Preparation method of an explosion-proof composite material

Abstract

The present application discloses a preparation method of an explosion-proof and protective composite material. By obtaining the real-time hot pressing temperature collected by a temperature sensor and using a data processing and analysis algorithm based on deep learning, the real-time hot pressing temperature is subjected to time series segmentation and local time series correlation, so as to automatically obtain a recommended value of the pressing pressure at the next time point according to the aggregation representation characteristics of each real-time hot pressing temperature in the local time series. In this way, by monitoring and analyzing the temperature change in the hot pressing process in real time, the pressing pressure can be dynamically adjusted according to these subtle time series changes to ensure the molding quality of the composite material, improve the consistency of product quality, and thus enhance the degree of automation in the hot pressing process of the material.

Description

Technical Field

[0001] The present application relates to the field of intelligent preparation, and more specifically, to a method for preparing an explosion-proof protective composite material. Background Art

[0002] With the increasing safety requirements in modern society, explosion-proof composite materials are playing an increasingly important role in military, civil construction, transportation and other fields. Such materials need to have good mechanical properties, impact resistance and temperature regulation capabilities to ensure stable safety protection performance in extreme environments.

[0003] Chinese patent CN111231461A discloses an explosion-proof protective composite material, a preparation method thereof and explosion-proof protective clothing, which uses a composite material layer woven with a Steiner minimum tree topological structure as an outer layer to enhance the stability and puncture and ballistic resistance of the material, and a flexible material layer filled with a phase change cold storage composition as an inner layer, which can relieve the stuffiness of the wearer when working and improve the wearing comfort, thereby comprehensively improving the explosion-proof protective performance of the composite material and the wearer's comfort experience.

[0004] In the preparation process of explosion-proof protective composite materials, the adhesive-coated materials need to be hot-pressed to ensure a strong connection between the composite structure layer and the flexible material layer, and to ensure the structural integrity and performance stability of the composite material. In particular, during the hot pressing process, precise control of pressure is the key to ensuring the quality of material molding. However, traditional control methods usually rely on experience and fixed operating parameters and lack a real-time feedback mechanism. This means that once the pressure parameters are set, it is difficult to dynamically adjust them according to changes in the actual process. In other words, with slight changes in production conditions, these fixed rules are difficult to guarantee the best control effect, and are prone to over- or under-compensation, thus affecting the consistency and reliability of product quality.

[0005] Therefore, an optimized method for preparing explosion-proof protective composite materials is desired. Summary of the invention

[0006] To solve the above technical problems, the present application is proposed. An embodiment of the present application provides a method for preparing an explosion-proof and protective composite material, which obtains the real-time hot pressing temperature collected by a temperature sensor and adopts a data processing and analysis algorithm based on deep learning to perform time series segmentation and local time series correlation on the real-time hot pressing temperature, so as to automatically obtain the recommended value of the pressing pressure at the next time point according to the aggregation representation characteristics of each real-time hot pressing temperature in the local time series. In this way, by real-time monitoring and analyzing the temperature changes during the hot pressing process, the pressing pressure can be dynamically adjusted according to these subtle time series changes to ensure the forming quality of the composite material, improve the consistency of product quality, and thus improve the automation degree during the hot pressing process of the material.

[0007] According to one aspect of the present application, there is provided a method for preparing an explosion-proof and protective composite material, which includes: using at least two high-performance fibers to perform interleaved weaving to form a composite structure layer; manufacturing a flexible substrate with a plurality of concealed holes, and injecting a phase change heat storage material into the plurality of concealed holes to obtain a flexible material layer containing the phase change heat storage material; placing the composite structure layer and the flexible material layer on both sides of a central layer coated with an adhesive to obtain an adhesive material, and performing a hot pressing treatment on the adhesive material at a temperature of 55 to 65 °C through a hot pressing device to obtain an explosion-proof and protective composite material;

[0008] Wherein, performing a hot pressing treatment on the adhesive material at a temperature of 55 to 65 °C through a hot pressing device, which is characterized by including:

[0009] Obtaining a time series of the real-time hot pressing temperature collected by a temperature sensor;

[0010] Performing sequence segmentation on the time series of the real-time hot pressing temperature to obtain a set of local time series of the real-time hot pressing temperature;

[0011] Respectively inputting each local time series of the real-time hot pressing temperature in the set of local time series of the real-time hot pressing temperature into a pressing temperature time series feature extractor to obtain a set of hot pressing temperature local time series correlation feature vectors;

[0012] Inputting the set of hot pressing temperature local time series correlation feature vectors into a time series node energy significance feature forward propagation optimizer to obtain a hot pressing temperature time series attenuation aggregation representation vector as a hot pressing temperature time series attenuation aggregation representation feature;

[0013] Based on the hot pressing temperature time series attenuation aggregation representation feature, obtaining an optimization instruction, and the optimization instruction is used to represent the recommended decoding value of the pressing pressure at the next time point.

