A PP base film aging device and process
Through the combination of deep neural network and feature fusion module, the electrostatic charge level of PP base film is monitored and analyzed in real time, and the subjectivity and inconsistency of manual detection in traditional maturation processes are solved, and the quality stability and production of PP base film are improved. efficiency.
Patent Information
- Application Number
- CN202410575876.2
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- Filing Date
- 2024-05-10
- Publication Date
- 2024-11-15
- Estimated Expiration
- 2044-05-10
AI Technical Summary
The traditional PP base film maturation process relies on manual detection of static charge levels, which has subjectivity and inconsistency, and cannot achieve real-time monitoring and feedback, affecting product quality and production efficiency.
The timing mode feature extractor and dynamic element-level feature fusion module based on deep neural network are used to combine temperature and humidity sensor data to monitor and analyze the static charge level in real time, and abnormal detection is performed through the responsive gamma correction fusion module and classifier.
Real-time monitoring and control of electrostatic charge levels is achieved, the quality stability of PP base film is improved, artificial errors are reduced, and the accurate identification and processing of electrostatic charge levels is ensured.
Smart Images

Figure CN118514249B_ABST
Abstract
Description
Technical Field
[0001] The invention relates to the field of polypropylene base film production, and in particular to a PP base film ripening device and process. Background Art
[0002] Static charge is a common and critical problem in the production of polypropylene (PP) based films. Specifically, during the production of PP based films, especially during the processing of polymerization, stretching and orientation, PP based films are easily affected by factors such as friction and stretching to generate static charges. The accumulation of these static charges may lead to an imbalance in the surface charge of the film material. If the static charge is not eliminated in a timely and effective manner, it may cause the unevenness of the film material and affect the quality and performance of the product. Therefore, the timely elimination of static charge is crucial to ensure the quality of PP based films. However, traditional methods for eliminating static charge include the use of ion air guns, static eliminators and other equipment, but these methods have certain limitations, such as the inability to completely eliminate static charge or the need for frequent operation.
[0003] As a new type of static charge elimination method, the aging process stores the PP base film under specific temperature and humidity conditions, utilizes the influence of temperature and humidity on static charge, promotes the dissipation and redistribution of static charge, and thus improves the quality stability of the product. Therefore, the aging process is currently an important step to ensure the stability of the performance of the PP base film.
[0004] However, the traditional PP-based film aging process usually relies on manual experience to detect abnormal fluctuations in the static charge level. Manual detection is subjective, and different operators may have different judgment criteria and preferences, resulting in inconsistent results. This subjective judgment may affect the accurate identification of abnormal fluctuations in the static charge level, and cannot guarantee the accuracy and stability of abnormal fluctuations in the static charge level. In addition, the manual detection method is usually intermittent, and it is impossible to achieve real-time monitoring and feedback of static charge level fluctuations, which may lead to abnormal situations not being discovered and handled in time, affecting product quality and production efficiency. Not only that, in the actual process of detecting abnormal fluctuations in the static charge level, no attention was paid to the impact of temperature and humidity on static charge, resulting in inaccurate judgment of abnormal static charge levels and inadequate timely handling, making it difficult to meet actual process requirements.
[0005] Therefore, an optimized PP-based film ripening process is desired. Summary of the invention
[0006] The present disclosure is made in view of the above problems. One object of the present disclosure is to provide a PP base film aging device and process.
[0007] The embodiment of the present disclosure provides a PP base film aging process, which includes:
[0008] Acquire a time series of a aging temperature and a time series of a aging humidity collected by a temperature sensor and a humidity sensor;
[0009] Acquire a time series of the electrostatic charge level of the aged PP base film collected by the electrostatic charge sensor;
[0010] Arranging the time series of the aging temperature and the time series of the aging humidity into a aging temperature time series input vector and a aging humidity time series input vector respectively according to the time dimension;
[0011] By using a time series pattern feature extractor based on a deep neural network, feature extraction is performed on the ripening temperature time series input vector and the ripening humidity time series input vector to obtain a ripening temperature time series associated feature vector and a ripening humidity time series associated feature vector;
[0012] Using a dynamic element-level feature fusion module to process the aging temperature time series associated feature vector and the aging humidity time series associated feature vector to obtain a aging process parameter time series fusion feature vector;
[0013] Arranging the time series of the electrostatic charge level according to the time dimension into an electrostatic charge level time series input vector and then passing it through the time series pattern feature extractor based on the deep neural network to obtain an electrostatic charge level time series correlation feature vector;
[0014] Performing responsiveness correction fusion analysis on the electrostatic charge level time series correlation feature vector and the aging process parameter time series fusion feature vector to obtain an electrostatic charge level-aging process time series response feature; and
[0015] Based on the electrostatic charge level-aging process timing response characteristics, it is determined whether there is an abnormal electrostatic charge level fluctuation.
[0016] For example, according to the PP-based film ripening process of an embodiment of the present disclosure, the temporal pattern feature extractor based on a deep neural network is a temporal pattern feature extractor based on a one-dimensional convolutional layer.
[0017] For example, according to the PP-based film aging process of an embodiment of the present disclosure, a dynamic element-level feature fusion module is used to process the aging temperature time series associated feature vector and the aging humidity time series associated feature vector to obtain a aging process parameter time series fusion feature vector, including:
[0018] The dynamic element-level feature fusion module is used to process the aging temperature time series associated feature vector and the aging humidity time series associated feature vector using the following element fusion formula to obtain the aging process parameter time series fusion feature vector; wherein the element fusion formula is:
[0019]
[0020] in, and are the time series correlation feature vector of the aging temperature and the time series correlation feature vector of the aging humidity, respectively, g is the time series fusion feature vector of the aging process parameters, concat(·,·) represents the cascade of vectors, t is the threshold value, t∈[0,1]], W g is the transformation matrix, b is the bias vector, and sigmoid(·) represents the sigmoid activation function.
[0021] For example, according to the PP-based film aging process of an embodiment of the present disclosure, the electrostatic charge level time series correlation feature vector and the aging process parameter time series fusion feature vector are subjected to responsiveness correction fusion analysis to obtain the electrostatic charge level-aging process time series response feature, including:
[0022] Performing feature optimization on the electrostatic charge level time series correlation feature vector and the aging process parameter time series fusion feature vector to obtain an optimized electrostatic charge level time series correlation feature vector and an optimized aging process parameter time series fusion feature vector; and
[0023] The optimized electrostatic charge level timing association feature vector and the optimized aging process parameter timing fusion feature vector are processed using a responsive gamma correction fusion module to obtain an electrostatic charge level-aging process timing response feature vector as the electrostatic charge level-aging process timing response feature.
