Intelligent distribution system and method for dyeing machine auxiliaries

Abstract

The present application relates to the field of intelligent distribution, and specifically discloses an intelligent distribution system and method for dyeing machine auxiliaries, which first obtains distribution accuracy influencing information collected from a database and auxiliary agent flow values ​​at multiple predetermined time points collected by a mass flow meter, and then uses deep learning technology to perform feature extraction and correlation analysis on the two, and finally obtains the dyeing machine auxiliary agent distribution pre-stop value through a decoder, thereby optimizing the distribution accuracy, reducing distribution errors, and improving the accuracy and stability of dyeing machine operation, thereby improving production efficiency and product quality.

Description

Technical Field

[0001] The present application relates to the field of intelligent distribution, and more specifically, to an intelligent distribution system and method for dyeing machine auxiliaries. Background Art

[0002] Dyeing machine auxiliaries are chemicals used in the dyeing process of textiles to improve dyeing results, increase dyeing uniformity, speed up dyeing, and protect fiber materials. These auxiliaries can include dispersants, emulsifiers, thickeners, acid-base regulators, etc., which make the dyeing process more efficient, stable, and environmentally friendly by regulating the interaction between dyes and textiles.

[0003] At present, the control of the delivery accuracy of dyeing machine auxiliaries requires technicians to conduct multiple delivery tests, record the error between the order delivery value and the actual delivery value, and then adjust the delivery accuracy of the auxiliaries. Considering that the dyeing machine auxiliaries delivery system usually involves multiple auxiliaries and multiple dyeing tanks, if these conditions are to be considered one by one, there will be thousands of possibilities. The workload of adjusting the delivery pre-stop value according to these combinations is very large and the accuracy is not high.

[0004] Therefore, an intelligent distribution system and method for dyeing machine auxiliaries is desired. Summary of the invention

[0005] In order to solve the above technical problems, the present application is proposed. The embodiment of the present application provides a dyeing machine auxiliary agent intelligent delivery system and method, which first obtains the delivery accuracy impact information collected from the database and the auxiliary agent flow values ​​at multiple predetermined time points collected by the mass flow meter, and then uses deep learning technology to perform feature extraction and correlation analysis on the two, and finally obtains the dyeing machine auxiliary agent delivery pre-stop value through a decoder, thereby optimizing the delivery accuracy to reduce the delivery error, improve the accuracy and stability of the dyeing machine operation, and thus improve production efficiency and product quality.

[0006] According to one aspect of the present application, a dyeing machine auxiliary agent intelligent distribution system is provided, which includes:

[0007] A dyeing agent auxiliary agent data acquisition module is used to acquire the delivery accuracy impact information collected by the database and the auxiliary agent flow values ​​at multiple predetermined time points collected by the mass flow meter;

[0008] A dyeing agent auxiliary agent data extraction module, used to extract a distribution accuracy impact information associated feature vector and an auxiliary agent flow feature vector from the distribution accuracy impact information collected by the database and the auxiliary agent flow values ​​at multiple predetermined time points collected by the mass flow meter;

[0009] The auxiliary agent prediction module is used to obtain the auxiliary agent delivery pre-stop value of the dyeing machine based on the delivery accuracy impact information association feature vector and the auxiliary agent flow feature vector.

[0010] According to another aspect of the present application, a method for intelligently distributing dyeing machine auxiliaries is provided, comprising:

[0011] Acquire the delivery accuracy impact information collected by the database and the auxiliary agent flow values ​​at multiple predetermined time points collected by the mass flow meter;

[0012] Extracting a distribution accuracy impact information associated feature vector and an additive flow feature vector from the distribution accuracy impact information collected by the database and the additive flow values ​​at a plurality of predetermined time points collected by the mass flow meter;

[0013] Based on the delivery accuracy influencing information association feature vector and the auxiliary agent flow feature vector, a dyeing machine auxiliary agent delivery pre-stop value is obtained.

[0014] Compared with the prior art, the present application provides an intelligent dyeing machine auxiliary agent distribution system and method, which first obtains the distribution accuracy influencing information collected from the database and the auxiliary agent flow values ​​at multiple predetermined time points collected by the mass flow meter, and then uses deep learning technology to perform feature extraction and correlation analysis on the two, and finally obtains the dyeing machine auxiliary agent distribution pre-stop value through a decoder, thereby optimizing the distribution accuracy, reducing distribution errors, and improving the accuracy and stability of dyeing machine operation, thereby improving production efficiency and product quality. BRIEF DESCRIPTION OF THE DRAWINGS

[0015] By describing the embodiments of the present application in more detail in conjunction with the accompanying drawings, the above and other purposes, features and advantages of the present application will become more apparent. The accompanying drawings are used to provide a further understanding of the embodiments of the present application and constitute a part of the specification. Together with the embodiments of the present application, they are used to explain the present application and do not constitute a limitation of the present application. In the accompanying drawings, the same reference numerals generally represent the same components or steps.

[0016] Figure 1 It is a block diagram of the intelligent distribution system of dyeing machine auxiliaries according to an embodiment of the present application.

[0017] Figure 2 It is a block diagram of a dyeing agent auxiliary data extraction module in the dyeing machine auxiliary intelligent distribution system according to an embodiment of the present application.

