of the towel rack;

[0147] Otherwise, the parameter data of the towel rack horizontal and vertical tube assembly device to be adjusted is adjusted once; after the adjustment is completed once, a first towel rack horizontal and vertical tube assembly device is obtained; specifically, the following steps are included:

[0148] S3321, build the first sparrow population f 1irepresents the i-th sparrow in the first sparrow population, f1′ represents the size of the first sparrow population; the maximum number of iterations of the first sparrow population is set to And the current number of iterations is are respectively recorded as the first maximum number of iterations and the first current number of iterations; the search space dimension of the first sparrow population is a′2;

[0149] S3322: assembling the equipment parameter data set to be adjusted Set the initial position of each sparrow in the first sparrow population to obtain the first initial position matrix as follows,

[0150]

[0151] in, represents the component of the initial position of the i-th sparrow in the first sparrow population in the j-th dimension; the calculation formula is as follows,

[0152]

[0153] In the formula, rand 1ij represents the random number generated in the jth dimension for the initial position of the i-th sparrow in the first sparrow population;

[0154] S3323: construct a fitness function of the first sparrow population according to the first success probability g′2 and the first success probability threshold g1′ as follows,

[0155]

[0156] Among them, α is a positive number, indicating the first correction parameter;

[0157] S3324, start iterating; in each round of iteration, according to the fitness function of the first sparrow population Calculating the fitness of each sparrow to obtain the best fitness and the corresponding best individual position; updating the position of each sparrow according to the best fitness and the corresponding best individual position;

[0158] when When , the iteration is stopped to obtain the first final position matrix; the fitness value of each position in the first final position matrix is ​​calculated to obtain the first final fitness value set; the final position corresponding to the largest fitness value in the first final fitness value set is selected as the first optimal position h 1i A component representing parameter data of the i-th type of the towel rack horizontal and vertical pipe assembly device to be adjusted in the first optimal position;

[0159] S3325, repeat g1 times to set the first optimal position And the parameter data set of the horizontal and vertical pipes to be assembled Input into the final KNN classification model to obtain a second non-operational result data set of the assembly equipment to be adjusted; calculate the probability of occurrence of the result data of the success type in the second non-operational result data set of the assembly equipment to be adjusted to obtain a second success probability g3′;

[0160] When g3′≥g1′, the first optimal position Substitute each data in into the towel rack horizontal and vertical tube assembly device to be adjusted to obtain the first towel rack horizontal and vertical tube assembly device; otherwise, continue the iterative operation of S3314 until g3′≥g1′, and obtain the first towel rack horizontal and vertical tube assembly device;

[0161] S34, assembling the device image data set to be adjusted Performing secondary adjustment on the first towel rack horizontal and vertical tube assembly device to obtain the final towel rack horizontal and vertical tube assembly device;

[0162] The S34 comprises the following steps:

[0163] S341, using the method of S14 to adjust the assembly equipment image data set to be adjusted Extract features from each image data in the image to obtain the feature data matrix h′ of the assembly equipment operation image to be adjusted; as follows:

[0164]

[0165] Among them, h i ' j The j-th type of feature data representing the collected image data of the i-th operation action of the assembly equipment to be adjusted; Represents the total number of feature types of the i-th operation action image;

[0166] S342, set the second repetition test number g2; repeat g2 times to input the assembly equipment operation image feature data matrix h′ to the final SVM classification model to obtain the first assembly equipment operability result data set to be adjusted represents the operational result data obtained by inputting the feature data matrix h′ of the assembly equipment operation image to be adjusted into the final SVM classification model for the i-th time;

[0167] S343, setting a second success probability threshold g′4; calculating the first assembly equipment operability result data set to be adjusted The probability of the occurrence of the result data of the success type is obtained to obtain the third success probability g5′;

[0168] When g5′≥g′4, there is no need to perform secondary adjustment on the first towel rack horizontal and vertical tube assembly device, and the final towel rack horizontal and vertical tube assembly device is obtained;

