Weaving production group shaft optimization system and method

Abstract

The invention belongs to the technical field of intelligent weaving, and discloses a weaving production group shaft optimization system and method, and the method comprises the steps: collecting order data; preprocessing the order data to obtain effective order data; obtaining a warp yarn temperature influence coefficient based on the constructed warp yarn production influence coefficient model, obtaining a weft yarn humidity influence coefficient based on the constructed weft yarn production influence coefficient model, and performing demand calculation through the warp yarn temperature influence coefficient, the weft yarn humidity influence coefficient and effective order data to obtain a warp yarn demand quantity and a weft yarn demand quantity; performing assembly shaft distribution based on the warp demand quantity and the weft demand quantity to obtain an optimal assembly shaft scheme; and the labor cost is greatly reduced.

Description

Technical Field

[0001] The present invention relates to the field of intelligent weaving technology, and more specifically, to a weaving production shaft optimization system and method. Background Art

[0002] The patent with application publication number CN118134144A discloses a production management method, system and storage medium for knitted fabrics. The method includes the following steps: obtaining the quantity of materials to be detected corresponding to the step to be detected; comparing the quantity of materials to be detected with a preset quantity, and determining whether the quantity of materials to be detected is similar to the preset quantity; if the quantity of materials to be detected is similar to the preset quantity, generating a corresponding transport signal, and sending the transport signal to a vehicle control module to control the transport vehicle to automatically transport the corresponding materials to be detected; by generating a transport signal and sending the transport signal to a vehicle control module to control the transport vehicle to automatically transport the corresponding materials to be detected, manual handling is reduced, time is saved, and the production efficiency of knitted fabrics is improved.

[0003] With the advancement of science and technology, the weaving industry is undergoing profound changes; digital transformation has become an important trend in the development of the industry. Through the application of industrial Internet technology, the production efficiency of the weaving industry has been significantly improved; for example, by increasing the machine operating rate from 60% to 90%, it not only promoted the increase in factory orders, but also enabled upstream manufacturers to accurately predict the amount of yarn required by downstream in the next week, thereby accurately producing and reducing operating costs; this digital transformation not only improved production efficiency, but also optimized resource allocation, enabling the weaving industry to adapt to market changes more quickly; intelligent production is also an important aspect of the background technology of the weaving industry; in modern textile factories, smart spindles, industrial robots and automated storage workshops have become the norm, and the application of these technologies has greatly improved production efficiency and product quality; in addition, the application of big data platforms has also brought revolutionary changes to the weaving industry; by building a textile industry big data platform, the digital transformation of textile enterprises is realized, industrial big data information is obtained, and these data are fed back for use in the platform's capacity allocation, thus forming a closed loop from raw material supply to finished product trade, and realizing the model of industrial digital reconstruction of resource allocation.

[0004] However, as the weaving industry becomes larger and larger, the number of product varieties increases, and labor becomes increasingly insufficient, the traditional shaft assembly method is time-consuming, labor-intensive, and experience-based, and is no longer applicable. When there are many order varieties, the traditional shaft assembly method will inevitably cause some production waste, and it is difficult to find the optimal solution for shaft assembly manually.

[0005] In view of this, the present invention proposes a weaving production shaft optimization system and method to solve the above problems. Summary of the invention

[0006] In order to overcome the above-mentioned defects of the prior art and to achieve the above-mentioned purpose, the present invention provides the following technical solutions: a system and method for optimizing the weaving production shaft group;

[0007] The weaving production shaft group optimization system described in the present invention includes: a data collection module for collecting order data;

[0008] Data processing module: pre-process order data to obtain valid order data;

[0009] Demand generation module: obtain the warp yarn temperature influence coefficient based on the constructed warp yarn production influence coefficient model, obtain the weft yarn humidity influence coefficient based on the constructed weft yarn production influence coefficient model, calculate the demand through the warp yarn temperature influence coefficient, weft yarn humidity influence coefficient and valid order data, and obtain the warp yarn demand and weft yarn demand;

[0010] Axis grouping scheme generation module: Axis grouping allocation is performed based on the warp yarn demand and weft yarn demand to obtain the optimal axis grouping scheme.

[0011] Furthermore, the order data includes: finished fabric length, fabric width, warp yarn density, weft yarn density, warp yarn spacing and order time.

