Quality tracing method and platform for polyester fiber production

Abstract

The invention discloses a quality tracing method and platform for polyester fiber production, and relates to the technical field of polyester fiber production, and the method comprises the steps: carrying out the sensor deployment and production sensing monitoring of N fiber production nodes, and obtaining the production sensing data flow of N nodes; performing mapping storage on the N node production sensing data streams based on N production node equipment blocks to obtain N node production data storage blocks; performing correlation analysis on the polyester fiber quality detection database, and constructing a polyester fiber quality anomaly detection forest; and performing polyester fiber production quality tracing based on the N node production data storage blocks and the polyester fiber quality anomaly detection forest. The technical effects that intelligent quality traceability and targeted quality anomaly detection of multi-polyester fiber production are achieved, the traceability integrity of production quality data is ensured, and then the quality traceability accuracy and traceability efficiency are improved are achieved.

Description

Technical Field

[0001] The present invention relates to the technical field of polyester fiber production, and in particular to a quality tracing method and platform for polyester fiber production. Background Art

[0002] As an important textile raw material, quality control and traceability management in the production process of polyester fiber are crucial to ensure product quality and consumer trust. However, existing traceability methods usually rely on paper records and manual management, which have problems such as low efficiency, high error rate, and incomplete data. With the rapid development of information technology, intelligent and automated traceability management solutions have gradually become an inevitable trend in the development of the industry. Summary of the invention

[0003] This application solves the technical problems of low intelligence level of quality traceability of polyester fiber production in the prior art, resulting in insufficient traceability accuracy and low traceability efficiency, by providing a quality traceability method and platform for polyester fiber production. The application achieves the technical effect of realizing intelligent quality traceability and targeted quality anomaly detection of multiple polyester fibers through data distributed block storage and fiber quality anomaly detection forest, ensuring the traceability integrity of production quality data, and thereby improving the accuracy and efficiency of quality traceability.

[0004] In view of the above problems, the present invention provides a quality traceability method and platform for polyester fiber production.

[0005] In a first aspect, the present application provides a quality traceability method for polyester fiber production, the method comprising: obtaining a polyester fiber production line, extracting production nodes of the polyester fiber production line, and obtaining N fiber production nodes; performing sensor deployment and production perception monitoring on the N fiber production nodes, and obtaining N node production perception data streams; performing distributed block deployment on the N fiber production nodes, and obtaining N production node equipment blocks; mapping and storing the N node production perception data streams based on the N production node equipment blocks, and obtaining N node production data storage blocks; collecting and obtaining a polyester fiber quality detection database, performing correlation analysis on the polyester fiber quality detection database, and constructing a polyester fiber quality anomaly detection forest; and tracing the polyester fiber production quality based on the N node production data storage blocks and the polyester fiber quality anomaly detection forest.

[0006] On the other hand, the present application also provides a quality traceability platform for polyester fiber production, the platform comprising: a production node extraction module, used to obtain a polyester fiber production line, extract production nodes from the polyester fiber production line, and obtain N fiber production nodes; a production perception monitoring module, used to perform sensor deployment and production perception monitoring on the N fiber production nodes, and obtain N node production perception data streams; a distributed block deployment module, used to perform distributed block deployment on the N fiber production nodes, and obtain N production node equipment blocks; a data mapping storage module, used to map and store the N node production perception data streams based on the N production node equipment blocks, and obtain N node production data storage blocks; an anomaly detection forest construction module, used to collect and obtain a polyester fiber quality detection database, perform correlation analysis on the polyester fiber quality detection database, and construct a polyester fiber quality anomaly detection forest; a production quality tracing module, used to trace the polyester fiber production quality based on the N node production data storage blocks and the polyester fiber quality anomaly detection forest.