[0169] Otherwise, the first towel rack horizontal and vertical tube assembly device is subjected to secondary adjustment, and after the secondary adjustment is completed, a final towel rack horizontal and vertical tube assembly device is obtained;

[0170] In S343, the first towel rack horizontal and vertical tube assembly device is subjected to secondary adjustment. After the secondary adjustment is completed, the final towel rack horizontal and vertical tube assembly device is obtained, which includes the following steps:

[0171] S3431, build the second sparrow population f 2i represents the i-th sparrow in the second sparrow population, f2′ represents the size of the second sparrow population; the maximum number of iterations of the second sparrow population is set to And the current number of iterations is are respectively recorded as the second maximum number of iterations and the second current number of iterations; the search space dimension of the second sparrow population is

[0172] S3432, setting the initial position of each sparrow in the second sparrow population according to the assembly equipment operation image feature data matrix h′ to obtain a second initial position matrix set represents the initial position matrix of the i-th sparrow in the second sparrow population; as follows,

[0173]

[0174] in, is a component of the kth type of feature data representing the collected image data of the jth operation action of the assembly device to be adjusted in the initial position matrix of the i-th sparrow in the second sparrow population;

[0175] S3433: construct a fitness function of the second sparrow population according to the second success probability threshold g′4 and the third success probability g5′ as follows,

[0176]

[0177] In the formula, β is a positive number, indicating the second correction parameter;

[0178] S3434, start iterative operation; in each round of iteration, calculate the fitness of each sparrow according to the fitness function g~2 of the second sparrow population, obtain the best fitness and the corresponding best individual position; update the position of each sparrow according to the best fitness and the corresponding best individual position;

[0179] when When , the iteration is stopped to obtain a second final position matrix set; the fitness value of each position in the second final position matrix set is calculated to obtain a second final fitness value set; the final position corresponding to the maximum fitness value in the second final fitness value set is selected as the second optimal position;

[0180] S3435, repeat g2 times to input the second optimal position into the final SVM classification model to obtain a second assembly equipment operability result data set to be adjusted; calculate the probability of occurrence of success type result data in the second assembly equipment operability result data set to be adjusted to obtain a fourth success probability g′6;

[0181] When g′6≥g′4, substitute each data in the second optimal position into the first towel rack horizontal and vertical tube assembly device to obtain the final towel rack horizontal and vertical tube assembly device; otherwise, continue the iterative operation of S3434 until g′6≥g′4 to obtain the final towel rack horizontal and vertical tube assembly device.

[0182] Embodiment 2

[0183] This embodiment discloses a towel rack horizontal and vertical pipe assembly control system based on big data, the system can implement the method of the above embodiment, including a parameter acquisition module, a feature extraction module, a verification module, a first classification model construction module, a second classification model construction module, a to-be-adjusted assembly equipment data acquisition module, a primary adjustment module and a secondary adjustment module;

[0184] The parameter acquisition module is used to collect multiple groups of non-operational parameter data and action image data when assembling the horizontal and vertical tubes of the towel rack, and obtain a horizontal and vertical tube parameter data matrix, an assembly equipment parameter data matrix, a first assembly result data set, an equipment operation image data matrix, and a second assembly result data set;

[0185] The feature extraction module is used to extract features from each image data in the device operation image data matrix to obtain the device operation image feature data matrix;

[0186] The verification module is used to perform a legality check on each horizontal and vertical tube parameter data in the horizontal and vertical tube parameter data matrix and each assembly equipment parameter data in the assembly equipment parameter data matrix;

[0187] The first classification model building module is used to train and test the initial KNN classification model using the verified horizontal and vertical pipe parameter data matrix, the verified assembly equipment parameter data matrix and the first assembly result data set to obtain the final KNN classification model;

[0188] The second classification model building module is used to train and test the initial SVM classification model using the device operation image feature data matrix and the second assembly result data set to obtain a final SVM classification model;

[0189] The assembly equipment to be adjusted data acquisition module is used to acquire parameters of the assembly equipment to be adjusted, and obtain an assembly equipment to be adjusted image data set and an assembly equipment to be adjusted parameter data set;

[0190] The primary adjustment module is used to iteratively adjust the non-operational parameters of the assembly equipment to be adjusted according to the classification result of the assembly equipment parameter data set to be adjusted by the final KNN classification model, so as to obtain the first towel rack horizontal and vertical tube assembly equipment;

[0191] The secondary adjustment module is used to iteratively adjust the operational parameters of the assembly equipment to be adjusted according to the classification results of the characteristics of the assembly equipment image data set to be adjusted by the final SVM classification model, so as to obtain the final towel rack horizontal and vertical tube assembly equipment.