[0012] Furthermore, the specific method of preprocessing the order data includes:

[0013] The order demand data is standardized by the Z-score algorithm to obtain standardized order data. The scale parameter set A is preset and the standardized order data is filtered and transformed. The formula for filtering transformation is: Where a represents the scale parameter, and a∈A, x i represents the i-th standardized order data, τ represents the translation parameter of the standardized order data, W a (x i ) represents the filtered value of the i-th standardized order data when the scale parameter is a, t represents the order time, e represents the base of the natural logarithm, and i represents the i-th standardized order data;

[0014] Calculate the dynamic adaptability weight of the standardized order data under different scale parameters. The calculation formula of the dynamic adaptability weight is: Among them, η a represents the dynamic adaptability weight corresponding to the scale parameter a, α is the adjustment factor, Var(W a (x i )) represents the variance of the filtered value of the i-th standardized order data when the scale parameter is a;

[0015] The filter value is weighted according to the dynamic adaptive weight. The weighted calculation formula is: in Represents the weighted sum of the filtered values ​​of the ith standardized order data after filtering transformation; the residual of the standardized order data is calculated based on the weighted sum, and the residual calculation formula is: where r i Represents the residual of the i-th standardized order data; sort all the calculated residuals from large to small, and obtain the median Mid of the residuals;

[0016] Preset the abnormal sensitivity parameter β and calculate the dynamic threshold. The calculation formula of the dynamic threshold is: R i =β×|r i -Mid|; where R i represents the dynamic threshold of the i-th standardized order data; the abnormality is eliminated through the dynamic threshold to obtain valid order data; the abnormality elimination method includes: when r i >R i When , the order data corresponding to the i-th standardized order data is marked as abnormal data, and all order data marked as abnormal data are removed.

[0017] Furthermore, the method for obtaining the warp yarn temperature influence coefficient includes:

[0018] Collect historical warp yarn data of group B, the historical warp yarn data includes historical ambient temperature and historical warp yarn temperature influence coefficient, build a warp yarn production influence coefficient model based on the historical warp yarn data, use historical ambient temperature as input of the warp yarn production influence coefficient model, use warp yarn temperature influence coefficient as output of the warp yarn production influence coefficient model, use historical warp yarn temperature influence coefficient as prediction target of the warp yarn production influence coefficient model, use root mean square error function as loss function of the warp yarn production influence coefficient model, use minimization of loss function value of the warp yarn production influence coefficient model as training target, and stop training when the loss function value of the warp yarn production influence coefficient model is less than or equal to the preset target loss value;

[0019] The warp yarn production influence coefficient model is a multiple regression model or a logistic regression model, and the root mean square error function formula is: Among them, RMSE is the root mean square error, Su is the total amount of historical warp data, and F b is the bth historical warp temperature influence coefficient, F' b is the temperature influence coefficient of the bth warp yarn;

[0020] The real-time ambient temperature is collected, and based on the constructed warp yarn production influence coefficient model, the real-time ambient temperature is used as input to obtain the warp yarn temperature influence coefficient.

[0021] Furthermore, the method for obtaining the weft yarn humidity influence coefficient includes:

[0022] Collecting group C of historical weft yarn data, the historical weft yarn data includes historical environmental humidity and historical weft yarn humidity influence coefficient, building a weft yarn production influence coefficient model based on the historical weft yarn data, taking the historical environmental humidity as the input of the weft yarn production influence coefficient model, taking the weft yarn humidity influence coefficient as the output of the weft yarn production influence coefficient model, taking the historical weft yarn humidity influence coefficient as the prediction target of the weft yarn production influence coefficient model, taking the recall rate function as the loss function of the weft yarn production influence coefficient model, taking minimizing the loss function value of the weft yarn production influence coefficient model as the training target, and stopping the training when the loss function value of the weft yarn production influence coefficient model is less than or equal to the preset target loss value;

[0023] The weft yarn production influence coefficient model is a linear regression model, and the recall function formula is:

[0024] Among them, Recall is the recall rate, TP is the number of samples correctly predicted by the weft yarn production influence coefficient model, and FN is the number of samples incorrectly predicted by the weft yarn production influence coefficient model;

[0025] The real-time ambient humidity is collected, and based on the constructed latitude production influence coefficient model, the real-time ambient humidity is used as input to obtain the weft yarn humidity influence coefficient.

[0026] Furthermore, the specific method of performing demand calculation includes:

[0027] The calculation formula for warp yarn demand is: Among them, D l represents the warp yarn demand corresponding to the lth valid order data, Wid represents the fabric width, indicating the lateral width of the fabric, Dmi represents the warp yarn density, Swi represents the warp yarn spacing, α T represents the warp yarn temperature influence coefficient, T represents the real-time ambient temperature, T0 represents the standard ambient temperature, and l represents the index of the valid order data;

[0028] The formula for calculating the demand for weft yarn is: Among them, E l represents the weft yarn demand corresponding to the lth valid order data, Len represents the length of the finished cloth, Wmi represents the weft yarn density, β H Represents the weft yarn humidity influence coefficient.

