Gradient co-extrusion control system for multilayer nanocomposite reinforced hose

Abstract

The application discloses a gradient co-extrusion forming control system of a multilayer nanocomposite reinforced hose, relates to the technical field of extrusion forming, and comprises the following modules: an extrusion temperature acquisition module, which synchronously acquires temperature sequences of a plurality of co-extrusion machine outlets and generates a temperature sequence set; a temperature gradient analysis module, which analyzes the temperature sequences based on a time mapping strategy and obtains an asynchronous fluctuation coefficient set; a three-dimensional tomography scanning module, which continuously scans the forming hose by means of X-ray CT and acquires a three-dimensional image sequence; a forming smoothness identification module, which identifies a smoothness deviation value in the image; and a parameter adjustment and forming control module, which adjusts the extrusion outlet parameters based on the temperature fluctuation and the deviation value and realizes co-extrusion forming control. Thus, the stability of the forming process of the multilayer nanocomposite reinforced hose and the quality of the forming hose are improved.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of extrusion molding, in particular to a gradient co-extrusion molding control system for a multi-layer nanocomposite reinforced hose. BACKGROUND

[0002] In the production process of the multi-layer nanocomposite reinforced hose, the traditional molding control method mainly focuses on the individual temperature control of each extrusion port and the overall quality detection of the molded hose, and often ignores the mutual influence of the temperatures at the extrusion ports of the multi-port co-extrusion machine and the dynamic characteristics of the temperature gradient changes during the molding process. At the same time, the existing molding control is usually based on a simple temperature feedback mechanism, and the collection and analysis of the extrusion temperature lack systematicness and depth, which makes it difficult to capture the potential fluctuation rules in the temperature sequence in time, resulting in inaccurate and timely adjustment of the extrusion parameters, thereby affecting the quality of the molded hose. SUMMARY

[0003] The present application provides a gradient co-extrusion molding control system for a multi-layer nanocomposite reinforced hose to solve the technical problems of single and extensive control strategy in the prior art, which affects the production stability and quality of the hose product, and achieves the technical effect of improving the stability of the molding process of the multi-layer nanocomposite reinforced hose and the quality of the molded hose.

[0004] The gradient co-extrusion molding control system for a multi-layer nanocomposite reinforced hose provided by the present application comprises:

[0005] An extrusion temperature sequence acquisition module is configured to configure a multi-port co-extrusion machine, synchronously and continuously acquire temperature sequences at the extrusion ports of a set of extrusion ports of the multi-port co-extrusion machine, and determine a set of extrusion temperature sequences.

[0006] A temperature gradient fluctuation analysis module is configured to perform iterative analysis on the set of extrusion temperature sequences based on a simultaneous sequence mapping iterative analysis strategy, and determine a set of co-extrusion temperature gradient asynchronous fluctuation coefficients.

[0007] A three-dimensional tomographic scanning module is configured to continuously scan the molded hose output by the multi-port co-extrusion machine using X-ray CT to obtain a sequence of three-dimensional tomograms.

[0008] A molding smoothness identification module is configured to traverse the sequence of three-dimensional tomograms to identify the molding smoothness, and determine a set of molding smoothness deviation values.

[0009] A parameter adjustment and molding control module is configured to adjust the extrusion parameters of the set of extrusion ports based on the set of co-extrusion temperature gradient asynchronous fluctuation coefficients and the set of molding smoothness deviation values, obtain a set of adjusted extrusion parameters, and control the co-extrusion molding of the set of extrusion ports based on the set of adjusted extrusion parameters.

[0010] In a feasible implementation, the extrusion temperature sequence set is iteratively analyzed based on a simultaneous sequence mapping iterative analysis strategy to determine a set of temperature gradient asynchronous fluctuation coefficients, and the temperature gradient fluctuation analysis module comprises:

[0011] A temperature sequence combination unit is configured to combine the extrusion temperature sequence set in a simultaneous sequence mapping manner to obtain an extrusion temperature mapping combination sequence.

[0012] A sequence gradient calculation unit is configured to calculate the difference between two extrusion temperatures associated with each other in the extrusion temperature mapping combination sequence to obtain a set of temperature association gradient sequences.

[0013] A gradient analysis unit is configured to iteratively analyze the set of temperature association gradient sequences to determine a set of temperature association gradient asynchronous fluctuation coefficients.

[0014] In a feasible implementation, the set of temperature association gradient sequences is iteratively analyzed to determine a set of temperature association gradient asynchronous fluctuation coefficients, and the execution steps of the gradient analysis unit comprise:

[0015] The fluctuation variance of the set of temperature association gradient sequences is calculated to obtain a first set of temperature association gradient fluctuation variances.

[0016] The set of temperature association gradient sequences is subjected to multi-scale convolution, and the multi-scale convolution results are iteratively interacted to obtain a second set of temperature association gradient fluctuation trend coefficients.

[0017] The first set of temperature association gradient fluctuation variances and the second set of temperature association gradient fluctuation trend coefficients are subjected to weighted calculation to determine a set of initial temperature association gradient asynchronous fluctuation coefficients.

[0018] The set of initial temperature association gradient asynchronous fluctuation coefficients is extracted based on the extrusion port set, and the extracted results are subjected to mean value calculation to determine a set of temperature association gradient asynchronous fluctuation coefficients.

[0019] In a feasible implementation, the set of temperature association gradient sequences is subjected to multi-scale convolution, and the multi-scale convolution results are iteratively interacted to obtain a second set of temperature association gradient fluctuation trend coefficients, and the execution steps comprise:

[0020] A first temperature association gradient sequence is extracted from the set of temperature association gradient sequences.

[0021] The first temperature association gradient sequence is subjected to multi-scale convolution, and the multi-scale convolution is iteratively interacted to determine a first second temperature association gradient fluctuation trend coefficient.

[0022] Traverse the temperature correlation gradient group sequence for analysis to determine the second temperature correlation gradient fluctuation trend coefficient set.

[0023] In a feasible implementation, the first temperature correlation gradient sequence is subjected to multi-scale convolution, and the multi-scale convolution is iteratively interacted to determine the first second temperature correlation gradient fluctuation trend coefficient, and the execution step includes:

[0024] The first temperature correlation gradient sequence is subjected to multi-scale convolution to obtain a first multi-scale convolution gradient feature set.

[0025] The first multi-scale convolution gradient feature set is iteratively interacted to obtain a first iteratively interacted convolution gradient feature.

[0026] The first iteratively interacted convolution gradient feature is subjected to trend coefficient analysis to obtain the first second temperature correlation gradient fluctuation trend coefficient.