[0192] In the description of this specification, the description with reference to the terms "one embodiment", "example", "specific example", etc. means that the specific features, structures, materials or characteristics described in conjunction with the embodiment or example are included in at least one embodiment or example of the invention. In this specification, the schematic representation of the above terms does not necessarily refer to the same embodiment or example. Moreover, the specific features, structures, materials or characteristics described can be combined in any one or more embodiments or examples in a suitable manner.

[0193] The preferred embodiments of the invention disclosed above are only used to help explain the invention. The preferred embodiments do not describe all the details in detail, nor do they limit the invention to the specific implementation methods described. Obviously, many modifications and changes can be made according to the content of this specification. This specification selects and specifically describes these embodiments in order to better explain the principles and practical applications of the invention, so that those skilled in the art can understand and use the invention well.

Claims

1. A towel rack horizontal and vertical pipe assembly control method based on big data, characterized in that: The following steps are involved: S1, the parameter collection module collects non-operational parameter data and action image data when assembling multiple groups of horizontal and vertical pipes of the towel rack; The feature extraction module extracts features from the action image data to obtain a device operation image feature data matrix; The verification module performs a legality verification on the non-operational parameter data; S2, the first classification model construction module uses the verified non-operational parameter data to construct the final KNN classification model; The second classification model building module uses the device operation image feature data matrix to build the final SVM classification model; S3, the assembly equipment to be adjusted data acquisition module collects the parameters of the assembly equipment to be adjusted to obtain an image data set of the assembly equipment to be adjusted and a parameter data set of the assembly equipment to be adjusted; The primary adjustment module iteratively adjusts the non-operational parameters of the assembly equipment to be adjusted to obtain the first towel rack horizontal and vertical tube assembly equipment; The secondary adjustment module iteratively adjusts the operational parameters of the assembly equipment to be adjusted to obtain the final towel rack horizontal and vertical tube assembly equipment.

2. The method for assembling and controlling the horizontal and vertical pipes of a towel rack based on big data according to claim 1, characterized in that: The S1 comprises the following steps: S11, collecting multiple sets of horizontal and vertical tube parameter data, assembly equipment parameter data and assembly failure data of the horizontal and vertical tubes of the towel rack assembled at historical moments, and obtaining a horizontal and vertical tube parameter data matrix, an assembly equipment parameter data matrix and a first assembly result data set respectively; S12, collecting multiple sets of image data of equipment during the assembly operation of assembling the horizontal and vertical tubes of the towel rack at historical moments and corresponding operational assembly fault type data, to obtain an equipment operation image data matrix and a second assembly result data set; S13, extracting features from each image in the device operation image data matrix to obtain a device operation image feature data matrix; S15. Perform a validity check on each horizontal and vertical tube parameter data in the horizontal and vertical tube parameter data matrix and each assembly equipment parameter data in the assembly equipment parameter data matrix to obtain a verified horizontal and vertical tube parameter data matrix and a verified assembly equipment parameter data matrix.

3. The method for assembling and controlling the horizontal and vertical pipes of a towel rack based on big data according to claim 2 is characterized in that: The S2 comprises the following steps: S21, constructing an initial KNN classification model, setting the number of nearest neighbors considered by the initial KNN classification model when determining the category of a data point to b; evenly dividing the value range of b to obtain a value set of the number of nearest neighbors; S22, training and testing the initial KNN classification model with different b values, and obtaining the optimal number of nearest neighbors and the final KNN classification model after the training and testing are completed; S23, constructing an initial SVM classification model; using the device operation image feature data matrix and the second assembly result data set to train and test the initial SVM classification model, after the training and testing are completed, obtaining the final SVM classification model.

