Fabric setting equipment running parameter optimization recommendation method, device, equipment and medium

Abstract

The invention discloses a fabric setting equipment running machine parameter optimization recommendation method and device, equipment and a medium, and the method comprises the steps: carrying out the preprocessing and standardization operation of collected fabric setting equipment running machine parameter data, extracting batch process characteristic data, calculating an optimal parameter value according to the process characteristic data of a stable interval, and carrying out the calculation of the optimal parameter value; and associating the optimal parameter value with fabric category information of a batch, matching the fabric category information with process data items in a process library when a new production batch is detected, if the matching is successful, recommending line machine parameter data, and if the matching is not successful, performing prediction through an equipment learning model. According to the technical scheme of the invention, the technological parameters can be quickly output through the system recommendation, and meanwhile, the quality problem caused by personal errors is reduced.

Description

Technical Field

[0001] The present invention relates to the technical field of machine learning, and in particular to a method, device, equipment and medium for optimizing and recommending machine parameters of a fabric shaping device. Background Art

[0002] In the textile industry, shaping equipment is a key post-processing equipment to ensure the dimensional stability and functionality of fabrics. The specifications of fabrics are controlled by preset parameters such as temperature, chain speed, overfeeding, etc. At present, most of the equipment has not realized equipment networking and data storage. Although some data collection equipment and shaping equipment central control systems have appeared on the market in recent years, the simple data does not completely solve the problem of maintaining and optimizing shaping process parameters. Dyeing factories lack a systematic solution to automatically collect and analyze data during the production process and effectively convert these data into process parameters. This results in a lot of time to set parameters every time a new product is encountered, and most of the time it depends on the experience of the shaping process master. Even for the same batch of fabrics, the shaping equipment parameters must be manually reset by the operator before each production. This operation method is not only inefficient, but also easily leads to inconsistent quality of each batch. Summary of the invention

[0003] The embodiments of the present invention provide a method, device, equipment and medium for optimizing and recommending machine parameters of a fabric setting device, aiming to solve the problems of low efficiency and unstable batch quality caused by the need to reset parameters for each production batch in the prior art.

[0004] In a first aspect, an embodiment of the present invention provides a method for optimizing and recommending machine parameters of a fabric setting device, the method comprising:

[0005] The machine parameter data of a certain production batch of fabric shaping equipment collected by the gateway is transmitted to the data warehouse;

[0006] Preprocessing and standardizing the machine parameter data in the data warehouse;

[0007] The machine parameter data is segmented according to the start time and end time of the batch, the process characteristic data of the batch is extracted, and the stable interval of the process characteristic data during operation is screened out according to the set time window;

[0008] Calculating an optimal parameter value according to the process characteristic data of the stable interval, associating the optimal parameter value with the fabric category information of the batch, forming a process data entry and storing it in a process library;

[0009] When a new production batch is detected, the fabric category information of the new batch input is matched with the process data entries in the process library. If the match is successful, the machine parameter data of the process data entries are recommended. If the match is unsuccessful, a prediction is performed through the equipment learning model, wherein the model learns the relationship between the machine parameter data of historical batches and the production effect, and outputs the recommended optimal machine parameter data.

[0010] In a second aspect, an embodiment of the present invention provides a device for optimizing and recommending machine parameters of a fabric setting device, the device comprising:

[0011] A data collection unit, used to transmit the machine parameter data of a certain production batch of fabric shaping equipment collected by the gateway to the data warehouse;

[0012] A data preprocessing unit, used for preprocessing and standardizing the machine parameter data in the data warehouse;

[0013] A data feature extraction unit, used to process the machine parameter data in sections according to the start time and end time of the batch, extract the process feature data of the batch, and filter out the stable interval of the process feature data during operation according to a set time window;

[0014] A data association unit, used for calculating an optimal parameter value according to the process characteristic data of the stable interval, and associating the optimal parameter value with the fabric category information of the batch to form a process data entry and store it in a process library;

[0015] The data recommendation unit is used to match the fabric category information of the new batch input with the process data entries in the process library when a new production batch is detected. If the match is successful, the machine parameter data of the process data entries is recommended. If the match is unsuccessful, prediction is performed through the equipment learning model, wherein the relationship between the machine parameter data of historical batches and the production effect is learned by the model, and the optimal machine parameter data is recommended.

[0016] In a third aspect, an embodiment of the present invention further provides a computer device, which includes a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein when the processor executes the computer program, the method for optimizing and recommending machine parameters of the fabric shaping equipment described in the first aspect is implemented.

[0017] In a fourth aspect, an embodiment of the present invention further provides a computer-readable storage medium, wherein the computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, the processor executes the method for optimizing the machine parameters of the fabric shaping equipment described in the first aspect.

[0018] The embodiment of the present invention provides a method, device, equipment and medium for optimizing and recommending the running parameters of a fabric shaping equipment. The method comprises: preprocessing and standardizing the collected running parameter data of the fabric shaping equipment, extracting the process characteristic data of the batch, calculating the optimal parameter value according to the process characteristic data in the stable interval, associating the optimal parameter value with the fabric category information of the batch, and when a new production batch is detected, matching the fabric category information with the process data entry in the process library, if the match is successful, recommending the running parameter data, and if the match is unsuccessful, predicting through the equipment learning model, and outputting the recommended optimal running parameter data. The above method, through process data feature extraction and associating the optimal parameter value with the fabric category information of the batch, can output the recommended optimal running parameter data by matching with the process data entry in the process library or predicting using the equipment learning model when a new batch is detected. The present application solves the problems of low efficiency and unstable batch quality caused by the need to reset parameters for new production batches. BRIEF DESCRIPTION OF THE DRAWINGS

[0019] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the accompanying drawings required for use in the description of the embodiments will be briefly introduced below. Obviously, the accompanying drawings described below are some embodiments of the present invention. For ordinary technicians in this field, other accompanying drawings can be obtained based on these accompanying drawings without paying any creative work.