[0027] In a feasible implementation, the first multi-scale convolution gradient feature set is iteratively interacted to obtain the first iteratively interacted convolution gradient feature, and the execution step includes:

[0028] Two multi-scale convolution gradient features are randomly extracted from the first multi-scale convolution gradient feature set, feature similarity calculation is performed using a cosine similarity calculation formula to obtain a feature similarity set.

[0029] The feature similarity set is subjected to normalization processing and added to an empty matrix to obtain an iteratively interacted matrix.

[0030] The iteratively interacted matrix and any one of the two multi-scale convolution gradient features are subjected to convolution operation to obtain a stage iteratively interacted convolution gradient feature.

[0031] By analogy, one multi-scale convolution gradient feature is again randomly extracted from the first multi-scale convolution gradient feature set, and is iteratively interacted with the stage iteratively interacted convolution gradient feature until the multi-scale convolution gradient features in the first multi-scale convolution gradient feature set are all interacted, and the first iteratively interacted convolution gradient feature is obtained.

[0032] In a feasible implementation, the first iteratively interacted convolution gradient feature is subjected to trend coefficient analysis to obtain the first second temperature correlation gradient fluctuation trend coefficient, and the execution step includes:

[0033] A trend coefficient analyzer is pre-constructed, wherein the trend coefficient analyzer is obtained after a framework based on a convolutional neural network is trained.

[0034] The trend coefficient analyzer is used to analyze the trend coefficient of the first iterative interactive convolution gradient feature, to obtain a first second temperature correlation gradient fluctuation trend coefficient.

[0035] In a feasible implementation, the three-dimensional tomogram sequence is traversed to identify the forming smoothness, and a forming smoothness deviation value set is determined. The execution steps of the forming smoothness identification module include:

[0036] An interlayer defect feature identifier is acquired.

[0037] The three-dimensional tomogram sequence is subjected to interlayer defect identification by using the interlayer defect feature identifier, to obtain an interlayer defect feature set sequence.

[0038] The interlayer defect feature set sequence is subjected to forming smoothness deviation value analysis, to determine K forming smoothness deviation values, where K is the number of composite layers of the forming hose minus 1 and is a positive integer.

[0039] Based on the correspondence between the extrusion port set and the composite layers of the forming hose, the K forming smoothness deviation values are matched, and the matching results are subjected to mean value processing, to obtain the forming smoothness deviation value set, where each forming smoothness deviation value corresponds to an extrusion port.

[0040] In a feasible implementation, the interlayer defect feature set sequence is subjected to forming smoothness deviation value analysis, to determine K forming smoothness deviation values, and the execution steps include:

[0041] The interlayer defect feature set sequence is extracted with the same layer as the extraction target, to obtain K extracted interlayer defect feature sequences.

[0042] The K extracted interlayer defect feature sequences are traversed to identify the forming smoothness, to obtain K forming smoothness sequences.

[0043] The K forming smoothness sequences are subjected to deviation identification by using the standard forming smoothness, to obtain K forming smoothness deviation value sequences, and the K forming smoothness deviation value sequences are subjected to intra-sequence mean value calculation, to obtain K forming smoothness deviation values.

[0044] In a feasible implementation, the execution steps of the parameter adjustment and forming control module include:

[0045] An extrusion parameter adjuster is pre-constructed.

[0046] The co-extrusion temperature gradient asynchronous fluctuation coefficient and the forming smoothness deviation value corresponding to each extrusion port extracted from the co-extrusion temperature gradient asynchronous fluctuation coefficient set and the forming smoothness deviation value set are analyzed by using the extrusion parameter adjuster, to obtain an adjusted extrusion parameter set.

[0047] The application discloses a gradient co-extrusion forming control system of a multi-layer nanocomposite reinforced hose, comprising: an extrusion temperature sequence acquisition module, configured to a multi-port co-extrusion machine, to synchronously and continuously acquire an outlet temperature sequence of each extrusion port of the multi-port co-extrusion machine, and to obtain an extrusion temperature sequence set; a temperature gradient fluctuation analysis module, configured to iteratively analyze the extrusion temperature sequence set based on a time sequence mapping iterative analysis strategy, to calculate and determine a co-extrusion temperature gradient asynchronous fluctuation coefficient set; a three-dimensional tomographic scanning module, configured to continuously three-dimensionally tomographically scan a formed hose output by the multi-port co-extrusion machine by using an X-ray CT technology, and to obtain a three-dimensional tomographic sequence; a forming smoothness identification module, configured to traverse the three-dimensional tomographic sequence, to identify and determine a forming smoothness deviation value set; and a parameter adjustment and forming control module, configured to adjust extrusion parameters of the extrusion port set based on the co-extrusion temperature gradient asynchronous fluctuation coefficient set and the forming smoothness deviation value set, to obtain an adjusted extrusion parameter set, and to implement co-extrusion forming control on each extrusion port set according to the adjusted extrusion parameter set. The gradient co-extrusion forming control system of the multi-layer nanocomposite reinforced hose solves the technical problem of single and extensive control strategy, which affects the production stability and quality of the hose product, and achieves the technical effect of improving the stability of the multi-layer nanocomposite reinforced hose forming process and the quality of the formed hose. BRIEF DESCRIPTION OF DRAWINGS

[0048] Figure 1 It is a structural schematic diagram of the gradient co-extrusion forming control system of the multi-layer nanocomposite reinforced hose.

[0049] Figure 2 It is a flowchart of determining K forming smoothness deviation values in the gradient co-extrusion forming control system of the multi-layer nanocomposite reinforced hose.

[0050] The reference signs are explained as follows: the extrusion temperature sequence acquisition module 11, the temperature gradient fluctuation analysis module 12, the three-dimensional tomographic scanning module 13, the forming smoothness identification module 14, and the parameter adjustment and forming control module 15. DETAILED DESCRIPTION

[0051] The above technical solutions will be described in detail below by combining with the drawings in the specification and specific embodiments, so as to better understand the above technical solutions. Obviously, the described embodiments are only part of the embodiments of the application, not all the embodiments of the application, and it should be understood that the application is not limited to the example embodiments for explaining the application. Based on the embodiments of the application, all other embodiments obtained by those skilled in the art without creative labor belong to the scope of protection of the application. In addition, it should be noted that, for convenience of description, only the parts related to the application are shown in the drawings, not all.

[0052] Embodiments, Figure 1 A structural schematic diagram of a gradient co-extrusion molding control system of a multi-layer nanocomposite reinforced hose of the present application, wherein the gradient co-extrusion molding control system of the multi-layer nanocomposite reinforced hose comprises:

[0053] An extrusion temperature sequence acquisition module 11 is configured to a multi-port co-extrusion machine, and continuously acquires temperature sequences at the outlets of the outlet set of the multi-port co-extrusion machine to determine an extrusion temperature sequence set.

