A surface enhancement treatment method for calcium silicate board

Abstract

The present application relates to the field of industrial processing systems, and specifically to a method for surface enhancement treatment of calcium silicate boards. It uses artificial intelligence technology based on deep learning to perform real-time monitoring and data analysis of oven temperature and feed speed, and mines out the dynamic change pattern of oven temperature and feed speed in the global time domain through local time series coding and feature forward propagation aggregation processing, and captures the complex correlation pattern between oven temperature and feed speed through fine-grained time series interactive response analysis of the two, so as to perform adaptive regulation of feed speed. In this way, reasonable control of the drying process can be achieved, dynamic matching between feed speed and drying temperature can be ensured, and the intelligent level of surface enhancement treatment of calcium silicate boards can be effectively improved.

Description

Technical Field

[0001] The present application relates to the field of industrial processing systems, and in particular to a surface enhancement processing method for a calcium silicate board. Background Art

[0002] Calcium silicate board is a material widely used in the construction and industrial fields. It is mostly made of powder of inorganic materials such as calcium silicate or calcium carbonate, and has excellent properties such as fire resistance, moisture resistance, sound insulation, and heat insulation. However, due to the relatively weak bonding force between its powders, when other materials need to be attached to its surface, the surface material is prone to peeling. In order to improve the physical properties, aesthetics and durability of calcium silicate board, it is usually necessary to perform surface enhancement treatment on it.

[0003] Specifically, in the process of surface gluing of calcium silicate board, the main purpose of the first gluing is to allow the glue to fully penetrate into the interior of the substrate and enhance the internal bonding force of the board; while the second gluing is to further enhance the bonding performance of the board surface on the basis of the first gluing to ensure close bonding with the surface material. In the second gluing process, a drying step is required to solidify the adhesive to ensure the chemical bonding between the adhesive and the board. In this step, the control of drying temperature and feed speed is crucial to ensure uniform curing of the adhesive and the surface quality of the calcium silicate board. However, traditional drying control methods are often based on manual experience to set fixed parameters, which is not only inefficient, but also difficult to adapt to the dynamic changes in production conditions, which can easily lead to unstable drying effects and affect the final quality of the calcium silicate board.

[0004] Therefore, an optimized surface enhancement treatment method for calcium silicate board is expected. Summary of the invention

[0005] This application is made in view of the above problems. One object of this application is to provide a surface enhancement treatment method for a calcium silicate board.

[0006] The embodiment of the present application provides a surface enhancement treatment method for a calcium silicate board, comprising: immersing the calcium silicate board in a silicon dioxide solution having a solid content of more than 95%, completely absorbing it until saturated, and then drying the immersed calcium silicate board to obtain a modified calcium silicate board; applying a first gluing process on any surface of the modified calcium silicate board, wherein the first gluing process includes one gluing process, one leveling penetration process, one drying process, one cooling process, and one polishing process; applying a second gluing process on the surface of the modified calcium silicate board after the first gluing process to obtain a solidified calcium silicate board surface, wherein the second gluing process includes a second gluing process, a second drying process, and a second polishing process, wherein the second drying process is performed on the surface of the modified calcium silicate board after the first gluing process, including:

[0007] Acquire the time queue of the real-time drying tunnel temperature collected by the temperature sensor and the time queue of the real-time feeding speed collected by the speed sensor;

[0008] Performing temporal coding based on the local time scale on the time queue of the real-time drying oven temperature and the time queue of the real-time feeding speed to obtain a sequence of local temporal correlation feature vectors of the real-time drying oven temperature and a sequence of local temporal correlation feature vectors of the real-time feeding speed;

[0009] Inputting the sequence of local time series associated feature vectors of the real-time drying oven temperature and the sequence of local time series associated feature vectors of the real-time feeding speed into the feature forward propagation model based on node energy decay perception to obtain a drying oven temperature time series significant aggregation representation vector and a feeding speed time series significant aggregation representation vector;

[0010] Performing feature interaction response analysis on the oven temperature time series significant aggregation representation vector and the feed speed time series significant aggregation representation vector to obtain an oven temperature-feed speed time series decoupling enhancement interaction representation vector;

[0011] Based on the drying tunnel temperature-feeding speed timing decoupling enhanced interaction representation vector, a control instruction of the feeding speed is determined.

[0012] For example, according to the surface enhancement treatment method of the calcium silicate board of the embodiment of the present application, the time queue of the real-time drying oven temperature and the time queue of the real-time feeding speed are subjected to temporal encoding based on the local time scale to obtain a sequence of local temporal correlation feature vectors of the real-time drying oven temperature and a sequence of local temporal correlation feature vectors of the real-time feeding speed, including:

[0013] The time queue of the real-time drying oven temperature and the time queue of the real-time feeding speed are divided into predetermined time scales and respectively input into a sequence encoder based on a 1D-CNN model to obtain a sequence of local temporal correlation feature vectors of the real-time drying oven temperature and a sequence of local temporal correlation feature vectors of the real-time feeding speed.