[0020] Figure 1 A diagram of the application environment of the method for optimizing the machine parameters of a fabric shaping device provided by an embodiment of the present invention;

[0021] Figure 2 A schematic diagram of a flow chart of a method for optimizing machine parameters of a fabric setting equipment provided by an embodiment of the present invention;

[0022] Figure 3 A structural diagram of a method for optimizing and recommending machine parameters of a fabric shaping device provided by an embodiment of the present invention;

[0023] Figure 4 A schematic diagram of a sub-flow chart of a method for optimizing and recommending machine parameters of a fabric setting equipment provided by an embodiment of the present invention;

[0024] Figure 5 A schematic diagram of another sub-flow of the method for optimizing and recommending machine parameters of a fabric setting device provided by an embodiment of the present invention;

[0025] Figure 6 A diagram of a recommended device for optimizing machine parameters of a fabric shaping device provided by an embodiment of the present invention;

[0026] Figure 7 A schematic block diagram of a computer device provided in an embodiment of the present invention. DETAILED DESCRIPTION

[0027] 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 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.

[0028] It should be understood that when used in this specification and the appended claims, the terms "include" and "comprises" indicate the presence of described features, integers, steps, operations, elements and / or components, but do not exclude the presence or addition of one or more other features, integers, steps, operations, elements, components and / or combinations thereof.

[0029] It should also be understood that the terms used in this specification of the present invention are only for the purpose of describing specific embodiments and are not intended to limit the present invention. As used in the specification of the present invention and the appended claims, unless the context clearly indicates otherwise, the singular forms "a", "an" and "the" are intended to include plural forms.

[0030] It should be further understood that the term "and / or" used in the present description and the appended claims refers to any and all possible combinations of one or more of the associated listed items, and includes these combinations.

[0031] In order to facilitate the understanding of the present application scheme, the application environment applicable to the embodiments of the present application is first introduced.

[0032] Figure 1 FIG. 1 is a schematic diagram of an application scenario of a method for optimizing machine parameters of a fabric setting device in an embodiment. Figure 1As shown, the application environment includes a gateway 1000 for collecting the row machine parameter data of the fabric setting equipment, a database 2000-200N, a server 3000 and a display terminal 4000. The fabric setting equipment row machine parameter optimization recommendation method is applied to the server 3000. The server 3000 can exchange data with the database 2000-200N through the network and save it in the data processing server. The server 3000 can be a cloud server or a local server. The row machine parameters output by the server calculation and processing are displayed in the display terminal 4000. The user can adjust the data. If the confirmation is completed, it can be sent to the fabric setting equipment through the gateway 1000. The PLC of the fabric setting equipment controls the oven temperature, the row machine chain speed, the upper and lower overfeeding and other control equipment operations according to the input parameters, and outputs the recommended optimal row machine parameter data through the process library, which solves the problem of low efficiency and unstable batch quality caused by the need to reset the parameters for new production batches.

[0033] The following specific embodiments are used to describe in detail the technical solution of the present application and how the technical solution of the present application solves the above technical problems. The following specific embodiments can be combined with each other, and the same or similar concepts or processes may not be repeated in some embodiments.

[0034] The embodiments of the present application will be described below in conjunction with the accompanying drawings.

[0035] Figure 2 Schematic diagram of a method for optimizing and recommending parameters of a fabric shaping device in an embodiment, wherein the method for optimizing and recommending parameters of a fabric shaping device can be executed on a local server or a cloud server. The cloud server can be implemented in software and / or hardware, and can be composed of two or more physical entities or one physical entity.

[0036] Specifically, Figure 2 and Figure 3 As shown, the method for optimizing the recommended parameters of the fabric setting equipment may include the following steps:

[0037] S110, transmitting the machine parameter data of a certain production batch of fabric shaping equipment collected by the gateway to the data warehouse.

[0038] At present, most shaping equipment does not have the ability to communicate with other systems. A data acquisition gateway is installed on the shaping equipment. The gateway is responsible for real-time acquisition of process parameter data during the production process of the shaping equipment. In one embodiment, the gateway is connected to the shaping equipment through a specific interface protocol (such as Modbus, OPC UA, etc.) to achieve data acquisition. The amount of data generated during the operation of the shaping equipment is huge, with an average of 150-250 IO points. The acquisition frequency is as fast as 1 second and as high as 60 seconds. Otherwise, the accuracy will be lost during the data acquisition process.

[0039] Specifically, the collected data include key process parameters such as shaping equipment speed, temperature, humidity, tension, wind speed, upper and lower overfeeding, etc. The data collection frequency is set between 1 second and 30 seconds to ensure the accuracy and timeliness of data collection.

[0040] Specifically, the collected real-time data is transmitted to the data warehouse through the gateway. In order to ensure the stability and reliability of data transmission, the gateway must have data caching and breakpoint resumption functions to ensure that data will not be lost when the network is unstable.

[0041] It should be noted that a data warehouse is a subject-oriented, integrated, relatively stable data collection that reflects historical changes and is used to support management decisions. A data warehouse can store a large amount of historical data, including structured data, semi-structured data, and unstructured data. Whether it is an enterprise's order data, equipment data, energy data, or process formula data, it can all be stored and managed in a data warehouse.

[0042] The data warehouse function includes data preprocessing, data cleaning, etc., and performs quality inspection and cleaning on the data entering the data warehouse to ensure the accuracy, completeness and consistency of the data. For example, check whether the data has missing values, abnormal values, duplicate values ​​and other problems, and perform corresponding processing, regularly evaluate and monitor the quality of the data in the data warehouse, and promptly discover and solve data quality problems. The data of equipment operation is uploaded to the data warehouse, solving the problem of data recording and retention.

[0043] S120, preprocessing and standardizing the machine parameter data in the data warehouse.