[0054] Specifically, the multi-port co-extrusion machine is a device that can simultaneously process multiple materials and co-extrude these materials through different extrusion outlets to form a multi-layer or multi-component structure. The outlet set refers to all extrusion outlets of such an extruder, each of which can be independently controlled for extruding different materials or different proportions of material mixtures.

[0055] Specifically, the extrusion temperature sequence set refers to the temperature data sequence continuously acquired from each extrusion outlet during production, which is crucial for the correct processing of materials and the quality of the final product. In other words, the temperature of the extrusion outlet needs to be accurately controlled to ensure that the layers of materials can be smoothly combined without obvious cross sections, thereby ensuring the stability and functional complementarity between the layers.

[0056] Specifically, the temperature of each extrusion outlet of the multi-port co-extrusion machine is continuously and synchronously acquired, for example, by installing temperature sensors at each extrusion outlet to monitor and record the temperature of the extruded material in real time, continuously collecting and storing as a temperature sequence set for subsequent analysis and adjustment of extrusion parameters to ensure that each layer of material is processed under its optimal temperature conditions.

[0057] A temperature gradient fluctuation analysis module 12 is configured to iteratively analyze the extrusion temperature sequence set based on a simultaneous sequence mapping iterative analysis strategy to determine a co-extrusion temperature gradient asynchronous fluctuation coefficient set.

[0058] Specifically, the co-extrusion temperature gradient asynchronous fluctuation coefficient is a quantitative representation of the fluctuation degree of the different extrusion outlets in the multi-port co-extrusion process, which is not synchronized in the temperature gradient change dimension, in other words, through the co-extrusion temperature gradient asynchronous fluctuation coefficient set, the difference and fluctuation of the temperature change of each extrusion outlet can be represented.

[0059] In some embodiments, based on the simultaneous sequence mapping iterative analysis strategy, the extrusion temperature sequence set is iteratively analyzed to determine the co-extrusion temperature gradient asynchronous fluctuation coefficient set, and the temperature gradient fluctuation analysis module comprises:

[0060] a temperature sequence combination unit configured to combine the extrusion temperature sequence set by time sequence mapping to obtain an extrusion temperature mapping combination sequence; a sequence gradient calculation unit configured to calculate the difference between two extrusion temperatures associated with temperature in the extrusion temperature mapping combination sequence to obtain an extrusion temperature associated gradient group sequence; and a gradient analysis unit configured to perform iterative analysis on the temperature associated gradient group sequence to determine a temperature associated gradient asynchronous fluctuation coefficient set.

[0061] Specifically, the extrusion temperature mapping combination sequence reflects the temperature data of each extrusion port at different time points. The difference between the two extrusion temperatures associated with temperature is the temperature gradient described above. The asynchronous fluctuation coefficient of the temperature associated gradient indicates the non-uniformity and instability of temperature control.

[0062] Specifically, first, the temperature sequence combination unit integrates the extrusion temperature data by time sequence mapping technology, i.e., combines the temperature values of each layer based on the same timestamp to form an extrusion temperature mapping combination sequence. Each combination sequence represents a snapshot of the extrusion temperature of different extrusion ports at a certain time point, providing a data basis for subsequent gradient calculation.

[0063] Specifically, the sequence gradient calculation unit is responsible for iterating the extrusion temperature item pairs associated with temperature between adjacent layers in the above temperature mapping combination sequence, and calculating the difference to obtain an extrusion temperature associated gradient group sequence. For example:

[0064] Assuming that the co-extrusion structure contains multiple layers (first layer to Nth layer), the temperature difference between the first layer and the second layer, the second layer and the third layer, …, the (N-1)th layer and the Nth layer is calculated in turn to form the corresponding gradient value. Among them, preferably, for the extrusion port corresponding to the middle layer (such as the second layer), it is associated with the previous layer and the next layer, so its gradient associated item may be two, and the mean value of the two gradients needs to be calculated to reflect the comprehensive fluctuation of the temperature associated gradient of the layer; in contrast, the extrusion port of the outermost layer or the innermost layer is usually associated with a gradient in one direction, so the single gradient value is directly used.

[0065] Specifically, the gradient analysis unit is configured to use a time sequence mapping iterative analysis strategy to perform iterative analysis on the above temperature associated gradient group sequence, i.e., to observe the gradient change trend within each time slice and to make a horizontal comparison between multiple time slices to quantify the asynchronous fluctuation coefficient of each extrusion port temperature associated gradient; wherein the temperature associated gradient asynchronous fluctuation coefficient is used to measure the collaborative consistency of the temperature control state between layers. The larger the value, the more unstable the temperature control response of the involved extrusion port, which is prone to cause sudden changes in material viscosity during the molding process, poor layer adhesion, and other problems.

[0066] Specifically, the final output temperature-related gradient asynchronous fluctuation coefficient set can be used as a basis for adjusting the control precision of the co-extrusion temperature control system, that is, when the asynchronous fluctuation coefficient of a certain layer is high, the temperature control closed-loop control frequency needs to be increased or more precise PID parameters need to be used to improve the interlayer coordination and stability of the extrusion molding.

[0067] Through the above temperature gradient fluctuation analysis mechanism, the fluctuation characteristics of the temperature control state between different layers in the co-extrusion structure are accurately modeled and iteratively tracked, which helps to identify abnormal temperature control risk areas in advance and provides a decision basis for adaptive precision configuration of the temperature control system, ensuring the consistency of the molding and the stability of the performance of the multi-layer composite structure in the dynamic co-extrusion process.

[0068] In some implementations, the temperature-related gradient group sequence is iteratively analyzed to determine a temperature-related gradient asynchronous fluctuation coefficient set, and the execution steps of the gradient analysis unit include:

[0069] The fluctuation variance of the temperature-related gradient group sequence is calculated to obtain a first temperature-related gradient fluctuation variance group; the temperature-related gradient group sequence is subjected to multi-scale convolution, and the multi-scale convolution results are iteratively interacted to obtain a second temperature-related gradient fluctuation trend coefficient group; the first temperature-related gradient fluctuation variance group and the second temperature-related gradient fluctuation trend coefficient group are subjected to weighted calculation to determine an initial temperature-related gradient asynchronous fluctuation coefficient group; the initial temperature-related gradient asynchronous fluctuation coefficient group is extracted based on the extrusion port set, and the extracted results are subjected to mean value calculation to determine a temperature-related gradient asynchronous fluctuation coefficient set.