[0007] One or more technical solutions provided in this application have at least the following technical effects or advantages: Due to the extraction of production nodes of polyester fiber production line, N fiber production nodes are obtained, and then sensors are deployed and production perception monitoring is performed on the N fiber production nodes, and N node production perception data streams are obtained. At the same time, distributed block deployment is performed on the N fiber production nodes to obtain N production node equipment blocks, and the N node production perception data streams are mapped and stored based on the N production node equipment blocks to obtain N node production data storage blocks, and the polyester fiber quality detection database is subjected to correlation analysis to construct a polyester fiber quality anomaly detection forest, and a technical solution for polyester fiber production quality tracing based on the N node production data storage blocks and the polyester fiber quality anomaly detection forest is comprehensively implemented. Then, the intelligent quality traceability and targeted quality anomaly detection of multiple polyester fiber production are realized through data distributed block storage and fiber quality anomaly detection forest, and the integrity of production quality data traceability is ensured, thereby improving the technical effect of quality traceability accuracy and traceability efficiency.

[0008] The above description is only an overview of the technical solution of the present application. In order to more clearly understand the technical means of the present application, it can be implemented in accordance with the contents of the specification. In order to make the above and other purposes, features and advantages of the present application more obvious and easy to understand, the specific implementation methods of the present application are listed below. BRIEF DESCRIPTION OF THE DRAWINGS

[0009] Figure 1 This is a flow chart of a quality traceability method for polyester fiber production in this application; Figure 2This is a schematic structural diagram of a quality traceability platform for polyester fiber production in this application.

[0010] Explanation of the accompanying drawings: production node extraction module 11, production perception monitoring module 12, distributed block deployment module 13, data mapping storage module 14, anomaly detection forest construction module 15, production quality traceability module 16. DETAILED DESCRIPTION

[0011] This application solves the technical problems of low intelligence level of quality traceability of polyester fiber production in the prior art, resulting in insufficient traceability accuracy and low traceability efficiency, by providing a quality traceability method and platform for polyester fiber production. The application achieves the technical effect of realizing intelligent quality traceability and targeted quality anomaly detection of multiple polyester fibers through data distributed block storage and fiber quality anomaly detection forest, ensuring the traceability integrity of production quality data, and thereby improving the accuracy and efficiency of quality traceability.

[0012] In order to make the purpose, technical solution and advantages of the present application more clearly understood, the present application is further described in detail below in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present application and are not used to limit the present application.

[0013] The present application is described below in conjunction with the drawings in the present application.

[0014] Embodiment 1, as Figure 1 As shown, the present application provides a quality traceability method for polyester fiber production, the method comprising: Step S1: Acquire a polyester fiber production line, extract production nodes from the polyester fiber production line, and obtain N fiber production nodes.

[0015] Specifically, in order to realize the intelligent quality traceability of polyester fiber production, first, obtain the polyester fiber production line, which is a complex and sophisticated system including multiple production nodes. Extract the key production nodes of the polyester fiber production line to obtain the corresponding N fiber production nodes, where N is a positive integer greater than or equal to 2. The extracted fiber production nodes may include: raw material preparation nodes, melt conveying and filtration nodes, spinning nodes, drafting nodes, heat setting nodes, winding and cutting nodes, etc., as the data monitoring basis for subsequent polyester fiber production, thereby ensuring the quality traceability of fiber production.

[0016] Step S2: Deploy sensors and perform production perception monitoring on the N fiber production nodes to obtain production perception data streams of the N nodes.

[0017] Furthermore, the step of obtaining N nodes to produce perception data streams also includes: Perform production quality impact analysis on the N fiber production nodes respectively to obtain N production node quality impact indicator sets; perform sensor deployment analysis on the N fiber production nodes based on the N production node quality impact indicator sets to obtain N node sensor deployment parameter sets; perform sensor network deployment on the N fiber production nodes according to the N node sensor deployment parameter sets to obtain N node perception sensor networks; use the N node perception sensor networks to perform production perception monitoring on the polyester fiber production line to obtain N node production perception data streams.

[0018] Furthermore, the steps of obtaining the N node sensor deployment parameter sets in this application also include: Based on the N production node quality influencing indicator sets, the monitoring requirements of the N fiber production nodes are analyzed to determine the monitoring requirement parameters of the N production nodes; according to the N production node monitoring requirement parameters, sensor parameter analysis is performed on the N fiber production nodes to obtain N production node sensor characteristic parameters; N distribution area information of the N fiber production nodes is acquired, and sensor layout analysis is performed based on the N distribution area information to obtain N production node sensor layout parameters; based on the N production node sensor characteristic parameters and the N production node sensor layout parameters, the N node sensor deployment parameter set is obtained by combining them.