[0044] It should be noted that in the data warehouse, data preprocessing and standardization operations are very important links. The main purpose is to improve data quality and ensure data consistency and availability.

[0045] In one embodiment, if Figure 4 As shown, step S120 includes sub-steps S121 and S123.

[0046] S121, the preprocessing operation includes removing redundant data, filling missing values, correcting abnormal values, and filtering extreme abnormal data caused by sensor failure or noise interference.

[0047] It should be noted that there may be exactly the same records in the data set, which may be caused by data input errors, repeated data collection, etc. For example, in a data list, there may be multiple records representing the same device and the same collection point, and the recording time and value of these data are exactly the same. It also includes near-duplicate data: some data records may be very similar in some fields, but not exactly the same, such as the speed of fan 1 and fan 2, etc. Furthermore, it also includes irrelevant data. Some data fields may be irrelevant to specific analysis tasks. These fields can be removed to reduce the amount of data and improve the efficiency of data analysis. Methods for removing redundant data include: deduplication processing, data similarity calculation and feature selection, etc.

[0048] Specifically, in a data set, some fields of some data records may have missing values. Missing values ​​may be caused by errors in the data collection process, data loss, data not recorded, etc. Generally, by adopting constant filling, mean filling, median filling, regression filling and interpolation filling methods, this embodiment adopts the mean filling method; outliers refer to data points that are significantly different from other data in the data set. Outliers may be caused by data collection errors, data input errors, abnormal data fluctuations, etc. This embodiment adopts the method of replacing outliers and smoothing. By using moving average method, exponential smoothing method and other methods for smoothing, the outliers are gradually approached to the level of normal data.

[0049] Furthermore, a reasonable threshold is set to determine whether the data is extremely abnormal. For example, for temperature sensor data, if the temperature value exceeds three times the standard deviation of the normal range, it can be considered as extremely abnormal data. Corresponding to the extremely abnormal data, this embodiment adopts Kalman filtering, based on the state space model, to estimate the state of the system through two steps of prediction and update, specifically including establishing the state equation and observation equation of the system, and determining the initial state and covariance matrix. At each time step, prediction and update operations are performed. The prediction step predicts the state at the current moment based on the state estimate and system model at the previous moment; the update step corrects the predicted value based on the observed value at the current moment.

[0050] Furthermore, for sensor data, the sensor's measured value can be used as the observed value, and the real state of the system can be used as the state to be estimated. It can handle noise and anomalies in dynamic systems and has a good tracking ability for the changing trend of data;

[0051] S123. The standardization operation is to standardize the preprocessed data according to the process characteristics of different equipment to ensure the comparability and consistency of data between different equipment and batches.

[0052] It should be noted that the process characteristics of different equipment include equipment type, model, function, working principle, technical parameters, etc. For some classified data or discrete data, encoding standardization can be carried out so that the data of different equipment have the same encoding method, ensuring that the data of different equipment have a unified format, including data type, field length, precision, etc.

[0053] S130, segmenting the machine parameter data according to the start time and end time of the batch, extracting the process characteristic data of the batch, and screening out the stable interval of the process characteristic data during operation according to the set time window.

[0054] S140, calculating the optimal parameter value according to the process characteristic data in the stable interval, associating the optimal parameter value with the fabric category information of the batch, forming a process data entry and storing it in the process library.

[0055] In one embodiment, if Figure 5 As shown, step S140 includes sub-steps S141 and S143.

[0056] S141, extracting the value with the highest frequency of occurrence in the stable interval as the optimal parameter value of the process characteristic data.

[0057] In one of the embodiments, the corresponding equipment name is confirmed by the batch information of the operation of the shaping equipment, and the data collection point of the corresponding equipment is found through the equipment name and the start time and end time of the batch. These collection points are combined with the start and end time of the batch and the equipment name to obtain the process characteristic data of each production batch extracted from the data collected by these equipment within a specific time range. The process characteristic data includes fabric type, composition, pre-setting CPI, pre-setting WPI, etc. Different characteristic data can be defined according to different situations. According to the set time window, the data of the entire production process is analyzed, and the range with the smallest residual of each collection point is statistically calculated, and this range is the stable interval during the operation.

[0058] In this embodiment, Table 1 is an information table of the production batches of each shaping equipment within a certain period of time, including batch number, equipment name, process, start time, end time, fabric composition, target parameters before the start of shaping and parameters after the end of shaping. In one embodiment, the batch with batch number 24101207 is selected for step description. In the batch number 24101207, the shaping equipment is F03, the production process is finished product shaping, the start time is 2024 / 10 / 175:26, the end time is 2024 / 10 / 176:43, and the fabric composition is 68.2% lyocell fiber + 29.3% acrylic fiber + 2.9% spandex.

[0059] In one of the embodiments, according to the equipment name F03, the start time 2024 / 10 / 175:26 and the end time 2024 / 10 / 176:43, the collection points corresponding to the time interval and equipment name are searched in the database. There are many collection points for the equipment, including the temperature of each oven, wind speed, weft straightening, expansion, width and other parameters, as shown in Table 2. Table 2 is a list of collection points such as width and chain speed of equipment F03. Even if they are all process characteristic data of width, each oven has a corresponding collection point, and the data of each collection point is not the same. In this embodiment, the width7_SET acquisition point is selected for illustration. Table 3 is the time series data of the device F03_width7_SET acquisition point. As shown in Table 3, the width7_SET acquisition point has a total of 8 data in the time interval from 2024 / 10 / 1714:58 to 2024 / 10 / 1715:05, with the smallest residual. Compared with other intervals, the residual is the smallest, so it is selected as a stable interval. In the stable interval, the value is 174, and this value appears most frequently, so 174 is used as the optimization parameter value of the process characteristic data of the width of this batch.