[0024] For example, according to the PP base film aging process of an embodiment of the present disclosure, the electrostatic charge level time series correlation feature vector and the aging process parameter time series fusion feature vector are feature optimized to obtain an optimized electrostatic charge level time series correlation feature vector and an optimized aging process parameter time series fusion feature vector, including:
[0025] respectively calculating weighting coefficients of the electrostatic charge level time series correlation feature vector and the aging process parameter time series fusion feature vector to obtain a first weighting coefficient and a second weighting coefficient; and
[0026] The first weighting coefficient and the second weighting coefficient are used as weighting factors to perform weighted optimization on the electrostatic charge level timing association feature vector and the aging process parameter timing fusion feature vector to obtain the optimized electrostatic charge level timing association feature vector and the optimized aging process parameter timing fusion feature vector.
[0027] For example, according to the PP base film aging process of an embodiment of the present disclosure, the optimized electrostatic charge level timing association feature vector and the optimized aging process parameter timing fusion feature vector are processed using a responsive gamma correction fusion module to obtain an electrostatic charge level-aging process timing response feature vector as the electrostatic charge level-aging process timing response feature, including:
[0028] The responsive gamma correction fusion module is used to process the optimized electrostatic charge level timing association feature vector and the optimized aging process parameter timing fusion feature vector using the following responsive gamma correction fusion formula to obtain the electrostatic charge level-aging process timing response feature vector; wherein the responsive gamma correction fusion formula is:
[0029]
[0030] in, is the eigenvalue of the ith position in the optimized electrostatic charge level time series associated eigenvector, is the eigenvalue of the i-th position in the time series fusion feature vector of the optimized ripening process parameters, A, B, C and D are the adjustment hyperparameters, v i is the characteristic value of the i-th position in the electrostatic charge level-aging process timing response characteristic vector.
[0031] For example, according to the PP-based film aging process of an embodiment of the present disclosure, determining whether there is an abnormal static charge level fluctuation based on the static charge level-aging process timing response characteristics includes:
[0032] The electrostatic charge level-aging process time series response feature vector is passed through a classifier-based abnormality monitor to obtain a monitoring result, and the monitoring result is used to indicate whether there is an abnormal fluctuation in the electrostatic charge level.
[0033] For example, according to the PP-based film aging process of an embodiment of the present disclosure, the static charge level-aging process time series response feature vector is passed through a classifier-based abnormality monitor to obtain a monitoring result, and the monitoring result is used to indicate whether there is an abnormal static charge level fluctuation, including:
[0034] Using the fully connected layer of the classifier-based anomaly monitor to fully connect encode the static charge level-ripening process time series response feature vector to obtain an encoded classification feature vector; and
[0035] The encoded classification feature vector is input into the Softmax classification function of the classifier-based anomaly monitor to obtain the monitoring result.
[0036] The embodiment of the present disclosure also provides a PP base film ripening device, which includes:
[0037] A temperature and humidity acquisition module, used to acquire the time series of the aging temperature and the time series of the aging humidity acquired by the temperature sensor and the humidity sensor;
[0038] An electrostatic charge acquisition module, used to acquire a time series of electrostatic charge levels of the cured PP base film acquired by an electrostatic charge sensor;
[0039] A temperature and humidity vectorization module, used for arranging the time series of the aging temperature and the time series of the aging humidity into a aging temperature time series input vector and a aging humidity time series input vector according to the time dimension respectively;
[0040] A temperature and humidity feature extraction module, used for performing feature extraction on the ripening temperature time series input vector and the ripening humidity time series input vector respectively through a time series pattern feature extractor based on a deep neural network to obtain a ripening temperature time series associated feature vector and a ripening humidity time series associated feature vector;
[0041] A feature fusion module, used for processing the aging temperature time series associated feature vector and the aging humidity time series associated feature vector using a dynamic element level feature fusion module to obtain a aging process parameter time series fusion feature vector;
[0042] An electrostatic charge feature extraction module, used for arranging the time series of the electrostatic charge level according to the time dimension into an electrostatic charge level time series input vector and then passing it through the time series pattern feature extractor based on the deep neural network to obtain an electrostatic charge level time series correlation feature vector;
[0043] a responsiveness correction fusion analysis module, configured to perform responsiveness correction fusion analysis on the electrostatic charge level time series correlation feature vector and the maturation process parameter time series fusion feature vector to obtain an electrostatic charge level-maturation process time series response feature; and
[0044] The abnormality analysis module is used to determine whether there is an abnormal static charge level fluctuation based on the static charge level-ripening process timing response characteristics.
[0045] For example, according to the PP-based film ripening device of an embodiment of the present disclosure, the temporal pattern feature extractor based on a deep neural network is a temporal pattern feature extractor based on a one-dimensional convolutional layer.
[0046] According to the PP base film aging device and process of the embodiment of the present disclosure, during the aging process, the temperature sensor and humidity sensor are used to monitor and collect the aging temperature and aging humidity data in real time, and the data processing and analysis algorithm is used at the back end to perform a time-series correlation and collaborative analysis of the aging temperature and the aging humidity, and a responsive correlation and fusion with the time-series characteristics of the static charge level of the PP base film being aging is performed to detect the fluctuation of the static charge level, so as to determine whether the static charge level of the PP base film is abnormal. In this way, the static charge level in the aging process can be monitored and controlled in real time, so as to better utilize the influence of temperature and humidity on static charge to perform aging treatment, so as to promote the dissipation and redistribution of static charge, realize the elimination of static charge stress, and thus improve the quality stability of the PP base film. BRIEF DESCRIPTION OF THE DRAWINGS
[0047] In order to more clearly illustrate the technical solutions of the embodiments of the present disclosure, the following briefly introduces the drawings of the embodiments of the present disclosure. Obviously, the drawings described below only relate to some embodiments of the present disclosure, and are not intended to limit the present disclosure.