4. The method for assembling and controlling the horizontal and vertical pipes of a towel rack based on big data according to claim 3 is characterized in that: The S22 comprises the following steps: S221, normalizing the verified horizontal and vertical pipe parameter data matrix and the verified assembly equipment parameter data matrix to obtain a processed horizontal and vertical pipe parameter data matrix and a processed assembly equipment parameter data matrix; S222, setting a first training data ratio; dividing the processed horizontal and vertical tube parameter data matrix, the processed assembly equipment parameter data matrix, and the first assembly result data set according to the first training data ratio to obtain a horizontal and vertical tube parameter training data matrix, a horizontal and vertical tube parameter test data matrix, an assembly equipment parameter training data matrix, an assembly equipment parameter test data matrix, a first assembly result training data set, and a first assembly result test data set; S223, setting a first training error threshold; inputting the horizontal and vertical pipe parameter training data matrix, the assembly equipment parameter training data matrix and the first assembly result training data set into the initial KNN classification model for training operation; during the training process, when the training error is less than the first training error threshold, stopping the training to obtain a trained initial KNN classification model; S224, set the first test accuracy threshold c1 and the initial optimal nearest neighbor number numerical database; input the horizontal and vertical tube parameter test data matrix, the assembly equipment parameter test data matrix and the first assembly result test data set into the trained initial KNN classification model for testing; after the test is completed, obtain the first test accuracy c1′; when c1′≥c1, store the first test accuracy c1′ in the optimal nearest neighbor number numerical database; and perform the next round of traversal, repeat S222, S223 and S234; when c1′<c1, perform the next round of traversal, repeat S222, S223 and S234; S225. After the traversal is completed, a final optimal nearest neighbor number numerical database is obtained; the value with the largest value of the nearest neighbor number in the final optimal nearest neighbor number numerical database is selected as the optimal nearest neighbor number; the optimal nearest neighbor number is substituted into the initial KNN classification model to obtain a final KNN classification model.

5. The method for assembling and controlling the horizontal and vertical pipes of a towel rack based on big data according to claim 4 is characterized in that: The S23 comprises the following steps: S231, setting a second training data ratio; dividing the device operation image feature data matrix and the second assembly result data set according to the second training data ratio to obtain a device operation image feature training data matrix, a device operation image feature test data matrix, a second assembly result training data set, and a second assembly result test data set; S232, setting a second training error threshold; inputting the device operation image feature training data matrix and the second assembly result training data set into the initial SVM classification model for training; during the training process, when the training error is less than the second training error threshold, stopping the training to obtain a trained SVM classification model; S233, set a first test accuracy threshold c2; input the device operation image feature test data matrix and the second assembly result test data set into the trained SVM classification model for testing; after the test is completed, obtain the first test accuracy c′2; when c′2≥c2, use the trained SVM classification model as the final SVM classification model; otherwise, continue the training operation of S232 until c′2≥c2 to obtain the final SVM classification model.

6. The method for controlling and assembling horizontal and vertical pipes of a towel rack based on big data according to claim 5, characterized in that: The S3 comprises the following steps: S31, collecting data on the assembly equipment of the horizontal and vertical pipes of the towel rack to be adjusted and the horizontal and vertical pipes to be assembled, and performing legality verification and normalization processing on the collected data in the manner of S15 and S221 to obtain parameter data sets of the assembly equipment to be adjusted and parameter data sets of the horizontal and vertical pipes to be assembled; S32, collecting image data of each operation action when the horizontal and vertical tube assembly device of the towel rack to be adjusted is assembled according to the operation action set, to obtain an image data set of the assembly device to be adjusted; S33, adjusting the towel rack horizontal and vertical tube assembly device to be adjusted once according to the assembly device parameter data set to be adjusted and the horizontal and vertical tube parameter data set to be assembled, to obtain a first towel rack horizontal and vertical tube assembly device; S34, performing secondary adjustment on the first towel rack horizontal and vertical tube assembly device according to the assembly device image data set to be adjusted, to obtain a final towel rack horizontal and vertical tube assembly device.