[0060] In another embodiment, the autoregressive integrated moving average model ARIMA can also be used to predict the optimal value, ARIMA (p, d, q), where p represents the order of autoregressive (AR), d represents the difference (number of differences), and q represents the order of moving average (MA). AR (p) component: MA(q) component: y(t)=c+θ1*ε(t-1)+θ2*ε(t-2)+...+θp*ε(tq)+ε(t), where, are the autoregressive coefficients, θ1, θ2, ..., θp are the moving average coefficients, c is a constant, and ε(t) is the white noise error term.

[0061] Table 1 Production batch information

[0062]

[0063]

[0064] Table 2F03 Equipment data collection points

[0065] Device Name Collection point value Display value unit F03 width7_SET 174 174 cm F03 width6_SET 174 174 cm F03 width5_SET 174 174 cm F03 width4_SET 174 174 cm F03 width3_SET 174 174 cm F03 width2_SET 174 174 cm F03 width1_SET 159 159 cm F03 main_chain_SET 400 40 m / min F03 plaiter_diff_SET 0 0 ％ F03 delivery_roller_diff_SET 0 0 ％

[0066] Table 3. Timing data of the acquisition point of device F03_width7_SET

[0067] Collection point Reporting time value width7_SET 2024 / 10 / 1715:05 174 width7_SET 2024 / 10 / 1715:04 174 width7_SET 2024 / 10 / 1715:03 174 width7_SET 2024 / 10 / 1715:02 174 width7_SET 2024 / 10 / 1715:01 174 width7_SET 2024 / 10 / 1715:00 174 width7_SET 2024 / 10 / 1714:59 174 width7_SET 2024 / 10 / 1714:58 174

[0068] S143, associating the optimal parameter value with the fabric category information of the batch, the fabric category information including fabric type, fabric composition, color and color depth, and classifying and itemizing the fabric category information and storing it in the process library.

[0069] In one of the embodiments, Table 4 is an example of a process data entry. As shown in Table 4, the process data entry includes batch number AB240908001, the process is blanking, the color is light blue, the machine model is A brand shaping equipment, the machine is 3#, the fabric composition of the produced and processed batch is 68.2% lyocell fiber + 29.3% acrylic fiber + 2.5% spandex, the machine speed is 36, the front exhaust is 30, the cold wind 1 is 30, the upper overfeed is 17, and the lower overfeed is 0. In addition, the process data entry also includes many parameters, which are not listed one by one in this application. The process data entries can be classified and stored in the process library according to categories such as color, fabric composition or machine name, which is helpful for process personnel to quickly retrieve.

[0070] Table 4 Process data entries

[0071]

[0072] S150. When a new production batch is detected, the fabric category information of the new batch is matched with the process data entry in the process library. If the match is successful, the machine parameter data of the process data entry is recommended.

[0073] In one of the embodiments, when a new production batch is detected, the system automatically starts a matching procedure between the fabric category information and the process data entries in the process library.

[0074] Specifically, first, obtain the fabric category information of the new batch input. It should be noted that this information may include characteristic parameters such as fabric color, dark, medium and light classification, fiber composition, target weight, target width, etc. Further, compare these parameters with each process data entry in the process library one by one. The process data entries in the process library usually contain detailed production process information, such as machine parameter data, finishing process, etc.

[0075] In one embodiment, during the matching process, if the fabric category information of a new batch is highly consistent with the characteristic parameters of a certain process data entry, it is determined that the match is successful. Once the match is successful, the system will immediately recommend the machine parameter data of the process data entry. The machine parameter data may include key parameters such as the chain speed of the equipment, the cloth feeding setting, and the width. These parameters are crucial to ensure the quality and efficiency of the fabric during the production process.

[0076] In this way, suitable machine parameter recommendations can be quickly provided for new production batches, reducing trial and error costs and production time, and improving production efficiency and product quality. At the same time, with the continuous accumulation of production batches and the continuous improvement of the process library, the accuracy of matching and the effectiveness of recommendations will continue to improve.

[0077] Further, according to setting thresholds for target fabric type component content and target process parameters in the new production batch, traverse the process data entries in the process library according to the thresholds to recommend the closest machine parameter data;

[0078] Table 5 is the batch information to be produced and processed. In one embodiment, as shown in Table 5, the fabric composition input in the model is 68.2% lyocell fiber + 29.3% acrylic fiber + 2.5% spandex, the target width is 165, the target weight is 190 and other parameters. Assuming the same process data category in the process library, accurate matching can be performed based on the input parameters.

[0079] In another embodiment, if there are no exactly identical data categories, existing processes can be directly matched based on preset rules, such as the same or similar fiber types and process requirements. For example, the following rules are set: filter data whose fabric components contain spandex and account for 0-5% and whose pre-setting CPI is 70-75; process personnel can judge from experience that process data whose fabric components contain spandex and account for 0-5% and whose pre-setting CPI is 120-125 is also applicable to this production batch whose fabric components are: 68.2% lyocell fiber + 29.3% acrylic fiber + 2.5% spandex, and whose pre-setting CPI is 72. Table 6 is the model output parameters. According to the matching rules, the process parameters shown in Table 6 are input, including the required cloth seal of 163, the required grammage of 162, the required density CPI of 65, the required density WPI of 33, the machine speed of 36, the front exhaust of 30, the cold air of 130, the upper overfeed of 17, the lower overfeed of 0 and other parameters. These parameters have been tested in production and do not affect the production quality. They can be used for stable batch production. When receiving a new batch task, the system directly recommends the use of them to improve the efficiency of parameter setting.