[0048] Figure 1 A schematic diagram of the application architecture of the PP base film ripening process in an embodiment of the present disclosure is shown;
[0049] Figure 2 A flow chart of a PP base film ripening process in an embodiment of the present disclosure is shown;
[0050] Figure 3 A flowchart of sub-step S570 of the PP base film ripening process in an embodiment of the present disclosure is shown;
[0051] Figure 4 A flowchart of sub-step S571 of the PP base film ripening process in an embodiment of the present disclosure is shown;
[0052] Figure 5 The schematic diagram of the structure of the PP base film ripening device in the embodiment of the present disclosure is shown;
[0053] Figure 6 An application scenario diagram of the PP base film ripening process in an embodiment of the present disclosure is shown;
[0054] Figure 7 A schematic diagram showing a storage medium according to an embodiment of the present disclosure; and
[0055] Figure 8 A flow chart of a PP-based film ripening process according to another embodiment of the present disclosure is shown. DETAILED DESCRIPTION
[0056] The technical solutions in the embodiments of the present disclosure will be described clearly and completely below in conjunction with the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present disclosure, rather than all embodiments. Based on the embodiments of the present disclosure, all other embodiments obtained by ordinary technicians in this field without creative work also fall within the scope of protection of the present disclosure.
[0057] The terms used in this specification are those general terms currently widely used in the art in consideration of the functions of the present disclosure, but these terms may vary according to the intention of a person of ordinary skill in the art, precedents, or new technologies in the art. In addition, specific terms may be selected by the applicant, and in this case, their detailed meanings will be described in the detailed description of the present disclosure. Therefore, the terms used in the specification should not be understood as simple names, but rather as an overall description based on the meaning of the terms and the present disclosure.
[0058] Although the present disclosure makes various references to certain modules in the system according to embodiments of the present disclosure, any number of different modules can be used and run on a user terminal and / or server. The modules are illustrative only, and different aspects of the systems and methods can use different modules.
[0059] Flowcharts are used in this disclosure to illustrate the operations performed by the system according to the embodiments of the present disclosure. It should be understood that the preceding or following operations are not necessarily performed precisely in order. On the contrary, various steps may be processed in reverse order or simultaneously as required. At the same time, other operations may also be added to these processes, or one or more operations may be removed from these processes.
[0060] Figure 1 A schematic diagram of the application architecture of the PP-based film ripening process in an embodiment of the present disclosure is shown, including a server 100 and a terminal device 200.
[0061] The terminal device 200 and the server 100 can be connected via the Internet to achieve mutual communication. Optionally, the above-mentioned Internet uses standard communication technology and / or protocol. The Internet is usually the Internet, but it can also be any network, including but not limited to any combination of a local area network (LAN), a metropolitan area network (MAN), a wide area network (WAN), a mobile, wired or wireless network, a dedicated network or a virtual private network. In some embodiments, the data exchanged through the network is represented by technologies and / or formats including Hyper Text Markup Language (HTML), Extensible Markup Language (XML), etc. In addition, conventional encryption technologies such as Secure Socket Layer (SSL), Transport Layer Security (TLS), Virtual Private Network (VPN), Internet Protocol Security (IPsec) can also be used to encrypt all or some links. In other embodiments, customized and / or dedicated data communication technologies can also be used to replace or supplement the above-mentioned data communication technologies.
[0062] The server 100 can provide various network services for the terminal device 200, wherein the server 100 can be a single server, a server cluster consisting of several servers, or a cloud computing center. Specifically, the server 100 may include a processor 110 (Center Processing Unit, CPU), a memory 120, an input device 130, and an output device 140, etc. The input device 130 may include a keyboard, a mouse, a touch screen, etc., and the output device 140 may include a display device, such as a liquid crystal display (Liquid Crystal Display, LCD), a cathode ray tube (Cathode Ray Tube, CRT), etc.
[0063] The memory 120 may include a read-only memory (ROM) and a random access memory (RAM), and provides the processor 110 with program instructions and data stored in the memory 120. In the embodiment of the present disclosure, the memory 120 may be used to store the program of the PP base film aging process in the embodiment of the present disclosure.
[0064] The processor 110 calls the program instructions stored in the memory 120, and the processor 110 is used to execute the steps of any PP base film aging process in the embodiments of the present disclosure according to the obtained program instructions.
[0065] In addition, the application architecture diagram in the embodiment of the present disclosure is intended to more clearly illustrate the technical solution in the embodiment of the present disclosure, and does not constitute a limitation on the technical solution provided in the embodiment of the present disclosure. Of course, for other application architectures and business applications, the technical solution provided in the embodiment of the present disclosure is also applicable to similar problems.
[0066] The following is a non-limiting description of the PP-based film aging process provided according to at least one embodiment of the present disclosure through several examples or embodiments. As described below, different features in these specific examples or embodiments can be combined with each other without conflicting with each other to obtain new examples or embodiments, and these new examples or embodiments also fall within the scope of protection of the present disclosure.
[0067] It should be understood that due to changes in environmental conditions, such as changes in temperature and humidity, or errors in process parameters, the traditional detection method of static charge may be unstable and inconsistent, resulting in fluctuations in product quality and limiting the application of the aging process. In addition, it is also considered that during the aging process, specific temperature and humidity conditions are crucial for the elimination of static charge. Appropriate temperature can promote the thermal motion of molecules inside the material, which helps to redistribute and dissipate the charge. At the same time, suitable humidity can reduce the accumulation of static electricity on the surface of the material and prevent the accumulation of charge.
[0068] Based on this, in the technical solution of the present application, a PP base film aging process is proposed, which can use temperature sensors and humidity sensors to monitor and collect aging temperature and aging humidity data in real time during the aging process, and use data processing and analysis algorithms at the back end to perform a coordinated analysis of the time series correlation of the aging temperature and the aging humidity, and perform a responsive correlation fusion with the time series characteristics of the static charge level of the PP base film being aging to detect the fluctuation of the static charge level, so as to determine whether the static charge level of the PP base film is abnormal. In this way, the static charge level in the aging process can be monitored and controlled in real time, so as to better utilize the influence of temperature and humidity on static charge to perform aging treatment, so as to promote the dissipation and redistribution of static charge, realize the elimination of static charge stress, and thus improve the quality stability of the PP base film.
[0069] Figure 2 1 shows a flow chart of the PP base film aging process in the embodiment of the present disclosure. For example, the PP base film aging process can be executed by a server, which can be Figure 1 The server 100 shown in FIG. Figure 2As shown, the PP base film aging process according to the embodiment of the present disclosure includes the following steps: S510, acquiring the time series of aging temperature and the time series of aging humidity collected by the temperature sensor and the humidity sensor; S520, acquiring the time series of the static charge level of the PP base film being aging collected by the static charge sensor; S530, arranging the time series of the aging temperature and the time series of the aging humidity into a aging temperature time series input vector and a aging humidity time series input vector according to the time dimension; S540, performing feature extraction on the aging temperature time series input vector and the aging humidity time series input vector respectively through a time series pattern feature extractor based on a deep neural network to obtain a aging temperature time series associated feature vector and a aging humidity time series associated feature vector; S550 , use the dynamic element-level feature fusion module to process the aging temperature time series correlation feature vector and the aging humidity time series correlation feature vector to obtain the aging process parameter time series fusion feature vector; S560, arrange the time series of the static charge level into a static charge level time series input vector according to the time dimension and then pass it through the time series pattern feature extractor based on the deep neural network to obtain the static charge level time series correlation feature vector; S570, perform responsiveness correction fusion analysis on the static charge level time series correlation feature vector and the aging process parameter time series fusion feature vector to obtain the static charge level-aging process time series response feature; and, S580, based on the static charge level-aging process time series response feature, determine whether there is an abnormal static charge level fluctuation.
