A data-driven based glove machine knitting efficiency optimization method and device

Abstract

The application discloses a glove knitting machine needle knitting efficiency optimization method and device based on data driving, first determines the maximum position error threshold allowed by the glove knitting machine, collects the cross-moving time and position error of the main shaft machine head under different working parameters, and thus establishes a data sample library; then, a first mathematical regression model between the cross-moving time and the working parameters is established, and a second mathematical regression model between the position error and the working parameters is established, and the first and second mathematical regression models are fitted based on the data sample library, and the first and second mathematical regression models with the highest fitting degree are obtained; finally, the adaptive bat algorithm is used to optimize the working parameters of the first and second mathematical regression models, and the optimal working parameters of the glove knitting machine are obtained. The application optimizes the working parameters of the glove knitting machine, and improves the production efficiency of the glove knitting machine.

Description

Technical Field

[0001] This application belongs to the field of glove knitting technology, specifically relating to a data-driven method and apparatus for optimizing glove knitting efficiency. Background Technology

[0002] In fully automatic knitting glove machines, the spindle head is the core component for glove knitting. Its motion characteristics are mainly high speed, high acceleration and deceleration start and stop, and small displacement repetitive motion. This characteristic will cause positional errors in the spindle head during lateral movement, which will affect glove quality and production efficiency.

[0003] Typically, glove production workshops have hundreds of machines, each requiring the spindle head to perform hundreds or even thousands of lateral movements to complete the knitting of a glove. If the single lateral movement time of the spindle head can be reduced as much as possible within the allowable range of positional error, the knitting time of each glove will be greatly shortened, and the overall production efficiency of the workshop will be significantly improved.

[0004] However, currently, knitted glove manufacturers mainly rely on manual experience to adjust the maximum speed of the spindle head, often setting the servo parameters of all equipment to the same constant value, which fails to maximize the operating efficiency of different equipment. Summary of the Invention

[0005] The purpose of this application is to propose a data-driven method and apparatus for optimizing the knitting efficiency of glove knitting machines, specifically addressing the problem of optimizing the production efficiency of glove knitting machines.

[0006] To achieve the above objectives, the technical solution of this application is as follows:

[0007] A data-driven method for optimizing the knitting efficiency of a glove machine includes:

[0008] The maximum allowable position error threshold of the glove machine was determined, and the lateral movement time and position error of the spindle head under different operating parameters were collected to establish a data sample library;

[0009] A first mathematical regression model is established between lateral movement time and working parameters, and a second mathematical regression model is established between position error and working parameters;

[0010] The first and second mathematical regression models were fitted based on the data sample library to obtain the first and second mathematical regression models with the highest fit.

[0011] The adaptive bat algorithm was used to optimize the operating parameters of the first and second mathematical regression models to obtain the optimal operating parameters of the glove machine.

[0012] Furthermore, the operating parameters include the servo position gain K. LAcceleration / deceleration time t a And the target velocity V.

[0013] Furthermore, establishing the first mathematical regression model between the lateral shift time and the operating parameters includes:

[0014] Given a constant target velocity V, establish the gain K at different positions during the lateral movement time t. L Down and acceleration time t a Mathematical regression model Φ t1 The formula is expressed as:

[0015]

[0016] In the formula, a t1 b t1 C t1 For constant parameters, To calculate the relationship between K and the transverse time t L and t a The regression function;

[0017] Given acceleration time t a Under the condition that the transverse time t is constant, establish the gain K at different positions. L The mathematical relationship between the target velocity V and Φ t2 The formula is expressed as:

[0018]

[0019] In the formula, a t2 b t2 C t2 For constant parameters, To calculate the relationship between K and the transverse time t L The regression function of V;

[0020] The first mathematical regression model between the transverse time t and the operating parameters is obtained by synthesis, and the formula is expressed as:

[0021]

[0022] In the formula, C t =C t1 +C t2 This is a constant parameter.