[0019] Specifically, the N fiber production nodes are deployed with sensors and production perception monitoring is performed. First, the production quality impact analysis is performed on the N fiber production nodes respectively to determine the key factors for the overall production quality. Historical data, process parameters and expert opinions can be collected to form a preliminary set of N production node quality impact indicators, which are used to show the quality impact indicators of each production node, such as the thermal stability index and structural size index of the heat setting node. The quality impact indicators corresponding to different fiber production nodes are also different, which can be used as the basis for sensor data monitoring.

[0020] Based on the N production node quality impact indicator sets, the N fiber production nodes are subjected to sensor deployment analysis. First, the monitoring requirements of the N fiber production nodes are analyzed based on the N production node quality impact indicator sets, and the N production node monitoring requirement parameters are determined. The N production node monitoring requirement parameters include the monitoring target type, monitoring range and monitoring accuracy of each production node. According to the N production node monitoring requirement parameters, the sensor parameters of the N fiber production nodes are analyzed, and the compatibility, reliability and service life of the sensors are considered. The N production node sensor characteristic parameters that meet the requirements are selected, including the measurement range, accuracy, response time, stability and other characteristic parameters of the sensors of each production node. Then, the N distribution area information of the N fiber production nodes is obtained, including the spatial position, environmental conditions, area size, etc., and the sensor layout analysis is performed based on the N distribution area information to ensure that the sensors can fully cover the key areas of the production nodes and accurately perceive the environmental information, and obtain the corresponding N production node sensor layout parameters, including the sensor layout position, layout density, topological structure, etc.

[0021] Based on the characteristic parameters of the N production node sensors and the layout parameters of the N production node sensors, the N node sensor deployment parameter sets are combined to determine, including detailed information such as sensor type, quantity, location, and connection method. The sensor network of the N fiber production nodes is deployed according to the N node sensor deployment parameter sets to ensure the stability and reliability of the sensor network, and necessary debugging and testing are performed to obtain the N node perception sensor network to achieve real-time data collection and transmission. The N node perception sensor network is used to perform production perception monitoring on the polyester fiber production line, and the production perception data stream of each production node is obtained in real time, including parameters such as temperature, humidity, pressure, and flow, and the corresponding N node production perception data streams are obtained. Comprehensive monitoring and data perception of the N fiber production nodes in the polyester fiber production line are achieved, the comprehensiveness of polyester fiber production data collection is improved, and a strong guarantee is provided for production quality traceability.

[0022] Step S3: Distributed block deployment is performed on the N fiber production nodes to obtain N production node equipment blocks.

[0023] Specifically, N fiber production nodes are deployed with distributed blocks, and the topology of the blockchain network is designed, including the node connection method, data transmission path, etc., in order to build a production node equipment network based on blockchain technology to ensure that each production node has sufficient computing resources (such as CPU, memory, storage space) and network bandwidth. Each fiber production node is deployed with a corresponding production node equipment block, and the deployed blockchain network is tested, including functional testing, performance testing, and security testing, to ensure the stability and reliability of the blockchain network, as well as the data consistency and security between nodes, so as to achieve transparency, traceability, and security of production data.

[0024] Step S4: Map and store the N node production perception data streams based on the N production node device blocks to obtain N node production data storage blocks.

[0025] Furthermore, the steps of obtaining N nodes to produce data storage blocks in this application also include: Performing attribute analysis on the N fiber production nodes to obtain attribute information of the N production nodes, and constructing a node coding system based on the attribute information of the N production nodes; encoding the N production node equipment blocks based on the node coding system to determine N node block identification codes; mapping and storing the N node production-aware encrypted data streams into the N production node equipment blocks according to the N node block identification codes to obtain the N node production data storage blocks.