[0070] Specifically, in the technical solution of the present application, first, the time series of the aging temperature and the time series of the aging humidity collected by the temperature sensor and the humidity sensor are obtained, and the time series of the static charge level of the PP base film collected by the static charge sensor is obtained. Then, considering that the aging temperature and the aging humidity have a time series correlation relationship and a dynamic change law in the time dimension, in order to capture the dynamic characteristics of the aging temperature and the aging humidity in the time series, it is necessary to first arrange the time series of the aging temperature and the time series of the aging humidity into a aging temperature time series input vector and a aging humidity time series input vector according to the time dimension, so as to retain the data distribution information of the aging temperature and aging humidity data in the time dimension.
[0071] Then, the aging temperature time series input vector and the aging humidity time series input vector are subjected to feature mining in a time series pattern feature extractor based on a one-dimensional convolutional layer to respectively extract the time series patterns and associated feature information of the aging temperature and the aging humidity in the time dimension, thereby obtaining a aging temperature time series associated feature vector and a aging humidity time series associated feature vector.
[0072] Correspondingly, in step S540, the temporal pattern feature extractor based on a deep neural network is a temporal pattern feature extractor based on a one-dimensional convolutional layer.
[0073] Then, it is considered that during the aging process, temperature and humidity conditions are crucial for monitoring and controlling the timing changes of electrostatic charge. Specifically, temperature can promote the thermal motion of molecules inside the material, which helps to redistribute and dissipate the charge. At the same time, humidity can reduce the accumulation of static electricity on the surface of the material and prevent the accumulation of charge. In addition, there is a mutual influence and timing correlation between the aging temperature and the aging humidity, and this correlation has an important impact on the quality and characteristics of the final product. Therefore, in order to capture the timing interaction and correlation feature information between the aging temperature timing characteristics and the aging humidity timing characteristics, so as to obtain more comprehensive information and better feature representation, and provide useful information for further data processing and decision-making, in the technical solution of the present application, the dynamic element-level feature fusion module is further used to process the aging temperature timing correlation feature vector and the aging humidity timing correlation feature vector to obtain the aging process parameter timing fusion feature vector. It should be understood that the dynamic element-level feature fusion module can fuse the aging temperature time series features and the aging humidity time series features based on dynamic elements, thereby using the implicit correlation and synergistic features between the aging temperature time series features and the aging humidity time series features to perform collaborative characterization of the aging process parameters in the time dimension, so as to obtain a more comprehensive and representative aging process parameter time series fusion feature vector. In this way, the fused aging process parameter time series fusion feature vector can help reveal the temporal dynamic correlation relationship and change trend between the aging temperature and the aging humidity, and help understand the temporal change law and mutual influence between the two, so as to obtain a more representative and rich feature representation.
[0074] Accordingly, in step S550, the dynamic element-level feature fusion module is used to process the aging temperature time series correlation feature vector and the aging humidity time series correlation feature vector to obtain the aging process parameter time series fusion feature vector, including: using the dynamic element-level feature fusion module to process the aging temperature time series correlation feature vector and the aging humidity time series correlation feature vector with the following element fusion formula to obtain the aging process parameter time series fusion feature vector; wherein the element fusion formula is:
[0075]
[0076] in, and are the time series correlation feature vector of the aging temperature and the time series correlation feature vector of the aging humidity, respectively, gis the time series fusion feature vector of the aging process parameters, concat(·,·) represents the cascade of vectors, t is the threshold value, t∈[0,1]], W g is the transformation matrix, b is the bias vector, and sigmoid(·) represents the sigmoid activation function.
[0077] Furthermore, for the electrostatic charge level, there is also a dynamic change law of time series in the time dimension. Therefore, in order to capture the time series change pattern and trend of the electrostatic charge level, in the technical solution of the present application, the time series of the electrostatic charge level is arranged according to the time dimension as a electrostatic charge level time series input vector, and then feature mining is performed in the time series pattern feature extractor based on the one-dimensional convolutional layer to extract the time series correlation feature information of the electrostatic charge level data in the time dimension, so as to obtain the electrostatic charge level time series correlation feature vector.
[0078] It should be understood that both the electrostatic charge level and the aging process parameters are important factors affecting product quality. At the same time, during the aging process, the aging temperature and aging humidity will affect the accumulation effect of the charge, and affect the redistribution and dissipation effect of the charge, and this influence effect is nonlinear. Therefore, in order to explore the relationship and mutual influence between the aging process parameters and the electrostatic charge level, so as to better demonstrate the nonlinear correlation and response between them to more accurately detect abnormal fluctuations in the electrostatic charge level, in the technical solution of the present application, a responsive gamma correction fusion module is further used to process the electrostatic charge level time series correlation feature vector and the aging process parameter time series fusion feature vector to obtain the electrostatic charge level-aging process time series response feature vector. Through the processing of the responsive gamma correction fusion module, the nonlinear responsive correlation and interaction between the time series characteristics of the electrostatic charge level and the element fusion characteristics between the aging temperature and the aging humidity can be used to better demonstrate the correlation and mutual influence between them, help reveal the nonlinear response details between the electrostatic charge level and the aging process parameters, thereby improving the fine-grained characterization strength of the response process, and providing more useful information for further analysis and decision-making.
[0079] Accordingly, in step S570, if Figure 3As shown, a responsive correction fusion analysis is performed on the electrostatic charge level timing association feature vector and the maturing process parameter timing fusion feature vector to obtain the electrostatic charge level-maturing process timing response feature, including: S571, feature optimization is performed on the electrostatic charge level timing association feature vector and the maturing process parameter timing fusion feature vector to obtain the optimized electrostatic charge level timing association feature vector and the optimized maturing process parameter timing fusion feature vector; and, S572, the optimized electrostatic charge level timing association feature vector and the optimized maturing process parameter timing fusion feature vector are processed using a responsive gamma correction fusion module to obtain the electrostatic charge level-maturing process timing response feature vector as the electrostatic charge level-maturing process timing response feature.
