Quick matching system and matching method for adjusting washers based on data driving

Abstract

The invention provides an adjusting washer rapid matching system and matching method based on data driving. The matching system comprises an acquisition subsystem which measures and acquires the machining size of a machined part and the crosslinking size of an assembled part; the calculation subsystem carries out secondary processing on the collected data, and an adjusting washer matching scheme is preliminarily obtained through calculation and grouping; the analysis subsystem carries out matching in an adjustment washer matching database according to the adjustment washer matching scheme, gives out a pre-judgment conclusion of the performance indexes of the actuator, and carries out adjustment and optimization according to the pre-judgment conclusion; the recommendation subsystem lights an indicating lamp according to the specification and the number of the adjusting washers obtained by the analysis subsystem so as to remind a person to select the adjusting washers according to a calculation result; the diagnosis subsystem is used for monitoring the test verification process of the actuator adjusting washers and determining a pre-judgment conclusion whether the grouping specification and the number of the adjusting washers conform to each other or not according to a test verification result; and updating an adjustment washer matching database in the analysis subsystem.

Description

Technical Field

[0001] This invention relates to, but is not limited to, the field of assembly and debugging technology, specifically to a data-driven rapid selection system and method for adjusting washers. Background Technology

[0002] The actuator series products are mainly used in conjunction with the high-lift systems of aircraft. Their function is to meet the transmission requirements of different parts of the aircraft. Key performance indicators include transmission efficiency, static friction, and product clearance. Because the actuator products have similar structures and functions, different sizes of adjusting washers need to be added to key parts of the products to meet the varying performance requirements of different aircraft.

[0003] Currently, manual selection of adjusting washers is based on operational experience, followed by product testing for physical verification. This lacks theoretical basis, relies excessively on manual experience, involves repeated disassembly and assembly processes, and results in inconsistent verification results. Summary of the Invention

[0004] The purpose of this invention is to provide a data-driven rapid selection system and method for adjusting washers, in order to solve the problems of excessive reliance on manual experience, lack of theoretical basis, repeated random disassembly and assembly processes, and poor consistency of test results in the existing actuator assembly process, where adjusting washers are selected manually based on operational experience.

[0005] The technical solution of the present invention: The present invention provides a data-driven rapid selection system for adjustment washers, comprising: a data acquisition subsystem 1, a calculation subsystem 2, an analysis subsystem 3, a recommendation subsystem 4, and a diagnostic subsystem 5 connected in sequence;

[0006] The acquisition subsystem 1 is used to measure and acquire the machining dimensions of machined parts and the cross-linking dimensions of assembled parts using digital measuring tools, and then upload them to the computing subsystem 2.

[0007] The calculation subsystem 2 is used to perform secondary processing on the data collected by the acquisition subsystem 1, and to perform calculations and grouping according to the given size chain and grouping logic to initially derive the adjustment washer selection scheme. Each selection scheme includes the grouping specifications and quantity of adjustment washers.

[0008] The analysis subsystem 3 is used to call the existing actuator adjustment washer selection database within the analysis subsystem 3, match the adjustment washer selection scheme obtained by the calculation subsystem 2 in the adjustment washer selection database, and give a preliminary conclusion on the actuator performance index. Based on the preliminary conclusion on the actuator performance index, the adjustment and optimization are performed, and the optimized grouping specifications and quantity of adjustment washers are obtained through analysis.

[0009] The recommendation subsystem 4 includes a washer lighting device, which has multiple washer storage boxes and an indicator light on each washer storage box. Each washer storage box is used to hold an adjustment washer of a certain specification. The recommendation subsystem 4 is used to light up the indicator light according to the specification and quantity of the adjustment washer obtained from the analysis subsystem 3 to remind the personnel to select the adjustment washer according to the calculation results.

[0010] The diagnostic subsystem 5 is used to monitor the testing and verification process of the actuator adjusting washers, determine whether the grouping specifications and quantity of the adjusting washers meet the prediction conclusion of the analysis subsystem 3 based on the testing and verification results, and update the adjusting washer selection database in the analysis subsystem 3.

[0011] Optionally, in the data-driven quick-fit system for adjusting washers described above,

[0012] The machining dimensions of the machined parts include: part height X1, part width X2, part length X3, drive shaft length X4, and part hole inner diameter X5; the cross-linking dimensions of the assembled parts include: component height X6, stepped hole depth X7, and bearing runout X8.

[0013] The data acquisition subsystem 1 includes: a height measuring instrument, a digital caliper, and a digital micrometer;

[0014] The height measuring instrument is used to collect the part height X1, part thickness X2, part length X3 of the machined part, and the component height X6 of the assembled part.

[0015] The digital caliper is used to collect the transmission shaft length X4 and the stepped hole depth X7 of the machined parts;

[0016] The digital micrometer is used to collect the inner diameter of the hole of the machined part (X5) and the bearing runout of the assembled part (X8).