[0018] Figure 3 It is a block diagram of the accuracy impact information feature extraction unit in the intelligent distribution system of dyeing machine auxiliaries according to an embodiment of the present application.

[0019] Figure 4It is a block diagram of the accuracy-affecting information feature encoding subunit in the intelligent distribution system of dyeing machine auxiliaries according to an embodiment of the present application.

[0020] Figure 5 It is a flow chart of the intelligent distribution method of dyeing machine auxiliaries according to an embodiment of the present application.

[0021] Figure 6 is a block diagram of an electronic device according to an embodiment of the present application. DETAILED DESCRIPTION

[0022] Below, the exemplary embodiments according to the present application will be described in detail with reference to the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present application, rather than all the embodiments of the present application, and it should be understood that the present application is not limited to the exemplary embodiments described here.

[0023] Exemplary Systems

[0024] Figure 1 FIG. 1 is a block diagram of an intelligent distribution system for dyeing machine auxiliaries according to an embodiment of the present application. Figure 1 As shown, according to the embodiment of the present application, the intelligent distribution system 100 for dyeing machine auxiliaries includes: a dyeing agent auxiliary data acquisition module 110, which is used to obtain the distribution accuracy influencing information collected by the database and the auxiliary flow values ​​at multiple predetermined time points collected by the mass flow meter; a dyeing agent auxiliary data extraction module 120, which is used to extract the distribution accuracy influencing information associated feature vector and the auxiliary flow feature vector from the distribution accuracy influencing information collected by the database and the auxiliary flow values ​​at multiple predetermined time points collected by the mass flow meter; an auxiliary prediction module 130, which is used to obtain the dyeing machine auxiliary distribution pre-stop value based on the distribution accuracy influencing information associated feature vector and the auxiliary flow feature vector.

[0025] In the above-mentioned dyeing machine auxiliary intelligent delivery system 100, the dyeing agent auxiliary data acquisition module 110 is used to obtain the delivery accuracy impact information collected by the database and the auxiliary agent flow value of multiple predetermined time points collected by the mass flow meter, wherein the delivery accuracy impact information collected by the database includes the variety of auxiliary agents, the operating frequency of the delivery pump, the delivery pipeline length of different blanking ports and the compressed gas pressure. It should be understood that dyeing agent auxiliary agents are chemical substances added in the textile dyeing process, and their functions include improving dyeing effect, enhancing dyeing uniformity and brightness, adjusting hue and depth, and improving dyeing durability and stability. These auxiliary agents help to enhance the combination of dyes and fibers, improve dyeing efficiency and quality, and reduce energy consumption and pollutant emissions at the same time, playing an indispensable role in textile dyeing. At present, ensuring the delivery accuracy of dyeing machine auxiliary agents requires technicians to perform multiple tests, record the error between the order delivery value and the actual value, and adjust the delivery accuracy. This process is time-consuming and has limited accuracy. In the technical solution of the present application, the distribution accuracy data is collected by the database and the auxiliary agent flow value at multiple time points is collected by the mass flow meter, and the optimization of the auxiliary agent distribution pre-stop value of the dyeing machine is realized by combining the deep learning technology. Among them, the distribution accuracy influencing information in the database includes key parameters such as the type of auxiliary agent, the operating frequency of the distribution pump, the length of the distribution pipeline of different drop ports, and the compressed air pressure. This information reflects the characteristics and requirements of different auxiliary agents in the distribution process. For example, different types of auxiliary agents may require different distribution methods and conditions. The operating frequency of the distribution pump directly affects the flow stability of the auxiliary agent, while the pipeline length and air pressure intensity affect the delivery speed and uniformity of the auxiliary agent. On the other hand, the auxiliary agent flow values ​​at multiple predetermined time points collected by the mass flow meter provide real-time auxiliary agent usage data. By monitoring the auxiliary agent flow at different time points, the system can understand the consumption of auxiliary agents, distribution stability, and possible abnormal conditions. In this way, the distribution accuracy influencing information in the database and the auxiliary agent flow data collected by the mass flow meter complement each other and together constitute the basis of the intelligent distribution system of dyeing machine auxiliaries, thereby realizing accurate monitoring and prediction of the auxiliary agent distribution process, improving distribution efficiency, saving costs, and ensuring the stability and quality of the dyeing process.

[0026] In the above-mentioned dyeing machine auxiliary agent intelligent distribution system 100, the dyeing agent auxiliary agent data extraction module 120 is used to extract the distribution accuracy impact information associated feature vector and the auxiliary agent flow feature vector from the distribution accuracy impact information collected by the database and the auxiliary agent flow values ​​at multiple predetermined time points collected by the mass flow meter. It should be understood that the process of extracting the distribution accuracy impact information associated feature vector and the auxiliary agent flow feature vector converts the original data into a form that can be recognized and utilized by the system, providing key input information for the dyeing machine auxiliary agent intelligent distribution system. The extraction of these feature vectors not only helps the system understand the influencing factors and laws of auxiliary agent distribution, but also provides effective data support for the system's prediction and decision-making, thereby improving the accuracy and efficiency of dyeing machine auxiliary agent distribution.