[0080] Among them, in step S572, the optimized static charge level timing association feature vector and the optimized maturing process parameter timing fusion feature vector are processed using a responsive gamma correction fusion module to obtain a static charge level-maturing process timing response feature vector as the static charge level-maturing process timing response feature, including: using the responsive gamma correction fusion module to process the optimized static charge level timing association feature vector and the optimized maturing process parameter timing fusion feature vector using the following responsive gamma correction fusion formula to obtain the static charge level-maturing process timing response feature vector; wherein the responsive gamma correction fusion formula is:
[0081]
[0082] in, is the eigenvalue of the ith position in the optimized electrostatic charge level time series associated eigenvector, is the eigenvalue of the i-th position in the time series fusion feature vector of the optimized ripening process parameters, A, B, C and D are the adjustment hyperparameters, v i is the characteristic value of the i-th position in the electrostatic charge level-aging process timing response characteristic vector.
[0083] In industrial production, it is very important to monitor the static charge level of the product, because the abnormal fluctuation of the static charge level may affect the quality and safety of the product. Then, the static charge level-ripening process timing response feature vector is passed through an abnormal monitor based on a classifier to obtain a monitoring result, and the monitoring result is used to indicate whether there is an abnormal fluctuation of the static charge level. In other words, the responsive characteristic information between the static charge level timing characteristics and the aging process parameter timing synergistic characteristics is used for classification processing, so as to perform abnormal detection of static charge level fluctuations, that is, through the analysis of the monitor, it can be known whether the current static charge level data is within the normal range. If the monitoring results show that there is an abnormality, it is necessary to take timely measures to adjust the production process or product quality control to avoid the occurrence of potential problems and ensure the stability of the PP base film. In this way, the static charge level in the aging process can be monitored and controlled in real time, so as to better utilize the influence of temperature and humidity on static charge for aging treatment, so as to promote the dissipation and redistribution of static charge, realize the elimination of static charge stress, and thus improve the quality stability of the PP base film.
[0084] Accordingly, in step S580, based on the electrostatic charge level-ripening process timing response characteristics, it is determined whether there is an abnormal electrostatic charge level fluctuation, including: passing the electrostatic charge level-ripening process timing response characteristic vector through a classifier-based anomaly monitor to obtain a monitoring result, and the monitoring result is used to indicate whether there is an abnormal electrostatic charge level fluctuation.
[0085] Specifically, the static charge level-ripening process timing response feature vector is passed through a classifier-based anomaly monitor to obtain a monitoring result, and the monitoring result is used to indicate whether there is an abnormal static charge level fluctuation, including: using the fully connected layer of the classifier-based anomaly monitor to fully connect encode the static charge level-ripening process timing response feature vector to obtain an encoded classification feature vector; and inputting the encoded classification feature vector into the Softmax classification function of the classifier-based anomaly monitor to obtain the monitoring result.
[0086] It should be understood that the role of the classifier is to use the given categories and known training data to learn classification rules and classifiers, and then classify (or predict) unknown data. Logistic regression and SVM are often used to solve binary classification problems. For multi-class classification problems, logistic regression or SVM can also be used, but multiple binary classifications are required to form a multi-classification. However, this is prone to errors and is not efficient. Commonly used multi-classification methods include the Softmax classification function.
[0087] In the technical solution described above, the static charge level time series association feature vector and the aging process parameter time series fusion feature vector respectively express the time series association characteristics of the position-by-position association between the static charge level and the aging temperature-aging humidity. Therefore, considering the source time series distribution differences between the static charge level, aging temperature, and aging humidity, the static charge level time series association feature vector and the aging process parameter time series fusion feature vector also have the distinctiveness of time series feature expression.
[0088] In this way, when the responsive gamma correction fusion module is used to process the static charge level timing association feature vector and the maturation process parameter timing fusion feature vector to obtain the static charge level-maturing process timing response feature vector, and the static charge level-maturing process timing response feature vector is classified by the abnormality monitor based on the classifier, it is expected to improve the class probability certainty of the static charge level timing association feature vector and the maturation process parameter timing fusion feature vector under the feature distribution distinguishability, thereby improving the static charge level-maturing process timing response feature vector for the predetermined class probability response fusion expression effect of the classifier-based abnormality monitor.
[0089] Based on this, the applicant of the present application respectively calculates the weighted coefficients of the electrostatic charge level time series correlation feature vector and the aging process parameter time series fusion feature vector.
[0090] Accordingly, in one example, in step S571, Figure 4 As shown, the electrostatic charge level timing association feature vector and the maturing process parameter timing fusion feature vector are feature optimized to obtain the optimized electrostatic charge level timing association feature vector and the optimized maturing process parameter timing fusion feature vector, including: S5711, respectively calculating the weighting coefficients of the electrostatic charge level timing association feature vector and the maturing process parameter timing fusion feature vector to obtain the first weighting coefficient and the second weighting coefficient; and, S5712, using the first weighting coefficient and the second weighting coefficient as weighting factors to perform weighted optimization on the electrostatic charge level timing association feature vector and the maturing process parameter timing fusion feature vector to obtain the optimized electrostatic charge level timing association feature vector and the optimized maturing process parameter timing fusion feature vector.
[0091] Wherein, in step S5711, respectively calculating the weighted coefficients of the electrostatic charge level time series correlation feature vector and the maturation process parameter time series fusion feature vector to obtain the first weighted coefficient and the second weighted coefficient, including: respectively calculating the weighted coefficients of the electrostatic charge level time series correlation feature vector and the maturation process parameter time series fusion feature vector according to the following coefficient calculation formula to obtain the first weighted coefficient and the second weighted coefficient; wherein the coefficient calculation formula is:
[0092]
[0093]
[0094] Among them, v 1-max and v 2-max are the maximum eigenvalues of the electrostatic charge level time series correlation feature vector V1 and the aging process parameter time series fusion feature vector V2, respectively. 1-j and v 2-j is the j-th eigenvalue of the electrostatic charge level time series correlation feature vector V1 and the aging process parameter time series fusion feature vector V2, L is the length of the feature vector, log represents the logarithm with base 2, exp[·] represents the exponential operation of the value, the exponential operation of the value represents the calculation of the natural exponential function value with the value as the power, and α is the weight hyperparameter, ω1 is the first weighting coefficient, and ω2 is the second weighting coefficient.