[0070] Specifically, the gradient analysis unit is used to deeply model the fluctuation and trend of the temperature-related gradient group sequence, and the gradient asynchronous characteristics are extracted through composite statistical analysis and time series convolution, and the steps include:

[0071] First, the temperature-related gradient group sequence is traversed, and for each pair of interlayer gradients with temperature correlation, the fluctuation variance is calculated within the time series range to quantify the fluctuation intensity of the gradient over time; the obtained first temperature-related gradient fluctuation variance group reflects the stability index of the temperature control correlation between different extrusion layers, and the greater the fluctuation variance, the more significant the dynamic instability of the gradient.

[0072] Then, the temperature-related gradient group sequence is subjected to multi-scale convolution processing, and through the introduction of a sliding weighting mechanism with different window lengths, the temperature control change characteristics from short-term dramatic changes to long-term trends are captured.

[0073] Then, the convolution results at multiple scales are iteratively processed to extract the interaction correlation and trend coupling between the temperature change trends at different levels, and the obtained second temperature correlation gradient fluctuation trend coefficient set represents the consistency degree of the temperature control trend across scales. For example, if the trend coupling degree is low, it indicates that the temperature control strategy is out of balance at different time scales.

[0074] Further, the two index groups (i.e., the first fluctuation variance group and the second trend coefficient group) are weighted to fuse the temperature fluctuation intensity and trend coupling information in two dimensions. For example, the weighting method can be performed according to a preset empirical parameter based on a fluctuation influence weight factor. The initial temperature correlation gradient asynchronous fluctuation coefficient set generated by the above steps is used to express the gradient asynchronous fluctuation degree between the levels under the current forming condition.

[0075] Further, the gradient mapping relationship corresponding to each extrusion port is extracted from the initial temperature correlation gradient asynchronous fluctuation coefficient set, wherein the gradient items generated by the upper and lower adjacent layers of each extrusion port are usually 1 to 2, and the extraction result needs to be averaged to measure the overall asynchronous fluctuation state of the extrusion port. The final temperature correlation gradient asynchronous fluctuation coefficient set is used to develop the temperature control precision configuration strategy of the subsequent control module.

[0076] Through the above steps, the variance analysis and trend extraction of the temperature correlation gradient are cooperatively fused and modeled, which not only can identify the risk of severe fluctuations in a short time, but also can identify the deviation trend of long-term temperature control coordination. It provides strong quantitative index support for the adaptive adjustment logic of the control system, which helps to improve the temperature control precision and interlayer adhesion stability in the multi-layer gradient co-extrusion process.

[0077] In some implementations, the temperature correlation gradient group sequence is subjected to multi-scale convolution, and the multi-scale convolution results are iteratively interacted to obtain a second temperature correlation gradient fluctuation trend coefficient set. The execution steps include:

[0078] A first temperature correlation gradient sequence is extracted from the temperature correlation gradient group sequence. The first temperature correlation gradient sequence is subjected to multi-scale convolution, and the multi-scale convolution is iteratively interacted to determine a first second temperature correlation gradient fluctuation trend coefficient. The temperature correlation gradient group sequence is traversed and analyzed to determine a second temperature correlation gradient fluctuation trend coefficient set.

[0079] Specifically, random selection is performed in the temperature correlation gradient group sequence to obtain a first temperature correlation gradient sequence, such as a temperature correlation gradient sequence of any extrusion port, which is used to reflect the temperature gradient data of the upper and lower extrusion layers corresponding to the extrusion port at a specific time point in the presence of a temperature coupling relationship. Then, multi-scale convolution operations (such as a short-time window of 3 steps, a medium-time window of 7 steps, and a long-time window of 15 steps) are performed on the sequence to analyze the change characteristics of the temperature gradient at different scales and capture fluctuation patterns at micro and macro levels. The convolution kernel can be set as a weighted average or a high-pass / low-pass function to extract fluctuation patterns or trend directions at different scales, and the convolution results at each scale represent gradient trend responses at different granularities.

[0080] Further, the convolution results at different scales are input into an iterative fusion mechanism (such as a weighted fusion or an integration algorithm based on a time sequence correlation coefficient) to improve the coordinated explanation capability between local fluctuations and overall trends while preserving the change characteristics at different scales until a stable second temperature correlation gradient fluctuation trend coefficient is determined. Meanwhile, temperature correlation gradient sequences corresponding to multiple extrusion ports in the temperature correlation gradient group sequence are traversed, and multiple second temperature correlation gradient fluctuation trend coefficients obtained are integrated to obtain a complete second temperature correlation gradient fluctuation trend coefficient set.

[0081] The above steps introduce multi-scale convolution and trend iterative fusion mechanisms, which can not only identify sensitive fluctuations at small time scales but also perceive long-period trend imbalances, thereby constructing the trend dynamic perception capability of the gradient temperature control system.

[0082] In some implementations, the first temperature correlation gradient sequence is subjected to multi-scale convolution, and the multi-scale convolution is iteratively interacted to determine a first second temperature correlation gradient fluctuation trend coefficient, and the execution steps include:

[0083] The first temperature correlation gradient sequence is subjected to multi-scale convolution to obtain a first multi-scale convolution gradient feature set, the first multi-scale convolution gradient feature set is iteratively interacted to obtain a first iteratively interacted convolution gradient feature, and the first iteratively interacted convolution gradient feature is subjected to trend coefficient analysis to obtain a first second temperature correlation gradient fluctuation trend coefficient.

[0084] Specifically, multi-scale convolution is a signal processing technique used to analyze the characteristics of data at different scales or resolutions. Iterative interaction refers to repeatedly exchanging information between multiple processing steps to gradually optimize or refine the results and identify and predict the trend of temperature fluctuations.

[0085] Specifically, first, the first temperature correlation gradient sequence is extracted from the temperature correlation gradient group sequence. This sequence contains temperature gradient data of different layers at a specific time point, and multiple time windows are selected as convolution scales. The sliding convolution operation is performed on the first temperature correlation gradient sequence, and each convolution scale corresponds to a trend extraction result of a time resolution. A first multi-scale convolution gradient feature set is obtained, which includes a local trend feature set at multiple different granularities.