[0027] Figure 2 FIG. 1 is a block diagram of a dyeing agent data extraction module in a dyeing machine auxiliary agent intelligent distribution system according to an embodiment of the present application. Figure 2 As shown, in a specific embodiment of the present application, the dyeing agent auxiliary data extraction module 120 includes: an accuracy impact information feature extraction unit 121, which is used to extract features from the delivery accuracy impact information collected by the database to obtain the delivery accuracy impact information associated feature vector; an auxiliary agent flow feature extraction unit 122, which is used to extract features from the auxiliary agent flow values ​​at multiple predetermined time points collected by the mass flow meter to obtain the auxiliary agent flow feature vector. It should be understood that the delivery accuracy impact information covers many key parameters, such as auxiliary agent variety, delivery pump frequency, pipeline length, air pressure intensity, etc. These parameters directly affect the delivery effect and stability of the auxiliary agent. Through feature extraction, the system can convert these parameters into numerical feature vectors, so that the system can better understand and compare the auxiliary agent characteristics under different delivery conditions. The extraction of these delivery accuracy impact information associated feature vectors enables the system to more efficiently process and utilize the delivery parameter data, and realize the refined management and optimization of the auxiliary agent delivery process. Through these feature vectors, the system can realize real-time monitoring, prediction and adjustment of the auxiliary agent delivery process, thereby improving delivery efficiency, reducing costs, and ensuring the stability and quality of the dyeing process. The extraction of these feature vectors provides important technical support and guarantee for the intelligent development of the dyeing industry.

[0028] Furthermore, the additive flow value is an important data reflecting the use of additives. The changes in flow values ​​at different time points can reveal the rules and trends of additive use. Through feature extraction, the system can convert these raw flow data into representative and distinguishing feature vectors, providing more accurate and effective data support for the system. The extraction of additive flow feature vectors not only helps the system to conduct a comprehensive analysis of the use of additives, but also provides an important basis for the system's intelligent decision-making. By extracting features from additive flow data, the system can achieve real-time monitoring and prediction of additive use, timely detect and handle distribution anomalies, and ensure the stability and quality of the dyeing process.

[0029] Figure 3 FIG. 1 is a block diagram of a precision-affecting information feature extraction unit in a dyeing machine auxiliary agent intelligent distribution system according to an embodiment of the present application. Figure 3 As shown, in a specific embodiment of the present application, the accuracy impact information feature extraction unit 121 includes: an accuracy impact information word segmentation subunit 1211, which is used to segment the delivery accuracy impact information collected by the database to obtain a delivery accuracy impact information input word sequence; an accuracy impact information feature encoding subunit 1212, which is used to feature encode the delivery accuracy impact information input word sequence to obtain the delivery accuracy impact information associated feature vector. It should be understood that the delivery accuracy impact information may contain various key words and phrases describing delivery conditions and effects, such as auxiliary agent type, delivery frequency, temperature impact, etc. Through word segmentation, the system can divide this information into different words or phrases to construct a word sequence of delivery accuracy impact information. The word segmentation process can adopt different methods, such as rule-based word segmentation, statistical-based word segmentation, or deep learning-based word segmentation. Through word segmentation, the system can convert the delivery accuracy impact information into a series of meaningful vocabulary units.

[0030] Furthermore, the input word sequence of information affecting delivery accuracy contains key words and phrases that describe delivery conditions and effects. Through feature encoding, the system can convert these text information into numerical feature vectors so that the computer can understand and process them. Among them, the process of feature encoding involves the operation of mapping text information into vector space. Common encoding methods include bag-of-words model, TF-IDF (term frequency-inverse document frequency), Word2Vec, BERT, etc. These methods can convert the words in the word sequence into dense numerical vector representations to capture the semantics and association information between words. Through feature encoding, the system can use the feature vector associated with information affecting delivery accuracy as the input of the machine learning model to achieve modeling and analysis of factors affecting delivery accuracy.

[0031] Figure 4 FIG. 1 is a block diagram of a precision-affecting information feature encoding subunit in a dyeing machine auxiliary agent intelligent distribution system according to an embodiment of the present application. Figure 4 As shown, in a specific embodiment of the present application, the accuracy impact information feature encoding subunit 1212 includes: an accuracy impact word sequence semantic encoding secondary subunit 1212-1, which is used to pass the delivery accuracy impact information input word sequence through the delivery accuracy impact information semantic encoder based on the converter to obtain multiple delivery accuracy impact information feature vectors; the accuracy impact attention encoding secondary subunit 1212-2, which is used to perform attention encoding on the multiple delivery accuracy impact information feature vectors to obtain the delivery accuracy impact information association feature vector. It should be understood that the delivery accuracy impact information input word sequence contains key words and phrases describing delivery conditions and effects. Through the converter-based semantic encoder, the system can convert this information into semantically rich feature vectors to capture more semantic information and context associations. Among them, the converter-based semantic encoder, such as the Transformer model, has a strong modeling ability and can effectively learn the semantic associations between text sequences. Through this encoder, the system can convert the delivery accuracy impact information input into multiple feature vectors, each of which contains rich semantic information. These feature vectors can capture the complex semantic relationships in the delivery accuracy impact information, which helps the system to more comprehensively understand the impact mechanism of delivery conditions on accuracy. The obtained multiple delivery accuracy impact information feature vectors can be used to construct a more accurate and powerful text representation, providing the system with more semantic information and contextual associations. Through the feature vectors obtained by the converter-based semantic encoder, the system can better achieve in-depth understanding and mining of delivery accuracy impact information, providing strong support for the intelligent development and optimization of the system. Specifically, the Bert model of the converter-based delivery accuracy impact information semantic encoder is used to perform global context semantic encoding on the delivery accuracy impact information input word sequence to obtain multiple delivery accuracy impact information feature vectors.