[0095] Specifically, the time series characteristics of the static charge level time series correlation feature vector V1 and the maturing process parameter time series fusion feature vector V2 are respectively represented by a majority voting mechanism in the form of time series crowdsourcing to seek the maximum expected distribution of the static charge level time series correlation feature vector V1 and the maturing process parameter time series fusion feature vector V2 relative to the class probability space under the distribution subordinate probability model. In this way, the static charge level time series correlation feature vector V1 and the maturing process parameter time series fusion feature vector V2 are weighted by coefficients ω1 and ω2 to optimize the static charge level time series correlation feature vector V1 and the maturing process parameter time series fusion feature vector V2. By using the time series fusion feature vector V2 of the curing process parameters, the distribution of the static charge level time series correlation feature vector V1 and the curing process parameter time series fusion feature vector V2 can be jointly estimated relative to the class probability space based on the average information confidence of the eigenvalues of the static charge level time series correlation feature vector V1 and the curing process parameter time series fusion feature vector V2, so as to improve the class probability certainty relative to the class probability space while maintaining the distinguishability of the feature distribution of the static charge level time series correlation feature vector V1 and the curing process parameter time series fusion feature vector V2, thereby improving the fusion expression effect of the static charge level-curing process time series response feature vector for the predetermined class probability response of the classifier-based abnormality monitor and improving the accuracy of the classification result. In this way, the influence of temperature and humidity on static charge can be used to monitor the fluctuation of static charge level in real time, so as to judge whether the static charge level of the PP base film is abnormal, so as to take corresponding measures to promote the dissipation and redistribution of static charge, eliminate static charge stress, and improve the quality stability of the PP base film.
[0096] Based on the above embodiments, see Figure 5The figure is a schematic diagram of the structure of a PP base film curing device 800 in an embodiment of the present disclosure. The PP base film curing device 800 includes: a temperature and humidity acquisition module 810, which is used to obtain the time series of the curing temperature and the time series of the curing humidity collected by the temperature sensor and the humidity sensor; an electrostatic charge acquisition module 820, which is used to obtain the time series of the electrostatic charge level of the cured PP base film collected by the electrostatic charge sensor; a temperature and humidity vectorization module 830, which is used to arrange the time series of the curing temperature and the time series of the curing humidity into a curing temperature time series input vector and a curing humidity time series input vector according to the time dimension; a temperature and humidity feature extraction module 840, which is used to extract features from the curing temperature time series input vector and the curing humidity time series input vector respectively through a time series pattern feature extractor based on a deep neural network to obtain a curing temperature time series associated feature vector and a curing humidity time series associated feature vector; a feature fusion module 850 , used to use the dynamic element-level feature fusion module to process the aging temperature time series correlation feature vector and the aging humidity time series correlation feature vector to obtain the aging process parameter time series fusion feature vector; the static charge feature extraction module 860, used to arrange the time series of the static charge level into a static charge level time series input vector according to the time dimension and then pass it through the time series pattern feature extractor based on the deep neural network to obtain the static charge level time series correlation feature vector; the responsiveness correction fusion analysis module 870, used to perform responsiveness correction fusion analysis on the static charge level time series correlation feature vector and the aging process parameter time series fusion feature vector to obtain the static charge level-aging process time series response feature; and, the abnormality analysis module 880, used to determine whether there is an abnormal static charge level fluctuation based on the static charge level-aging process time series response feature.
[0097] In one example, in the above-mentioned PP-based film aging device 800, the temporal pattern feature extractor based on a deep neural network is a temporal pattern feature extractor based on a one-dimensional convolutional layer.
[0098] Here, those skilled in the art will appreciate that the specific functions and operations of the various modules in the PP base film ripening device 800 have been described in detail above. Figures 2 to 4 The PP base film aging process has been described in detail, and therefore, its repeated description will be omitted.
[0099] Figure 6 FIG. 1 is an application scenario diagram of the PP base film ripening process according to an embodiment of the present disclosure. Figure 6 As shown, in this application scenario, first, the time series of the aging temperature and the time series of the aging humidity collected by the temperature sensor and the humidity sensor are obtained (for example, Figure 6D1 as shown in FIG. 1 ) and the time series of the electrostatic charge level of the aged PP base film collected by the electrostatic charge sensor (e.g., Figure 6 Then, the time series of the aging temperature, the time series of the aging humidity and the time series of the electrostatic charge level are input to a server (for example, Figure 6 In S) shown in , the server is able to use the PP base film aging algorithm to process the time series of the aging temperature, the time series of the aging humidity and the time series of the static charge level to obtain a monitoring result indicating whether there is an abnormal fluctuation in the static charge level.
[0100] Based on the above embodiment, another exemplary embodiment of an electronic device is provided in the embodiments of the present disclosure. In some possible implementations, the electronic device in the embodiments of the present disclosure may include a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor can implement the steps of the PP base film aging process in the above embodiment when executing the program.
[0101] For example, the electronic device is disclosed in the present invention Figure 1 Taking the server 100 in the example as an example, the processor in the electronic device is the processor 110 in the server 100, and the memory in the electronic device is the memory 120 in the server 100.
[0102] An embodiment of the present disclosure also provides a computer-readable storage medium. Figure 7 FIG. 1 is a schematic diagram of a computer-readable storage medium 1000 according to an embodiment of the present disclosure. Figure 7 As shown, the computer-readable storage medium 1000 stores computer-executable instructions 1001. When the computer-executable instructions 1001 are executed by the processor, the PP-based film ripening process according to the embodiment of the present disclosure described with reference to the above figures can be executed. The computer-readable storage medium includes, but is not limited to, for example, volatile memory and / or non-volatile memory. The volatile memory may include, for example, a random access memory (RAM) and / or a cache memory (cache), etc. The non-volatile memory may include, for example, a read-only memory (ROM), a hard disk, a flash memory, etc.
[0103] The embodiment of the present disclosure also provides a computer program product or a computer program, which includes computer instructions stored in a computer-readable storage medium. A processor of a computer device reads the computer instructions from the computer-readable storage medium, and the processor executes the computer instructions, so that the computer device performs the PP base film aging process according to the embodiment of the present disclosure.
[0104] Further, refer to Figure 8 As shown, in another embodiment of the present disclosure, a PP base film aging process is also provided, wherein the aging process refers to the curing and aging process.