[0080] Table 5 Model input parameters

[0081]

[0082] Table 6 Model output parameters

[0083] Output Parameters value describe require_cover 163 Request for machine cloth seal require_weight 162 Required machine weight require_cpi 65 Required output density CPI require_wpi 33 Required output density WPI speed 36 Machine speed front_exhaust 30 Front convulsion cold_one 30 Cold Wind 1 in_fabric 10 Insert cloth super_feeding 17 Overfeed under_feeding 0 Overfeed cold_two 30 Cold Wind 2 padder 20 Rolling car weft_traightener 20 Weft straightening bar_pressure 5 Roller pressure expansion_one 79 Roller 1 expansion console_expansion 79 Operating table expansion mop_rope 20 Mop rope temperature_five 200 Temperature 5(℃) upwind_onetwo 98 12 boxes advantage (%) downwind_onetwo 100 12 boxes downwind (%) widthdeviation_one -10 Width deviation 1(cm) widthdeviation_two 0 Width deviation 2(cm) width_reference 168 Width standard 6 (cm)

[0084] S160. If the match is unsuccessful, prediction is performed through the equipment learning model, wherein the model learns the relationship between the machine parameter data of historical batches and the production effect, and outputs the recommended optimal machine parameter data.

[0085] In one embodiment, when the optimal machine parameter data is obtained through a relatively conventional matching method, no successful matching result that meets the current needs is found. Based on the existing standard parameter templates, parameter configurations that are directly available under specific conditions in the past, etc., due to the differences between the current production scenarios, product requirements, and the situations in the existing matching library, it is impossible to directly obtain the appropriate machine parameters through this matching method.

[0086] It should be noted that the system will collect a large number of historical batches of machine parameter data, which covers various parameters set during the operation of the shaping equipment in different production batches in the past, such as the speed, temperature, pressure, overfeed and other specific parameter values ​​of the shaping equipment. At the same time, the production effect data corresponding to each historical batch must also be recorded. The production effect can be measured by a variety of indicators, such as the product's gram weight, bursting point, width and other key performance indicators of the product.

[0087] The collected historical batch machine parameter data is used as input features, the corresponding production effect data is used as the target output, and the model is built using the equipment learning algorithm. The model will try to find the inherent correlation pattern between different machine parameters and production effects by analyzing these large numbers of data points. For example, it may find that when the speed of the shaping equipment is within a certain range and the temperature is maintained in another suitable range, the product quality pass rate will be relatively high; or when the chain speed and temperature meet a certain relationship, the production efficiency can reach the optimal level.

[0088] When faced with a new production batch that requires machine parameters to be determined, some basic information related to the new batch (such as product type, production scale, and other factors that may affect the selection of machine parameters) and the initial state of the current shaping equipment are provided as input to the trained equipment learning model.

[0089] The model analyzes and predicts new situations based on the relationship between the machine parameters and production results of the historical batches learned before, and outputs a set of recommended optimal machine parameter data. For example, the model may output specific parameter recommendations such as how many revolutions per minute the shaping equipment speed should be set to, how many degrees Celsius the temperature should be controlled at, and how many degrees the secret should be set to, so that the operator can set the equipment according to these recommended parameters and start the production of the new batch.

[0090] Furthermore, the device learning model includes one or more combinations of random forest, gradient boosting decision tree and neural network methods.

[0091] Random forest is an ensemble learning algorithm based on decision trees. It builds multiple independent decision trees and lets them predict the data separately. Finally, it combines the prediction results of these decision trees to get the final prediction conclusion. Each decision tree in the random forest is built based on a random sampling sample of the training data (with replacement sampling, i.e. Bootstrapping sampling) and some randomly selected features. The purpose of this is to introduce randomness so that each decision tree is different and avoid overfitting.

[0092] When predicting new data, the data is input into each decision tree for prediction. For classification problems, a voting method is usually used, that is, each decision tree outputs a category, and finally the number of occurrences of each category is counted, and the category with the most occurrences is selected as the final prediction result; for regression problems, the prediction values ​​of each decision tree are averaged to obtain the final prediction value.

[0093] Gradient Boosting Decision Tree (GBDT) GBDT is to gradually improve the prediction results by iteratively training a series of weak learners (usually decision trees) and fitting the residuals of the previous round of models as the core idea. Each round of iteration will train a new decision tree based on the residuals generated by the current model prediction, and then add the newly trained decision tree to the existing model set, and update the prediction value of the entire model according to a certain learning rate.

[0094] Neural networks are trained using the back-propagation algorithm. First, the input data is passed into the network and processed by each layer of neurons to obtain the predicted value. Then, based on the difference between the predicted value and the true value (loss function), the error information is passed from the output layer to the input layer layer by layer through the back-propagation algorithm, and the connection weights between the neurons in each layer are adjusted to continuously reduce the loss function and finally reach a better weight configuration, so that the model can accurately predict the data.

[0095] In one embodiment, the data is first processed using random forest to obtain a preliminary prediction result or feature representation, and then the result or feature representation is used as input to a gradient boosting decision tree or neural network for further processing and prediction. In this way, the anti-overfitting ability of random forest and the role of preliminary feature screening can be fully utilized to provide a basis for subsequent more accurate predictions.

[0096] In another embodiment, several models, including random forest, gradient boosting decision tree and neural network, are trained simultaneously, and then their prediction results are integrated according to different rules. For example, for classification problems, a voting method can be adopted, that is, the categories output by each model are counted separately, and the category with the most occurrences is selected as the final prediction result; for regression problems, the prediction values ​​of each model can be weighted averaged, and different weights can be set according to factors such as the performance of each model to obtain the final prediction value.

[0097] Furthermore, after obtaining the recommended optimal machine parameter data, the machine parameter data is sent to the fabric shaping equipment for execution. The accuracy of the machine parameter data is evaluated by the fabric production effect after the execution. If the evaluation result is unqualified, manual parameter adjustment is performed, and the manually adjusted data is updated to the process library. The production effect includes product quality and production stability.