[0032] Furthermore, attention encoding is performed on multiple feature vectors of information affecting delivery accuracy, and the attention mechanism is used to strengthen the correlation between different feature vectors, highlight important information, and more accurately capture the key features of factors affecting delivery accuracy. Through attention encoding, the system can perform weighted combination of multiple feature vectors of information affecting delivery accuracy, so that the model can focus on features that have a greater impact on delivery accuracy, and improve the system's efficiency in capturing and utilizing key information. The attention mechanism can dynamically adjust the weights according to the interaction and importance between the input feature vectors, allowing the system to more flexibly handle the correlation between different features, thereby better understanding the complexity and diversity of information affecting delivery accuracy.

[0033] In a specific embodiment of the present application, the precision impact attention encoding secondary subunit 1212-2 includes: passing the multiple delivery precision impact information feature vectors through the delivery precision impact information feature encoder based on the attention mechanism to obtain multiple delivery precision impact information attention feature vectors; associating the multiple delivery precision impact information attention feature vectors to obtain the delivery precision impact information association feature vector. It should be understood that through the delivery precision impact information feature encoder based on the attention mechanism, the system can perform weighted combination of multiple feature vectors to obtain multiple delivery precision impact information attention feature vectors. These attention feature vectors can better capture the key features and correlations in the delivery precision impact information, and help the system to more comprehensively understand the impact mechanism of delivery conditions on accuracy. The attention mechanism enables the model to adaptively adjust the importance of different feature vectors, thereby more effectively utilizing information and improving the system's modeling accuracy of delivery precision impact factors. The multiple delivery precision impact information attention feature vectors obtained contain richer semantic information and correlation, which helps the system to more accurately predict and optimize delivery accuracy during the delivery process.

[0034] Furthermore, by associating multiple attention feature vectors of information affecting delivery accuracy, the system can comprehensively consider the correlation between different feature vectors, so as to better capture the interaction and influence mechanism between the factors affecting delivery accuracy. The generation process of the associated feature vector can improve the system's understanding and modeling effect of the factors affecting delivery accuracy by integrating the information of each attention feature vector, strengthening the expression of important features, and weakening the influence of secondary features. The obtained associated feature vectors of information affecting delivery accuracy can provide the system with a more representative and comprehensive feature representation. These associated feature vectors not only reflect the complex relationship between the various factors affecting delivery accuracy, but can also be used to optimize the system's delivery strategy and improve delivery accuracy and efficiency.

[0035] In a specific embodiment of the present application, the additive flow feature extraction unit 122 includes: arranging the additive flow values ​​at multiple predetermined time points collected by the mass flow meter according to the time dimension to obtain an additive flow input vector; passing the additive flow input vector through an additive flow timing encoder to obtain the additive flow feature vector. It should be understood that by arranging the additive flow values ​​at multiple predetermined time points according to the time dimension, the system can convert the time series data into a one-dimensional vector form, which is convenient for the machine learning model to process and analyze. The generation process of the additive flow input vector can capture the changes in the additive flow at different time points, helping the system to better understand the dynamic characteristics and laws of the additive flow. This vectorized processing method of time series data helps to extract feature information of the time dimension and provide the system with richer and more comprehensive input features. The obtained additive flow input vector contains the additive flow information at different time points, reflecting the historical changes in the use of additives.

[0036] Furthermore, through the additive flow time series encoder, the system can effectively extract and abstract the features of the additive flow input vector, thereby capturing the time series characteristics and changing patterns of the additive flow. The time series encoder can model and learn time series data through various neural network structures or sequence modeling techniques, extract important features and patterns, and encode them into more representative and high-dimensional feature vectors. These additive flow feature vectors not only contain the basic information of the additive flow data, but also contain the dynamic changes and trends in the time dimension, providing the system with a more comprehensive and in-depth representation of the additive flow characteristics. Through the processing of the additive flow time series encoder, the system can better mine and utilize the information in the additive flow data, improve the system's modeling and analysis capabilities for the additive flow, and realize intelligent management and control of the additive flow. Specifically, the auxiliary agent flow input vector is fully-connectedly encoded using the fully-connected layer of the auxiliary agent flow timing encoder to extract high-dimensional implicit features of the eigenvalues ​​at each position in the auxiliary agent flow input vector; and the auxiliary agent flow input vector is one-dimensionally encoded using the one-dimensional convolutional layer of the auxiliary agent flow timing encoder to extract high-dimensional implicit correlation features of the correlation between the eigenvalues ​​at each position in the auxiliary agent flow input vector.

[0037] In the above-mentioned dyeing machine auxiliary agent intelligent distribution system 100, the auxiliary agent prediction module 130 is used to obtain the dyeing machine auxiliary agent distribution pre-stop value based on the distribution accuracy impact information association feature vector and the auxiliary agent flow feature vector. It should be understood that by associating the distribution accuracy impact information association feature vector and the auxiliary agent flow feature vector, the system can further optimize the auxiliary agent distribution strategy and improve the distribution accuracy and efficiency after considering the impact of distribution accuracy on the operation of the dyeing machine. The calculation of the dyeing machine auxiliary agent distribution pre-stop value is based on this comprehensive consideration, aiming to predict future auxiliary agent distribution needs, so that the system can make adjustments and optimizations in advance, thereby ensuring that the dyeing machine can complete the auxiliary agent distribution work on time and in quantity during operation. This comprehensive consideration and prediction capability helps to improve the production efficiency of the dyeing machine, reduce production costs, and improve production quality and reliability.