[0029] Furthermore, the specific steps of performing group axis allocation include:

[0030] An axle assembly optimization model is constructed, which includes the target optimization function Mdest and constraints. An optimization solution is performed based on the axle assembly optimization model to obtain the optimal axle assembly solution.

[0031] Furthermore, the objective optimization function

[0032] Where λ1 represents the warp balance coefficient, λ2 represents the weft balance coefficient; N j represents the total amount of warp yarn provided by the jth group of axes, N max Represents the maximum number of warps for a single device; Q j represents the total amount of weft yarn provided by the jth group of axes, Q max represents the maximum number of weft yarns in a single group of axes, m represents the total number of axes, j represents the index of the axis, and n represents the amount of valid order data;

[0033] The constraints include: a total warp yarn quantity constraint, a total weft yarn quantity constraint, a warp yarn demand constraint, and a weft yarn demand constraint;

[0034] The total warp yarn constraint is N j ≤N max ; The total amount of weft yarn is constrained by Q j ≤Q max ; The warp yarn demand constraint is The weft yarn demand constraint is

[0035] Furthermore, the specific steps of performing the optimization solution include:

[0036] Step 1: preset K groups of random axis grouping schemes, each group of axis grouping schemes is represented by a chromosome, and the genes in the chromosome correspond to the axis grouping allocation of each valid order data;

[0037] Step 2: For each chromosome, the fitness of each chromosome is calculated according to the target optimization function, and the parent chromosome is selected by the tournament selection algorithm according to the fitness value;

[0038] Step 3: According to the parent chromosome, the Uth gene is preset as the exchange node, and the genes of the parent chromosome after the exchange node are exchanged to generate the daughter chromosome;

[0039] Step 4: Using a random mutation algorithm, randomly select a gene from the offspring chromosome and replace it with the gene at the corresponding position of the pre-stored chromosome;

[0040] Step 5: Calculate the fitness of the offspring chromosome according to the target optimization function. When the fitness value of the offspring chromosome is greater than the fitness value of the pre-stored chromosome, replace the pre-stored chromosome with the offspring chromosome as the new pre-stored chromosome. When the maximum number of iterations has not been reached and the fitness value of the pre-stored chromosome is less than the preset fitness threshold, return to step 2 to continue iterating. When the maximum number of iterations is reached or the fitness value of the pre-stored chromosome is greater than the preset fitness threshold, stop the iteration and output the pre-stored chromosome as the optimal axis combination solution.

[0041] The method for optimizing weaving production shaft assembly of the present invention comprises: S1, collecting order data;

[0042] S2. Preprocess the order data to obtain valid order data;

[0043] S3, obtaining the warp yarn temperature influence coefficient based on the constructed warp yarn production influence coefficient model, obtaining the weft yarn humidity influence coefficient based on the constructed weft yarn production influence coefficient model, performing demand calculation through the warp yarn temperature influence coefficient, the weft yarn humidity influence coefficient and the valid order data, and obtaining the warp yarn demand and the weft yarn demand;

[0044] S4. Perform axis grouping allocation based on warp yarn demand and weft yarn demand to obtain the optimal axis grouping solution.

[0045] The technical effects and advantages of the weaving production shaft optimization system and method of the present invention are as follows:

[0046] The present invention ensures the validity and accuracy of input data by preprocessing order data, reduces the impact of errors on the final shaft assembly plan, and improves the stability and reliability of the entire system; by building a model to predict the impact of ambient temperature and humidity on the demand for warp and weft yarns, the actual demand for warp and weft yarns is calculated more accurately, reducing waste and unnecessary adjustments in the production process; by calculating the demand for warp and weft yarns and combining an optimization algorithm to generate an optimal shaft assembly plan, dynamically adapt to different production needs, ensure maximum production efficiency, and can be adjusted according to actual conditions, avoiding production problems caused by unreasonable shaft assembly and reducing production costs. BRIEF DESCRIPTION OF THE DRAWINGS

[0047] Figure 1 It is a schematic diagram of a weaving production shaft group optimization system of the present invention;

[0048] Figure 2 It is a schematic diagram of a weaving production shaft optimization method of the present invention. DETAILED DESCRIPTION