[0014] For example, according to the surface enhancement treatment method of the calcium silicate board of the embodiment of the present application, the sequence of the local time series correlation feature vectors of the real-time drying oven temperature and the sequence of the local time series correlation feature vectors of the real-time feed speed are input into the feature forward propagation model based on node energy decay perception to obtain the drying oven temperature time series significant aggregation representation vector and the feed speed time series significant aggregation representation vector, including:

[0015] Based on the mean and variance of each real-time drying tunnel temperature local time series correlation feature vector in the sequence of the real-time drying tunnel temperature local time series correlation feature vector, calculating the energy significant descriptor of each real-time drying tunnel temperature local time series correlation feature vector to obtain a sequence of energy significant descriptors;

[0016] Extracting the timestamp of each real-time drying oven temperature local time series correlation feature vector in the sequence of the real-time drying oven temperature local time series correlation feature vector to obtain a sequence of timestamps;

[0017] The real-time drying oven temperature local time series association feature vector corresponding to the current time point in the sequence of the real-time drying oven temperature local time series association feature vectors is used as the current real-time drying oven temperature local time series association feature vector, and the real-time drying oven temperature local time series association feature vectors corresponding to other time points are used as the historical real-time drying oven temperature local time series association feature vectors, so as to obtain a sequence of the current real-time drying oven temperature local time series association feature vector and the historical real-time drying oven temperature local time series association feature vector;

[0018] Calculating the energy time decay factor of each historical real-time drying oven temperature local time series associated feature vector in the sequence of the historical real-time drying oven temperature local time series associated feature vector relative to the current real-time drying oven temperature local time series associated feature vector to obtain a sequence of energy time decay factors;

[0019] Using the sequence of energy significant descriptors as positive adjustment factors and the sequence of energy time decay factors as negative adjustment factors, the sequence of local time series correlation feature vectors of the historical real-time oven temperature is subjected to time series propagation aggregation to obtain a time series aggregation feature vector of the historical node real-time oven temperature;

[0020] The real-time oven temperature time series aggregation feature vector of the historical node and the local time series association feature vector of the current real-time oven temperature are fused to obtain the oven temperature time series significant aggregation representation vector.

[0021] For example, according to the surface enhancement treatment method of the calcium silicate board of the embodiment of the present application, based on the mean and variance of each real-time drying oven temperature local time series correlation feature vector in the sequence of the real-time drying oven temperature local time series correlation feature vector, the energy significant descriptor of each real-time drying oven temperature local time series correlation feature vector is calculated to obtain a sequence of energy significant descriptors, including:

[0022] The expected value of the fourth power of the difference between each eigenvalue in the local time series associated eigenvalue vector of the real-time drying tunnel temperature and its eigenmean is calculated, and the expected value is divided by the square of the eigenvariance of the local time series associated eigenvalue vector of the real-time drying tunnel temperature to obtain the energy significant descriptor.

[0023] For example, according to the surface enhancement treatment method of the calcium silicate board of the embodiment of the present application, the energy time decay factor of each historical real-time drying oven temperature local time series correlation feature vector in the sequence of the historical real-time drying oven temperature local time series correlation feature vector is calculated relative to the current real-time drying oven temperature local time series correlation feature vector to obtain a sequence of energy time decay factors, including:

[0024] Calculate the timestamp difference between the historical real-time drying oven temperature local time series associated feature vector and the current real-time drying oven temperature local time series associated feature vector;

[0025] The timestamp difference is rounded down and divided by a preset time decay period and then squared, and the square result is multiplied by a preset decay rate to obtain an initial energy decay coefficient;

[0026] An exponential function value with a natural constant as the base and the initial energy decay coefficient as the exponent is calculated to obtain the energy time decay factor.

[0027] For example, according to the surface enhancement treatment method of the calcium silicate board of the embodiment of the present application, the characteristic interaction response analysis is performed on the oven temperature time series significant aggregation representation vector and the feed speed time series significant aggregation representation vector to obtain the oven temperature-feed speed time series decoupling enhancement interaction representation vector, including:

[0028] The oven temperature time series significant aggregation representation vector and the feed speed time series significant aggregation representation vector are input into a feature response aggregation module based on feature fine-grained decoupling to obtain the oven temperature-feed speed time series decoupling enhanced interaction representation vector.