[0086] Specifically, then, the gradient features at each scale in the above feature set are compared and fused one by one in an iterative interaction manner. For example, the multi-scale features are interactively evolved by using weighted superposition, mutual correlation weighted average, maximum response value screening, etc. to form the fused trend feature output, i.e., the first iterative interaction convolution gradient feature. This feature can better represent the dominant factor of the trend change in the real temperature control system. For example, Table 1 is a processing example:

[0087] Table 1 Temperature data and multi-scale convolution feature extraction and processing table

[0088] Serial number Original temperature Convolutional feature (scale 1) Convolutional feature (scale 2) Interaction feature (weighted average) formula Interaction feature value 1 25 - - - - 2 27 26.2 26.5 0.6×26.2+0.4×26.5 26.32 3 26 26.8 27.8 0.6×26.8+0.4×27.8 27.28 4 28 28 29.1 0.6×28.0+0.4×29.1 28.64 5 30 29.6 30.5 0.6×29.6+0.4×30.5 29.92 6 31 30.8 30.3 0.6×30.8+0.4×30.3 30.68 7 29 30.2 28.9 0.6×30.2+0.4×28.9 29.56 8 27 28.4 27.4 0.6×28.4+0.4×27.4 27.84 9 26 26.8 - - - 10 25 - - - -

[0089] The scale 1 convolution kernel is [0.2, 0.6, 0.2], and 8 convolution feature values are obtained.

[0090] The scale 2 convolution kernel is [0.1, 0.2, 0.4, 0.2, 0.1], and 7 convolution feature values are obtained.

[0091] The lengths of the two scale convolution features are different, so the interaction processing is only performed on the first 7 data points where both have values.

[0092] The weight is set to scale 1 weight 0.6 and scale 2 weight 0.4.

[0093] The first iterative interaction convolution gradient feature sequence is: [26.32, 27.28, 28.64, 29.92, 30.68, 29.56, 27.84].

[0094] Further, the first iterative interaction convolution gradient feature is analyzed for trend by using linear fitting slope, local fluctuation direction consistency ratio, or trend stability function, etc. to obtain a quantitative index as a first second temperature correlation gradient fluctuation trend coefficient, which is used to measure the trend fluctuation intensity of the gradient sequence at multiple scales. In other words, the larger the trend coefficient value, the more significant the fluctuation trend, and the higher the requirement of the system for the dynamic adaptation of the extrusion outlet temperature control strategy.

[0095] Through the above process, multi-scale and integrated analysis of the trend of key gradient fluctuations in the extrusion process can be achieved, which not only enhances the sensitivity of the system to local disturbances, but also improves the response capability to trend anomalies, assists the temperature control system in dynamically adjusting the response weight or control strategy of each layer, and thus effectively improves the molding consistency and interface stability of the product.

[0096] In some implementations, the first multi-scale convolution gradient feature set is iteratively interacted to obtain a first iteratively interacted convolution gradient feature, and the execution step includes:

[0097] Two multi-scale convolution gradient features are randomly extracted from the first multi-scale convolution gradient feature set, and a cosine similarity calculation formula is used to calculate the feature similarity, to obtain a feature similarity set. The feature similarity set is normalized and added to an empty matrix to obtain an iterative interaction matrix. The iterative interaction matrix and any one of the two multi-scale convolution gradient features are convolved to obtain a stage iteratively interacted convolution gradient feature. Similarly, a multi-scale convolution gradient feature is randomly extracted from the first multi-scale convolution gradient feature set again, and is iteratively interacted with the stage iteratively interacted convolution gradient feature until the multi-scale convolution gradient features in the first multi-scale convolution gradient feature set are all interacted, to obtain the first iteratively interacted convolution gradient feature.

[0098] Specifically, the multi-scale convolution gradient features include temperature gradient change rate features, local temperature extreme value features, and temperature fluctuation frequency features. The temperature gradient change rate feature refers to the change speed of the temperature gradient over time, the local temperature extreme value feature refers to the highest or lowest temperature reached in a specific region, and the temperature fluctuation frequency feature describes the frequency of temperature change. These features help to comprehensively analyze different dimensions of temperature change. The iterative interaction matrix is a matrix for storing and updating feature similarity, which helps to gradually integrate features during iteration.

[0099] Specifically, first, two multi-scale convolution gradient features are randomly extracted from the first multi-scale convolution gradient feature set, each feature in the set can include temperature gradient change rate features, local temperature extreme value features, temperature fluctuation frequency features, and the like. Then, the similarity index between the two extracted features is calculated based on each feature dimension, for example, the cosine similarity calculation formula is used to calculate the vector similarity of the temperature gradient change rate sequence, the extreme point distribution vector, and the frequency domain fluctuation amplitude vector, respectively, to obtain the feature similarity set between the two features in different feature dimensions.

[0100] Specifically, then, the feature similarity set is normalized and filled into the initialized empty matrix to obtain an iterative interaction matrix representing the current interaction process. This matrix can represent the degree of interaction between different dimensional features, laying the foundation for subsequent feature fusion.

[0101] Further, the obtained iterative interaction matrix is convolved with any one of the two multi-scale convolution gradient features to obtain a stage iterative interaction convolution gradient feature of the current stage. This feature combines the structural change trend and similarity features of the initial two features, and has certain statistical generalization ability.

[0102] Further, on this basis, a new feature is randomly extracted from the first multi-scale convolution gradient feature set, and the same similarity calculation, interaction matrix construction and convolution fusion processing are performed with the stage iterative interaction convolution gradient feature obtained in the previous step. Repeat the above process until all multi-scale convolution gradient features in the set participate in iterative interaction, and finally obtain the first iterative interaction convolution gradient feature.

[0103] Through the above process, the multi-scale convolution gradient features can be effectively integrated and optimized, so that the contribution of each feature is fully considered and utilized. Through the use of cosine similarity, the similarity between features is accurately measured, and normalization and iterative interaction matrix ensure the stability and consistency of the calculation, thereby providing a reliable basis for subsequent temperature control and product quality optimization.

[0104] In some implementations, the first iterative interaction convolution gradient feature is analyzed for trend coefficients to obtain a first second temperature correlation gradient fluctuation trend coefficient, and the execution step includes:

[0105] A trend coefficient analyzer is pre-constructed, wherein the trend coefficient analyzer is obtained after training a framework based on a convolutional neural network; the trend coefficient analyzer is used to analyze the first iterative interaction convolution gradient feature for trend coefficients to obtain a first second temperature correlation gradient fluctuation trend coefficient.

[0106] Optionally, a trend coefficient analyzer is first pre-constructed. The analyzer can use a convolutional neural network (CNN) as the core to construct a framework, and is trained on a large number of historical temperature gradient sequences and their labeled fluctuation trend coefficient data sets to extract the feature structures of various temperature fluctuation patterns. The analyzer constructed by the convolutional neural network has deep feature learning and local change perception ability, can adapt to complex temperature fluctuation structures in the multi-layer gradient co-extrusion process, and has strong trend recognition accuracy.