[0023] Furthermore, establishing a second mathematical regression model between position error and operating parameters includes:

[0024] Given a constant target velocity V, establish the position error e and its gain K at different positions. L Down and acceleration time t a Mathematical relationship Φ e1The formula is expressed as:

[0025]

[0026] In the formula, a e1 b e1 C e1 For constant parameters, To calculate the position error e with respect to K L and t a The regression function;

[0027] Given acceleration time t a Under the condition that the position error e remains unchanged, establish the position gain K at different positions. L The mathematical relationship between the target velocity V and Φ e2 The formula is expressed as:

[0028]

[0029] In the formula, a e2 b e2 C e2 For constant parameters, To calculate the position error e with respect to K L The regression function of V;

[0030] The second mathematical regression model, which combines the position error e with the operating parameters, is expressed as:

[0031]

[0032] In the formula, C e =C e1 +C e2 This is a constant parameter.

[0033] Furthermore, the adaptive bat algorithm is used to optimize the operating parameters of the first and second mathematical regression models to obtain the optimal operating parameters of the glove machine, including:

[0034] S4.1 Algorithm Parameter Design: The working parameters are described as individual bat b i The bat population size is N, and i = 1, 2, ..., N; the initial value of the impulse loudness A is A0, the initial value of the impulse rate r is r0, and the range of the impulse frequency f is [f min ,f max ], f min For the minimum frequency pulse, f min The maximum frequency pulse is defined; the maximum number of iterations for algorithm optimization is set to T. max ;

[0035] S4.2 Population Initialization: For N bat individuals b iThe initial design is performed using the values ​​of , and an initial population is obtained by spreading the constrained range of working parameters in d-dimensional space.

[0036] S4.3 Fitness Calculation: Calculated using the fitness function S... f Calculate the fitness of each bat and find the bat individual b with the optimal fitness function. Gbest :

[0037] S f =f t +αf e ;

[0038] Where α represents the position error model f e The penalty coefficient for the constraint;

[0039] S4.4, Bat Individual Update: At the current iteration number t c Less than the maximum number of iterations T max At that time, the current t of the bat is updated using the pulse frequency update formula, flight speed update formula, and individual bat update formula in the following formula. c Speed ​​during the next iteration and individual

[0040]

[0041] In the formula, i represents an individual in the population, and i = 1, 2, ... N; f i Represents individual bat b i The pulse frequency value, where β is a uniformly distributed random number. For random inertia weights, and They are respectively The minimum and maximum values, ψ∈[0,1] are random numbers, ρ is an adjustment factor greater than 1, and b Gbest It is the bat individual with the best fitness function among all bat groups;

[0042] S4.5, Bat Individual Optimization: At the tth c In this iteration, based on the optimal bat individual b Gbest Randomly search for new bat individuals b i :

[0043] Generate random number R num1 ∈[0,1], judge Whether it holds true depends on the impulse loudness. Calculate individual bat b i The value is:

[0044]

[0045] In the formula, ε is a random number distributed within [0,1]. Subsequently, a random number R is generated. num2 ∈[0,1], judge Whether this is true, and whether individual bat b i fitness function S f (b i () Less than the optimal bat individual b Gbest fitness function S f (b Gbest ), that is, S f (b i )<S f (b Gbest If b is the optimal bat individual, then b is updated to be b. i To achieve individual bat optimization;

[0046] At the same time, update the bat individual b in the next iteration according to the following formula. i Pulse loudness and pulse rate

[0047]

[0048] In the formula, μ and γ are constants, and exp represents the natural constant;

[0049] S4.6 Solving for the global optimum of a bat: Calculate the fitness function values ​​of all bats and find the bat with the smallest fitness function value, i.e., the globally optimum bat b. Gbest ;

[0050] S4.7 Algorithm Termination Condition Judgment: Determine if the number of iterations has reached the maximum number of iterations T. max If the condition is met, the iteration stops and the optimal solution is output; otherwise, jump to S4.4 for the next iteration until the termination condition is met. Finally, the globally optimal bat individual b can be solved. Gbest This yields the optimal operating parameters for the glove machine.

[0051] This application also proposes a data-driven glove machine knitting efficiency optimization device, including a processor and a memory storing a number of computer instructions, which, when executed by the processor, implement the steps of the above-described data-driven glove machine knitting efficiency optimization method.