[0038] In a specific embodiment of the present application, the auxiliary agent prediction module 130 includes: fusing the distribution accuracy impact information associated feature vector and the auxiliary agent flow feature vector to obtain a dyeing machine auxiliary agent distribution feature vector; expressing the dyeing machine auxiliary agent distribution feature vector in an equivalent model parameter space based on class probability to obtain an optimized dyeing machine auxiliary agent distribution feature vector; passing the optimized dyeing machine auxiliary agent distribution feature vector through a decoder to obtain a decoded value, and the decoded value is used to represent the dyeing machine auxiliary agent distribution pre-stop value. It should be understood that the distribution accuracy impact information associated feature vector contains important features related to distribution accuracy, such as distribution error, distribution stability, etc., while the auxiliary agent flow feature vector contains the time series change characteristics of the auxiliary agent flow, such as flow fluctuation, trend, etc. Fusion of the distribution accuracy impact information associated feature vector and the auxiliary agent flow feature vector can enable the system to more comprehensively consider the impact of distribution accuracy on the auxiliary agent flow when evaluating the dyeing machine auxiliary agent distribution. By combining these two types of feature information, the dyeing machine auxiliary agent distribution feature vector can not only reflect the time series characteristics of the auxiliary agent flow, but also take into account the impact of distribution accuracy, thereby improving the comprehensive grasp of the dyeing machine auxiliary agent distribution status. The fused feature vector of dyeing machine auxiliary distribution can provide the system with richer and more accurate feature representation, which helps the system better understand and analyze the characteristics and laws of dyeing machine auxiliary distribution. Such feature vectors can not only be used to monitor and diagnose the distribution status of dyeing machine auxiliary, but also help the system to realize intelligent control and optimization of the dyeing machine auxiliary distribution process. By fusing the information correlation feature vector affecting distribution accuracy and the auxiliary flow feature vector, the system can grasp the characteristics and status of dyeing machine auxiliary distribution more comprehensively and accurately, and improve distribution efficiency and quality.

[0039] In particular, in the technical solution of the present application, it is considered that there may be data quality problems in collecting the distribution accuracy affecting information from the database and the auxiliary agent flow value from the mass flow meter. The distribution accuracy affecting information recorded in the database and the auxiliary agent flow value collected by the mass flow meter may contain outliers or noise data, which may have an adverse effect on the dyeing machine auxiliary agent distribution feature vector during the feature extraction and fusion process. When the distribution accuracy affecting information feature vector and the auxiliary agent flow feature vector are fused, if there are outliers or noise data in them, the final dyeing machine auxiliary agent distribution feature vector may be affected. The feature fusion process may be sensitive to outliers or noise data, causing these outliers or noise data to have a significant impact in the feature vector, reducing the accuracy and robustness of the feature vector. If the distribution accuracy affecting information and the auxiliary agent flow value are missing or incomplete, the feature extraction and fusion process may not be able to effectively handle the missing data, thereby introducing uncertainty. Missing data may cause some feature vectors to lack accurate information in the dyeing machine auxiliary agent distribution feature vector, making the outliers or noise data in the feature vector more prominent. In order to solve the problem of outliers or noise data in the dyeing machine auxiliary agent distribution feature vector, in the technical solution of the present application, the dyeing machine auxiliary agent distribution feature vector is expressed in an equivalent model parameter space based on class probability.

[0040] Specifically, the dyeing machine auxiliary agent distribution feature vector is expressed in an equivalent model parameter space based on class probability to obtain an optimized dyeing machine auxiliary agent distribution feature vector, including: passing the dyeing machine auxiliary agent distribution feature vector through a softmax activation function to obtain a class probability vector; calculating the joint probability distribution of the dyeing machine auxiliary agent distribution feature vector relative to the class probability vector to obtain a class probability-guided dyeing machine auxiliary agent distribution feature vector; passing the class probability-guided dyeing machine auxiliary agent distribution feature vector through a model parameter latent space mapper based on a fully connected layer to obtain a dyeing machine auxiliary agent distribution feature vector in an equivalent model parameter space;

[0041] The following optimization formula is used to perform feature interaction on the dyeing machine auxiliary agent distribution feature vector and the dyeing machine auxiliary agent distribution feature vector in the equivalent model parameter space to obtain the optimized dyeing machine auxiliary agent distribution feature vector, wherein the optimization formula is:

[0042]

[0043] Among them, v1 represents the dyeing machine auxiliary agent distribution feature vector in the equivalent model parameter space, v2 represents the dyeing machine auxiliary agent distribution feature vector, W is the transformation matrix, and b is the bias parameter; sigmoid represents the activation function, concat(v1, v2) represents the connection feature, t∈[0, 1] is a threshold value used to adaptively balance the importance of the fusion information from v1 and v2, and v represents the optimized dyeing machine auxiliary agent distribution feature vector.