[0014] Compared with the prior art, a preparation method of an explosion-proof and protective composite material provided by the present application obtains the real-time hot pressing temperature collected by a temperature sensor, and uses a data processing and analysis algorithm based on deep learning to perform time series segmentation and local time series correlation on the real-time hot pressing temperature, so as to automatically obtain the recommended value of the pressing pressure at the next time point according to the aggregation representation characteristics of each real-time hot pressing temperature in the local time series. In this way, by monitoring and analyzing the temperature change in the hot pressing process in real time, the pressing pressure can be dynamically adjusted according to these subtle time series changes to ensure the forming quality of the composite material, improve the consistency of product quality, and thus improve the automation degree in the hot pressing process of the material. BRIEF DESCRIPTION OF THE DRAWINGS

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

[0016] Figure 1 It is a flowchart of a preparation method of an explosion-proof and protective composite material according to an embodiment of the present application;

[0017] Figure 2 It is a schematic diagram of data flow of a preparation method of an explosion-proof and protective composite material according to an embodiment of the present application;

[0018] Figure 3 It is a flowchart of sub-step S3 of a preparation method of an explosion-proof and protective composite material according to an embodiment of the present application. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0019] Next, exemplary embodiments of the present application will be described in detail with reference to the accompanying drawings. Obviously, the described embodiments are only a part of the embodiments of the present application, rather than all the embodiments of the present application. It should be understood that the present application is not limited by the exemplary embodiments described herein.

[0020] As shown in the present application and the claims, unless the context clearly indicates an exception, words such as "a", "an", "one", and / or "the" are not specifically singular, but may also include the plural. Generally speaking, the terms "comprising" and "including" only indicate the inclusion of the clearly identified steps and elements, and these steps and elements do not constitute an exclusive list. The method or device may also include other steps or elements.

[0021] Although this application makes various references to certain modules in the system according to the embodiments of this application, any number of different modules can be used and run on the user terminal and / or the server. The modules are merely illustrative, and different aspects of the system and method can use different modules.

[0022] Flowcharts are used in this application to illustrate the operations performed by the system according to the embodiments of this application. It should be understood that the previous or following operations are not necessarily executed precisely in sequence. On the contrary, various steps can be processed in reverse order or simultaneously as needed. At the same time, other operations can also be added to these processes, or one or several steps can be removed from these processes.

[0023] Hereinafter, example embodiments according to this application will be described in detail with reference to the accompanying drawings. Obviously, the described embodiments are only a part of the embodiments of this application, rather than all the embodiments of this application. It should be understood that this application is not limited by the example embodiments described herein.

[0024] Traditional control methods usually rely on experience and fixed operating parameters and lack a real-time feedback mechanism. This means that once the pressure parameters are set, it is difficult to make dynamic adjustments according to changes in the actual process. That is, with the slightest changes in production conditions, these fixed rules are difficult to ensure the best control effect, and overcompensation or undercompensation is likely to occur, thus affecting the consistency and reliability of product quality. Therefore, an optimized method for preparing an explosion-proof and protective composite material is desired.

[0025] In the technical solution of this application, a method for preparing an explosion-proof and protective composite material is proposed. Figure 1 FIG. is a flowchart of a method for preparing an explosion-proof and protective composite material according to an embodiment of this application. Figure 2 FIG. is a schematic diagram of data flow of a method for preparing an explosion-proof and protective composite material according to an embodiment of this application. As Figure 1 and Figure 2 shown, the method for preparing an explosion-proof and protective composite material according to the embodiment of this application includes the steps of: S1, using at least two high-performance fibers for interleaved weaving to form a composite structure layer; S2, making a flexible substrate containing a plurality of concealed holes, and injecting a phase change heat storage material into the plurality of concealed holes to obtain a flexible material layer containing a phase change heat storage material; S3, respectively placing the composite structure layer and the flexible material layer on both sides of a central layer coated with an adhesive to obtain an adhesive material, and performing a hot pressing treatment on the adhesive material at a temperature of 55 to 65 °C by a hot pressing device to obtain an explosion-proof and protective composite material.

[0026] In particular, for the S1, at least two types of high-performance fibers are interwoven to form a composite structure layer. It should be understood that a composite structure layer with a complex structure and high toughness can be created by the interweaving method. This structure can effectively resist impact loads and fracture propagation, thereby improving the toughness and impact resistance of the composite material. Different types of fibers have different mechanical properties, and interweaving can utilize their synergistic effect to enhance the explosion resistance of the composite material.

[0027] In particular, for the S2, a flexible substrate containing multiple concealed holes is fabricated, and a phase change heat storage material is injected into the multiple concealed holes to obtain a flexible material layer containing the phase change heat storage material. Among them, the phase change heat storage material has the ability to absorb or release a large amount of heat within a specific temperature range. Injecting it into the concealed holes of the flexible substrate can effectively store cold energy.