[0107] Further, the first iteration of the interactive convolution gradient feature obtained previously is input into the trend coefficient analyzer. The trend coefficient analyzer performs trend strength modeling and quantitative prediction on the input feature through multi-layer convolution extraction, feature pooling, and regression mapping operations, and finally outputs a numerical indicator representing the current temperature-related gradient fluctuation trend, i.e., the first second temperature-related gradient fluctuation trend coefficient. Since the first iteration of the interactive convolution gradient feature has fused multi-scale temperature gradient information and its similarity interaction information, it contains comprehensive local gradient fluctuation, frequency characteristics, and change trend parameters, which is conducive to improving the accuracy of trend recognition.

[0108] Through the above trend analysis mechanism, the stability and adaptability of temperature fluctuation trend extraction can be improved, and the failure problem of traditional artificial threshold setting or single fitting method under complex working conditions can be avoided, thereby providing a more reliable reference basis for subsequent dynamic adjustment of temperature control accuracy.

[0109] The three-dimensional tomography scanning module 13 is configured to perform continuous three-dimensional scanning on the formed hose output by the multi-port co-extrusion machine using X-ray CT to obtain a three-dimensional tomography sequence.

[0110] Specifically, by performing three-dimensional scanning on the formed hose output by the multi-port co-extrusion machine, an internal three-dimensional tomography sequence of the hose can be obtained. The three-dimensional tomography sequence reflects the material distribution and structural characteristics of the cross section of the produced multi-layer nanocomposite reinforced hose, and can detect whether there are defects such as bubbles, cracks, or material unevenness inside the hose, which helps to adjust production parameters in a timely manner and ensure product quality.

[0111] The forming smoothness recognition module 14 is configured to traverse the three-dimensional tomography sequence to recognize the forming smoothness and determine a forming smoothness deviation value set.

[0112] Specifically, the forming smoothness recognition module is configured to analyze the three-dimensional tomography sequence to evaluate the surface and internal smoothness of the hose. The forming smoothness deviation value set represents the deviation of the surface and internal structure of the hose from the ideal state, which is affected by the extrusion pressure and extrusion rate.

[0113] Specifically, the module traverses the three-dimensional tomography sequence to recognize the smoothness features in each tomography, such as protrusions, depressions, or irregular textures, through image processing algorithms. Then, the deviation values between the smoothness feature sequence and the ideal smoothness are calculated respectively to form a deviation value set to reflect any irregularities or defects that occur during the production process.

[0114] In some embodiments, the forming smoothness recognition module traverses the three-dimensional tomography sequence to recognize the forming smoothness and determine a forming smoothness deviation value set. The execution steps of the forming smoothness recognition module include:

[0115] An interlayer defect feature identifier is obtained; the interlayer defect feature identifier is used to identify layered defects in the three-dimensional tomographic sequence to obtain an interlayer defect feature set sequence; based on the interlayer defect feature set sequence, molding smoothness deviation value analysis is performed to determine K molding smoothness deviation values, where K is the number of composite layers of the molding hose minus 1 and is a positive integer; based on the correspondence between the extrusion port set and the composite layers of the molding hose, the K molding smoothness deviation values ​​are matched, and the matching results are averaged to obtain the molding smoothness deviation value set, where each molding smoothness deviation value corresponds to one extrusion port.

[0116] Specifically, an interlaminar defect feature recognizer is a tool based on image processing and pattern recognition technologies that can automatically detect the presence of defect features, such as pores, cracks, and delamination, between layers in a tomographic image. For example, this interlaminar defect feature recognizer can be a pre-trained image recognizer based on YOLO or R-CNN.

[0117] Specifically, each set in the interlayer defect feature set sequence corresponds to an interlayer defect feature in a tomographic image. Based on the interlayer defect feature set sequence, the degree of smoothness deviation between each layer of the molded hose can be quantitatively analyzed, thereby determining K molding smoothness deviation values. Here, K is the number of composite layers of the molded hose minus 1, and is a positive integer, because there is an interface between each two adjacent layers, and its smoothness needs to be evaluated.

[0118] Specifically, the system first acquires a pre-trained interlayer defect feature recognizer, processes the 3D tomographic sequence layer by layer, identifies defect features at each interface, and constructs a sequence of interlayer defect feature sets. Then, based on this sequence, it calculates indicators such as the rate of change of interface curvature, surface fitting residual, and normalized high-frequency interference amplitude to obtain K corresponding molding smoothness deviation values. Finally, combining a pre-defined one-to-one mapping relationship between the set of extrusion ports and the composite layer, each deviation value is associated with a specific extrusion port and averaged to output a set of molding smoothness deviation values ​​for analysis and control.

[0119] Through the above process, the molding smoothness deviation of each layer can be systematically analyzed and quantified. By comparing with the standard, it can be ensured that the identified deviations accurately reflect the actual problems in production. In particular, the calculation of the mean within the sequence further improves the stability and reliability of the results, which helps to achieve more precise quality control and production optimization.

[0120] In some implementations, such as Figure 2 As shown, based on the interlayer defect feature set sequence, molding smoothness deviation value analysis is performed to determine K molding smoothness deviation values. The steps include:

[0121] extracting, from the interlayer defect feature set sequence, K interlayer defect feature sequences belonging to the same layer, each of which contains interlayer defect feature information related to the layer; identifying molding smoothness of the K interlayer defect feature sequences to obtain K molding smoothness sequences; identifying deviation of the K molding smoothness sequences from a standard molding smoothness to obtain K molding smoothness deviation value sequences, and calculating the mean value of each of the K molding smoothness deviation value sequences to obtain K molding smoothness deviation values.

[0122] Specifically, the standard molding smoothness refers to a reference value or ideal state of the molding smoothness of each layer preset according to production process requirements, which is used for comparison with the actual molding smoothness.

[0123] Specifically, first, K interlayer defect feature sequences belonging to the same layer are extracted from the interlayer defect feature set sequence, each of which contains interlayer defect feature information related to the layer. For example, for a five-layer molded hose, K = 5-1 = 4, i.e., four interlayer defect feature sequences of the interlayer interface need to be extracted. Then, the K interlayer defect feature sequences are traversed, and a predefined molding smoothness evaluation algorithm (such as a surface roughness calculation-based or defect feature weight-based comprehensive evaluation method) is used to identify the molding smoothness of each sequence to obtain K molding smoothness sequences. Each molding smoothness sequence contains multiple data points, representing the molding smoothness of the layer at different production times.

[0124] Further, the deviation of each molding smoothness sequence from the standard molding smoothness is identified, i.e., the difference between each data point and the standard value is calculated (such as by Euclidean distance), to obtain K molding smoothness deviation value sequences. Finally, the mean value of each molding smoothness deviation value sequence is calculated to obtain K molding smoothness deviation values.