[0052] This application proposes a data-driven method and apparatus for optimizing the knitting efficiency of glove knitting machines. Addressing the problem of optimizing glove knitting machine production efficiency, it collects operational data from multiple glove knitting machines under various operating parameters to establish a mathematical model describing the relationship between position error and lateral movement time based on multivariate nonlinear regression. Then, an adaptive bat algorithm is used to optimize system parameters for lateral movement time under the constraints of the position error model. Based on the obtained optimal parameters, the parameters of the glove knitting machine spindle head servo are configured. This application optimizes the operating parameters of glove knitting machines, thereby improving their production efficiency. Attached Figure Description

[0053] Figure 1 This is a flowchart of the data-driven glove machine knitting efficiency optimization method of this application.

[0054] Figure 2 The diagram shows the iterative results of the adaptive bat algorithm in the embodiments of this application. Detailed Implementation

[0055] To make the objectives, technical solutions, and advantages of this application clearer, the following detailed description is provided in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative and not intended to limit the scope of this application.

[0056] In one embodiment, such as Figure 1 As shown, a data-driven method for optimizing the knitting efficiency of glove machines is proposed, including:

[0057] Step S1: Determine the maximum allowable position error threshold of the glove machine, and collect the lateral movement time and position error of the spindle head under different working parameters to establish a data sample library.

[0058] Different models of glove machines have different maximum position error thresholds. For a glove machine to be optimized, its maximum allowable position error threshold can be determined according to its model.

[0059] The operating parameters of the glove machine include the servo position gain K. L Acceleration / deceleration time t a And the target velocity V.

[0060] The spindle head's operating data is collected under different working parameters, including traverse time t and position error e. The data collection method can be orthogonal experimental design, which involves setting a series of parameter combinations at equal intervals within the range of working parameter values ​​and collecting operating data for each parameter combination. After collecting the operating data, a data sample library is established, consisting of operating data under different working parameters.

[0061] This application used a 7-needle glove machine in the experiment, determined the positional error threshold of this model to be 2mm, and collected the operating data of the spindle head of the five finger and palm areas (there are 9 parts in the glove) under different operating parameters to establish a data sample library. The data collection method was orthogonal experiment, that is, a series of parameter combinations were set at equal intervals within the range of operating parameter values, and a total of 210 sets of operating data under parameter combinations were collected. Among them, the position gain K L The value range is 35–55 rad / s, and the acceleration / deceleration time t a The value range is 30 to 60 ms, and the value range of the target velocity V is 350 r / min to 600 r / min.

[0062] Step S2: Establish a first mathematical regression model between lateral movement time and working parameters, and establish a second mathematical regression model between position error and working parameters.

[0063] This application establishes the relationship between the transverse time t and the operating parameters (position gain K). L Acceleration / deceleration time t a Mathematical regression model f between the target velocity (V) and the target velocity (V) t And establish the relationship between position error e and operating parameters (position gain K) L Acceleration / deceleration time t a Mathematical regression model f between the target velocity (V) and the target velocity (V) e .

[0064] The establishment of the first mathematical regression model between the lateral movement time and the working parameters includes:

[0065] S2.1.1. Given a constant target velocity V, establish the gain K at different positions during the lateral movement time t. L Down and acceleration time t a Mathematical regression model Φ t1 The formula is expressed as:

[0066]

[0067] In the formula, a t1 b t1 C t1 For constant parameters, To calculate the relationship between K and the transverse time t L and t a The regression function.

[0068] S2.1.2, at a given acceleration time t a Under the condition that the transverse time t is constant, establish the gain K at different positions. L The mathematical relationship between the target velocity V and Φ t2 The formula is expressed as:

[0069]

[0070] In the formula, a t2 b t2 C t2 For constant parameters, To calculate the relationship between K and the transverse time t L The regression function of V.

[0071] S2.1.3. The first mathematical regression model between the transverse shift time t and the working parameters is obtained, and the formula is expressed as:

[0072]

[0073] In the formula, C t =C t1 +C t2 This is a constant parameter.