[0028] In particular, for the S3, the composite structure layer and the flexible material layer are respectively placed on both sides of a central layer coated with an adhesive to obtain an adhesive material, and the adhesive material is hot-pressed at a temperature of 55 to 65 °C by a hot-pressing device to obtain an explosion-proof and protective composite material. In a specific example of the present application, as Figure 3 shown, the S3 includes: S31, obtaining a time series of the real-time hot-pressing temperature collected by a temperature sensor; S32, performing sequence segmentation on the time series of the real-time hot-pressing temperature to obtain a set of local time series of the real-time hot-pressing temperature; S33, respectively inputting each local time series of the real-time hot-pressing temperature in the set of local time series of the real-time hot-pressing temperature into a pressing temperature time series feature extractor to obtain a set of hot-pressing temperature local time series correlation feature vectors; S34, inputting the set of hot-pressing temperature local time series correlation feature vectors into a time series node energy significance feature forward propagation optimizer to obtain a hot-pressing temperature time series attenuation aggregation representation vector as a hot-pressing temperature time series attenuation aggregation representation feature; S35, based on the hot-pressing temperature time series attenuation aggregation representation feature, obtaining an optimization instruction, and the optimization instruction is used to represent the recommended decoding value of the pressing pressure at the next time point.

[0029] Specifically, for S31 and S32, obtain the time series of the real-time hot pressing temperature collected by the temperature sensor; and perform sequence segmentation on the time series of the real-time hot pressing temperature to obtain a set of local time series of the real-time hot pressing temperature. Considering that the temperature of the real-time hot pressing shows different change trends and fluctuation conditions in different local time periods in the time series of the real-time hot pressing temperature. Based on this, in order to better capture and carefully analyze the change characteristic information of the local hot pressing temperature in the real-time hot pressing temperature, in the technical solution of this application, the time series of the real-time hot pressing temperature is segmented based on a predetermined time scale to segment the entire time series according to the time window size, obtaining a set of local time series of the real-time hot pressing temperature.

[0030] Specifically, for S33, input each local time series of the real-time hot pressing temperature in the set of local time series of the real-time hot pressing temperature into the pressing temperature time series feature extractor respectively to obtain a set of local time series correlation feature vectors of the hot pressing temperature. Considering that there are different correlation relationships among the local time series of the real-time hot pressing temperature in different local time spans in the set of local time series of the real-time hot pressing temperature, and the Bi-LSTM model can simultaneously process the information in the past and future of the time series data, is good at modeling long-term and short-term time series dependencies, and can better extract the complex time series features contained in the temperature series. Therefore, in the technical solution of this application, each local time series of the real-time hot pressing temperature in the set of local time series of the real-time hot pressing temperature is input into the pressing temperature time series feature extractor based on the Bi-LSTM model respectively to obtain a set of local time series correlation feature vectors of the hot pressing temperature. In particular, the essence of the pressing temperature time series feature extractor based on the Bi-LSTM model is the Bi-LSTM model.

[0031] Specifically, in S34, the set of local temporal correlation feature vectors of the hot pressing temperature is input into the temporal node energy significance feature forward propagation optimizer to obtain the hot pressing temperature temporal decay aggregation representation vector as the hot pressing temperature temporal decay aggregation representation feature. Considering that the significance and importance shown in each local temporal correlation feature of the hot pressing temperature in the set of local temporal correlation feature vectors of the hot pressing temperature are different, and in order to be able to assign corresponding attention to each local temporal correlation feature vector of the hot pressing temperature, so as to more comprehensively represent and aggregate the hot pressing temperature features in the entire time domain, thereby highlighting the features of key time points. In the technical solution of the present application, the set of local temporal correlation feature vectors of the hot pressing temperature is input into the temporal node energy significance feature forward propagation optimizer to obtain the hot pressing temperature temporal decay aggregation representation vector. It is worth mentioning that the temporal node energy significance feature forward propagation optimizer quantifies the significance feature information of each node and strengthens and aggregates the significance of each local temporal correlation feature of the hot pressing temperature based on the law that energy decays over time.