[0017] Optionally, in the data-driven rapid selection system for adjusting washers described above, the calculation subsystem 2 performs calculations and grouping according to a given size chain and grouping logic as follows:

[0018] The dimension chain is:

[0019] Calculate the thickness of adjusting washer 1: Y1 = (X5 - X3 - X6 - X7) ÷ 2;

[0020] Calculate the thickness of adjusting washer 2: Y2=(X1+X2-X4-X8)÷2;

[0021] The grouping logic is as follows:

[0022] Based on the calculated thickness of the adjusting shims, select adjusting shims of the corresponding thickness and quantity. The thickness difference between the adjusting shims used in the same position should not exceed 0.2mm, and the number of adjusting shims should not exceed 5.

[0023] Optionally, in the data-driven quick-fit system for adjusting washers described above, the actuator performance indicators include transmission efficiency, static friction, and product clearance (i.e., axial runout clearance of the product); the analysis subsystem 3 derives the following predictive conclusions regarding the actuator performance indicators:

[0024] a) Based on the actuator performance index selection model library, the mapping relationship between Y1 and transmission efficiency K is obtained;

[0025] The actuator transmission efficiency K is negatively correlated with Y1. The transmission efficiency K = -2Y1 + B1, where -2 is the coefficient of the transmission efficiency mapping model and B1 is the bias term of the model.

[0026] b) Based on the actuator performance index matching model library, the mapping relationship between Y2 and static friction force F is obtained;

[0027] The static friction force F of the actuator is positively correlated with Y2. The static friction force F = 2.6Y2 + B2, where 2.6 is the coefficient of the static friction force mapping model and B2 is the bias term of the model.

[0028] c) Based on the actuator performance index matching model library, obtain the mapping relationship between Y1, Y2 and product clearance C;

[0029] The product clearance C is nonlinearly related to Y1 and Y2, and the product clearance C = f(Y1, Y2).

[0030] Optionally, in the data-driven quick-fit system for adjusting washers described above,

[0031] The recommendation subsystem 4 also includes: a mis-picking sensor identification device installed in each gasket storage box, used to provide a warning when the specifications and quantity of the originally selected adjustment gasket do not match those calculated by the calculation subsystem 3.

[0032] Optionally, the data-driven quick matching system for adjustment washers described above further includes a database subsystem 6 connected to the diagnostic subsystem 5 and the analysis subsystem 3, respectively.

[0033] The database subsystem 6 is used to identify the mapping relationship between the collected parameters and the actuator performance indicators using mathematical statistical analysis methods before executing the actuator adjustment washer selection scheme, and to construct an actuator adjustment washer selection database using mathematical analysis models; so as to analyze the actuator adjustment washer selection scheme during the adjustment washer selection process through the actuator adjustment washer selection database, and to perform preventive diagnosis of actuator performance.

[0034] Secondly, embodiments of the present invention also provide a method for rapid selection and fitting of adjusting washers, wherein the method is performed on an actuator using a data-driven rapid selection and fitting system for adjusting washers as described in any of the preceding claims, comprising:

[0035] Step 1: Use digital measuring tools to measure and collect the machining dimensions of machined parts and the cross-linking dimensions of assembled parts;

[0036] Step 2: The collected data is processed in a secondary manner. According to the given dimension chain formula and grouping logic, calculations and grouping are performed to preliminarily derive the adjustment washer selection scheme. Each selection scheme includes: the grouping specifications and quantity of adjustment washers.

[0037] Step 3: Call the existing actuator adjustment washer selection data in the analysis subsystem, compare and preliminarily obtain the adjustment washer selection scheme, conduct item-by-item comparison and analysis, and give the preliminary conclusion of the actuator performance index; based on the preliminary conclusion of the actuator performance index, adjust and optimize the grouping specifications and quantity of adjustment washers until the preliminary conclusion meets the expected performance index of the actuator.

[0038] Step 4: The recommendation subsystem, based on the grouping specifications and quantity of the adjusting washers obtained in Step 3, guides the staff to select adjusting washers according to the determined grouping specifications and quantity through indicator lights, and records and manages the selection process of adjusting washers to prevent errors and correct deviations.

[0039] Step 5: After assembling the actuator with adjusting washers, automatically monitor the testing and verification process of the actuator adjusting washers, retrieve the test verification data, determine whether the grouping specifications and quantity of adjusting washers meet the prediction conclusion of the analysis subsystem, and update the actuator adjusting washer selection database in the analysis subsystem.

[0040] Optionally, the data-driven rapid selection method for adjusting washers described above is characterized in that the construction method of the actuator adjusting washer selection database invoked in step 3 includes:

[0041] S1. Collect historical data on the selection of a first preset number of actuator adjustment washers as the training set for the adjustment washer selection database; collect the machining dimensions of machined parts and the cross-linking dimensions of assembled parts in the training set, as well as the actuator performance indicators, and initially construct the actuator adjustment washer selection database.

[0042] S2. Using the Z-Score method, the standardized distances of the machining dimensions of the parts, the standardized distances of the assembly cross-linking dimensions, and the standardized distances of the actuator performance indicators in the gasket selection database are calculated to identify outliers and remove abnormal data points.