[0026] Specifically, the N node production perception data streams are mapped and stored based on the N production node equipment blocks. First, the N fiber production nodes are subjected to attribute analysis to obtain the corresponding N production node attribute information, including the function, performance, production type, etc. of the node. And according to the N production node attribute information, a node coding system is designed and constructed. The node coding system is used to encode and identify each production node according to the production node attribute, so as to uniquely identify each production node and facilitate data retrieval and management. The N production node equipment blocks are encoded based on the node coding system, and the generated node code is associated with the equipment block to form a node block identification code, and the unique N node block identification codes after encoding are determined. According to the N node block identification codes, the N node production perception encrypted data streams are mapped and stored in the N production node equipment blocks, and the N node production data storage blocks are obtained after data storage, each block containing information such as the production perception data stream and equipment block identification of the corresponding production node. The production-aware data stream is mapped and stored in the corresponding production node device block according to the node block identification code, ensuring the integrity and consistency of the data, as well as the stability and reliability of the storage system, and achieving traceability and security of production data.

[0027] Step S5: Acquire a polyester fiber quality detection database, perform association analysis on the polyester fiber quality detection database, and construct a polyester fiber quality anomaly detection forest.

[0028] Furthermore, the steps of constructing a polyester fiber quality anomaly detection forest in this application also include: Acquire polyester fiber attribute factor information, classify and identify the polyester fiber quality detection database according to the polyester fiber attribute factor information, and obtain a multi-attribute factor parameter quality detection data set; perform abnormal feature identification on the multi-attribute factor parameter quality detection data set respectively, and obtain a multi-attribute factor parameter quality abnormality feature set; perform feature association analysis on the multi-attribute factor parameter quality abnormality feature set respectively, and obtain a multi-attribute factor parameter abnormality detection tree set; identify and network the multi-attribute factor parameter abnormality detection tree set to construct the polyester fiber quality abnormality detection forest.

[0029] Furthermore, the step of obtaining a multi-attribute factor parameter anomaly detection tree set in this application further includes: Perform primary feature extraction on the multi-attribute factor parameter quality abnormality feature set to determine the multi-attribute factor parameter primary abnormality feature set; perform feature association analysis on the multi-attribute factor parameter quality abnormality feature set based on the multi-attribute factor parameter primary abnormality feature set to obtain a multi-attribute factor parameter abnormality feature association degree set; screen the multi-attribute factor parameter quality abnormality feature set according to the multi-attribute factor parameter abnormality feature association degree set to obtain a multi-attribute factor parameter secondary abnormality feature set; based on the multi-attribute factor parameter secondary abnormality feature set, perform feature cascade construction on the multi-attribute factor parameter quality abnormality feature set in sequence to obtain the multi-attribute factor parameter abnormality detection tree set.

[0030] Specifically, a polyester fiber quality inspection database is collected from various sources (such as production lines, laboratory tests, customer feedback, etc.), and the data should contain various attribute factor information of polyester fibers, such as strength, elongation, work of rupture, abrasion resistance, moisture regain, fiber diameter, etc., as well as the corresponding abnormal generation reasons. The polyester fiber quality inspection database is subjected to association analysis, and firstly, polyester fiber attribute factor information is obtained, and the polyester fiber attribute factor information is used to distinguish different polyester fiber types, such as morphological structure, dyeability and application attributes, etc. The polyester fiber quality inspection database is then classified and identified according to the polyester fiber attribute factor information, and the inspection data of the same attribute factor type are integrated into one category, that is, the data set is divided into different attribute factor type subsets, and the corresponding multi-attribute factor parameter quality inspection data set is obtained.

[0031] The abnormal features of the multi-attribute factor parameter quality detection data set are identified respectively, and statistical methods (such as 3σ principle, box plot, etc.) or machine learning algorithms (such as isolation forest, LOF, etc.) are used to identify data outliers, and then analyze and obtain its abnormal pattern characteristics and corresponding abnormal generation cause characteristics, so as to integrate and obtain the multi-attribute factor parameter quality abnormal feature set. The multi-attribute factor parameter quality abnormal feature set is respectively subjected to feature association analysis, and firstly, the multi-attribute factor parameter quality abnormal feature set is subjected to primary feature extraction to determine the multi-attribute element parameter primary abnormal feature set with significant abnormal features, and the multi-attribute element parameter primary abnormal feature set includes significant major categories of abnormal pattern characteristics of multi-attribute element parameter types, such as appearance quality abnormality, physical property abnormality, dyeing property abnormality, etc.