[0044] Here, the equivalent model parameter space expression based on class probability is performed on the dyeing machine auxiliary distribution feature vector, which allows the use of class probability to describe the model parameters, thereby providing a probabilistic explanation of the model prediction results. Then, the probability-based expression takes into account the distribution of the data, making the model more robust and better able to handle outliers and noisy data. At the same time, since it utilizes the distribution information of the data, it will also improve the performance of the decoder. In addition, this method improves the interpretability of the model because the class probability is more intuitive and easy to understand, which helps to explain the model's prediction results more clearly. At the same time, the expression based on class probability can also quantify the uncertainty of the model prediction, which is crucial for risk assessment and decision making. Finally, the expression based on class probability also helps to explore the model parameter space more effectively because the class probability provides the gradient information of the objective function, making the search of the parameter space more efficient.

[0045] Furthermore, the decoder is used to convert the optimized dyeing machine auxiliary agent distribution feature vector into a decoded value. The core of this process is to convert the abstract feature vector information into a specific dyeing machine auxiliary agent distribution pre-stop value, thereby realizing the accurate prediction and control of the dyeing machine auxiliary agent distribution process. The role of the decoder here is similar to a translator, which translates the abstract feature vector information into an operational pre-stop value to provide guidance and reference for the actual operation of the dyeing machine. The decoded value obtained by the decoder can provide timely and accurate information for production managers, helping them to adjust and optimize the auxiliary agent distribution strategy of the dyeing machine and improve production efficiency and product quality. This feature vector-based decoding value representation method effectively combines scientific analysis with production practice, and provides important support and guarantee for the accurate control of dyeing machine auxiliary agent distribution.

[0046] In summary, the embodiment of the present application first obtains the delivery accuracy influencing information collected from the database and the auxiliary agent flow values ​​at multiple predetermined time points collected by the mass flow meter, and then uses deep learning technology to perform feature extraction and correlation analysis on the two, and finally obtains the dyeing machine auxiliary agent delivery pre-stop value through the decoder, thereby optimizing the delivery accuracy to reduce delivery errors, improve the accuracy and stability of dyeing machine operation, and thus improve production efficiency and product quality.

[0047] As described above, the dyeing machine auxiliary agent intelligent distribution system 100 according to the embodiment of the present application can be implemented in various terminal devices, such as a server deployed with a dyeing machine auxiliary agent intelligent distribution algorithm. In one example, the dyeing machine auxiliary agent intelligent distribution system 100 can be integrated into the terminal device as a software module and / or a hardware module. For example, the dyeing machine auxiliary agent intelligent distribution system 100 can be a software module in the operating system of the terminal device, or can be an application developed for the terminal device; of course, the dyeing machine auxiliary agent intelligent distribution system 100 can also be one of the many hardware modules of the terminal device.

[0048] Alternatively, in another example, the dyeing machine auxiliary agent intelligent distribution system 100 and the terminal device may also be separate devices, and the dyeing machine auxiliary agent intelligent distribution system 100 may be connected to the terminal device via a wired and / or wireless network and transmit interactive information in accordance with an agreed data format.

[0049] Exemplary Methods

[0050] Figure 5 FIG. 1 is a flow chart of a method for intelligently distributing dyeing machine auxiliaries according to an embodiment of the present application. Figure 5 As shown, according to the intelligent delivery method of dyeing machine auxiliaries in the embodiment of the present application, it includes: S110, acquiring the delivery accuracy influencing information collected by the database and the auxiliary agent flow values ​​at multiple predetermined time points collected by the mass flow meter; S120, extracting the delivery accuracy influencing information associated feature vector and the auxiliary agent flow feature vector from the delivery accuracy influencing information collected by the database and the auxiliary agent flow value at multiple predetermined time points collected by the mass flow meter; S130, obtaining the dyeing machine auxiliary agent delivery pre-stop value based on the delivery accuracy influencing information associated feature vector and the auxiliary agent flow feature vector.

[0051] Here, those skilled in the art can understand that the specific operations of each step in the above-mentioned dyeing machine auxiliary agent intelligent delivery method have been referred to above. Figures 1 to 4 The invention has been introduced in detail in the description of the intelligent distribution system of dyeing machine auxiliaries, and therefore, its repeated description will be omitted.

[0052] Exemplary Electronic Devices

[0053] Below, reference Figure 6 To describe an electronic device according to an embodiment of the present application.

[0054] like Figure 6As shown, the electronic device 10 includes an input device 11, an input interface 12, a central processing unit 13, a memory 14, an output interface 15, an output device 16, and a bus 17. The input interface 12, the central processing unit 13, the memory 14, and the output interface 15 are connected to each other through the bus 17, and the input device 11 and the output device 16 are connected to the bus 17 through the input interface 14 and the output interface 15, respectively, and then connected to other components of the electronic device 10.

[0055] Specifically, the input device 11 receives input information from the outside and transmits the input information to the central processing unit 13 through the input interface 12; the central processing unit 13 processes the input information based on the computer executable instructions stored in the memory 14 to generate output information, temporarily or permanently stores the output information in the memory 14, and then transmits the output information to the output device 16 through the output interface 15; the output device 16 outputs the output information to the outside of the electronic device 10 for user use.

[0056] In one embodiment, Figure 6 The electronic device 10 shown can be implemented as a network device, which may include: a memory configured to store programs; a processor configured to run the programs stored in the memory to execute any one of the intelligent distribution methods for dyeing machine auxiliaries described in the above embodiments.