[0032] In an embodiment of the present application, the set of local temporal correlation feature vectors of the hot pressing temperature is input into a temporal node energy significance feature forward propagation optimizer to obtain a hot pressing temperature temporal decay aggregation representation vector as a hot pressing temperature temporal decay aggregation representation feature, including: First, based on the mean and standard deviation of each local temporal correlation feature vector of the hot pressing temperature in the set of local temporal correlation feature vectors of the hot pressing temperature, calculate the energy significance descriptor of each local temporal correlation feature vector of the hot pressing temperature to obtain a sequence of hot pressing temperature energy significance descriptors, so as to quantify the importance degree of the hot pressing temperature at different time points; Then, extract the timestamps of each local temporal correlation feature vector in the sequence of local temporal correlation feature vectors of the hot pressing temperature to obtain a sequence of hot pressing temperature timestamps; that is, extract the timestamps from each local temporal correlation feature vector of the hot pressing temperature to retain the dynamic law of the hot pressing temperature change over time. Subsequently, use the local temporal correlation feature vector of the hot pressing temperature corresponding to the current time point in the set of local temporal correlation feature vectors of the hot pressing temperature as the current node feature vector and use the local temporal correlation feature vectors of the hot pressing temperature corresponding to other time points as historical node feature vectors to obtain a sequence of the current local temporal correlation feature vector of the hot pressing temperature and the historical local temporal correlation feature vectors of the hot pressing temperature; By regarding the node feature vector at the current time point in the sequence as the current local temporal correlation feature vector of the hot pressing temperature and the others as the historical local temporal correlation feature vectors of the hot pressing temperature to distinguish the current and historical node feature vectors, so as to better capture the key impact of the current temperature change on the final pressure. Further, calculate the energy time decay factor of each historical local temporal correlation feature vector in the sequence of historical local temporal correlation feature vectors of the hot pressing temperature relative to the current local temporal correlation feature vector of the hot pressing temperature to obtain a sequence of hot pressing temperature energy time decay factors; Here, by calculating the energy time decay factor of each historical local temporal correlation feature vector, the relative importance of the historical feature vector is reduced, and the influence of the current feature is highlighted. Then, use the sequence of hot pressing temperature energy significance descriptors as a positive adjustment factor and use the sequence of hot pressing temperature energy time decay factors as a negative adjustment factor to perform temporal propagation aggregation on the sequence of historical local temporal correlation feature vectors of the hot pressing temperature to obtain a historical hot pressing temperature temporal propagation aggregation representation feature vector; Finally, fuse the historical hot pressing temperature temporal propagation aggregation representation feature vector and the current local temporal correlation feature vector of the hot pressing temperature to obtain the hot pressing temperature temporal decay aggregation representation vector.

[0033] Among them, the process of calculating the energy significant descriptor of each hot pressing temperature local time series correlation feature vector based on the mean and standard deviation of each hot pressing temperature local time series correlation feature vector in the set of hot pressing temperature local time series correlation feature vectors to obtain a sequence of hot pressing temperature energy significant descriptors includes: calculating the mean and standard deviation of the hot pressing temperature local time series correlation feature vector respectively to obtain the hot pressing temperature local time series correlation feature mean and the hot pressing temperature local time series correlation feature standard deviation; calculating the position-by-position difference between the hot pressing temperature local time series correlation feature vector and the hot pressing temperature local time series correlation feature mean to obtain the hot pressing temperature local time series correlation difference vector; calculating the fourth power of each eigenvalue in the hot pressing temperature local time series correlation difference vector to obtain the hot pressing temperature local time series correlation modulation difference vector; calculating the expected value of the hot pressing temperature local time series correlation modulation difference vector to obtain the hot pressing temperature local time series correlation expected value; dividing the hot pressing temperature local time series correlation expected value by the value of the fourth power of the hot pressing temperature local time series correlation feature standard deviation to obtain the hot pressing temperature energy significant descriptor.

[0034] More specifically, the process of calculating the energy-time decay factor of each historical hot pressing temperature local time series correlation feature vector in the sequence of the historical hot pressing temperature local time series correlation feature vectors relative to the current hot pressing temperature local time series correlation feature vector to obtain the sequence of the hot pressing temperature energy-time decay factors includes: respectively subtracting the hot pressing temperature timestamp corresponding to the current hot pressing temperature local time series correlation feature vector from the hot pressing temperature timestamps corresponding to the respective historical hot pressing temperature local time series correlation feature vectors and then taking the floor to obtain the sequence of the historical hot pressing temperature time span values; calculating the square of the element-wise division of the sequence of the historical hot pressing temperature time span values by the time decay inverse scaling parameter to obtain the sequence of the historical hot pressing temperature time span squared modulation values, and then multiplying the sequence of the historical hot pressing temperature time span squared modulation values by the decay rate positive scaling parameter element-wise to obtain the sequence of the hot pressing temperature time series span energy decay coefficients; taking each hot pressing temperature time series span energy decay coefficient in the sequence of the hot pressing temperature time series span energy decay coefficients as the exponent power and calculating the natural exponential function value with the natural constant e as the base to obtain the sequence of the hot pressing temperature energy-time decay factors. And, using the sequence of the hot pressing temperature energy significant descriptors as the forward adjustment factor and the sequence of the hot pressing temperature energy-time decay factors as the reverse adjustment factor, performing time series propagation aggregation on the sequence of the historical hot pressing temperature local time series correlation feature vectors to obtain the historical hot pressing temperature time series propagation aggregation representation feature vector, including: dividing the sequence of the hot pressing temperature energy significant descriptors by the sequence of the hot pressing temperature energy-time decay factors element-wise to obtain the sequence of the hot pressing temperature adjustment factors, and calculating the element-wise weighted sum of the sequence of the hot pressing temperature adjustment factors and the sequence of the historical hot pressing temperature local time series correlation feature vectors to obtain the historical hot pressing temperature time series propagation aggregation representation feature vector.