[0098] In one of the embodiments, after obtaining the recommended optimal machine parameter data, three process delivery methods can be used. The first process delivery method: through the intelligent production scheduling page of the design configuration, the shaping equipment of the entire factory can be planned and scheduled. During the scheduling process, the process formula can be bound uniformly, and the process parameters can be delivered to the corresponding machine; on the production scheduling page, click Add Order to trigger the script program to enter the batch, machine, and process binding page; click different setting areas to trigger different script programs to retrieve synchronized database data. Finally, the batch, machine, and process are bound together; after clicking Confirm to add, the script program will save the delivery record in the database; the process parameters are delivered to the corresponding machine through the configured machine connection information. The second process delivery method: through the new delivery button in the process formula function page, directly enter the process parameter delivery page, set the batch number, select the process sheet in the process library, and click Confirm to trigger the script program for the batch number-bound process sheet to be delivered, and the delivery record will be saved in the corresponding database of the system. The third process delivery method: When the shaping equipment operation terminal enters the batch number bound to the delivered process, selects the corresponding process step, and clicks on the execution machine, the relevant trigger instructions are sent to the system through data acquisition, and then the process parameters are triggered to be delivered and written into the program script of the shaping equipment controller. Through the communication protocol link, the relevant parameters can be written into the shaping equipment PLC to complete the entire process delivery operation.

[0099] After the fabric production is completed, the production effect is comprehensively evaluated according to the evaluation indicators within the factory to determine the accuracy of the machine parameter data. In one embodiment, if the cloth seal after washing / finished product cloth seal is between 0.96 and 1.04, it is qualified, otherwise it is unqualified; if the gram weight after washing / standard gram weight is between 1.05 and 1.16, it is qualified, otherwise it is unqualified. If it is evaluated as qualified, it is saved in the process library and continued to be called in the subsequent same production batches. If it is evaluated as unqualified, the parameters are adjusted. The production effect is evaluated at an interval time period until it is evaluated as qualified. The adjusted parameters are updated in the process library in real time.

[0100] In the method for optimizing and recommending the running parameters of fabric shaping equipment provided in an embodiment of the present invention, the collected running parameter data of the fabric shaping equipment are preprocessed and standardized, the process characteristic data of the batch are extracted, the optimal parameter value is calculated based on the process characteristic data in the stable interval, and the optimal parameter value is associated with the fabric category information of the batch. When a new production batch is detected, the fabric category information is matched with the process data entry in the process library. If the match is successful, the running parameter data is recommended. If the match is unsuccessful, the equipment learning model is used to predict and output the recommended optimal running parameter data. The present invention can quickly output process parameters through system recommendation, and at the same time reduce quality problems caused by human errors.

[0101] The embodiment of the present invention also provides a device for optimizing and recommending parameters of a fabric shaping device. The device for optimizing and recommending parameters of a fabric shaping device can be configured in a user terminal or a server. The device for optimizing and recommending parameters of a fabric shaping device is used to execute any embodiment of the aforementioned method for optimizing and recommending parameters of a fabric shaping device. Figure 6 , Figure 6 A schematic block diagram of a device for optimizing and recommending machine parameters of a fabric setting equipment provided by an embodiment of the present invention.

[0102] like Figure 6 As shown, the device 200 for optimizing and recommending parameters of a fabric setting equipment comprises a data acquisition unit 201 , a data preprocessing unit 203 , a data feature extraction unit 205 , a data association unit 207 and a data recommendation unit 209 .

[0103] The data collection unit 201 is used to transmit the machine parameter data of a certain production batch of fabric shaping equipment collected by the gateway to the data warehouse.

[0104] The data preprocessing unit 203 is used to preprocess and standardize the machine parameter data in the data warehouse.

[0105] In a specific embodiment, the data preprocessing unit 203 includes removing redundant data, filling missing values, correcting abnormal values, filtering extreme abnormal data due to sensor failure or noise interference, and standardizing the preprocessed data according to the process characteristics of different equipment to ensure that the data between different equipment and batches are comparable and consistent;

[0106] The data feature extraction unit 205 is used to segment the machine parameter data according to the start time and end time of the batch, extract the process feature data of the batch, and filter out the stable interval of the process feature data during operation according to the set time window.

[0107] In a specific embodiment, the data feature extraction unit 205 includes a sub-unit: a stable interval calculation unit, which is used to obtain the name of the processing equipment corresponding to the batch, retrieve all the collection points of the processing equipment in the data warehouse, and calculate the range with the smallest residual of each collection point within the time range limited by the batch start time and the batch end time as the stable interval.

[0108] The data association unit 207 is used to calculate the optimal parameter value according to the process characteristic data in the stable interval, associate the optimal parameter value with the fabric category information of the batch, form a process data entry and store it in the process library.

[0109] In a specific embodiment, the data feature extraction unit 207 includes subunits: an optimal parameter value calculation unit, which is used to extract the numerical value with the highest frequency of occurrence in the stable interval as the optimal parameter value of the process feature data; a classification storage unit: used to associate the optimal parameter value with the fabric category information of the batch, the fabric category information includes fabric type, fabric composition, color and color depth, and the fabric category information is classified and stored in the process library.

[0110] The data recommendation unit 209 is used to match the fabric category information of the new batch input with the process data entries in the process library when a new production batch is detected. If the match is successful, the machine parameter data of the process data entries is recommended. If the match is unsuccessful, prediction is performed through the equipment learning model, wherein the relationship between the machine parameter data of historical batches and the production effect is learned by the model, and the optimal machine parameter data is recommended.

[0111] In a specific embodiment, the data recommendation unit 209 includes subunits: a data matching unit, which sets a threshold value based on the target fabric type component content and target process parameter in the new production batch, traverses the process data entries in the process library according to the threshold value, and recommends the closest machine parameter data. A device learning unit, which is used to recommend parameter data using one or more combinations of random forest, gradient boosting decision tree and neural network methods.