[0043] S3. Calculate the dispersion of all collected data Xi using scatter plots, and calculate the linear relationship between any collected data Xi and performance index Yi using covariance. Calculate the Pearson coefficient r between different parameters using the following formula:

[0044]

[0045] Where r represents the Pearson coefficient, and Xi and Yi represent the data values ​​of the collected data and performance indicators, respectively. and Let σX and σY represent the average values ​​of the two collected data and the performance index, respectively; n represents the sample size; and σX and σY represent the standard deviation of the collected data and the standard deviation of the actuator performance index, respectively.

[0046] S4 captures the nonlinear relationships in the training set data and constructs an actuator adjustment washer selection database by establishing a multi-factor coupled mapping scenario. Specifically, the collected data is used as the network input Xi, and the actuator product performance indicators, namely transmission efficiency, static friction, and product clearance, are used as the network output. Continuous simulation optimization is performed to complete model training and result prediction.

[0047] S5, recollect the historical data of the first preset number of sets of adjustment washers as an extended set of the adjustment washer selection database, construct the adjustment washer selection capability coefficient calculation matrix, realize the nonlinear transformation from input data to output actuator performance index, and expand the function of actuator performance optimization model; use cross-entropy loss to calculate the difference between the predicted value and the true value, and update the proportional coefficient and bias value through backpropagation algorithm;

[0048] S6, collect a second preset number of sets of standard data for adjusting washers as a validation set, and perform simulation verification on the adjusting washer selection database; after training the adjusting washer selection database of the actuator with the validation set data, obtain the root mean square error value of the generalization model of adjusting washer selection.

[0049] The beneficial effects of this invention are as follows: This invention provides a data-driven rapid selection system and method for adjusting washers. In the rapid selection process, firstly, basic data on actuator adjusting washers is collected using digital measuring tools to ensure data accuracy and validity; secondly, a computer system is used to perform logical calculations on the collected data to preliminarily determine the key parameters (group specifications and quantity) of the adjusting washers; thirdly, an existing adjusting washer selection database is used to analyze the group specifications and quantity of adjusting washers, determine whether the values ​​meet expectations, and make adjustments and optimizations; fourthly, based on the analysis results, a recommendation subsystem guides personnel to select the appropriate adjusting washer specifications and quantity; and fifthly, a diagnostic subsystem, combined with actuator test verification results, corrects and improves the actuator adjusting washer selection database. The rapid selection system and corresponding method for adjusting washers provided by this invention have the following beneficial effects:

[0050] 1) Utilize historical assembly data to establish a process model for rapid selection and matching of actuator adjustment washers, optimize the process flow for selecting and matching adjustment washers, solidify process parameters, and improve product quality stability.

[0051] 2) Predict key performance indicators of actuators through data models, and adjust production parameters based on the prediction results to improve actuator quality and reliability and reduce physical iterative verification;

[0052] 3) After the actuator fails, potential faults can be identified through data analysis, targeted troubleshooting can be carried out, maintenance costs can be reduced, and the service life of the actuator can be increased.

[0053] 4) By establishing actuator data models, deconstructing actuator assembly experience, and analyzing actuator physical laws, a knowledge graph-driven digital expert guidance system is built.

[0054] 5) The process innovation paradigm will shift from physical testing and verification to virtual simulation optimization, enabling data-driven design space exploration and agile iterative development, and driving enterprise digital transformation. Attached Figure Description

[0055] The accompanying drawings are provided to further understand the technical solutions of the present invention and constitute a part of the specification. They are used together with the embodiments of this application to explain the technical solutions of the present invention and do not constitute a limitation on the technical solutions of the present invention.

[0056] Figure 1 A schematic diagram of the system architecture of a data-driven rapid selection system for adjustment washers provided in an embodiment of the present invention;

[0057] Figure 2 This is a schematic diagram illustrating the construction process of the actuator performance index database in an embodiment of the present invention;

[0058] Figure 3This is the mapping logic relationship of the actuator adjustment washer selection scheme obtained by using the quick selection method for adjustment washers provided in the embodiments of the present invention. Detailed Implementation

[0059] To make the objectives, technical solutions, and advantages of the present invention clearer, the embodiments of the present invention will be described in detail below with reference to the accompanying drawings. It should be noted that, unless otherwise specified, the embodiments and features described in this application can be arbitrarily combined with each other.

[0060] To address the numerous problems associated with the current actuator assembly process, where adjusting washers are selected manually based on operational experience—namely, over-reliance on manual experience, lack of theoretical basis, repetitive and random disassembly / assembly processes, and poor consistency of test results—this invention provides a data-driven rapid adjusting washer selection system and method. By collecting data from the adjusting washer selection process, statistical methods are used to identify key parameters, and machine learning is employed to establish an adjusting washer selection database. This guides the optimization of adjusting washer selection data, shortens the selection time, improves the consistency of product performance indicators, and provides theoretical and empirical evidence for process optimization and preventative diagnostics.