[0032] Based on the multi-attribute element parameter primary abnormal feature set, the multi-attribute factor parameter quality abnormal feature set is subjected to feature association analysis. The correlation coefficient, mutual information, chi-square test and other methods can be used to evaluate the correlation between features, including generation cause association analysis and subdivided abnormal pattern association analysis, to obtain a multi-attribute element parameter abnormal feature association degree set with other abnormal features. The multi-attribute factor parameter quality abnormal feature set is screened according to the multi-attribute element parameter abnormal feature association degree set, and those features with a high degree of association with the multi-attribute element parameter primary abnormal feature set are retained to obtain a multi-attribute element parameter secondary abnormal feature set, including generation cause association features and subdivided secondary abnormal pattern features with strong association therewith. Based on the multi-attribute element parameter secondary abnormal feature set, the multi-attribute factor parameter quality abnormal feature set is sequentially subjected to feature cascade construction, that is, the above-mentioned feature association analysis step and feature screening step are repeated on the basis of the multi-attribute element parameter secondary abnormal feature set, and then the corresponding multi-attribute factor parameter abnormality detection tree set is obtained by adding associated features step by step, so as to be used for abnormal feature detection of polyester fiber quality of a specific attribute factor parameter type, including abnormal pattern and abnormal generation cause. The multi-attribute factor parameter anomaly detection tree set is attribute-identified and network-integrated according to the attribute factor parameter type, and multiple anomaly detection trees are integrated into an overall detection model, that is, a polyester fiber quality anomaly detection forest is constructed to detect abnormal characteristics of polyester fiber quality of different attribute factor parameter types. Through the cascade construction of the polyester fiber quality anomaly detection forest, the abnormal characteristics in the polyester fiber quality can be automatically identified, and the quality anomaly detection of polyester fibers of different attribute factor parameter types can be carried out in a targeted manner, thereby improving the accuracy of quality anomaly detection.

[0033] Step S6: Trace the polyester fiber production quality based on the N node production data storage blocks and the polyester fiber quality anomaly detection forest.

[0034] Furthermore, the polyester fiber production quality traceability based on the N node production data storage blocks and the polyester fiber quality anomaly detection forest, the application steps also include: According to the attribute information of the target polyester fiber, a target fiber quality anomaly detection tree is obtained from the polyester fiber quality anomaly detection forest; the abnormal quality detection data of the target polyester fiber is obtained, and the abnormal quality detection data is analyzed based on the target fiber quality anomaly detection tree to obtain fiber abnormality feature information; based on the N-node production data storage blocks, the fiber abnormality feature information is traced back to production data association to obtain the target fiber abnormal production perception data stream.

[0035] Specifically, the polyester fiber production quality is traced based on the N node production data storage blocks and the polyester fiber quality anomaly detection forest. First, the target polyester fiber with unqualified production quality is obtained, and then the target fiber quality anomaly detection tree with the same attribute factor type is called from the polyester fiber quality anomaly detection forest according to the attribute information of the target polyester fiber, so as to perform quality anomaly feature detection on the target polyester fiber in a targeted manner. The abnormal quality detection data of the target polyester fiber is obtained, and these data may come from real-time monitoring, sampling detection or customer feedback during the production process. The abnormal quality detection data is analyzed based on the target fiber quality anomaly detection tree, and the fiber abnormality feature information is output, including the abnormal pattern type of the target polyester fiber and the corresponding abnormal generation cause characteristics.

[0036] Based on the N node production data storage blocks, the fiber abnormal characteristic information is traced for production data association, that is, the fiber abnormal characteristic information is analyzed for associated production nodes to obtain its corresponding associated production node set, and the node production data related to these features is searched in the production data storage block through the associated production node set and the fiber abnormal characteristic information to obtain the target fiber abnormal production perception data stream associated with its abnormal quality characteristics. Through the association traceability analysis, it is possible to construct a full-chain production data flow from raw materials to finished products, especially those parts related to abnormal characteristics, which helps to trace and locate the links where the abnormalities occur, analyze the causes of the abnormalities, and subsequently propose improvement measures. Through data distributed block storage and fiber quality anomaly detection forest, intelligent quality traceability and targeted quality anomaly detection of multi-polyester fiber production can be realized to ensure the integrity of production quality data traceability, thereby improving the accuracy and efficiency of quality traceability.