[0049] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0050] Example 1

[0051] See also Figure 1 As shown, a weaving production shaft group optimization system described in this embodiment includes: a data collection module: collecting order data;

[0052] Data processing module: pre-process order data to obtain valid order data;

[0053] Demand generation module: obtain the warp yarn temperature influence coefficient based on the constructed warp yarn production influence coefficient model, obtain the weft yarn humidity influence coefficient based on the constructed weft yarn production influence coefficient model, calculate the demand through the warp yarn temperature influence coefficient, weft yarn humidity influence coefficient and valid order data, and obtain the warp yarn demand and weft yarn demand;

[0054] Shaft assembly scheme generation module: allocates shaft assembly based on warp yarn demand and weft yarn demand to obtain the optimal shaft assembly scheme;

[0055] Each module is connected by wired and / or wireless means to achieve data transmission between modules;

[0056] Order data includes: finished fabric length, fabric width, warp density, weft density, warp spacing and order time;

[0057] The methods for preprocessing order data include:

[0058] The order demand data is standardized by the Z-score algorithm, the differences between different data magnitudes are eliminated, and standardized order data is obtained. The scale parameter set A is preset, and the standardized order data is filtered and transformed to retain the main trend of the standardized order data. The formula for filtering transformation is:

[0059] Where a represents the scale parameter, and a∈A, x i represents the i-th standardized order data, τ represents the translation parameter of the standardized order data, which controls the position of the standardized order data in the time domain after filtering and transformation, and W a (x i ) represents the filtered value of the i-th standardized order data when the scale parameter is a, t represents the order time, e represents the base of the natural logarithm, and i represents the i-th standardized order data;

[0060] Calculate the dynamic adaptability weight of standardized order data under different scale parameters to improve the adaptability and accuracy of the processing results. The calculation formula of the dynamic adaptability weight is:

[0061] Among them, η a Represents the dynamic adaptive weight corresponding to the scale parameter a. α is the adjustment factor used to control the weight difference between different scales. The larger the value, the more obvious the weight difference. Var(W a (x i )) represents the variance of the filtered value of the i-th standardized order data when the scale parameter is a;

[0062] The filter value is weighted according to the dynamic adaptive weight. The weighted calculation formula is: in, Represents the weighted sum of the filtered values ​​of the ith standardized order data after filtering and transformation by all scale parameters; the residual of the standardized order data is calculated based on the weighted sum, and the residual calculation formula is: where r i Represents the residual of the i-th standardized order data; sort all the calculated residuals from large to small, and obtain the median Mid of the residuals;

[0063] Preset the abnormal sensitivity parameter β and calculate the dynamic threshold. The calculation formula of the dynamic threshold is: R i =β×|r i -Mid|; where R i represents the dynamic threshold of the i-th standardized order data; the dynamic threshold is used to eliminate anomalies and obtain valid order data; when r i >R i When , the order data corresponding to the i-th standardized order data is marked as abnormal data, and all order data marked as abnormal data are removed;

[0064] The construction methods of the warp yarn production influence coefficient model include:

[0065] Collect historical warp yarn data of group B, the historical warp yarn data includes historical ambient temperature and historical warp yarn temperature influence coefficient, build a warp yarn production influence coefficient model based on the historical warp yarn data, use historical ambient temperature as input of the warp yarn production influence coefficient model, use warp yarn temperature influence coefficient as output of the warp yarn production influence coefficient model, use historical warp yarn temperature influence coefficient as prediction target of the warp yarn production influence coefficient model, use root mean square error function as loss function of the warp yarn production influence coefficient model, use minimization of loss function value of the warp yarn production influence coefficient model as training target, and stop training when the loss function value of the warp yarn production influence coefficient model is less than or equal to the preset target loss value;

[0066] The warp yarn production influence coefficient model is a multiple regression model or a logistic regression model, and the root mean square error function formula is: Where RMSE is the root mean square error, Su is the total amount of historical warp data, and F b is the bth historical warp temperature influence coefficient, F' b is the temperature influence coefficient of the bth warp yarn;

[0067] The historical warp yarn temperature influence coefficient is obtained by collecting the historical predicted warp yarn demand data and the historical actual warp yarn demand data under the same historical ambient temperature of group P, taking the historical warp yarn temperature influence coefficient as the optimization variable, and solving the optimal historical warp yarn temperature influence coefficient through the Bayesian optimization algorithm;

[0068] The construction methods of the weft yarn production influence coefficient model include:

[0069] Collecting group C of historical weft yarn data, the historical weft yarn data includes historical environmental humidity and historical weft yarn humidity influence coefficient, building a weft yarn production influence coefficient model based on the historical weft yarn data, taking the historical environmental humidity as the input of the weft yarn production influence coefficient model, taking the weft yarn humidity influence coefficient as the output of the weft yarn production influence coefficient model, taking the historical weft yarn humidity influence coefficient as the prediction target of the weft yarn production influence coefficient model, taking the recall rate function as the loss function of the weft yarn production influence coefficient model, taking minimizing the loss function value of the weft yarn production influence coefficient model as the training target, and stopping the training when the loss function value of the weft yarn production influence coefficient model is less than or equal to the preset target loss value;

[0070] The weft yarn production influence coefficient model is a linear regression model, and the recall function formula is:

[0071] Where Recall is the recall rate, TP is the number of samples correctly predicted by the weft yarn production influence coefficient model, and FN is the number of samples incorrectly predicted by the weft yarn production influence coefficient model;

[0072] The historical weft yarn humidity influence coefficient is obtained by collecting the historical predicted weft yarn demand data and the historical actual weft yarn demand data under the same historical environmental humidity of group P, taking the historical weft yarn humidity influence coefficient as the optimization variable, and solving the optimal historical weft yarn humidity influence coefficient through the Bayesian optimization algorithm;

[0073] Ways to perform demand calculations include:

[0074] The real-time ambient temperature is collected, and based on the constructed warp yarn production influence coefficient model, the real-time ambient temperature is used as input to obtain the warp yarn temperature influence coefficient; the influence of ambient temperature on the warp yarn demand is considered to ensure the accuracy of the production plan; based on the warp yarn temperature influence coefficient and valid order data, the warp yarn demand is calculated. The calculation formula for the warp yarn demand is: Among them, D l represents the warp yarn demand corresponding to the lth valid order data, Wid represents the fabric width, indicating the lateral width of the fabric, Dmi represents the warp yarn density, Swi represents the warp yarn spacing, α T represents the warp yarn temperature influence coefficient, T represents the real-time ambient temperature, T0 represents the standard ambient temperature, and l represents the index of the valid order data;

[0075] The real-time ambient humidity is collected, and based on the constructed latitude production impact coefficient model, the real-time ambient humidity is used as input to obtain the weft yarn humidity impact coefficient; based on the weft yarn humidity impact coefficient and valid order data, the weft yarn demand is calculated. The calculation formula for the weft yarn demand is:

[0076] Among them, E l represents the weft yarn demand corresponding to the lth valid order data, Len represents the length of the finished cloth, Wmi represents the weft yarn density, β H Represents the influence coefficient of weft yarn humidity; considering the influence of environmental factors, the influence coefficient is dynamically adjusted through the model to reduce production errors caused by environmental changes;

[0077] The specific methods for generating the optimal axis combination solution include:

[0078] Define the target optimization function Mdest,

[0079] Where λ1 represents the warp balance coefficient, λ2 represents the weft balance coefficient; N j represents the total amount of warp yarn provided by the jth group of axes, N max Represents the maximum number of warps for a single device (device parameters are obtained by reading device information); Q j represents the total amount of weft yarn provided by the jth group of axes, Q max Represents the maximum number of weft yarns for a single group of axes (device parameters are obtained by reading device information), m represents the total number of axes, j represents the axis, and n represents the amount of valid order data;

[0080] The constraints of the objective optimization function include: total warp yarn quantity constraint, total weft yarn quantity constraint, warp yarn demand constraint and weft yarn demand constraint.

[0081] The total warp yarn constraint is N j ≤N max ; The total amount of weft yarn is constrained by Q j ≤Q max ; The warp yarn demand constraint is The weft yarn demand constraint is

[0082] The warp balance coefficient and weft balance coefficient are adjusted according to different production demands to ensure the flexibility and balance of the production plan under different demands. When the matching degree of the warp demand is greater than or equal to the matching degree of the weft demand, λ1 is greater than λ2; when the matching degree of the warp demand is less than the matching degree of the weft demand, λ1 is less than λ2; the matching degree of the warp demand is the consistency between the warp demand and the actual number of warp yarns provided, and the matching degree of the weft demand is the consistency between the weft demand and the actual number of weft yarns provided;

[0083] The target optimization function is optimized and solved. The specific steps of the optimization solution include:

[0084] Step 1: preset K groups of random axis grouping schemes, each group of axis grouping schemes is represented by a chromosome, and the genes in the chromosome correspond to the axis grouping allocation of each valid order data;

[0085] Step 2: For each chromosome, the fitness of each chromosome is calculated according to the target optimization function. The chromosome with a larger fitness value is better. The parent chromosome is selected through the tournament selection algorithm according to the fitness value.