[0061] Figure 1 A schematic diagram of the system architecture of a data-driven rapid selection system for adjusting washers provided in an embodiment of the present invention; as shown. Figure 1 As shown, the components of the quick selection system for adjusting washers provided by the present invention include: a data acquisition subsystem 1, a calculation subsystem 2, an analysis subsystem 3, a recommendation subsystem 4, and a diagnostic subsystem 5 connected in sequence.

[0062] In this embodiment of the invention, the acquisition subsystem 1 mainly uses digital measuring tools to measure and acquire the machining dimensions of machined parts and the cross-linking dimensions of assembled parts, and automatically uploads the acquired parameters to the calculation subsystem 2.

[0063] The calculation subsystem 2 in this embodiment of the invention is used to perform secondary processing on the data collected by the acquisition subsystem 1, perform calculations and grouping according to the given size chain and grouping logic, and initially derive the adjustment washer selection scheme. Each selection scheme includes the grouping specifications and quantity of adjustment washers.

[0064] In this embodiment of the invention, the analysis subsystem 3 can use MySQL as a data management platform. It calls the existing actuator adjustment washer selection database within the system, matches the adjustment washer selection scheme derived by the calculation subsystem 2 against the adjustment washer selection database, and provides a preliminary judgment on the actuator performance indicators. This preliminary judgment on the actuator performance indicators can guide manual adjustments and optimizations of the adjustment washer selection scheme, and analyze the optimized grouping specifications and quantity of adjustment washers.

[0065] The recommendation subsystem 4 in this embodiment of the invention includes a washer lighting device, which has multiple washer storage boxes and an indicator light on each washer storage box. Each washer storage box is used to hold an adjustment washer of a certain specification. The recommendation subsystem 4 is used to light up the indicator light according to the specification and quantity of the adjustment washer obtained by the analysis subsystem 3 to remind the personnel to select the adjustment washer according to the calculation results.

[0066] The diagnostic subsystem 5 in this embodiment of the invention is used to monitor the testing and verification process of the actuator adjusting washers, determine whether the grouping specifications and quantity of the adjusting washers meet the prediction conclusion of the analysis subsystem 3 based on the testing and verification results, and update the adjusting washer selection database in the analysis subsystem 3.

[0067] The testing and verification process for actuator adjusting washers is as follows: The actuator and the selected adjusting washer are installed on a test bench, a set torque and speed are applied, and the actuator performance indicators are tested, including: transmission efficiency, static friction, and product clearance. In this embodiment of the invention, the diagnostic subsystem 5 updates the adjusting washer selection database in the analysis subsystem 3 to improve the database, further enhancing the early warning and correction capabilities of the rapid adjusting washer selection system for abnormal data.

[0068] In one implementation of this invention, the machining dimensions of the machined parts collected by the acquisition subsystem 1 include: part height X1, part width X2, part length X3, drive shaft length X4, and part hole inner diameter X5; the cross-linking dimensions of the assembled parts include: component height X6, stepped hole depth X7, and bearing runout X8.

[0069] In this implementation, in order to measure and acquire the above-mentioned acquisition parameters X1 to X8, the acquisition subsystem 1 includes: a height gauge, a digital caliper, and a digital micrometer.

[0070] Among them, the height measuring instrument is used to collect the part height X1, part thickness X2, part length X3 of the machined parts, and the component height X6 of the assembled parts.

[0071] Digital calipers are used to measure the length of the drive shaft (X4) and the depth of the stepped hole (X7) of machined parts.

[0072] Digital micrometer, used to collect data on the inner diameter of machined parts (X5) and the bearing runout of assembled parts (X8).

[0073] It should be noted that the acquisition subsystem 1 in this embodiment of the invention should be able to acquire as many relevant parameters as possible, collect all relevant parameters, and realize digital acquisition and automatic uploading to reduce measurement errors.

[0074] In one implementation of this invention, the computational subsystem 2 performs computation and grouping according to a given size chain and grouping logic as follows:

[0075] On the one hand, the size chain is:

[0076] Calculate the thickness of adjusting washer 1: Y1=(X5-X3-X6-X7)÷2; (1)

[0077] Calculate the thickness of adjusting washer 2: Y2=(X1+X2-X4-X8)÷2; (2)

[0078] On the other hand, the grouping logic is:

[0079] Based on the calculated thickness of the adjusting shims, select adjusting shims of the corresponding thickness and quantity. The thickness difference between the adjusting shims used in the same position should not exceed 0.2mm, and the number of adjusting shims should not exceed 5.

[0080] Furthermore, based on the dimensional parameters of the machined and assembled parts to be collected by the acquisition subsystem 1, and the preliminary grouping results of the adjusting washers, the analysis subsystem 3 in this embodiment of the invention draws the following preliminary conclusions regarding the actuator performance indicators:

[0081] a) Based on the actuator performance index selection model library, the mapping relationship between Y1 and transmission efficiency K is obtained;

[0082] The product's transmission efficiency K is negatively correlated with Y1. Transmission efficiency K = -2Y1 + B1, where -2 is the coefficient of the transmission efficiency mapping model and B1 is the bias term of the model.