[0074] The establishment of the second mathematical regression model between the position error and the operating parameters includes:

[0075] S2.2.1. Given a constant target velocity V, establish the position error e and its gain K at different positions. L Down and acceleration time t a Mathematical relationship Φ e1 The formula is expressed as:

[0076]

[0077] In the formula, a e1 b e1 C e1 For constant parameters, To calculate the position error e with respect to K L and t a The regression function.

[0078] S2.2.2, at a given acceleration time t a Under the condition that the position error e remains unchanged, establish the position gain K at different positions. L The mathematical relationship between the target velocity V and Φ e2 The formula is expressed as:

[0079]

[0080] In the formula, a e2 b e2 C e2 For constant parameters, To calculate the position error e with respect to K L The regression function of V.

[0081] S2.2.3. The second mathematical regression model, which integrates the position error e with the working parameters, is expressed as:

[0082]

[0083] In the formula, C e =C e1 +C e2 This is a constant parameter.

[0084] Step S3: Fit the established first and second mathematical regression models based on the data sample library to obtain first and second mathematical regression models with high fitting degree.

[0085] This step is based on the established mathematical regression model f between the lateral movement time t and the position error e, derived from the data sample database. t and f e Perform fitting modeling.

[0086] In one specific embodiment, the established first and second mathematical regression models are fitted based on the data sample library to obtain the first and second mathematical regression models with the highest fit, including:

[0087] S3.1 Select a commonly used multivariate linear model f line Multivariate nonlinear quadratic model f quad and the multivariate nonlinear logarithmic model f ln As a foundation for modeling, it is represented as follows:

[0088]

[0089] In the formula, y i It is the dependent variable, x i1 ,…,x in α is the independent variable. i0 ,α i1 ,…,α in is the regression coefficient.

[0090] S3.2, Using the model fit goodness R 2 The regression model f for the lateral movement time t and position error e t and f e The significance test is performed, and the formula is expressed as follows:

[0091]

[0092] In the formula, SSR is the regression sum of squares, SST is the residual sum of squares, and y is the actual observed value. These are the fitted values. R is the average of the actual observed values. 2The value of ranges from 0 to 1. A larger value indicates a better significance and higher fit of the constructed regression model. Conversely, a smaller value indicates poorer significance and lower fit.

[0093] S3.3. Based on the data sample library, use the regression model f listed in S3.1. line f quad and f ln And the trapezoidal acceleration / deceleration formula, for the mathematical regression model f of lateral movement time t and position error e. t and f e Perform fitting modeling and solve for the specific regression model expression. and model parameter a t1 b t1 a t2 b t2 C t a e1 b e1 a e2 b e2 C e Finally, the goodness-of-fit Rfit in S3.2 is used. 2 Indicator, select R 2 The highest regression model yields the regression equation f. t and f e .

[0094] Step S4: Use the adaptive bat algorithm to optimize the working parameters of the first and second mathematical regression models to obtain the optimal working parameters of the glove machine.

[0095] This embodiment uses the adaptive bat algorithm to regress the regression model f on the lateral movement time t and the position error e. t and f e Operating parameters (position gain K) are selected. L Acceleration / deceleration time t a Optimization of target speed (V). The optimal individual bat b is found by adjusting the bat's flight speed v, pulse rate r, pulse frequency f, and pulse loudness A. Gbest Thus, the optimal solution to the optimization problem is obtained. The design process of the adaptive bat algorithm is as follows:

[0096] S4.1 Algorithm Parameter Design: The working parameters are described as individual bat b i The bat population size is N, and i = 1, 2, ..., N; the initial value of the impulse loudness A is A0, the initial value of the impulse rate r is r0, and the range of the impulse frequency f is [f min ,f max ], f min For the minimum frequency pulse, f minThe maximum frequency pulse is defined; the maximum number of iterations for algorithm optimization is set to T. max .

[0097] S4.2 Population Initialization: For N bat individuals b i The initial design is performed using the values ​​of , and an initial population is obtained by spreading the constrained range of working parameters in d-dimensional space.