[0037] In summary, the quality traceability method for polyester fiber production provided by this application has the following technical effects: Due to the extraction of production nodes of polyester fiber production line, N fiber production nodes are obtained, and then sensors are deployed and production perception monitoring is performed on the N fiber production nodes, and N node production perception data streams are obtained. At the same time, distributed block deployment is performed on the N fiber production nodes to obtain N production node equipment blocks, and the N node production perception data streams are mapped and stored based on the N production node equipment blocks to obtain N node production data storage blocks, and the polyester fiber quality detection database is subjected to correlation analysis to construct a polyester fiber quality anomaly detection forest, and a technical solution for polyester fiber production quality tracing based on the N node production data storage blocks and the polyester fiber quality anomaly detection forest is comprehensively implemented. Then, the intelligent quality traceability and targeted quality anomaly detection of multiple polyester fiber production are realized through data distributed block storage and fiber quality anomaly detection forest, and the integrity of production quality data traceability is ensured, thereby improving the technical effect of quality traceability accuracy and traceability efficiency.

[0038] Embodiment 2: Based on the same inventive concept as the quality traceability method for polyester fiber production in the above embodiment, the present invention also provides a quality traceability platform for polyester fiber production, such as Figure 2 As shown, the platform includes: A production node extraction module 11 is used to obtain a polyester fiber production line, extract production nodes from the polyester fiber production line, and obtain N fiber production nodes; a production perception monitoring module 12 is used to perform sensor deployment and production perception monitoring on the N fiber production nodes, and obtain N node production perception data streams; a distributed block deployment module 13 is used to perform distributed block deployment on the N fiber production nodes, and obtain N production node equipment blocks; a data mapping storage module 14 is used to map and store the N node production perception data streams based on the N production node equipment blocks, and obtain N node production data storage blocks; an anomaly detection forest construction module 15 is used to collect and obtain a polyester fiber quality detection database, perform correlation analysis on the polyester fiber quality detection database, and construct a polyester fiber quality anomaly detection forest; a production quality tracing module 16 is used to trace the polyester fiber production quality based on the N node production data storage blocks and the polyester fiber quality anomaly detection forest.

[0039] Furthermore, the platform is also used to perform the following steps: Perform production quality impact analysis on the N fiber production nodes respectively to obtain N production node quality impact indicator sets; perform sensor deployment analysis on the N fiber production nodes based on the N production node quality impact indicator sets to obtain N node sensor deployment parameter sets; perform sensor network deployment on the N fiber production nodes according to the N node sensor deployment parameter sets to obtain N node perception sensor networks; use the N node perception sensor networks to perform production perception monitoring on the polyester fiber production line to obtain N node production perception data streams.

[0040] Furthermore, the platform is also used to perform the following steps: Based on the N production node quality influencing indicator sets, the monitoring requirements of the N fiber production nodes are analyzed to determine the monitoring requirement parameters of the N production nodes; according to the N production node monitoring requirement parameters, sensor parameter analysis is performed on the N fiber production nodes to obtain N production node sensor characteristic parameters; N distribution area information of the N fiber production nodes is acquired, and sensor layout analysis is performed based on the N distribution area information to obtain N production node sensor layout parameters; based on the N production node sensor characteristic parameters and the N production node sensor layout parameters, the N node sensor deployment parameter set is obtained by combining them.

[0041] Furthermore, the platform is also used to perform the following steps: Performing attribute analysis on the N fiber production nodes to obtain attribute information of the N production nodes, and constructing a node coding system based on the attribute information of the N production nodes; encoding the N production node equipment blocks based on the node coding system to determine N node block identification codes; mapping and storing the N node production-aware encrypted data streams into the N production node equipment blocks according to the N node block identification codes to obtain the N node production data storage blocks.

[0042] Furthermore, the platform is also used to perform the following steps: Acquire polyester fiber attribute factor information, classify and identify the polyester fiber quality detection database according to the polyester fiber attribute factor information, and obtain a multi-attribute factor parameter quality detection data set; perform abnormal feature identification on the multi-attribute factor parameter quality detection data set respectively, and obtain a multi-attribute factor parameter quality abnormality feature set; perform feature association analysis on the multi-attribute factor parameter quality abnormality feature set respectively, and obtain a multi-attribute factor parameter abnormality detection tree set; identify and network the multi-attribute factor parameter abnormality detection tree set to construct the polyester fiber quality abnormality detection forest.