[0086] Step 3: According to the parent chromosome, the Uth gene is preset as the exchange node, and the genes of the parent chromosome after the exchange node are exchanged to generate the daughter chromosome;

[0087] Step 4: Using a random mutation algorithm, randomly select a gene from the offspring chromosome and replace it with the gene at the corresponding position of the pre-stored chromosome (preset initial solution);

[0088] Step 5: Calculate the fitness of the offspring chromosome according to the target optimization function. When the fitness value of the offspring chromosome is greater than the fitness value of the pre-stored chromosome, replace the pre-stored chromosome with the offspring chromosome as the new pre-stored chromosome. When the maximum number of iterations is not reached and the fitness value of the pre-stored chromosome is less than the preset fitness threshold, return to step 2 to continue iterating. When the maximum number of iterations is reached or the fitness value of the pre-stored chromosome is greater than the preset fitness threshold, stop iterating and output the pre-stored chromosome as the optimal axis combination scheme.

[0089] The fitness calculation method includes: based on the target optimization function, the fitness is calculated using the reverse proportional fitness function. The fitness calculation formula is: Where MinF represents the value of the target optimization function, Fit represents the fitness, and ε is a very small positive number used to avoid a division by zero error caused by a MinF that is too small;

[0090] Through optimization and multiple iterations, the final output shaft assembly solution is ensured to be the global optimal one, avoiding the trap of local optimal solution, and finding the optimal shaft assembly solution quickly and effectively to meet the ever-changing needs and environmental conditions in production;

[0091] This embodiment ensures the validity and accuracy of input data by preprocessing order data, reduces the impact of errors on the final axis assembly plan, and improves the stability and reliability of the entire system; by building a model to predict the impact of ambient temperature and humidity on the demand for warp and weft yarns, the actual demand for warp and weft yarns is calculated more accurately, reducing waste and unnecessary adjustments in the production process; by calculating the demand for warp and weft yarns and combining the optimization algorithm to generate the optimal axis assembly plan, dynamically adapt to different production needs, ensure maximum production efficiency, and can be adjusted according to actual conditions, avoiding production problems caused by unreasonable axis assembly and reducing production costs.

[0092] Example 2

[0093] See also Figure 2 As shown, the part not described in detail in this embodiment can be found in the description of embodiment 1, which provides a method for optimizing the axis group of weaving production, including: S1, collecting order data;

[0094] S2. Preprocess the order data to obtain valid order data;

[0095] S3, obtaining the warp yarn temperature influence coefficient based on the constructed warp yarn production influence coefficient model, obtaining the weft yarn humidity influence coefficient based on the constructed weft yarn production influence coefficient model, performing demand calculation through the warp yarn temperature influence coefficient, the weft yarn humidity influence coefficient and the valid order data, and obtaining the warp yarn demand and the weft yarn demand;

[0096] S4. Perform axis grouping allocation based on warp yarn demand and weft yarn demand to obtain the optimal axis grouping solution.

[0097] Example 3

[0098] This embodiment discloses an electronic device, including a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the computer program, the operation mode of the weaving production group axis optimization method provided above is implemented.

[0099] Since the electronic device introduced in this embodiment is an electronic device used to implement a weaving production shaft optimization method in the embodiment of the present application, based on the weaving production shaft optimization method introduced in the embodiment of the present application, the technical personnel of the field can understand the specific implementation of the electronic device of the present embodiment and its various variations, so how the electronic device implements the method in the embodiment of the present application is not described in detail here. As long as the technical personnel of the field implement the electronic device used in the weaving production shaft optimization method in the embodiment of the present application, it belongs to the scope of protection of the present application.

[0100] The above formulas are all dimensionless and numerical calculations. The formula is a formula for the most recent real situation obtained by collecting a large amount of data and performing software simulation. The preset parameters and thresholds in the formula are set by technicians in this field according to actual conditions.

[0101] The above is only a preferred embodiment of the present invention, and the protection scope of the present invention is not limited to the above embodiments. All technical solutions under the concept of the present invention belong to the protection scope of the present invention. It should be pointed out that for ordinary technical users in this technical field, some improvements and modifications without departing from the principle of the present invention should also be regarded as the protection scope of the present invention.