[0098] S4.3 Fitness Calculation: Calculated using the fitness function S... f Calculate the fitness of each bat and find the bat individual b with the optimal fitness function. Gbest :

[0099] S f =f t +αf e (9)

[0100] Where α represents the position error model f e The penalty coefficient for the constraint.

[0101] S4.4, Bat Individual Update: At the current iteration number t c Less than the maximum number of iterations T max At that time, the current t of the bat is updated using the pulse frequency update formula, flight speed update formula, and individual bat update formula in the following formula. c Speed ​​during the next iteration and individuals

[0102]

[0103] In the formula, i represents an individual in the population, and i = 1, 2, ... N; f i Represents individual bat b i The pulse frequency value, where β is a uniformly distributed random number. For random inertia weights, and They are respectively The minimum and maximum values, ψ∈[0,1] are random numbers, ρ is an adjustment factor greater than 1, and b Gbest It is the bat individual with the optimal fitness function among all bat groups.

[0104] S4.5, Bat Individual Optimization: At the tth c In this iteration, based on the optimal bat individual b Gbest Randomly search for new bat individuals b i Specifically: Generate random numbers R num1 ∈[0,1], judge Whether it holds true depends on the impulse loudness. Calculate individual bat bi The value is:

[0105]

[0106] In the formula, ε is a random number distributed within [0,1]. Then, a random number R is generated. num2 ∈[0,1], judge Whether this is true, and whether individual bat b i fitness function S f (b i () Less than the optimal bat individual b Gbest fitness function S f (b Gbest ), that is, S f (b i )<S f (b Gbest If b is the optimal bat individual, then b is updated to be b. i This enables the optimization of individual bats.

[0107] At the same time, update the bat individual b in the next iteration according to the following formula. i Pulse loudness and pulse rate

[0108]

[0109] In the formula, μ and γ are constants, and exp represents the natural constant.

[0110] S4.6 Solving for the global optimum of a bat: Calculate the fitness function values ​​of all bats and find the bat with the smallest fitness function value, i.e., the globally optimum bat b. Gbest .

[0111] S4.7 Algorithm Termination Condition Judgment: Determine if the number of iterations has reached the maximum number of iterations T. max If the condition is met, the iteration stops and the optimal solution is output; otherwise, it jumps to S4.4 for the next iteration until the termination condition is met. Finally, the globally optimal bat individual b can be solved. Gbest This yields the optimal operating parameters for the glove machine: position gain K. L Acceleration / deceleration time t a And the target velocity V.

[0112] The results of the iterative experiments of the adaptive bat algorithm are as follows: Figure 2As shown, in practice, to minimize the lateral movement time under the constraint of satisfying the position error, only the relationship between the lateral movement time and the number of iterations is given. In this embodiment, the algorithm converges to the global optimum, and the shortest lateral movement time is 227.9ms. Therefore, when the current glove machine is knitting the five fingers and palms, the system operating parameters can be set to the optimal operating parameters corresponding to the above iterative optimization results. Then, the machine can achieve the shortest knitting time while satisfying the position error threshold (i.e., without causing damage).

[0113] This application also verified the technical solution through experiments, and obtained the solution through fitting. For logarithmic models, For linear models, For logarithmic models, As a reciprocal model, the regression model f t After substituting the data into the database and performing fitting, we can obtain a. t1 b t1 a t2 b t2 C t The values ​​of the parameters and the regression equation f for the transverse shift time t. t Calculate the goodness of fit R of the model 2 f when it is 0.998 t for:

[0114]

[0115] The actual values ​​of the transverse time and the transverse time model f under 30 different working parameters were used. t Comparing the predicted and actual values, the maximum absolute value of the error between the predicted and actual values ​​is 1.4 ms, and the mean is 0.64 ms. The results demonstrate the effectiveness of the shift-time model f. t The accuracy.