[0043] Furthermore, the platform is also used to perform the following steps: Perform primary feature extraction on the multi-attribute factor parameter quality abnormality feature set to determine the multi-attribute factor parameter primary abnormality feature set; perform feature association analysis on the multi-attribute factor parameter quality abnormality feature set based on the multi-attribute factor parameter primary abnormality feature set to obtain a multi-attribute factor parameter abnormality feature association degree set; screen the multi-attribute factor parameter quality abnormality feature set according to the multi-attribute factor parameter abnormality feature association degree set to obtain a multi-attribute factor parameter secondary abnormality feature set; based on the multi-attribute factor parameter secondary abnormality feature set, perform feature cascade construction on the multi-attribute factor parameter quality abnormality feature set in sequence to obtain the multi-attribute factor parameter abnormality detection tree set.

[0044] Furthermore, the platform is also used to perform the following steps: According to the attribute information of the target polyester fiber, a target fiber quality anomaly detection tree is obtained from the polyester fiber quality anomaly detection forest; the abnormal quality detection data of the target polyester fiber is obtained, and the abnormal quality detection data is analyzed based on the target fiber quality anomaly detection tree to obtain fiber abnormality feature information; based on the N-node production data storage blocks, the fiber abnormality feature information is traced back to production data association to obtain the target fiber abnormal production perception data stream.

[0045] The foregoing Figure 1 The various variations and specific examples of a quality traceability method for polyester fiber production in Example 1 are also applicable to a quality traceability platform for polyester fiber production in this embodiment. Through the above detailed description of a quality traceability method for polyester fiber production, those skilled in the art can clearly know the implementation method of a quality traceability platform for polyester fiber production in this embodiment, so for the sake of brevity of the specification, it will not be described in detail here.

[0046] This specification and the accompanying drawings are merely exemplary illustrations of the present application, but the protection scope of the present application is not limited thereto. It should be noted that any technician familiar with the technical field can easily think of changes or replacements within the technical scope disclosed in the present application, which should be included in the protection scope of the present application. In some cases, the actions or steps recorded in the present application can be performed in an order different from that in the embodiment and the desired results can still be achieved. In addition, the process depicted in the accompanying drawings does not necessarily require the specific order or continuous order shown to achieve the desired results. In some embodiments, multitasking and parallel processing are also possible or may be advantageous.

Claims

1. A quality traceability method for polyester fiber production, characterized in that: The method comprises: Obtain a polyester fiber production line, extract production nodes from the polyester fiber production line, and obtain N fiber production nodes; Deploy sensors and perform production perception monitoring on the N fiber production nodes to obtain production perception data streams of the N nodes; Performing distributed block deployment on the N fiber production nodes to obtain N production node equipment blocks; Mapping and storing the N node production perception data streams based on the N production node device blocks to obtain N node production data storage blocks; Acquire a polyester fiber quality detection database, perform correlation analysis on the polyester fiber quality detection database, and construct a polyester fiber quality anomaly detection forest; The polyester fiber production quality is traced based on the N node production data storage block and the polyester fiber quality anomaly detection forest.

2. A quality traceability method for polyester fiber production according to claim 1, characterized in that: The obtaining of N node production perception data streams includes: Performing production quality impact analysis on the N fiber production nodes respectively to obtain a set of quality impact indicators of the N production nodes; Performing sensor deployment analysis on the N fiber production nodes based on the N production node quality impact indicator sets to obtain N node sensor deployment parameter sets; Deploy a sensor network on the N fiber production nodes according to the N node sensor deployment parameter set to obtain an N node perception sensor network; The N-node perception sensor network is used to perform production perception monitoring on the polyester fiber production line to obtain N-node production perception data streams.

3. A quality traceability method for polyester fiber production according to claim 2, characterized in that: The step of obtaining the N node sensor deployment parameter sets includes: Based on the N production node quality impact indicator sets, the N fiber production nodes are subjected to monitoring demand analysis to determine the N production node monitoring demand parameters; Performing sensor parameter analysis on the N fiber production nodes according to the monitoring requirement parameters of the N production nodes to obtain characteristic parameters of the N production node sensors; Acquire N distribution area information of the N fiber production nodes, perform sensor layout analysis based on the N distribution area information, and obtain N production node sensor layout parameters; Based on the characteristic parameters of the N production node sensors and the deployment parameters of the N production node sensors, the N node sensor deployment parameter sets are obtained by combination.