[0035] In summary, in the above embodiments, inputting the set of the hot pressing temperature local time series correlation feature vectors into the time series node energy significance feature forward propagation optimizer to obtain the hot pressing temperature time series decay aggregation representation vector as the hot pressing temperature time series decay aggregation representation feature includes: inputting the set of the hot pressing temperature local time series correlation feature vectors into the time series node energy significance feature forward propagation optimizer and processing it with the following forward propagation formula to obtain the hot pressing temperature time series decay aggregation representation vector; where the forward propagation formula is:

[0036] X = {x1, x2,... x k-1 , x k}

[0037]

[0038]

[0039]

[0040]

[0041] Among them, X is the set of local temporal correlation feature vectors of the hot pressing temperature, and x k-1 and x k are the (k - 1)-th and k-th local temporal correlation feature vectors of the hot pressing temperature in the set of local temporal correlation feature vectors of the hot pressing temperature respectively. k represents the number of feature vectors in the set of local temporal correlation feature vectors of the hot pressing temperature. x ij is the eigenvalue at each position in the i-th local temporal correlation feature vector of the hot pressing temperature in the set of local temporal correlation feature vectors of the hot pressing temperature. E(·) is to calculate the expected value, and μ i and σ i 2 are the mean and variance of the i-th local temporal correlation feature vector of the hot pressing temperature respectively. L is the number of eigenvalues in each local temporal correlation feature vector of the hot pressing temperature. is the energy significance descriptor of the k-th local temporal correlation feature vector of the hot pressing temperature. is the energy significance descriptor of the i-th local temporal correlation feature vector of the hot pressing temperature. t k and t i respectively represent the timestamps of the k-th and i-th local temporal correlation feature vectors of the hot pressing temperature. is the floor operation. δ is the time decay inverse scaling parameter, β is the decay rate positive scaling parameter, θ and γ are weight hyperparameters. exp(·) represents the exponential function value with the natural constant e as the base. f c is the temporal decay aggregation representation vector of the hot pressing temperature.

[0042] Specifically, in S35, an optimization instruction is obtained based on the thermocompression temperature time-series attenuation aggregation representation feature, and the optimization instruction is used to represent the recommended decoded value of the pressing pressure at the next time point. In a specific example of the present application, the thermocompression temperature time-series attenuation aggregation representation vector is input into a decoder-based pressing pressure optimizer to obtain an optimization instruction, and the optimization instruction is used to represent the recommended decoded value of the pressing pressure at the next time point. That is, classification processing is performed on the thermocompression temperature time-series attenuation aggregation representation obtained by performing energy significant propagation aggregation on the set of thermocompression temperature local time-series correlation feature vectors, so as to automatically obtain the recommended value of the pressing pressure at the next time point. In this way, by real-time monitoring and analyzing the temperature changes during the thermocompression process, the pressing pressure can be dynamically adjusted according to these subtle time-series changes to ensure the forming quality of the composite material, improve the consistency of product quality, and thus enhance the automation degree during the material thermocompression process.

[0043] Here, the set of thermocompression temperature local time-series correlation feature vectors respectively express the short-range-long-range two-way time-series correlation features of the real-time thermocompression temperature in the local time domain. Therefore, when performing forward propagation optimization of the global time domain features based on the energy significance of time-series nodes, considering the time-series offset in the entire time domain caused by the difference in the energy significance of time-series nodes due to the difference in time-series features in each local time domain, the regression understanding imbalance of the selective local time domain feature propagation state space representation in the distribution field with respect to the global time domain time-series semantic feature representation is caused. Therefore, it is desired to further improve the regression understanding consistency of the full-time domain aggregation sequence of the thermocompression temperature time-series attenuation aggregation representation vector, so as to improve the accuracy of the optimization instruction obtained by inputting it into the decoder-based pressing pressure optimizer.

[0044] Therefore, when the present application considers inputting the thermocompression temperature time-series attenuation aggregation representation vector into a decoder-based pressing pressure optimizer to obtain an optimization instruction, the thermocompression temperature time-series attenuation aggregation representation vector is optimized, and the optimization process includes:

[0045] Determine the number of zero eigenvalues in the thermocompression temperature time-series attenuation aggregation representation vector, and calculate the reciprocal of the logarithm to the base 2 of the number of zero eigenvalues and the exponential value to the base of the natural constant of the reciprocal of the number of zero eigenvalues respectively to obtain the first thermocompression temperature time-series attenuation aggregation representation value and the second thermocompression temperature time-series attenuation aggregation representation value:

[0046] α = [log2(n)] -1

[0047] β = exp[(n) -1

[0048] ​Wherein, n is the number of zero eigenvalues in the time-sequence decay aggregation representation vector of the hot pressing temperature, α is the first time-sequence decay aggregation representation value of the hot pressing temperature, β is the second time-sequence decay aggregation representation value of the hot pressing temperature, and exp(·) is the exponential function with the natural constant as the base;

[0049] Specifically, it can be obtained by subtracting the zero norm of the time-sequence decay aggregation representation vector of the hot pressing temperature from the length of the time-sequence decay aggregation representation vector of the hot pressing temperature, that is:

[0050] n = L - ||V||0

[0051] Wherein, V represents the time-sequence decay aggregation representation vector of the hot pressing temperature, L represents the length of the time-sequence decay aggregation representation vector of the hot pressing temperature, ||V||0 represents the zero norm of the time-sequence decay aggregation representation vector of the hot pressing temperature, and n is the number of zero eigenvalues in the time-sequence decay aggregation representation vector of the hot pressing temperature;

[0052] Then, calculate the power function of each eigenvalue of the time-sequence decay aggregation representation vector of the hot pressing temperature with the reciprocal of the number of zero eigenvalues as the exponent to obtain the time-sequence decay aggregation structure state vector of the hot pressing temperature, that is:

[0053]

[0054] Wherein, represents the power function of each eigenvalue of the time-sequence decay aggregation representation vector of the hot pressing temperature with the reciprocal of the number of zero eigenvalues as the exponent, and V1 represents the time-sequence decay aggregation structure state vector;

[0055] Perform a dot product of the time-sequence decay aggregation structure state vector and the first time-sequence decay aggregation representation value of the hot pressing temperature to obtain the time-sequence decay aggregation information representation vector, that is:

[0056] V2 = α ⊙ V1

[0057] Wherein, V1 represents the time-sequence decay aggregation structure state vector, α is the first time-sequence decay aggregation representation value of the hot pressing temperature, ⊙ is the dot product by position, and V2 represents the time-sequence decay aggregation information representation vector;

[0058] Perform a dot product of the autocorrelation matrix of the time-sequence decay aggregation structure state vector with the second time-sequence decay aggregation representation value of the hot pressing temperature and the weight hyperparameter respectively to obtain the time-sequence decay aggregation regression understanding matrix, that is:

[0059]

[0060] Wherein, V is the time-sequence attenuation aggregation representation vector of the hot pressing temperature, w is the weight hyperparameter, n is the number of zero eigenvalues in the time-sequence attenuation aggregation representation vector of the hot pressing temperature, and β is the second time-sequence attenuation aggregation representation value of the hot pressing temperature. represents calculating the power function of each eigenvalue of the time-sequence attenuation aggregation representation vector of the hot pressing temperature with the reciprocal of the number of zero eigenvalues as the exponent, T represents the transpose of the vector, and ⊙ is the element-wise multiplication. is the matrix multiplication, and M is the time-sequence attenuation aggregation regression understanding matrix of the hot pressing temperature.

[0061] Multiplying the time-sequence attenuation aggregation information representation vector of the hot pressing temperature by the time-sequence attenuation aggregation regression understanding matrix of the hot pressing temperature to obtain an optimized time-sequence attenuation aggregation representation vector of the hot pressing temperature, that is:

[0062]

[0063] Here, V2 represents the time-sequence attenuation aggregation information representation vector of the hot pressing temperature. is the matrix multiplication, M is the time-sequence attenuation aggregation regression understanding matrix of the hot pressing temperature, and V′ is the optimized time-sequence attenuation aggregation representation vector of the hot pressing temperature. Finally, input the optimized time-sequence attenuation aggregation representation vector of the hot pressing temperature into the press pressure optimizer based on the decoder to obtain an optimization instruction.

[0064] Based on this, by modeling the zero dimension of the vector field of the time-sequence attenuation aggregation representation vector of the hot pressing temperature as the global field redundancy dependence, the effective long-range modeling of the time-sequence attenuation aggregation representation vector of the hot pressing temperature under the field linear complexity is realized. In this way, during the feature regression process of the time-sequence attenuation aggregation representation vector of the hot pressing temperature, the structured correlation self-distillation between the information representation and the class regression understanding is carried out in the selective feature enhancement state space based on the feature distribution, so as to promote the remote dynamic regression understanding balance of its feature convergence while maintaining the overall information complexity of the feature distribution of the time-sequence attenuation aggregation representation vector of the hot pressing temperature, thereby improving the regression understanding consistency between the feature regression and the feature extraction process, and improving the accuracy of the optimization instruction obtained by inputting the time-sequence attenuation aggregation representation vector of the hot pressing temperature into the press pressure optimizer based on the decoder. In this way, by real-time monitoring and analyzing the temperature changes during the hot pressing process, the press pressure can be dynamically adjusted according to these subtle time-sequence changes to ensure the forming quality of the composite material, improve the consistency of the product quality, and thus improve the automation degree during the hot pressing process of the material.

[0065] In summary, the preparation method of the explosion-proof and protective composite material according to the embodiments of the present application is elucidated. By obtaining the real-time hot pressing temperature collected by the temperature sensor and adopting a data processing and analysis algorithm based on deep learning, the real-time hot pressing temperature is subjected to time series segmentation and local time series correlation, so as to automatically obtain the recommended value of the pressing pressure at the next time point according to the aggregation representation characteristics of each real-time hot pressing temperature in the local time series. In this way, by monitoring and analyzing the temperature change in the hot pressing process in real time, the pressing pressure can be dynamically adjusted according to these subtle time series changes to ensure the forming quality of the composite material, improve the consistency of product quality, and thus enhance the automation degree in the hot pressing process of the material.