[0116] Experimental results It is a quadratic model. For linear models, For logarithmic models, For a logarithmic model, the regression model f e After substituting the data into the database and performing fitting, we can obtain a. e1 b e1 a e2 b e2 C e The regression equation f for the isoparameter values ​​and the position error e e Calculate the goodness of fit R of the model 2 f when it is 0.972 e for:

[0117]

[0118] The actual values ​​of position error and the position error model f under 30 different working parameters were used. e Comparing the predicted and actual values, the maximum absolute value of the error between the predicted and actual values ​​is 0.094 mm, and the mean is 0.0443 mm. The results demonstrate the effectiveness of the position error model f. e The accuracy.

[0119] Using the above method, the optimal working parameters for the remaining glove parts can be calculated separately. Finally, by substituting these optimal working parameters into the glove machine, the working efficiency of the spindle head can be optimized. To demonstrate that the optimized working parameters have higher efficiency, they are compared with manually set parameters. Furthermore, to verify the effectiveness of this optimization method on other models of the same type, four 7-needle glove machines were selected for testing. The test results are shown in Table 1.

[0120] Table 1

[0121]

[0122] As shown in Table 1, the optimized parameters have higher weaving efficiency, and the working efficiency can be improved by approximately 5.2% compared to manually set parameters.

[0123] In one embodiment, this application also provides a data-driven glove machine knitting efficiency optimization device, including a processor and a memory storing a plurality of computer instructions, which, when executed by the processor, implement the steps of the above-described data-driven glove machine knitting efficiency optimization method.

[0124] Specific limitations regarding the data-driven glove machine knitting efficiency optimization device can be found in the limitations of the data-driven glove machine knitting efficiency optimization method described above, and will not be repeated here. The aforementioned data-driven glove machine knitting efficiency optimization device can be implemented entirely or partially through software, hardware, or a combination thereof. It can be embedded in or independent of the processor in a computer device in hardware form, or stored in the memory of a computer device in software form, so that the processor can call and execute the operations corresponding to the above modules.

[0125] The memory and processor are electrically connected directly or indirectly to enable data transmission or interaction. For example, these components can be electrically connected to each other via one or more communication buses or signal lines. The memory stores a computer program that can run on the processor, which implements the network topology layout method in this embodiment of the invention by running the computer program stored in the memory.

[0126] The memory may be, but is not limited to, Random Access Memory (RAM), Read Only Memory (ROM), Programmable Read-Only Memory (PROM), Erasable Programmable Read-Only Memory (EPROM), Electrically Erasable Programmable Read-Only Memory (EEPROM), etc. The memory stores the program, and the processor executes the program upon receiving an execution instruction.

[0127] The processor may be an integrated circuit chip with data processing capabilities. The aforementioned processor can be a general-purpose processor, including a Central Processing Unit (CPU), a Network Processor (NP), etc. It can implement or execute the methods, steps, and logic block diagrams disclosed in the embodiments of this invention. The general-purpose processor can be a microprocessor or any conventional processor.

[0128] The embodiments described above are merely illustrative of several implementation methods of this application, and while the descriptions are relatively specific and detailed, they should not be construed as limiting the scope of the invention patent. It should be noted that those skilled in the art can make various modifications and improvements without departing from the concept of this application, and these all fall within the protection scope of this application. Therefore, the protection scope of this patent application should be determined by the appended claims.

Claims

1. A data-driven method for optimizing the knitting efficiency of a glove machine, characterized in that, The data-driven glove machine knitting efficiency optimization method includes: The maximum allowable position error threshold of the glove machine was determined, and the lateral movement time and position error of the spindle head under different working parameters were collected to establish a data sample library. A first mathematical regression model is established between traverse time and operating parameters, and a second mathematical regression model is established between position error and operating parameters, wherein the operating parameters include servo position gain. Acceleration and deceleration time and target speed ; The first and second mathematical regression models were fitted based on the data sample library to obtain the first and second mathematical regression models with the highest fit. The adaptive bat algorithm was used to optimize the working parameters of the first and second mathematical regression models to obtain the optimal working parameters of the glove machine. The establishment of the first mathematical regression model between the lateral movement time and the working parameters includes: Given target speed Establish the transverse time under the condition that it remains unchanged. Gain at different locations Down and acceleration time Mathematical regression model The formula is expressed as: ； In the formula, , , For constant parameters, , To calculate the lateral movement time Time about and The regression function; At a given acceleration time Establish the transverse time under the condition that it remains unchanged. Gain at different locations Downward and target speed mathematical relationship The formula is expressed as: ； In the formula, , , For constant parameters, , To calculate the lateral movement time Time about and The regression function; The transverse time is obtained by combining the results. The first mathematical regression model between the working parameters and the operating parameters is expressed as follows: ； In the formula, This is a constant parameter.