[0125] The above steps play a role in quality quantitative evaluation in the entire gradient co-extrusion molding control system, and by converting interlayer defect features into specific molding smoothness deviation values, accurate quantitative basis is provided for subsequent extrusion parameter adjustment.

[0126] The parameter adjustment and molding control module 15 is configured to adjust the extrusion parameters of the extrusion port set based on the co-extrusion temperature gradient asynchronous fluctuation coefficient set and the molding smoothness deviation value set to obtain an adjusted extrusion parameter set, and perform co-extrusion molding control on the extrusion port set according to the adjusted extrusion parameter set.

[0127] Specifically, the parameter adjustment and molding control module 15 is responsible for adjusting the parameters of each extrusion port based on the co-extrusion temperature gradient asynchronous fluctuation coefficient set and the molding smoothness deviation value set to ensure that the parameters of each extrusion port can optimally control the melt temperature and the molding process.

[0128] In some embodiments, the parameter adjustment and the execution step of the molding control module include:

[0129] A pre-constructed extrusion parameter adjuster is used to analyze the co-extrusion temperature gradient asynchronous fluctuation coefficient and the molding smoothness deviation value corresponding to each extrusion port extracted from the co-extrusion temperature gradient asynchronous fluctuation coefficient set and the molding smoothness deviation value set, and obtain an adjusted extrusion parameter set.

[0130] Specifically, the extrusion parameter adjuster is constructed based on a mathematical model or a neural network model, which can predict and adjust the extrusion parameters according to the input co-extrusion temperature gradient asynchronous fluctuation coefficient and the molding smoothness deviation value. In other words, such an adjuster establishes a mapping relationship between the temperature gradient fluctuation, the molding quality and the specific operation parameters of the extruder, so that the control process is more accurate and automated.

[0131] Specifically, in the production process of the multi-layer nanocomposite reinforced hose, the co-extrusion temperature gradient asynchronous fluctuation coefficient obtained by the gradient analysis unit and the smoothness deviation value in the molding process are input into the pre-constructed extrusion parameter adjuster, the data are analyzed and the corresponding extrusion parameter adjustment suggestion is output. For example, if the temperature gradient fluctuation coefficient of an extrusion port is high, it indicates that the temperature control at this position needs to be optimized, and the adjuster may suggest reducing the heating temperature or adjusting the heating rate. Similarly, if the molding smoothness deviation value is large, the adjuster may suggest changing the extrusion speed or pressure to achieve a better molding effect.

[0132] Through the above process, using the pre-constructed extrusion parameter adjuster can significantly improve the automation and accuracy in the production process, thereby improving the consistency of the product and reducing the waste rate. In addition, it also helps to reduce human operation errors, improve production efficiency and reduce production cost. By adjusting and optimizing the production parameters in real time, it can ensure that the production process is more stable, thereby improving the product quality.

[0133] In summary, the gradient co-extrusion molding control system of the multi-layer nanocomposite reinforced hose provided by the present application has the following technical effects:

[0134] The extrusion temperature sequence acquisition module is used for configuring a multi-port co-extrusion machine, synchronously and continuously collecting temperature sequences at the outlets of each extrusion port of the multi-port co-extrusion machine, and obtaining an extrusion temperature sequence set; the temperature gradient fluctuation analysis module is based on a time sequence mapping iterative analysis strategy, iteratively analyzes the extrusion temperature sequence set, calculates and determines a co-extrusion temperature gradient asynchronous fluctuation coefficient set; the three-dimensional tomographic scanning module uses X-ray CT technology to continuously perform three-dimensional tomographic scanning on a formed hose output by the multi-port co-extrusion machine, and obtains a three-dimensional tomographic sequence; the forming smoothness identification module iterates through the three-dimensional tomographic sequence, identifies and determines a forming smoothness deviation value set; the parameter adjustment and forming control module is based on the co-extrusion temperature gradient asynchronous fluctuation coefficient set and the forming smoothness deviation value set, adjusts the extrusion parameters of the extrusion port set, obtains an adjusted extrusion parameter set, and accordingly implements co-extrusion forming control on each extrusion port set, thereby realizing the technical effect of improving the stability of the multi-layer nanometer composite reinforced hose forming process and the quality of the formed hose.

[0135] It should be understood that the disclosed embodiments and the above description enable those skilled in the art to implement the present application. Meanwhile, the present application is not limited to the above-mentioned part of the embodiments, and it should be understood that those skilled in the art can still modify the technical solutions recorded in the foregoing embodiments or make equivalent replacement for some technical features; and such modifications or replacements do not make the essence of the corresponding technical solutions deviate from the spirit and scope of the technical solutions of the embodiments of the present application, and should be included in the protection scope of the present application.

Claims

1. A gradient co-extrusion molding control system for a multi-layer nanocomposite reinforced hose, characterized by, The method comprises the following steps: an extrusion temperature sequence acquisition module is configured to configure a multi-port co-extrusion machine, synchronously and continuously acquire temperature sequences at the outlets of the outlet set of the multi-port co-extrusion machine, and determine a set of extrusion temperature sequences; a temperature gradient fluctuation analysis module is configured to iteratively analyze the set of extrusion temperature sequences based on a simultaneous time mapping iterative analysis strategy, and determine a set of co-extrusion temperature gradient asynchronous fluctuation coefficients; a three-dimensional tomographic scanning module is configured to continuously scan the molded hose output by the multi-port co-extrusion machine using X-ray CT, and obtain a three-dimensional tomographic sequence; a molded smoothness identification module is configured to identify the molded smoothness by traversing the three-dimensional tomographic sequence, and determine a set of molded smoothness deviation values; a parameter adjustment and molding control module is configured to adjust the extrusion parameters of the outlet set based on the set of co-extrusion temperature gradient asynchronous fluctuation coefficients and the set of molded smoothness deviation values, obtain a set of adjusted extrusion parameters, and control the co-extrusion molding of the outlet set according to the set of adjusted extrusion parameters.

2. The gradient co-extrusion molding control system for a multi-layer nanocomposite reinforced hose according to claim 1, wherein The set of extrusion temperature sequences is iteratively analyzed based on a simultaneous time mapping iterative analysis strategy, and a set of co-extrusion temperature gradient asynchronous fluctuation coefficients is determined. The temperature gradient fluctuation analysis module comprises: a temperature sequence combination unit configured to simultaneously map and combine the set of extrusion temperature sequences, and obtain an extrusion temperature mapping combination sequence; a sequence gradient calculation unit configured to traverse each extrusion temperature mapping combination in the extrusion temperature mapping combination sequence, calculate the difference between two extrusion temperatures with temperature correlation, and obtain a set of extrusion temperature correlation gradient sequences; a gradient analysis unit configured to iteratively analyze the set of temperature correlation gradient sequences, and determine a set of temperature correlation gradient asynchronous fluctuation coefficients.