[0105] Specifically, in the raw material storage, feeding and pneumatic suction process, the polypropylene resin (PP) raw material is in granular form. After unpacking, it is fed into the granular bin in the raw material supply system and transported to the screw melt extrusion system by the pneumatic suction conveying device. In the melt filtration and extrusion die process, the raw material is heated from low temperature to high temperature by the conveyor in the screw melt extrusion system in sequence, and the heating temperature is 200-250 degrees. The temperature is gradually increased to become a fully plasticized melt, which is passed through a metering pump and a filter, and then injected into a T-hanger die for extrusion, and a sheet-like melt is extruded through the lip of the die. In the chilled sheet casting process, it is cooled into a solid sheet by the chilled roller in the sheet casting machine system. The chilled roller can quickly cool the PP melt flowing out of the die head in a viscous flow state to below its glass transition temperature (80-110°C), forming a glassy sheet with uniform thickness. The temperature of the chilled roller surface is precisely controlled within the range of ±0.5°C. Cooling water at about 30°C is passed through the roller. The casting is indirectly cooled by the chilled roller and then drops to a specific crystallization temperature (below 60°C). A fully automatic air knife device is added to the non-roller-attached surface of the casting to ensure that both sides of the casting obtain the same crystallization temperature. In the thickness measurement and cross-cutting process, measuring and controlling the film thickness and its uniformity is an important technology for achieving ultra-thin thickness of high-temperature resistant capacitor films. The present invention uses an X-ray thickness measuring device with a thickness resolution of 0.01μm to measure the thickness. After the thickness is qualified, the next process is entered. In the synchronous stretching process, the diaphragm formed by the casting enters the synchronous stretching device, and the two sides of the thick sheet are clamped with a synchronous "chain clamp". It is stretched in the oven of the synchronous stretching machine. In the preheating zone, the air is heated electrically to evenly heat the diaphragm to the accurate stretching temperature. In the stretching zone, biaxial stretching is performed by width expansion, and then it undergoes a shaping process (preheating temperature is about 170°C, stretching temperature is about 160°C, and shaping temperature is about 170°C). In the film thickness measurement process, the present disclosure uses an X-ray thickness measuring device with a thickness resolution of 0.01 μm on the production line to measure the film thickness, and enters the next process after meeting the product thickness requirements. In the traction and trimming process, the base film produced by the above process is sent to the traction equipment for cooling and transportation. During the traction process, the thickness of the base film is continuously measured by an X-ray measuring instrument with a thickness resolution of 0.01 μm. The measurement data is fed back to the computer automatic control system (TEC) and the die lip gap is adjusted in time to package the film thickness and accuracy; then the edge of the "chain clamp" clamping part is removed by a trimmer. In the corona treatment process, in order to increase the surface tension strength of the film, improve the wettability of the polymer, and meet the use requirements, the present disclosure uses a corona treatment method to treat the film surface. The principle of corona treatment is to pass the film between two electrodes with high voltage. The high voltage ionizes the air between the electrodes, generates electron flow between the electrodes, and generates polarity on the surface of the film. In the winding and mid-inspection process, the film is wound by a banana roller flattening method of the winder, and samples are taken and tested in the detection center. In the aging process, aging treatment refers to storing the film roll at a specific temperature and humidity to eliminate the static charge stress.In the slitting process, a slitting machine is used to cut the film roll into small rolls of certain specifications (500-1800mm), and then packaged and stored. In the final inspection and balancing process, the inspection center performs sampling and final inspection. In the packaging and finished product storage process, a baler is used to arrange the roll core, plywood and plug on the product, load it with a pallet, and transport it to the finished product warehouse by forklift.
[0106] Those skilled in the art will appreciate that the contents disclosed in this disclosure may be subject to various modifications and improvements. For example, the various devices or components described above may be implemented by hardware, or by software, firmware, or a combination of some or all of the three.
[0107] In addition, although the present disclosure makes various references to certain units in the system according to embodiments of the present disclosure, any number of different units may be used and run on the client and / or server. The units are illustrative only, and different aspects of the systems and methods may use different units.
[0108] Those skilled in the art will appreciate that all or part of the steps in the above method can be completed by instructing related hardware through a program, and the program can be stored in a computer-readable storage medium, such as a read-only memory, a disk, or an optical disk. Optionally, all or part of the steps in the above embodiment can also be implemented using one or more integrated circuits. Accordingly, each module / unit in the above embodiment can be implemented in the form of hardware or in the form of a software functional module. The present disclosure is not limited to any particular form of combination of hardware and software.
[0109] Unless otherwise defined, all terms (including technical and scientific terms) used herein have the same meaning as commonly understood by one of ordinary skill in the art to which the present disclosure belongs. It should also be understood that terms such as those defined in common dictionaries should be interpreted as having a meaning consistent with their meaning in the context of the relevant technology and should not be interpreted in an idealized or highly formal sense, unless explicitly defined as such herein.
[0110] The above is an explanation of the present disclosure and should not be considered as a limitation thereof. Although several exemplary embodiments of the present disclosure have been described, it will be readily understood by those skilled in the art that many modifications may be made to the exemplary embodiments without departing from the novel teachings and advantages of the present disclosure. Therefore, all such modifications are intended to be included within the scope of the present disclosure as defined in the claims. It should be understood that the above is an explanation of the present disclosure and should not be considered as being limited to the specific embodiments disclosed, and modifications to the disclosed embodiments and other embodiments are intended to be included within the scope of the appended claims. The present disclosure is defined by the claims and their equivalents.
Claims
1. A PP base film aging process, characterized in that: include: Acquire a time series of a aging temperature and a time series of a aging humidity collected by a temperature sensor and a humidity sensor; Acquire a time series of the electrostatic charge level of the aged PP base film collected by the electrostatic charge sensor; Arranging the time series of the aging temperature and the time series of the aging humidity into a aging temperature time series input vector and a aging humidity time series input vector respectively according to the time dimension; By using a time series pattern feature extractor based on a deep neural network, feature extraction is performed on the ripening temperature time series input vector and the ripening humidity time series input vector to obtain a ripening temperature time series associated feature vector and a ripening humidity time series associated feature vector; Using a dynamic element-level feature fusion module to process the aging temperature time series associated feature vector and the aging humidity time series associated feature vector to obtain a aging process parameter time series fusion feature vector; Arranging the time series of the electrostatic charge level according to the time dimension into an electrostatic charge level time series input vector and then passing it through the time series pattern feature extractor based on the deep neural network to obtain an electrostatic charge level time series correlation feature vector; Performing responsiveness correction fusion analysis on the electrostatic charge level time series correlation feature vector and the aging process parameter time series fusion feature vector to obtain an electrostatic charge level-aging process time series response feature; as well as Based on the electrostatic charge level-aging process time series response characteristics, determining whether there is an abnormal electrostatic charge level fluctuation; The temporal pattern feature extractor based on deep neural network is a temporal pattern feature extractor based on one-dimensional convolutional layer; The dynamic element-level feature fusion module is used to process the aging temperature time series associated feature vector and the aging humidity time series associated feature vector to obtain a aging process parameter time series fusion feature vector, including: The dynamic element-level feature fusion module is used to process the aging temperature time series associated feature vector and the aging humidity time series associated feature vector using the following element fusion formula to obtain the aging process parameter time series fusion feature vector; wherein the element fusion formula is: ; in, and are the time series correlation feature vector of the aging temperature and the time series correlation feature vector of the aging humidity, respectively, is the time series fusion feature vector of the maturation process parameters, represents the concatenation of vectors, is the threshold value, , is the transformation matrix, is the bias vector, express Activation function.