[0112] In a specific embodiment, the fabric shaping equipment line parameter optimization recommendation device 200 also includes a data sending unit, which sends the line parameter data to the fabric shaping equipment for execution; an effect evaluation unit: evaluates the accuracy of the line parameter data through the fabric production effect after execution. If the evaluation result is unqualified, manual parameter adjustment is performed, and the manually adjusted data is updated to the process library. The production effect includes product quality and production stability.

[0113] The device for optimizing and recommending the parameters of the fabric shaping equipment provided in the embodiment of the present invention applies the above-mentioned method for optimizing and recommending the parameters of the fabric shaping equipment, preprocesses and standardizes the collected data of the parameters of the fabric shaping equipment, extracts the process characteristic data of the batch, calculates the optimal parameter value based on the process characteristic data in the stable interval, associates the optimal parameter value with the fabric category information of the batch, and when a new production batch is detected, matches the fabric category information with the process data entry in the process library. If the match is successful, the parameter data is recommended. If the match is unsuccessful, the equipment learning model is used for prediction, and the optimal recommended parameter data is output. Through the above-mentioned method, the process parameters can be quickly output through system recommendation, and the quality problems caused by human errors are reduced.

[0114] The above-mentioned device for optimizing and recommending parameters of the fabric setting equipment can be realized in the form of a computer program. The computer program can be used in Figure 7 Runs on the computer device shown.

[0115] See also Figure 7 , Figure 7 1 is a schematic block diagram of a computer device provided by an embodiment of the present invention. The computer device may be a user terminal or a server for executing a method for optimizing and recommending machine parameters of a fabric shaping device to detect defects in plants.

[0116] See also Figure 7 The computer device 500 includes a processor 502 , a memory and a network interface 505 connected via a system bus 501 , wherein the memory may include a storage medium 503 and an internal memory 504 .

[0117] The storage medium 503 can store an operating system 5031 and a computer program 5032. When the computer program 5032 is executed, the processor 502 can execute a method for optimizing and recommending machine parameters of a fabric setting device, wherein the storage medium 503 can be a volatile storage medium or a non-volatile storage medium.

[0118] The processor 502 is used to provide computing and control capabilities to support the operation of the entire computer device 500 .

[0119] The internal memory 504 provides an environment for the operation of the computer program 5032 in the storage medium 503. When the computer program 5032 is executed by the processor 502, the processor 502 can execute the method for optimizing the machine parameters of the fabric setting equipment.

[0120] The network interface 505 is used for network communication, such as providing data information transmission, etc. Those skilled in the art will understand that Figure 7 The structure shown in the figure is only a block diagram of a part of the structure related to the solution of the present invention, and does not constitute a limitation on the computer device 500 to which the solution of the present invention is applied. The specific computer device 500 may include more or fewer components than those shown in the figure, or combine certain components, or have a different arrangement of components.

[0121] The processor 502 is used to run the computer program 5032 stored in the memory to implement the corresponding functions in the above-mentioned method for optimizing the machine parameters of the fabric setting equipment.

[0122] Those skilled in the art will understand that Figure 7 The embodiments of the computer device shown in the figure do not constitute a limitation on the specific composition of the computer device. In other embodiments, the computer device may include more or fewer components than shown in the figure, or combine certain components, or arrange the components differently. For example, in some embodiments, the computer device may only include a memory and a processor. In such embodiments, the structure and function of the memory and the processor are the same as those of the embodiment of the present invention. Figure 7 The embodiments shown are consistent and will not be described again here.

[0123] It should be understood that in the embodiment of the present invention, the processor 502 may be a central processing unit (CPU), and the processor 502 may also be other general-purpose processors, digital signal processors (DSP), application-specific integrated circuits (ASIC), field-programmable gate arrays (FPGA) or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. Among them, the general-purpose processor may be a microprocessor or the processor may also be any conventional processor, etc.

[0124] In another embodiment of the present invention, a computer-readable storage medium is provided. The computer-readable storage medium may be a volatile or non-volatile computer-readable storage medium. The computer-readable storage medium stores a computer program, wherein when the computer program is executed by a processor, the steps included in the above-mentioned method for optimizing the recommended parameters of the fabric setting equipment are implemented.

[0125] Those skilled in the art can clearly understand that, for the convenience and simplicity of description, the specific working process of the above-described equipment, devices and units can refer to the corresponding process in the aforementioned method embodiment, and will not be repeated here. Those of ordinary skill in the art can appreciate that the units and algorithm steps of each example described in conjunction with the embodiments disclosed herein can be implemented with electronic hardware, computer software or a combination of the two. In order to clearly illustrate the interchangeability of hardware and software, the composition and steps of each example have been generally described in the above description according to the function. Whether these functions are executed in hardware or software depends on the specific application and design constraints of the technical solution. Professional and technical personnel can use different methods to implement the described functions for each specific application, but such implementation should not be considered to exceed the scope of the present invention.

[0126] In the several embodiments provided by the present invention, it should be understood that the disclosed equipment, devices and methods can be implemented in other ways. For example, the device embodiments described above are only schematic. For example, the division of the units is only a logical function division. There may be other division methods in actual implementation. Units with the same function may also be combined into one unit. For example, multiple units or components may be combined or integrated into another system, or some features may be ignored or not executed. In addition, the mutual coupling or direct coupling or communication connection shown or discussed may be an indirect coupling or communication connection through some interfaces, devices or units, or may be an electrical, mechanical or other form of connection.

[0127] The units described as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed on multiple network units. Some or all of the units may be selected according to actual needs to achieve the purpose of the embodiments of the present invention.

[0128] In addition, each functional unit in each embodiment of the present invention may be integrated into one processing unit, or each unit may exist physically separately, or two or more units may be integrated into one unit. The above-mentioned integrated unit may be implemented in the form of hardware or in the form of software functional units.