[0083] b) Based on the actuator performance index matching model library, the mapping relationship between Y2 and static friction force F is obtained;

[0084] The static friction force F of the product is positively correlated with Y2. The static friction force F = 2.6Y2 + B2, where 2.6 is the coefficient of the static friction force mapping model and B2 is the bias term of the model.

[0085] c) Based on the actuator performance index matching model library, obtain the mapping relationship between Y1, Y2 and product clearance C;

[0086] The product clearance C is nonlinearly related to Y1 and Y2, and the product clearance C = f(Y1, Y2).

[0087] In one implementation of this invention, the recommendation subsystem 4 further includes: a mis-picking sensor identification device installed in each gasket storage box, used to provide a warning when the specifications and quantity of the originally selected adjustment gasket do not match those calculated by the calculation subsystem 3.

[0088] In one implementation of this invention, the quick matching system for adjusting washers further includes a database subsystem 6 connected to the diagnostic subsystem 5 and the analysis subsystem 3 respectively, the database subsystem 6 being used to form an actuator adjusting washer matching database.

[0089] In this implementation, before executing the actuator adjustment washer selection scheme, the database subsystem 6 uses mathematical statistical analysis methods to identify the mapping relationship (linear / nonlinear) between the collected parameters (i.e., X1-X8) and the actuator performance indicators (transmission efficiency, static friction, product clearance). It then uses a mathematical analysis model to construct an actuator adjustment washer selection database. This database is used to analyze the actuator adjustment washer selection scheme during the adjustment washer selection process and to perform preventive diagnosis of actuator performance.

[0090] Based on the data-driven rapid selection system for adjusting washers provided in the embodiments of the present invention, the embodiments of the present invention also provide a rapid selection method for adjusting washers, characterized in that the rapid selection method for adjusting washers is performed on the actuator using the data-driven rapid selection system for adjusting washers as described in any one of claims 1 to 6, including:

[0091] Step 1: Use digital measuring tools to measure and collect the machining dimensions of machined parts and the cross-linking dimensions of assembled parts;

[0092] Step 2: The collected data is processed in a secondary manner. According to the given dimension chain formula and grouping logic, calculations and grouping are performed to preliminarily derive the adjustment washer selection scheme. Each selection scheme includes: the grouping specifications and quantity of adjustment washers.

[0093] Step 3: Call the existing actuator adjustment washer selection data in the analysis subsystem, compare and preliminarily derive the adjustment washer selection scheme, conduct item-by-item comparative analysis, and give the preliminary conclusion of the actuator performance indicators (transmission efficiency, static friction, product clearance); based on the preliminary conclusion of the actuator performance indicators, adjust and optimize the grouping specifications and quantity of adjustment washers until the preliminary conclusion meets the expected performance indicators of the actuator.

[0094] Step 4: The recommendation subsystem, based on the grouping specifications and quantity of the adjusting washers obtained in Step 3, guides the staff to select adjusting washers according to the determined grouping specifications and quantity through indicator lights, and records and manages the selection process of adjusting washers to prevent errors and correct deviations.

[0095] Step 5: After assembling the actuator with adjusting washers, automatically monitor the testing and verification process of the actuator adjusting washers, retrieve the test verification data, determine whether the grouping specifications and quantity of adjusting washers meet the prediction conclusion of the analysis subsystem, and update the actuator adjusting washer selection database in the analysis subsystem; thereby further improving the early warning and correction capability of the adjusting washer rapid selection system for abnormal data.

[0096] like Figure 3 The diagram shows the mapping logic of the actuator adjustment washer selection scheme obtained by using the quick selection method for adjustment washers provided in this embodiment of the invention.

[0097] S1. Collect historical data on the selection and matching of 2000 sets of actuator adjusting washers as the training set for the adjusting washer selection and matching database; collect the machining dimensions of machined parts and the cross-linking dimensions of assembled parts in the training set, as well as the actuator performance indicators, and initially construct the actuator adjusting washer matching data.

[0098] S2. Using the Z-Score method, the standardized distances of the machining dimensions of the parts, the standardized distances of the assembly cross-linking dimensions, and the standardized distances of the actuator performance indicators in the gasket selection database are calculated to identify outliers and remove abnormal data points.

[0099] S3. Calculate the dispersion of all collected data Xi using scatter plots, and calculate the linear relationship between any collected data Xi and performance index Yi using covariance. Calculate the Pearson coefficient r between different parameters using the following formula:

[0100]

[0101] Where r represents the Pearson coefficient, and Xi and Yi represent the data values ​​of the collected data and performance indicators, respectively. and Let σX and σY represent the average values ​​of the two collected data points and the performance index, respectively. Let n represent the sample size, and let σX and σY represent the standard deviation of the collected data and the standard deviation of the actuator performance index, respectively.

[0102] It should be noted that the Pearson coefficient ranges from -1 to 1; when r is close to 1, it indicates a strong positive correlation between σX and σY; when r is close to -1, it indicates a strong negative correlation between σX and σY. The collected data includes: machining dimensions of machined parts and cross-linking dimensions of assembled parts, specifically including: part height X1, part width X2, part length X3, drive shaft length X4, part hole inner diameter X5, assembly height X6, stepped hole depth X7, and bearing runout X8; actuator performance indicators include: transmission efficiency, static friction, and product clearance.