[0066] In another example, the preparation of the explosion-proof and protective composite material can be realized through the following steps:

[0067] 1. Material preparation: When the temperature is lower than 15°C, components A and B need to be heated and insulated, and the viscosity is controlled to be less than 2000 cps;

[0068] 2. Equipment debugging: During construction, the high-temperature and high-pressure impact mixing equipment specified by the company must be used.

[0069] 3. Check the hydraulic pressure of materials A and B (2500 - 3000 psi) before spraying. If the hydraulic pressure difference is more than 500 psi, it is necessary to relieve the pressure and discharge the material, and then restart the machine for debugging;

[0070] 4. The metal substrate should be treated to Sa2.5 level and coated with a special primer for matching. Ensure that the spraying surface is clean, dry, dust-free before construction, and shield the parts that do not need to be sprayed; other substrates are processed according to the actual situation.

[0071] 5. During construction, control the spraying distance at 50 - 100 cm, and the latter spray covers about 40% - 60% of the previous spray. The traveling speed is effectively controlled to prevent sagging.

[0072] The embodiments of the present disclosure have been described above. The above description is exemplary and not exhaustive, and is not limited to the disclosed embodiments. Many modifications and variations are obvious to those of ordinary skill in the art in the technical field without departing from the scope and spirit of the described embodiments. The selection of the terms used herein is intended to best explain the principles of the embodiments, the actual application, or the improvement of the technology in the market, or to enable other ordinary technical personnel in the technical field to understand the embodiments disclosed herein.

Claims

1. A preparation method of an explosion-proof composite material, comprising: using at least two high-performance fibers for interleaved weaving to form a composite structure layer; manufacturing a flexible substrate with a plurality of concealed holes, injecting a phase change heat storage material into the plurality of concealed holes to obtain a flexible material layer containing the phase change heat storage material; placing the composite structure layer and the flexible material layer on both sides of a central layer coated with an adhesive to obtain an adhesive material, and performing a hot pressing treatment on the adhesive material at a temperature of 55 to 65 °C through a hot pressing device to obtain the explosion-proof composite material; Among them, Performing a hot pressing treatment on the adhesive material at a temperature of 55 to 65 °C through a hot pressing device, characterized by comprising: Obtaining a time series of the real-time hot pressing temperature collected by a temperature sensor; Performing sequence segmentation on the time series of the real-time hot pressing temperature to obtain a set of local time series of the real-time hot pressing temperature; Respectively inputting each local time series of the real-time hot pressing temperature in the set of local time series of the real-time hot pressing temperature into a pressing temperature time series feature extractor to obtain a set of hot pressing temperature local time series correlation feature vectors; Inputting the set of hot pressing temperature local time series correlation feature vectors into a time series node energy significance feature forward propagation optimizer to obtain a hot pressing temperature time series decay aggregation representation vector as a hot pressing temperature time series decay aggregation representation feature; Based on the hot pressing temperature time series decay aggregation representation feature, obtaining an optimization instruction, where the optimization instruction is used to represent a recommended decoded value of the pressing pressure at the next time point; Among them, based on the hot pressing temperature time series decay aggregation representation feature, obtaining an optimization instruction includes: inputting the hot pressing temperature time series decay aggregation representation vector into a pressing pressure optimizer based on a decoder to obtain the optimization instruction.

2. The preparation method of the explosion-proof and protective composite material according to claim 1, characterized in that Performing sequence segmentation on the time series of the real-time hot pressing temperature to obtain a set of local time series of the real-time hot pressing temperature, including: performing sequence segmentation on the time series of the real-time hot pressing temperature based on a predetermined time scale to obtain the set of local time series of the real-time hot pressing temperature.

3. The preparation method of the explosion-proof and protective composite material according to claim 2, characterized in that, Respectively inputting each local time series of the real-time hot pressing temperature in the set of local time series of the real-time hot pressing temperature into a pressing temperature time series feature extractor to obtain a set of hot pressing temperature local time series correlation feature vectors, including: respectively inputting each local time series of the real-time hot pressing temperature in the set of local time series of the real-time hot pressing temperature into a pressing temperature time series feature extractor based on a Bi-LSTM model to obtain the set of hot pressing temperature local time series correlation feature vectors.