7. The method for controlling and assembling horizontal and vertical pipes of a towel rack based on big data according to claim 6, characterized in that: The S33 comprises the following steps: S331, setting a first repetition test number g1; repeating g1 times to input the parameter data set of the assembly equipment to be adjusted and the parameter data set of the horizontal and vertical pipes to be assembled into the final KNN classification model to obtain a first non-operational result data set of the assembly equipment to be adjusted; S332. Set a first success probability threshold g1′; calculate the probability of occurrence of success type result data in the non-operational result data set of the first assembly device to be adjusted, and obtain a first success probability g′2; when g′2≥g1′, there is no need to adjust the towel rack horizontal and vertical tube assembly device to be adjusted, and use the towel rack horizontal and vertical tube assembly device to be adjusted as the first towel rack horizontal and vertical tube assembly device; otherwise, adjust the parameter data of the towel rack horizontal and vertical tube assembly device to be adjusted; after the adjustment is completed, obtain the first towel rack horizontal and vertical tube assembly device.

8. The method for controlling and assembling horizontal and vertical pipes of a towel rack based on big data according to claim 7, characterized in that: The S34 comprises the following steps: S341, performing feature extraction on each image data in the image data set of the assembly equipment to be adjusted in the manner of S14, to obtain a feature data matrix of the operation image of the assembly equipment to be adjusted; S342, setting a second repetition test number g2; repeating g2 times to input the operation image feature data matrix of the assembly equipment to be adjusted into the final SVM classification model to obtain the first assembly equipment to be adjusted operability result data set; S343. Set a second success probability threshold g′4; calculate the probability of occurrence of the result data of the success type in the operational result data set of the first assembly equipment to be adjusted, and obtain the third success probability g5′; when g5′≥g′4, there is no need to perform a secondary adjustment on the first towel rack horizontal and vertical tube assembly equipment to obtain the final towel rack horizontal and vertical tube assembly equipment; otherwise, perform a secondary adjustment on the first towel rack horizontal and vertical tube assembly equipment, and after the secondary adjustment is completed, obtain the final towel rack horizontal and vertical tube assembly equipment.

9. The method for controlling and assembling horizontal and vertical pipes of a towel rack based on big data according to claim 8, characterized in that: In S343, the first towel rack horizontal and vertical tube assembly device is subjected to secondary adjustment. After the secondary adjustment is completed, the final towel rack horizontal and vertical tube assembly device is obtained, which includes the following steps: S3431, construct a second sparrow population; set the maximum number of iterations of the second sparrow population to And the current iteration number is Denoted as the second maximum number of iterations and the second current number of iterations respectively; S3432, setting the initial position of each sparrow in the second sparrow population according to the assembly equipment operation image feature data matrix h′ to obtain a second initial position matrix set; S3433, start iterative operation; when When , the iteration is stopped to obtain a second final position matrix set; the fitness value of each position in the second final position matrix set is calculated to obtain a second final fitness value set; the final position corresponding to the maximum fitness value in the second final fitness value set is selected as the second optimal position; S3434, repeat g2 times to input the second optimal position into the final SVM classification model to obtain the second operational result data set of the assembly equipment to be adjusted; calculate the probability of occurrence of the result data of the success type in the operational result data set of the second assembly equipment to be adjusted to obtain the fourth success probability g′6; when g′6≥g′4, substitute each data in the second optimal position into the first towel rack horizontal and vertical tube assembly equipment to obtain the final towel rack horizontal and vertical tube assembly equipment; otherwise, continue the iterative operation of S3434 until g′6≥g′4 to obtain the final towel rack horizontal and vertical tube assembly equipment.

10. A system for implementing the big data-based towel rack horizontal and vertical pipe assembly control method as described in any one of claims 1 to 9.