[0057] According to an embodiment of the present application, the process described above with reference to the flowchart can be implemented as a computer software program. For example, an embodiment of the present application includes a computer program product, which includes a computer program tangibly contained on a machine-readable medium, and the computer program includes a program code for executing the method shown in the flowchart. In such an embodiment, the computer program can be downloaded and installed from a network, and / or installed from a removable storage medium.

[0058] It will be appreciated by those skilled in the art that all or some of the steps, systems, and functional modules / units in the methods disclosed above may be implemented as software, firmware, hardware, and appropriate combinations thereof. In hardware implementations, the division between the functional modules / units mentioned in the above description does not necessarily correspond to the division of physical components; for example, a physical component may have multiple functions, or a function or step may be performed by several physical components in cooperation. Some or all physical components may be implemented as software executed by a processor, such as a central processing unit, a digital signal processor, or a microprocessor, or implemented as hardware, or implemented as an integrated circuit, such as an application-specific integrated circuit. Such software may be distributed on a computer-readable medium, which may include a computer storage medium (or non-transitory medium) and a communication medium (or temporary medium). As known to those skilled in the art, the term computer storage medium includes volatile and non-volatile, removable and non-removable media implemented in any method or technology for storing information (such as computer-readable instructions, data structures, program modules, or other data). Computer storage media include, but are not limited to, RAM, ROM, EEPROM, flash memory or other memory technology, CD-ROM, digital versatile disks (DVD) or other optical disk storage, magnetic cassettes, magnetic tapes, magnetic disk storage or other magnetic storage devices, or any other medium that can be used to store the desired information and can be accessed by a computer. In addition, it is well known to those skilled in the art that communication media typically contain computer-readable instructions, data structures, program modules, or other data in a modulated data signal such as a carrier wave or other transport mechanism, and may include any information delivery media.

[0059] It is to be understood that the above embodiments are merely exemplary embodiments used to illustrate the principles of the present application, but the present application is not limited thereto. For those skilled in the art, various modifications and improvements can be made without departing from the spirit and substance of the present application, and these modifications and improvements are also considered to be within the scope of protection of the present application.

Claims

1. An intelligent distribution system for dyeing machine auxiliaries, characterized in that: include: A dyeing agent auxiliary agent data acquisition module is used to acquire the delivery accuracy impact information collected by the database and the auxiliary agent flow values ​​at multiple predetermined time points collected by the mass flow meter; A dyeing agent auxiliary agent data extraction module, used to extract a distribution accuracy impact information associated feature vector and an auxiliary agent flow feature vector from the distribution accuracy impact information collected by the database and the auxiliary agent flow values ​​at multiple predetermined time points collected by the mass flow meter; An auxiliary agent prediction module, used for obtaining a pre-stop value of auxiliary agent delivery for the dyeing machine based on the distribution accuracy impact information correlation feature vector and the auxiliary agent flow feature vector; Wherein, the auxiliary agent prediction module includes: The distribution accuracy influencing information correlation feature vector and the auxiliary agent flow feature vector are merged to obtain the dyeing machine auxiliary agent distribution feature vector; The dyeing machine auxiliary agent distribution feature vector is expressed in an equivalent model parameter space based on class probability to obtain an optimized dyeing machine auxiliary agent distribution feature vector; Passing the optimized dyeing machine auxiliary agent distribution feature vector through a decoder to obtain a decoded value, wherein the decoded value is used to represent a pre-stop value for dyeing machine auxiliary agent distribution; The dyeing machine auxiliary agent distribution feature vector is expressed in an equivalent model parameter space based on class probability to obtain an optimized dyeing machine auxiliary agent distribution feature vector, including: The dyeing machine auxiliary agent distribution feature vector is passed through a softmax activation function to obtain a category probability vector; Calculating the joint probability distribution of the dyeing machine auxiliary agent distribution feature vector relative to the class probability vector to obtain a class probability guided dyeing machine auxiliary agent distribution feature vector; The class probability guides the dyeing machine auxiliary agent distribution feature vector through a model parameter latent space mapper based on a fully connected layer to obtain the dyeing machine auxiliary agent distribution feature vector in an equivalent model parameter space; Performing feature interaction on the dyeing machine auxiliary agent distribution feature vector under the equivalent model parameter space and the dyeing machine auxiliary agent distribution feature vector to obtain the optimized dyeing machine auxiliary agent distribution feature vector; Among them, performing feature interaction on the dyeing machine auxiliary agent distribution feature vector under the equivalent model parameter space and the dyeing machine auxiliary agent distribution feature vector to obtain the optimized dyeing machine auxiliary agent distribution feature vector includes: The dyeing machine auxiliary agent distribution feature vector and the dyeing machine auxiliary agent distribution feature vector in the equivalent model parameter space are subjected to feature interaction to obtain the optimized dyeing machine auxiliary agent distribution feature vector, wherein the optimization formula is: in, represents the dyeing machine auxiliary agent distribution feature vector in the equivalent model parameter space, represents the dyeing machine auxiliary agent distribution feature vector, is the transformation matrix, is the bias parameter; represents the activation function, Represents the connection characteristics, is a threshold value, represents the optimized dyeing machine auxiliary agent distribution feature vector.