2. The data-driven glove machine knitting efficiency optimization method according to claim 1, characterized in that, The establishment of the second mathematical regression model between position error and operating parameters includes: Given target speed Establish position error under the condition that it remains unchanged. Gain at different locations Down and acceleration time mathematical relationship The formula is expressed as: ； In the formula, , , For constant parameters, , To calculate position error Time about and The regression function; At a given acceleration time Establish position error under the condition that it remains unchanged. Gain at different locations Downward and target speed mathematical relationship The formula is expressed as: ； In the formula, , , For constant parameters, , To calculate position error Time about and The regression function; The combined position error The second mathematical regression model with respect to the working parameters is expressed as: ； In the formula, This is a constant parameter.

3. The data-driven glove machine knitting efficiency optimization method according to claim 1, characterized in that, The adaptive bat algorithm is used to optimize the operating parameters of the first and second mathematical regression models to obtain the optimal operating parameters of the glove machine, including: S4.1 Algorithm Parameter Design: The working parameters are described as individual bats. The size of the bat population is ,and =1,2,… ; Pulse loudness The initial value is Pulse rate The initial value is pulse frequency The range of values ​​is , The minimum frequency pulse, The maximum frequency pulse is defined; the maximum number of iterations for algorithm optimization is set to [value missing]. ; S4.2 Population Initialization: For Individual bats Initialize the design based on the values, within the constraints of the working parameters. A set of initial populations is obtained by diffusing and distributing them in dimensional space; S4.3 Fitness Calculation: Calculated using the fitness function. Calculate the fitness of each bat and find the bat with the best fitness function. : ； in, Representing the position error model The penalty coefficient for the constraint; S4.4, Bat Individual Update: At the current iteration number Less than the maximum number of iterations At that time, the current status of the bat is updated using the pulse frequency update formula, flight speed update formula, and individual bat update formula in the following equations. Speed ​​during the next iteration and individual : ； In the formula, Represents individuals in a population, and =1,2,… ; Represents individual bats The pulse frequency value, The random numbers are uniformly distributed. For random inertia weights, and They are respectively The minimum and maximum values, It is a random number. For adjustment factors greater than 1, It is the bat individual with the best fitness function among all bat groups; S4.5, Bat Individual Optimization: In the first... In the next iteration, based on the optimal bat individual Randomly search for new bat individuals : Generate random numbers ,judge Whether it holds true depends on the impulse loudness. Calculate individual bats The value is: ； In the formula, for Random numbers are generated from an internal distribution, and then random numbers are generated. ,judge Whether it is true, and the individual bat fitness function Smaller than the optimal bat individual fitness function ,Right now Then update the optimal bat individual as To achieve individual bat optimization; At the same time, update the individual bats in the next iteration according to the following formula. Pulse loudness and pulse rate : ； In the formula, and It is a constant. Represents the natural constant; S4.6 Solving for the global optimum of an individual bat: Calculate the fitness function values ​​of all individual bats and find the bat with the smallest fitness function value, which is the globally optimal bat. ; S4.7 Algorithm Termination Condition Judgment: Determine if the maximum number of iterations has been reached. If the condition is met, the iteration stops and the optimal solution is output; otherwise, it jumps to S4.4 for the next iteration until the termination condition is met. Finally, the globally optimal individual bat can be found. This yields the optimal operating parameters for the glove machine.

4. A data-driven glove knitting efficiency optimization device, comprising a processor and a memory storing a plurality of computer instructions, characterized in that, When the computer instructions are executed by the processor, they implement the steps of the method according to any one of claims 1 to 3.

Images