4. The preparation method of the explosion-proof and protective composite material according to claim 3, characterized in that, Inputting the set of hot pressing temperature local time series correlation feature vectors into a time series node energy significance feature forward propagation optimizer to obtain a hot pressing temperature time series decay aggregation representation vector, including: Based on the mean and standard deviation of each hot pressing temperature local time series correlation feature vector in the set of hot pressing temperature local time series correlation feature vectors, calculating an energy significant descriptor of each hot pressing temperature local time series correlation feature vector to obtain a sequence of hot pressing temperature energy significant descriptors; Extract the timestamps of each local temporal correlation feature vector of the hot pressing temperature in the sequence of local temporal correlation feature vectors of the hot pressing temperature to obtain a sequence of hot pressing temperature timestamps; Take the local temporal correlation feature vector of the hot pressing temperature corresponding to the current time point in the set of local temporal correlation feature vectors of the hot pressing temperature as the current node feature vector and take the local temporal correlation feature vectors of the hot pressing temperature corresponding to other time points as the historical node feature vectors to obtain a sequence of the current local temporal correlation feature vector of the hot pressing temperature and the historical local temporal correlation feature vectors of the hot pressing temperature; Calculate the energy-time decay factor of each historical local temporal correlation feature vector of the hot pressing temperature in the sequence of historical local temporal correlation feature vectors of the hot pressing temperature relative to the current local temporal correlation feature vector of the hot pressing temperature to obtain a sequence of energy-time decay factors of the hot pressing temperature; Using the sequence of the hot pressing temperature energy significant descriptors as the positive adjustment factor and the sequence of the energy-time decay factors of the hot pressing temperature as the negative adjustment factor, perform temporal propagation aggregation on the sequence of the historical local temporal correlation feature vectors of the hot pressing temperature to obtain a historical hot pressing temperature temporal propagation aggregation representation feature vector; Fuse the historical hot pressing temperature temporal propagation aggregation representation feature vector and the current local temporal correlation feature vector of the hot pressing temperature to obtain the hot pressing temperature temporal decay aggregation representation vector.

5. The preparation method of the explosion-proof and protective composite material according to claim 4, characterized in that, Based on the mean and standard deviation of each local temporal correlation feature vector of the hot pressing temperature in the set of local temporal correlation feature vectors of the hot pressing temperature, calculate the energy significant descriptor of each local temporal correlation feature vector of the hot pressing temperature to obtain a sequence of energy significant descriptors of the hot pressing temperature, including: Calculate the mean and standard deviation of the local temporal correlation feature vector of the hot pressing temperature respectively to obtain the mean of the local temporal correlation feature of the hot pressing temperature and the standard deviation of the local temporal correlation feature of the hot pressing temperature; Calculate the position-wise difference between the local temporal correlation feature vector of the hot pressing temperature and the mean of the local temporal correlation feature of the hot pressing temperature to obtain a local temporal correlation difference vector of the hot pressing temperature; Calculate the fourth power of each eigenvalue in the local temporal correlation difference vector of the hot pressing temperature to obtain a local temporal correlation modulation difference vector of the hot pressing temperature; Calculate the expected value of the local temporal correlation modulation difference vector of the hot pressing temperature to obtain the local temporal correlation expected value of the hot pressing temperature; Divide the value of the local temporal correlation expected value of the hot pressing temperature by the fourth power of the standard deviation of the local temporal correlation feature of the hot pressing temperature to obtain the energy significant descriptor of the hot pressing temperature.

6. The preparation method of the explosion-proof and protective composite material according to claim 5, wherein, Calculate the energy-time decay factor of each historical local temporal correlation feature vector of the hot pressing temperature in the sequence of historical local temporal correlation feature vectors of the hot pressing temperature relative to the current local temporal correlation feature vector of the hot pressing temperature to obtain a sequence of energy-time decay factors of the hot pressing temperature, including: Subtract the hot pressing temperature timestamps corresponding to the current hot pressing temperature local temporal correlation feature vectors from the hot pressing temperature timestamps corresponding to each of the historical hot pressing temperature local temporal correlation feature vectors respectively, and then round down to obtain a sequence of historical hot pressing temperature time span values; After calculating the squares of the element-wise division of the sequence of historical hot pressing temperature time span values by the time decay inverse scaling parameter to obtain a sequence of historical hot pressing temperature time span squared modulation values, multiply the sequence of historical hot pressing temperature time span squared modulation values by the decay rate positive scaling parameter element-wise to obtain a sequence of hot pressing temperature temporal span energy decay coefficients; Taking each hot pressing temperature temporal span energy decay coefficient in the sequence of hot pressing temperature temporal span energy decay coefficients as the exponent power, calculate the natural exponential function value with the natural constant e as the base to obtain a sequence of the hot pressing temperature energy time decay factors.

7. The preparation method of the explosion-proof composite material according to claim 6, characterized in that, Using the sequence of the hot pressing temperature energy significant descriptors as the forward adjustment factor and the sequence of the hot pressing temperature energy time decay factors as the reverse adjustment factor, perform temporal propagation aggregation on the sequence of the historical hot pressing temperature local temporal correlation feature vectors to obtain a historical hot pressing temperature temporal propagation aggregation representation feature vector, including: dividing the sequence of the hot pressing temperature energy significant descriptors by the sequence of the hot pressing temperature energy time decay factors element-wise to obtain a sequence of hot pressing temperature adjustment factors, and calculating the element-wise weighted sum of the sequence of the hot pressing temperature adjustment factors and the sequence of the historical hot pressing temperature local temporal correlation feature vectors to obtain the historical hot pressing temperature temporal propagation aggregation representation feature vector.

Images