[0129] If the integrated unit is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention is essentially or the part that contributes to the prior art, or all or part of the technical solution can be embodied in the form of a software product, and the computer software product is stored in a computer-readable storage medium, including a number of instructions for a computer device (which can be a personal computer, a server, or a network device, etc.) to perform all or part of the steps of the method described in each embodiment of the present invention. The aforementioned computer-readable storage medium includes: various media that can store program codes, such as a USB flash drive, a mobile hard disk, a read-only memory (ROM, Read-Only Memory), a magnetic disk or an optical disk.

[0130] The above is only a specific embodiment of the present invention, but the protection scope of the present invention is not limited thereto. Any technician familiar with the technical field can easily think of various equivalent modifications or replacements within the technical scope disclosed by the present invention, and these modifications or replacements should be included in the protection scope of the present invention. Therefore, the protection scope of the present invention shall be based on the protection scope of the claims.

Claims

1. A method for optimizing and recommending machine parameters of a fabric setting equipment, characterized in that: include: The machine parameter data of a certain production batch of fabric shaping equipment collected by the gateway is transmitted to the data warehouse; Preprocessing and standardizing the machine parameter data in the data warehouse; The machine parameter data is segmented according to the start time and end time of the batch, the process characteristic data of the batch is extracted, and the stable interval of the process characteristic data during operation is screened out according to the set time window; Calculating an optimal parameter value according to the process characteristic data of the stable interval, associating the optimal parameter value with the fabric category information of the batch, forming a process data entry and storing it in a process library; When a new production batch is detected, the fabric category information of the new batch input is matched with the process data entries in the process library. If the match is successful, the machine parameter data of the process data entries are recommended. If the match is unsuccessful, a prediction is performed through the equipment learning model, wherein the model learns the relationship between the machine parameter data of historical batches and the production effect, and outputs the recommended optimal machine parameter data.

2. The method for optimizing and recommending machine parameters of fabric setting equipment according to claim 1, characterized in that: The preprocessing and standardization operations of the machine parameter data in the data warehouse include: The preprocessing operation includes removing redundant data, filling missing values, correcting abnormal values, and filtering extreme abnormal data caused by sensor failure or noise interference; The standardization operation is to standardize the preprocessed data according to the process characteristics of different equipment, so as to ensure the comparability and consistency of data between different equipment and batches.

3. The method for optimizing and recommending machine parameters of fabric setting equipment according to claim 1, characterized in that: The machine parameter data is segmented according to the start time and end time of the batch, the process characteristic data of the batch is extracted, and the stable interval of the process characteristic data during operation is screened out according to the set time window, including: The name of the processing equipment corresponding to the batch is obtained, all the collection points of the processing equipment are retrieved in the data warehouse, and the range with the smallest residual of each collection point is counted as the stable interval within the time range limited by the batch start time and the batch end time.

4. The method for optimizing and recommending machine parameters of fabric setting equipment according to claim 1, characterized in that: Calculating an optimal parameter value according to the process characteristic data in the stable interval, associating the optimal parameter value with the fabric category information of the batch, forming a process data entry and storing it in a process library, including extracting the value with the highest frequency in the stable interval as the optimal parameter value of the process characteristic data; The optimal parameter value is associated with fabric category information of the batch, wherein the fabric category information includes fabric type, fabric composition, color and color depth, and the fabric category information is classified and itemized and stored in the process library.

5. The method for optimizing and recommending machine parameters of fabric setting equipment according to claim 1, characterized in that: When a new production batch is detected, the fabric category information of the new batch is matched with the process data entry in the process library. If the match is successful, the machine parameter data of the process data entry is recommended, including: Thresholds are set according to the target fabric type component content and target process parameters in the new production batch, and the process data entries in the process library are traversed according to the thresholds to recommend the closest machine parameter data.

6. The method for optimizing and recommending machine parameters of fabric setting equipment according to claim 5, characterized in that: If the match is unsuccessful, the prediction is performed through the equipment learning model, wherein the model learns the relationship between the machine parameter data of the historical batches and the production effect, and outputs the recommended optimal machine parameter data, including: The device learning model includes one or more combinations of random forest, gradient boosting decision tree and neural network methods.

7. The method for optimizing and recommending machine parameters of fabric setting equipment according to claim 1, characterized in that: After obtaining the recommended optimal machine parameter data, the machine parameter data is sent to the fabric shaping equipment for execution. The accuracy of the machine parameter data is evaluated by the fabric production effect after execution. If the evaluation result is unqualified, manual parameter adjustment is performed and the manually adjusted data is updated to the process library. The production effect includes product quality and production stability.

8. A device for optimizing and recommending machine parameters of fabric shaping equipment, characterized in that: The device comprises: A data collection unit, used to transmit the machine parameter data of a certain production batch of fabric shaping equipment collected by the gateway to the data warehouse; A data preprocessing unit, used for preprocessing and standardizing the machine parameter data in the data warehouse; A data segmentation and feature extraction unit, used to segment the machine parameter data according to the start time and end time of the batch, extract the process feature data of the batch, and filter out the stable interval of the process feature data during operation according to a set time window; A data analysis and association unit, used to calculate an optimal parameter value according to the process characteristic data of the stable interval, associate the optimal parameter value with the fabric category information of the batch, form a process data entry and store it in a process library; The parameter data recommendation unit is used to match the fabric category information of the new batch input with the process data items in the process library when a new production batch is detected. If the match is successful, the machine parameter data of the process data items are recommended. If the match is unsuccessful, prediction is performed through the equipment learning model, wherein the relationship between the machine parameter data of historical batches and the production effect is learned by the model, and the optimal machine parameter data is recommended.

9. A computer device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that: When the processor executes the computer program, the method for optimizing and recommending machine parameters of a fabric setting equipment according to any one of claims 1 to 7 is implemented.

10. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, the method for optimizing and recommending machine parameters of a fabric setting device as described in any one of claims 1 to 7 is implemented.

Images