2. The intelligent distribution system for dyeing machine auxiliaries according to claim 1, characterized in that: The dyeing agent auxiliary data extraction module comprises: An accuracy impact information feature extraction unit, used for extracting features from the delivery accuracy impact information collected from the database to obtain a feature vector associated with the delivery accuracy impact information; The auxiliary agent flow feature extraction unit is used to extract features from the auxiliary agent flow values ​​at multiple predetermined time points collected by the mass flow meter to obtain the auxiliary agent flow feature vector.

3. The intelligent distribution system for dyeing machine auxiliaries according to claim 2 is characterized in that: The accuracy impact information feature extraction unit includes: The precision impact information word segmentation subunit is used to segment the delivery precision impact information collected from the database to obtain a delivery precision impact information input word sequence; The accuracy impact information feature encoding subunit is used to perform feature encoding on the delivery accuracy impact information input word sequence to obtain the delivery accuracy impact information associated feature vector.

4. The intelligent distribution system for dyeing machine auxiliaries according to claim 3 is characterized in that: The precision-affecting information feature encoding subunit comprises: A precision-affecting word sequence semantic encoding secondary subunit is used to input the delivery precision-affecting information into a word sequence through a delivery precision-affecting information semantic encoder based on a converter to obtain a plurality of delivery precision-affecting information feature vectors; The precision impact attention encoding secondary sub-unit is used to perform attention encoding on the multiple delivery precision impact information feature vectors to obtain the delivery precision impact information associated feature vector.

5. The intelligent distribution system for dyeing machine auxiliaries according to claim 4, characterized in that: The precision affects the attention encoding secondary subunits, including: Passing the multiple delivery accuracy impact information feature vectors through a delivery accuracy impact information feature encoder based on an attention mechanism to obtain multiple delivery accuracy impact information attention feature vectors; The multiple delivery accuracy impact information attention feature vectors are associated to obtain the delivery accuracy impact information association feature vector.

6. The intelligent distribution system for dyeing machine auxiliaries according to claim 5, characterized in that: The auxiliary agent flow characteristic extraction unit comprises: Arranging the auxiliary agent flow values ​​at a plurality of predetermined time points collected by the mass flow meter according to the time dimension to obtain an auxiliary agent flow input vector; The auxiliary agent flow input vector is passed through an auxiliary agent flow timing encoder to obtain the auxiliary agent flow feature vector.

7. A method for intelligent distribution of dyeing machine auxiliaries, characterized in that: include: Acquire the delivery accuracy impact information collected by the database and the auxiliary agent flow values ​​at multiple predetermined time points collected by the mass flow meter; Extracting a distribution accuracy impact information associated feature vector and an additive flow feature vector from the distribution accuracy impact information collected by the database and the additive flow values ​​at a plurality of predetermined time points collected by the mass flow meter; Based on the distribution accuracy impact information association feature vector and the auxiliary agent flow feature vector, a dyeing machine auxiliary agent distribution pre-stop value is obtained; Wherein, based on the distribution accuracy impact information association feature vector and the auxiliary agent flow feature vector, the auxiliary agent distribution pre-stop value of the dyeing machine is obtained, including: The distribution accuracy influencing information correlation feature vector and the auxiliary agent flow feature vector are merged to obtain the dyeing machine auxiliary agent distribution feature vector; The dyeing machine auxiliary agent distribution feature vector is expressed in an equivalent model parameter space based on class probability to obtain an optimized dyeing machine auxiliary agent distribution feature vector; Passing the optimized dyeing machine auxiliary agent distribution feature vector through a decoder to obtain a decoded value, wherein the decoded value is used to represent a pre-stop value for dyeing machine auxiliary agent distribution; The dyeing machine auxiliary agent distribution feature vector is expressed in an equivalent model parameter space based on class probability to obtain an optimized dyeing machine auxiliary agent distribution feature vector, including: The dyeing machine auxiliary agent distribution feature vector is passed through a softmax activation function to obtain a category probability vector; Calculating the joint probability distribution of the dyeing machine auxiliary agent distribution feature vector relative to the class probability vector to obtain a class probability guided dyeing machine auxiliary agent distribution feature vector; The class probability guides the dyeing machine auxiliary agent distribution feature vector through a model parameter latent space mapper based on a fully connected layer to obtain the dyeing machine auxiliary agent distribution feature vector in an equivalent model parameter space; Performing feature interaction on the dyeing machine auxiliary agent distribution feature vector under the equivalent model parameter space and the dyeing machine auxiliary agent distribution feature vector to obtain the optimized dyeing machine auxiliary agent distribution feature vector; Among them, performing feature interaction on the dyeing machine auxiliary agent distribution feature vector under the equivalent model parameter space and the dyeing machine auxiliary agent distribution feature vector to obtain the optimized dyeing machine auxiliary agent distribution feature vector includes: The dyeing machine auxiliary agent distribution feature vector and the dyeing machine auxiliary agent distribution feature vector in the equivalent model parameter space are subjected to feature interaction to obtain the optimized dyeing machine auxiliary agent distribution feature vector, wherein the optimization formula is: in, represents the dyeing machine auxiliary agent distribution feature vector in the equivalent model parameter space, represents the dyeing machine auxiliary agent distribution feature vector, is the transformation matrix, is the bias parameter; represents the activation function, Represents the connection characteristics, is a threshold value, represents the optimized dyeing machine auxiliary agent distribution feature vector.

Images