[0103] S4 captures the nonlinear relationships in the training set data and constructs an actuator adjustment washer selection database by establishing a multi-factor coupled mapping scenario. Specifically, the collected data X1, X2, X3, X4, X5, X6, X7, and X8 are used as the network input Xi, and the actuator product performance indicators are used as the network output Yi. Continuous simulation optimization is performed to complete model training and result prediction.

[0104] S5. Collect 2000 sets of historical data on adjusting washer selection as an extended set of the adjusting washer selection database, construct the adjusting washer selection capability coefficient calculation matrix, realize the nonlinear transformation from input data to output actuator performance index, and expand the function of actuator performance optimization model; use cross-entropy loss to calculate the difference between predicted value and true value, and update the proportional coefficient and bias value through backpropagation algorithm.

[0105] S6. Collect 1000 sets of standard data for adjusting washers as a validation set and perform simulation verification on the adjusting washer selection database. After training the adjusting washer selection database of the actuator with the validation set data, obtain the root mean square error value of the generalization model of adjusting washer selection.

[0106] The root mean square error (RMSE) of the generalized model for adjusting washer selection obtained in this step is, for example, RMSE = 0.33, which meets the requirements for product performance prediction and has stronger scalability.

[0107] Based on the actuator adjusting washer selection database constructed in S1 to S6 above, during the process of actuator adjusting washer selection and actuator performance index analysis, the adjusting washer selection database is used to construct an actuator performance index database, identify key parameters, control parameter fluctuation range, and is used for actuator performance index optimization, fault diagnosis, and preventive maintenance.

[0108] The data-driven rapid selection system for adjusting washers provided in this invention is used to perform a rapid selection method for adjusting washers on actuators. During the rapid selection process, the following steps are taken: First, basic data on the actuator's adjusting washers is collected using digital measuring tools to ensure the data is accurate and valid. Second, a computer system is used to perform logical calculations on the collected data to preliminarily determine the key parameters (group specifications and quantity) of the adjusting washers. Third, an existing adjusting washer selection database is used to analyze the group specifications and quantity of the adjusting washers, determine whether the values ​​meet expectations, and make adjustments and optimizations. Fourth, based on the analysis results, a recommendation subsystem guides personnel to select the appropriate adjusting washer specifications and quantity. Fifth, a diagnostic subsystem, combined with actuator test verification results, corrects and improves the actuator adjusting washer selection database. The rapid selection system for adjusting washers and the corresponding selection method provided by this invention have the following beneficial effects:

[0109] 1) Utilize historical assembly data to establish a process model for rapid selection and matching of actuator adjustment washers, optimize the process flow for selecting and matching adjustment washers, solidify process parameters, and improve product quality stability.

[0110] 2) Predict key performance indicators of actuators through data models, and adjust production parameters based on the prediction results to improve actuator quality and reliability and reduce physical iterative verification;

[0111] 3) After the actuator fails, potential faults can be identified through data analysis, targeted troubleshooting can be carried out, maintenance costs can be reduced, and the service life of the actuator can be increased.

[0112] 4) By establishing actuator data models, deconstructing actuator assembly experience, and analyzing actuator physical laws, a knowledge graph-driven digital expert guidance system is built.

[0113] 5) The process innovation paradigm will shift from physical testing and verification to virtual simulation optimization, enabling data-driven design space exploration and agile iterative development, and driving enterprise digital transformation.

[0114] The following is an implementation example of an actuator performance index database:

[0115] Implementation Example 1, Actuator Transmission Efficiency Database:

[0116] The mathematical model between the transmission efficiency K and the measured parameter Xi is determined by using a regression model, specifically: K = f1(X1,…,Xi).

[0117] With a transmission efficiency K greater than 90% as the control objective, a transmission efficiency database is constructed. The core key measurement parameters are: X3, X5, X6, and X7. The model parameter ranges are as follows:

[0118] X3: Average value 12.12mm, fluctuation control range 0.02mm;

[0119] X5: Average value 12.89mm, fluctuation control range 0.06mm;

[0120] X6: Average value 0.21mm, fluctuation control range 0.01mm;

[0121] X7: Average value 0.39mm, fluctuation control range 0.15mm.

[0122] Implementation Example 2, Actuator Static Friction Database:

[0123] The mathematical model between static friction force F and measured parameter Xi is determined by using a regression model, specifically: F = f2(X1,…,Xi).

[0124] With a static friction torque of less than 0.5 Nm as the analysis objective, a static friction database was constructed. The core key measurement parameters are: X1, X2, X4, and X8. The model parameter ranges are as follows:

[0125] X1: Average value 0.39mm, fluctuation control range 0.01mm;

[0126] X2: Average value 0.72mm, fluctuation control range 0.14mm;

[0127] X4: Average value 0.35mm, fluctuation control range ±0.07mm;

[0128] X8: Average value 0.26mm, fluctuation control range ±0.07mm.