Claims

1. A weaving production shaft optimization system, characterized in that: include: Data collection module: collect order data; Data processing module: pre-process order data to obtain valid order data; Demand generation module: obtain the warp yarn temperature influence coefficient based on the constructed warp yarn production influence coefficient model, obtain the weft yarn humidity influence coefficient based on the constructed weft yarn production influence coefficient model, calculate the demand through the warp yarn temperature influence coefficient, weft yarn humidity influence coefficient and valid order data, and obtain the warp yarn demand and weft yarn demand; Axis grouping scheme generation module: Axis grouping allocation is performed based on the warp yarn demand and weft yarn demand to obtain the optimal axis grouping scheme.

2. A weaving production shaft optimization system according to claim 1, characterized in that: The order data includes: finished fabric length, fabric width, warp yarn density, weft yarn density, warp yarn spacing and order time.

3. A weaving production shaft optimization system according to claim 2, characterized in that: The specific method of preprocessing the order data includes: The order demand data is standardized by the Z-score algorithm to obtain standardized order data. The scale parameter set A is preset and the standardized order data is filtered and transformed. The formula for filtering transformation is: Where a represents the scale parameter, and a∈A, x i represents the i-th standardized order data, τ represents the translation parameter of the standardized order data, W a (x i ) represents the filtered value of the i-th standardized order data when the scale parameter is a, t represents the order time, e represents the base of the natural logarithm, and i represents the i-th standardized order data; Calculate the dynamic adaptability weight of the standardized order data under different scale parameters. The calculation formula of the dynamic adaptability weight is: Among them, η a represents the dynamic adaptability weight corresponding to the scale parameter a, α is the adjustment factor, Var(W a (x i )) represents the variance of the filtered value of the i-th standardized order data when the scale parameter is a; The filter value is weighted according to the dynamic adaptive weight. The weighted calculation formula is: in Represents the weighted sum of the filtered values ​​of the ith standardized order data after filtering transformation; the residual of the standardized order data is calculated based on the weighted sum, and the residual calculation formula is: where r i Represents the residual of the i-th standardized order data; sort all the calculated residuals from large to small, and obtain the median Mid of the residuals; Preset the abnormal sensitivity parameter β and calculate the dynamic threshold. The calculation formula of the dynamic threshold is: R i =β×|r i -Mid|; where R i represents the dynamic threshold of the i-th standardized order data; the abnormality is eliminated through the dynamic threshold to obtain valid order data; the abnormality elimination method includes: when r i >R i When , the order data corresponding to the i-th standardized order data is marked as abnormal data, and all order data marked as abnormal data are removed.

4. A weaving production shaft optimization system according to claim 3, characterized in that: The method for obtaining the warp yarn temperature influence coefficient includes: Collect historical warp yarn data of group B, the historical warp yarn data includes historical ambient temperature and historical warp yarn temperature influence coefficient, build a warp yarn production influence coefficient model based on the historical warp yarn data, use historical ambient temperature as input of the warp yarn production influence coefficient model, use warp yarn temperature influence coefficient as output of the warp yarn production influence coefficient model, use historical warp yarn temperature influence coefficient as prediction target of the warp yarn production influence coefficient model, use root mean square error function as loss function of the warp yarn production influence coefficient model, use minimization of loss function value of the warp yarn production influence coefficient model as training target, and stop training when the loss function value of the warp yarn production influence coefficient model is less than or equal to the preset target loss value; The warp yarn production influence coefficient model is a multiple regression model or a logistic regression model, and the root mean square error function formula is: Among them, RMSE is the root mean square error, Su is the total amount of historical warp data, and F b is the bth historical warp temperature influence coefficient, F' b is the temperature influence coefficient of the bth warp yarn; The real-time ambient temperature is collected, and based on the constructed warp yarn production influence coefficient model, the real-time ambient temperature is used as input to obtain the warp yarn temperature influence coefficient.

5. A weaving production shaft optimization system according to claim 4, characterized in that: The method for obtaining the weft yarn humidity influence coefficient includes: Collecting group C of historical weft yarn data, the historical weft yarn data includes historical environmental humidity and historical weft yarn humidity influence coefficient, building a weft yarn production influence coefficient model based on the historical weft yarn data, taking the historical environmental humidity as the input of the weft yarn production influence coefficient model, taking the weft yarn humidity influence coefficient as the output of the weft yarn production influence coefficient model, taking the historical weft yarn humidity influence coefficient as the prediction target of the weft yarn production influence coefficient model, taking the recall rate function as the loss function of the weft yarn production influence coefficient model, taking minimizing the loss function value of the weft yarn production influence coefficient model as the training target, and stopping the training when the loss function value of the weft yarn production influence coefficient model is less than or equal to the preset target loss value; The weft yarn production influence coefficient model is a linear regression model, and the recall function formula is: Among them, Recall is the recall rate, TP is the number of samples correctly predicted by the weft yarn production influence coefficient model, and FN is the number of samples incorrectly predicted by the weft yarn production influence coefficient model; The real-time ambient humidity is collected, and based on the constructed latitude production influence coefficient model, the real-time ambient humidity is used as input to obtain the weft yarn humidity influence coefficient.