4. A quality traceability method for polyester fiber production according to claim 1, characterized in that: The step of obtaining N nodes to produce data storage blocks includes: Performing attribute analysis on the N fiber production nodes to obtain attribute information of the N production nodes, and constructing a node coding system according to the attribute information of the N production nodes; Encoding the N production node device blocks based on the node coding system to determine N node block identification codes; The N node production-aware encrypted data streams are mapped and stored in the N production node device blocks according to the N node block identification codes to obtain the N node production data storage blocks.

5. A quality traceability method for polyester fiber production according to claim 1, characterized in that: The method for constructing a polyester fiber quality anomaly detection forest comprises: Acquire polyester fiber attribute factor information, classify and identify the polyester fiber quality detection database according to the polyester fiber attribute factor information, and obtain a multi-attribute factor parameter quality detection data set; Respectively performing abnormal feature recognition on the multi-attribute factor parameter quality detection data set to obtain a multi-attribute factor parameter quality abnormal feature set; Performing feature association analysis on the multi-attribute factor parameter quality anomaly feature sets respectively to obtain a multi-attribute factor parameter anomaly detection tree set; The multi-attribute factor parameter anomaly detection tree set is identified and networked to construct the polyester fiber quality anomaly detection forest.

6. A quality traceability method for polyester fiber production as claimed in claim 5, characterized in that: The step of obtaining a multi-attribute factor parameter anomaly detection tree set includes: Extracting primary features from the multi-attribute factor parameter quality abnormality feature set to determine the multi-attribute factor parameter primary abnormality feature set; Based on the primary abnormal feature set of the multi-attribute factor parameters, a feature association analysis is performed on the quality abnormal feature set of the multi-attribute factor parameters to obtain a multi-attribute factor parameter abnormal feature association degree set; The multi-attribute factor parameter quality abnormal feature set is screened according to the multi-attribute factor parameter abnormal feature association degree set to obtain a multi-attribute factor parameter secondary abnormal feature set; Taking the multi-attribute element parameter secondary anomaly feature set as a benchmark, the multi-attribute factor parameter quality anomaly feature set is sequentially constructed by feature cascading to obtain the multi-attribute factor parameter anomaly detection tree set.

7. A quality traceability method for polyester fiber production according to claim 1, characterized in that: The polyester fiber production quality tracing based on the N node production data storage block and the polyester fiber quality anomaly detection forest includes: According to the property information of the target polyester fiber, a target fiber quality anomaly detection tree is obtained by calling from the polyester fiber quality anomaly detection forest; Acquire abnormal quality detection data of the target polyester fiber, perform abnormal analysis on the abnormal quality detection data based on the target fiber quality abnormality detection tree, and obtain fiber abnormality feature information; Based on the N node production data storage blocks, the fiber abnormality feature information is traced back to production data association to obtain the target fiber abnormality production perception data stream.

8. A quality traceability platform for polyester fiber production, characterized in that: Used to implement a quality traceability method for polyester fiber production according to any one of claims 1 to 7, the platform comprises: A production node extraction module is used to obtain a polyester fiber production line, extract production nodes from the polyester fiber production line, and obtain N fiber production nodes; A production perception monitoring module, used to perform sensor deployment and production perception monitoring on the N fiber production nodes, and obtain production perception data streams of the N nodes; A distributed block deployment module, used to perform distributed block deployment on the N fiber production nodes to obtain N production node equipment blocks; A data mapping storage module, used for mapping and storing the N node production perception data streams based on the N production node device blocks to obtain N node production data storage blocks; An anomaly detection forest construction module is used to collect and obtain a polyester fiber quality detection database, perform association analysis on the polyester fiber quality detection database, and construct a polyester fiber quality anomaly detection forest; A production quality tracing module is used to trace the production quality of polyester fiber based on the N node production data storage blocks and the polyester fiber quality anomaly detection forest.

Images