[0129] Implementation Example 3, Actuator Product Clearance Database:

[0130] The mathematical model between the product clearance C and the measurement parameter Xi is determined by using a regression model, specifically: C = f3(X1,…,Xi).

[0131] With the goal of ensuring actuator clearance is no greater than 9′, an actuator clearance matching database is constructed. The core key measurement parameters are: X1, X2, X3, X5, and X7. The model parameter ranges are as follows:

[0132] X1: Average value 0.39mm, fluctuation control range 0.01mm;

[0133] X2: Average value 0.72mm, fluctuation control range 0.06mm;

[0134] X3: Average value 12.12mm, fluctuation control range 0.04mm;

[0135] X5: Average value 12.89mm, fluctuation control range 0.06mm;

[0136] X7: Average value 0.39mm, fluctuation control range 0.1mm.

[0137] While the embodiments disclosed in this invention are as described above, they are merely illustrative of the embodiments to facilitate understanding of the invention and are not intended to limit the invention. Any person skilled in the art to which this invention pertains may make any modifications and variations in the form and details of the implementation without departing from the spirit and scope disclosed herein; however, the scope of patent protection for this invention shall still be determined by the scope defined in the appended claims.

Claims

1. A data-driven rapid selection system for adjusting washers, characterized in that, include: The data acquisition subsystem (1), calculation subsystem (2), analysis subsystem (3), recommendation subsystem (4), and diagnosis subsystem (5) are connected in sequence. The acquisition subsystem (1) is used to measure and acquire the machining dimensions of machined parts and the cross-linking dimensions of assembled parts using digital measurement tools, and upload them to the calculation subsystem (2). The calculation subsystem (2) is used to perform secondary processing on the data collected by the acquisition subsystem (1), calculate and group according to the given size chain and grouping logic, and initially obtain the adjustment washer selection scheme. Each selection scheme includes the grouping specifications and quantity of adjustment washers. The analysis subsystem (3) is used to call the existing actuator adjustment washer selection database in the analysis subsystem (3), match the adjustment washer selection scheme obtained by the calculation subsystem (2) in the adjustment washer selection database, and give the preliminary conclusion of the actuator performance index. Based on the preliminary conclusion of the actuator performance index, the adjustment and optimization are carried out, and the optimized adjustment washer grouping specifications and quantity are obtained through analysis. The recommendation subsystem (4) includes a gasket lighting device, which has multiple gasket storage boxes and an indicator light on each gasket storage box. Each gasket storage box is used to hold an adjustment gasket of a certain size. The recommendation subsystem (4) is used to illuminate an indicator light to remind personnel to select the adjustment shims according to the calculation results based on the specifications and quantity of the adjustment shims obtained from the analysis subsystem (3); The diagnostic subsystem (5) is used to monitor the testing and verification process of the actuator adjusting washers, and to determine whether the grouping specifications and quantity of the adjusting washers meet the prediction conclusion of the analysis subsystem (3) based on the testing and verification results; and to update the adjusting washer selection database in the analysis subsystem (3).

2. The data-driven rapid selection system for adjusting washers according to claim 1, characterized in that, The machining dimensions of the machined parts include: part height X1, part width X2, part length X3, drive shaft length X4, and part hole inner diameter X5; the cross-linking dimensions of the assembled parts include: component height X6, stepped hole depth X7, and bearing runout X8. The data acquisition subsystem (1) includes: a height measuring instrument, a digital caliper, and a digital micrometer; The height measuring instrument is used to collect the part height X1, part thickness X2, part length X3 of the machined part, and the component height X6 of the assembled part. The digital caliper is used to collect the transmission shaft length X4 and the stepped hole depth X7 of the machined parts; The digital micrometer is used to collect the inner diameter of the hole of the machined part (X5) and the bearing runout of the assembled part (X8).

3. The data-driven rapid selection system for adjusting washers according to claim 2, characterized in that, The computational subsystem (2) performs computation and grouping according to the given size chain and grouping logic in the following manner: The dimension chain is: Calculate the thickness of adjusting washer 1: Y1 = (X5 - X3 - X6 - X7) ÷ 2; Calculate the thickness of adjusting washer 2: Y2=(X1+X2-X4-X8)÷2; The grouping logic is as follows: Based on the calculated thickness of the adjusting shims, select adjusting shims of the corresponding thickness and quantity. The thickness difference between the adjusting shims used in the same position should not exceed 0.2mm, and the number of adjusting shims should not exceed 5.

4. The data-driven rapid selection system for adjusting washers according to claim 3, characterized in that, The actuator performance indicators include transmission efficiency, static friction, and product clearance; the analysis subsystem (3) draws the following preliminary conclusions on the actuator performance indicators: a) Based on the actuator performance index selection model library, the mapping relationship between Y1 and transmission efficiency K is obtained; The actuator transmission efficiency K is negatively correlated with Y1. The transmission efficiency K = -2Y1 + B1, where -2 is the coefficient of the transmission efficiency mapping model and B1 is the bias term of the model. b) Based on the actuator performance index matching model library, the mapping relationship between Y2 and static friction force F is obtained; The static friction force F of the actuator is positively correlated with Y2. The static friction force F = 2.6Y2 + B2, where 2.6 is the coefficient of the static friction force mapping model and B2 is the bias term of the model. c) Based on the actuator performance index matching model library, obtain the mapping relationship between Y1, Y2 and product clearance C; The product clearance C is nonlinearly related to Y1 and Y2, and the product clearance C = f(Y1, Y2).