6. A weaving production shaft optimization system according to claim 5, characterized in that: The specific method of performing demand calculation includes: The calculation formula for warp yarn demand is: Among them, D l represents the warp yarn demand corresponding to the lth valid order data, Wid represents the fabric width, indicating the lateral width of the fabric, Dmi represents the warp yarn density, Swi represents the warp yarn spacing, α T represents the warp yarn temperature influence coefficient, T represents the real-time ambient temperature, T0 represents the standard ambient temperature, and l represents the index of the valid order data; The formula for calculating the demand for weft yarn is: Among them, E l represents the weft yarn demand corresponding to the lth valid order data, Len represents the length of the finished cloth, Wmi represents the weft yarn density, β H Represents the weft yarn humidity influence coefficient.

7. A weaving production shaft optimization system according to claim 6, characterized in that: The specific steps of performing group axis allocation include: Construct an axle assembly optimization model, which includes the target optimization function Mdest and constraints; perform optimization and solution based on the axle assembly optimization model to obtain the optimal axle assembly solution.

8. A weaving production shaft optimization system according to claim 7, characterized in that: The objective optimization function Where λ1 represents the warp balance coefficient, λ2 represents the weft balance coefficient; N j represents the total amount of warp yarn provided by the jth group of axes, N max Represents the maximum number of warps for a single device; Q j represents the total amount of weft yarn provided by the jth group of axes, Q max represents the maximum number of weft yarns in a single group of axes, m represents the total number of axes, j represents the index of the axis, and n represents the amount of valid order data; The constraints include: a total warp yarn quantity constraint, a total weft yarn quantity constraint, a warp yarn demand constraint, and a weft yarn demand constraint; The total warp yarn constraint is N j ≤N max ; The total amount of weft yarn is constrained by Q j ≤Q max ; The warp yarn demand constraint is The weft yarn demand constraint is 9. A weaving production shaft optimization system according to claim 8, characterized in that: The specific steps of performing the optimization solution include: Step 1: preset K groups of random axis grouping schemes, each group of axis grouping schemes is represented by a chromosome, and the genes in the chromosome correspond to the axis grouping allocation of each valid order data; Step 2: For each chromosome, the fitness of each chromosome is calculated according to the target optimization function, and the parent chromosome is selected by the tournament selection algorithm according to the fitness value; Step 3: According to the parent chromosome, the Uth gene is preset as the exchange node, and the genes of the parent chromosome after the exchange node are exchanged to generate the daughter chromosome; Step 4: Using a random mutation algorithm, randomly select a gene from the offspring chromosome and replace it with the gene at the corresponding position of the pre-stored chromosome; Step 5: Calculate the fitness of the offspring chromosome according to the target optimization function. When the fitness value of the offspring chromosome is greater than the fitness value of the pre-stored chromosome, replace the pre-stored chromosome with the offspring chromosome as the new pre-stored chromosome. When the maximum number of iterations has not been reached and the fitness value of the pre-stored chromosome is less than the preset fitness threshold, return to step 2 to continue iterating. When the maximum number of iterations is reached or the fitness value of the pre-stored chromosome is greater than the preset fitness threshold, stop the iteration and output the pre-stored chromosome as the optimal axis combination solution.

10. A method for optimizing weaving production shaft assembly, which is implemented based on a weaving production shaft assembly optimization system according to any one of claims 1 to 9, characterized in that: include: S1. Collect order data; S2. Preprocess the order data to obtain valid order data; S3, obtaining the warp yarn temperature influence coefficient based on the constructed warp yarn production influence coefficient model, obtaining the weft yarn humidity influence coefficient based on the constructed weft yarn production influence coefficient model, performing demand calculation through the warp yarn temperature influence coefficient, the weft yarn humidity influence coefficient and the valid order data, and obtaining the warp yarn demand and the weft yarn demand; S4. Perform axis grouping allocation based on warp yarn demand and weft yarn demand to obtain the optimal axis grouping solution.

Images