3. The gradient co-extrusion molding control system for a multi-layer nanocomposite reinforced hose according to claim 2, wherein The set of temperature correlation gradient sequences is iteratively analyzed, and a set of temperature correlation gradient asynchronous fluctuation coefficients is determined. The execution steps of the gradient analysis unit comprise: traverse and calculate the fluctuation variance of the set of temperature correlation gradient sequences to obtain a first set of temperature correlation gradient fluctuation variances; perform multi-scale convolution on the set of temperature correlation gradient sequences, and iteratively interact the multi-scale convolution results to obtain a second set of temperature correlation gradient fluctuation trend coefficients; perform weighted calculation on the first set of temperature correlation gradient fluctuation variances and the second set of temperature correlation gradient fluctuation trend coefficients to determine an initial set of temperature correlation gradient asynchronous fluctuation coefficients; extract the initial set of temperature correlation gradient asynchronous fluctuation coefficients based on the outlet set, and perform mean value calculation on the extraction results to determine a set of temperature correlation gradient asynchronous fluctuation coefficients.

4. The gradient co-extrusion molding control system for a multi-layer nanocomposite reinforced hose according to claim 3, wherein The set of temperature correlation gradient sequences is iteratively analyzed, and a set of temperature correlation gradient asynchronous fluctuation coefficients is determined. The execution steps of the gradient analysis unit comprise: extract a first temperature correlation gradient sequence from the set of temperature correlation gradient sequences; perform multi-scale convolution on the first temperature correlation gradient sequence, and iteratively interact the multi-scale convolution to determine a first second temperature correlation gradient fluctuation trend coefficient; traverse and analyze the set of temperature correlation gradient sequences to determine a second set of temperature correlation gradient fluctuation trend coefficients.

5. The gradient co-extrusion molding control system for a multi-layer nanocomposite reinforced hose of claim 4, wherein, The first temperature correlation gradient sequence is subjected to multi-scale convolution, and the multi-scale convolution is iteratively interacted to determine the first second temperature correlation gradient fluctuation trend coefficient. The execution steps include: The first temperature correlation gradient sequence is subjected to multi-scale convolution to obtain a first multi-scale convolution gradient feature set; Iterative interaction is performed on the first multi-scale convolution gradient feature set to obtain a first iterative interaction convolution gradient feature; The first iterative interaction convolution gradient feature is subjected to trend coefficient analysis to obtain a first second temperature correlation gradient fluctuation trend coefficient.

6. The gradient co-extrusion molding control system for a multi-layer nanocomposite reinforced hose of claim 5, wherein, Iterative interaction is performed on the first multi-scale convolution gradient feature set to obtain a first iterative interaction convolution gradient feature. The execution steps include: Two multi-scale convolution gradient features are randomly extracted from the first multi-scale convolution gradient feature set, and feature similarity calculation is performed using a cosine similarity calculation formula to obtain a feature similarity set; The feature similarity set is normalized and added to an empty matrix to obtain an iterative interaction matrix; Convolution operation is performed on the iterative interaction matrix and any one of the two multi-scale convolution gradient features to obtain a stage iterative interaction convolution gradient feature; By analogy, a multi-scale convolution gradient feature is again randomly extracted from the first multi-scale convolution gradient feature set, and iterative interaction is performed between the multi-scale convolution gradient feature and the stage iterative interaction convolution gradient feature until the multi-scale convolution gradient features in the first multi-scale convolution gradient feature set are all interacted, and the first iterative interaction convolution gradient feature is obtained.

7. The gradient co-extrusion molding control system for a multi-layer nanocomposite reinforced hose of claim 5, wherein, The first iterative interaction convolution gradient feature is subjected to trend coefficient analysis to obtain a first second temperature correlation gradient fluctuation trend coefficient. The execution steps include: A trend coefficient analyzer is pre-constructed, wherein the trend coefficient analyzer is obtained after training a framework based on a convolutional neural network; The trend coefficient analyzer is used to perform trend coefficient analysis on the first iterative interaction convolution gradient feature to obtain a first second temperature correlation gradient fluctuation trend coefficient.

8. The gradient co-extrusion molding control system for a multi-layer nanocomposite reinforced hose of claim 1, wherein, The three-dimensional fault map sequence is traversed to identify the molding smoothness deviation value set. The execution steps of the molding smoothness identification module include: An interlayer defect feature identifier is obtained; The three-dimensional fault map sequence is subjected to layered defect identification using the interlayer defect feature identifier to obtain an interlayer defect feature set sequence; The molding smoothness deviation value analysis is performed according to the interlayer defect feature set sequence to determine K molding smoothness deviation values, wherein K is the number of composite layers of the molding hose minus 1 and is a positive integer; Based on the correspondence between the extrusion port set and the composite layers of the molding hose, the K molding smoothness deviation values are matched, and the matching results are subjected to mean value processing to obtain the molding smoothness deviation value set, wherein each molding smoothness deviation value corresponds to an extrusion port.

9. The gradient co-extrusion molding control system for a multi-layer nanocomposite reinforced hose of claim 8, wherein, The molding smoothness deviation value analysis is performed according to the interlayer defect feature set sequence to determine K molding smoothness deviation values. The execution steps include: The interlayer defect feature set sequence is extracted with the same layer as the extraction target, and K extracted interlayer defect feature sequences are obtained; The K extracted interlayer defect feature sequences are traversed to identify the molding smoothness, and K molding smoothness sequences are obtained; The K molding smoothness sequences are identified for deviation using a standard molding smoothness, K molding smoothness deviation value sequences are obtained, and the K molding smoothness deviation value sequences are calculated for intrasequence mean values, and K molding smoothness deviation values are obtained.

10. The gradient co-extrusion molding control system for a multi-layer nanocomposite reinforced hose of claim 1, wherein, The execution steps of the parameter adjustment and molding control module include: A pre-constructed extrusion parameter adjuster; The co-extrusion temperature gradient asynchronous fluctuation coefficient and the molding smoothness deviation value corresponding to each extrusion port extracted from the co-extrusion temperature gradient asynchronous fluctuation coefficient set and the molding smoothness deviation value set are analyzed using the extrusion parameter adjuster, and an adjusted extrusion parameter set is obtained.

Images