5. The data-driven rapid selection system for adjusting washers according to claim 1, characterized in that, The recommendation subsystem (4) further includes: a mis-picking sensor identification device installed in each gasket storage box, which is used to provide a warning when the specifications and quantity of the originally selected adjustment gasket do not match those calculated by the calculation subsystem (3).

6. The data-driven rapid selection system for adjusting washers according to any one of claims 1 to 5, characterized in that, Also includes: A database subsystem (6) is connected to the diagnostic subsystem (5) and the analysis subsystem (3) respectively; The database subsystem (6) is used to identify the mapping relationship between the collected parameters and the actuator performance indicators using mathematical statistical analysis methods before executing the actuator adjustment washer selection scheme, and to construct an actuator adjustment washer selection database using mathematical analysis models; so as to analyze the actuator adjustment washer selection scheme during the adjustment washer selection process through the actuator adjustment washer selection database, and to perform preventive diagnosis of actuator performance.

7. A method for quickly selecting and matching adjusting washers, characterized in that, The method for quickly selecting and matching adjustment washers for actuators using the data-driven adjustment washer quick-match system as described in any one of claims 1 to 6 includes: Step 1: Use digital measuring tools to measure and collect the machining dimensions of machined parts and the cross-linking dimensions of assembled parts; Step 2: The collected data is processed in a secondary manner. According to the given dimension chain formula and grouping logic, calculations and grouping are performed to preliminarily derive the adjustment washer selection scheme. Each selection scheme includes: the grouping specifications and quantity of adjustment washers. Step 3: Call the existing actuator adjustment washer selection data in the analysis subsystem, compare and preliminarily obtain the adjustment washer selection scheme, conduct item-by-item comparison and analysis, and give the preliminary conclusion of the actuator performance index; based on the preliminary conclusion of the actuator performance index, adjust and optimize the grouping specifications and quantity of adjustment washers until the preliminary conclusion meets the expected performance index of the actuator. Step 4: The recommendation subsystem, based on the grouping specifications and quantity of the adjusting washers obtained in Step 3, guides the staff to select adjusting washers according to the determined grouping specifications and quantity through indicator lights, and records and manages the selection process of adjusting washers to prevent errors and correct deviations. Step 5: After assembling the actuator with adjusting washers, automatically monitor the testing and verification process of the actuator adjusting washers, retrieve the test verification data, determine whether the grouping specifications and quantity of adjusting washers meet the prediction conclusion of the analysis subsystem, and update the actuator adjusting washer selection database in the analysis subsystem.

8. The data-driven rapid selection method for adjusting washers according to claim 7, characterized in that, The construction method of the actuator adjustment washer selection database called in step 3 includes: S1. Collect historical data on the selection of a first preset number of actuator adjustment washers as the training set for the adjustment washer selection database; collect the machining dimensions of machined parts and the cross-linking dimensions of assembled parts in the training set, as well as the actuator performance indicators, and initially construct the actuator adjustment washer selection database. S2, calculate the standardized distance of the machining dimensions of the parts, the standardized distance of the assembly cross-linking dimensions, and the standardized distance of the actuator performance indicators in the adjustment washer selection database, identify outliers, and remove outlier data points; S3. Calculate the dispersion of all collected data Xi using scatter plots, and calculate the linear relationship between any collected data Xi and performance index Yi using covariance. Calculate the Pearson coefficient r between different parameters using the following formula: Where r represents the Pearson coefficient, and Xi and Yi represent the data values ​​of the collected data and performance indicators, respectively. and Let σX and σY represent the average values ​​of the two collected data and the performance index, respectively; n represents the sample size; and σX and σY represent the standard deviation of the collected data and the standard deviation of the actuator performance index, respectively. S4 captures the nonlinear relationships in the training set data and constructs an actuator adjustment washer selection database by establishing a multi-factor coupled mapping scenario; specifically, the collected data is used as the network input X. i Using the actuator product performance indicators—transmission efficiency, static friction, and product clearance—as network outputs, continuous simulation optimization is performed to complete model training and result prediction. S5, recollect the historical data of the first preset number of sets of adjustment washers as an extended set of the adjustment washer selection database, construct the adjustment washer selection capability coefficient calculation matrix to realize the nonlinear transformation from input data to output actuator performance index, and expand the function of actuator performance optimization model; use cross-entropy loss to calculate the difference between the predicted value and the true value, and update the proportional coefficient and bias value through backpropagation algorithm; S6, collect a second preset number of sets of standard data for adjusting washers as a validation set, and perform simulation verification on the adjusting washer selection database; after training the adjusting washer selection database of the actuator with the validation set data, obtain the root mean square error value of the generalization model of adjusting washer selection.

Images