2. The PP-based film aging process according to claim 1, characterized in that: Performing responsiveness correction fusion analysis on the electrostatic charge level time series correlation feature vector and the aging process parameter time series fusion feature vector to obtain an electrostatic charge level-aging process time series response feature, including: Performing feature optimization on the electrostatic charge level time series correlation feature vector and the aging process parameter time series fusion feature vector to obtain an optimized electrostatic charge level time series correlation feature vector and an optimized aging process parameter time series fusion feature vector; and The optimized electrostatic charge level timing association feature vector and the optimized aging process parameter timing fusion feature vector are processed using a responsive gamma correction fusion module to obtain an electrostatic charge level-aging process timing response feature vector as the electrostatic charge level-aging process timing response feature.
3. The PP-based film aging process according to claim 2, characterized in that: The electrostatic charge level time series correlation feature vector and the aging process parameter time series fusion feature vector are optimized to obtain an optimized electrostatic charge level time series correlation feature vector and an optimized aging process parameter time series fusion feature vector, including: respectively calculating weighting coefficients of the electrostatic charge level time series correlation feature vector and the aging process parameter time series fusion feature vector to obtain a first weighting coefficient and a second weighting coefficient; and The first weighting coefficient and the second weighting coefficient are used as weighting factors to perform weighted optimization on the electrostatic charge level timing association feature vector and the aging process parameter timing fusion feature vector to obtain the optimized electrostatic charge level timing association feature vector and the optimized aging process parameter timing fusion feature vector.
4. The PP-based film aging process according to claim 3, characterized in that: The optimized electrostatic charge level timing correlation feature vector and the optimized aging process parameter timing fusion feature vector are processed using a responsive gamma correction fusion module to obtain an electrostatic charge level-aging process timing response feature vector as the electrostatic charge level-aging process timing response feature, including: The responsive gamma correction fusion module is used to process the optimized electrostatic charge level timing association feature vector and the optimized aging process parameter timing fusion feature vector using the following responsive gamma correction fusion formula to obtain the electrostatic charge level-aging process timing response feature vector; wherein the responsive gamma correction fusion formula is: ; in, is the first one in the optimized electrostatic charge level time series correlation feature vector The eigenvalues at the positions, The first The feature values at each position, A, B, C and D are the hyperparameters for adjustment. is the first in the electrostatic charge level-aging process timing response characteristic vector The feature value at each position.
5. The PP-based film aging process according to claim 4, characterized in that: Based on the electrostatic charge level-aging process time series response characteristics, determining whether there is an abnormal electrostatic charge level fluctuation includes: The electrostatic charge level-aging process time series response feature vector is passed through a classifier-based abnormality monitor to obtain a monitoring result, and the monitoring result is used to indicate whether there is an abnormal fluctuation in the electrostatic charge level.
6. The PP-based film aging process according to claim 5, characterized in that: The electrostatic charge level-aging process time series response feature vector is passed through a classifier-based abnormality monitor to obtain a monitoring result, which is used to indicate whether there is an abnormal electrostatic charge level fluctuation, including: Using the fully connected layer of the classifier-based anomaly monitor to fully connect encode the static charge level-ripening process time series response feature vector to obtain an encoded classification feature vector; and The encoded classification feature vector is input into the Softmax classification function of the classifier-based anomaly monitor to obtain the monitoring result.
7. A PP base film ripening device, characterized in that: include: A temperature and humidity acquisition module, used to acquire the time series of the aging temperature and the time series of the aging humidity acquired by the temperature sensor and the humidity sensor; An electrostatic charge acquisition module, used to acquire a time series of electrostatic charge levels of the cured PP base film acquired by an electrostatic charge sensor; A temperature and humidity vectorization module, used for arranging the time series of the aging temperature and the time series of the aging humidity into a aging temperature time series input vector and a aging humidity time series input vector according to the time dimension respectively; A temperature and humidity feature extraction module, used for performing feature extraction on the ripening temperature time series input vector and the ripening humidity time series input vector respectively through a time series pattern feature extractor based on a deep neural network to obtain a ripening temperature time series associated feature vector and a ripening humidity time series associated feature vector; A feature fusion module, used for processing the aging temperature time series associated feature vector and the aging humidity time series associated feature vector using a dynamic element level feature fusion module to obtain a aging process parameter time series fusion feature vector; An electrostatic charge feature extraction module, used for arranging the time series of the electrostatic charge level according to the time dimension into an electrostatic charge level time series input vector and then passing it through the time series pattern feature extractor based on the deep neural network to obtain an electrostatic charge level time series correlation feature vector; A responsiveness correction fusion analysis module, used for performing responsiveness correction fusion analysis on the electrostatic charge level time series correlation feature vector and the aging process parameter time series fusion feature vector to obtain an electrostatic charge level-aging process time series response feature; as well as An abnormality analysis module, used to determine whether there is an abnormal static charge level fluctuation based on the static charge level-ripening process time series response characteristics; The temporal pattern feature extractor based on deep neural network is a temporal pattern feature extractor based on one-dimensional convolutional layer; The feature fusion module is used to: The dynamic element-level feature fusion module is used to process the aging temperature time series associated feature vector and the aging humidity time series associated feature vector using the following element fusion formula to obtain the aging process parameter time series fusion feature vector; wherein the element fusion formula is: ; in, and are the time series correlation feature vector of the aging temperature and the time series correlation feature vector of the aging humidity, respectively, is the time series fusion feature vector of the maturation process parameters, represents the concatenation of vectors, is the threshold value, , is the transformation matrix, is the bias vector, express Activation function.