[0029] For example, according to the surface enhancement treatment method of the calcium silicate board of the embodiment of the present application, the drying oven temperature time series significant aggregation representation vector and the feed speed time series significant aggregation representation vector are input into the feature response aggregation module based on feature fine-grained decoupling to obtain the drying oven temperature-feed speed time series decoupling enhanced interaction representation vector, including:

[0030] Decoupling the drying tunnel temperature time series significant aggregation representation vector and the feeding speed time series significant aggregation representation vector to obtain a sequence of drying tunnel temperature time series feature local representation vectors and a sequence of feeding speed time series feature local representation vectors;

[0031] Calculating the local semantic association interaction vectors between each corresponding group of the local representation vectors of the drying oven temperature time series features and the local representation vectors of the feeding speed time series features in the sequence of the local representation vectors of the drying oven temperature time series features and the sequence of the local representation vectors of the feeding speed time series features to obtain a sequence of the local semantic association interaction vectors of the drying oven temperature and the feeding speed;

[0032] Calculating the fine-grained attention weight of each oven temperature-feed speed local semantic association interaction vector in the sequence of oven temperature-feed speed local semantic association interaction vectors to obtain a sequence of fine-grained attention weights;

[0033] Inputting the sequence of fine-grained attention weights into a random dropout module to obtain a sequence of fine-grained attention weights after dropout processing;

[0034] The sequence of fine-grained attention weights after the deactivation process is used as the weight distribution, and the weighted sum of the sequence of the oven temperature-feeding speed local semantic association interaction vectors is calculated to obtain the oven temperature-feeding speed temporal decoupling reinforcement interaction representation vector.

[0035] For example, according to the surface enhancement treatment method of the calcium silicate board of the embodiment of the present application, the local semantic association interaction vector between each group of corresponding local representation vectors of the drying oven temperature timing characteristics and the local representation vectors of the feeding speed timing characteristics in the sequence of the drying oven temperature timing characteristics and the sequence of the local representation vectors of the feeding speed timing characteristics is calculated to obtain a sequence of local semantic association interaction vectors of the drying oven temperature-feeding speed, including:

[0036] Calculate the position point addition, position point subtraction and position point multiplication between the local representation vector of the drying tunnel temperature time series feature and the corresponding local representation vector of the feeding speed time series feature to obtain a first fusion result, a second fusion result and a third fusion result;

[0037] After the first fusion result, the second fusion result and the third fusion result are cascaded into a drying tunnel temperature-feed speed multi-scale fusion feature vector, one-dimensional convolution processing and maximum pooling processing are performed on it to obtain the drying tunnel temperature-feed speed local semantic association interaction vector.

[0038] For example, according to the surface enhancement treatment method of the calcium silicate board of an embodiment of the present application, the fine-grained attention weight of each drying oven temperature-feeding speed local semantic association interaction vector in the sequence of the drying oven temperature-feeding speed local semantic association interaction vector is calculated to obtain a sequence of fine-grained attention weights, including:

[0039] The drying oven temperature-feeding speed local semantic association interaction vector is multiplied by a weight coefficient matrix, and then added to the bias vector and passed through a tanh function to obtain a drying oven temperature-feeding speed local semantic association interaction encoding feature vector;

[0040] The attention score reference vector is multiplied by the oven temperature-feed speed local semantic association interaction encoding feature vector to obtain the fine-grained attention weight.

[0041] For example, according to the surface enhancement treatment method of the calcium silicate board of the embodiment of the present application, the control instruction of the feed speed is determined based on the drying oven temperature-feed speed timing decoupling enhancement interaction representation vector, including:

[0042] The drying tunnel temperature-feeding speed timing decoupling enhanced interaction representation vector is input into the classifier-based feeding speed controller to obtain the control instruction, and the control instruction is used to indicate whether the feeding speed at the next time point should increase, decrease or remain unchanged.

[0043] According to the surface enhancement treatment method of the calcium silicate board of the embodiment of the present application, it adopts artificial intelligence technology based on deep learning to monitor and analyze the oven temperature and feed speed in real time, and digs out the dynamic change pattern of the oven temperature and feed speed in the global time domain through local time series coding and feature forward propagation aggregation processing, and captures the complex correlation pattern between the oven temperature and feed speed by performing fine-grained time series interactive response analysis on the two, so as to perform adaptive control of the feed speed. In this way, reasonable control of the drying process can be achieved, ensuring the dynamic matching between the feed speed and the drying temperature, and effectively improving the intelligence level of the surface enhancement treatment of the calcium silicate board. BRIEF DESCRIPTION OF THE DRAWINGS

[0044] In order to more clearly illustrate the technical solutions of the embodiments of the present application, the drawings of the embodiments of the present application are briefly introduced below. Obviously, the drawings described below only relate to some embodiments of the present application, and are not intended to limit the present application.

[0045] Figure 1 A flow chart showing the secondary drying in the surface enhancement treatment method of the calcium silicate board in the embodiment of the present application is shown;

[0046] Figure 2 A flowchart of sub-step S530 of the surface enhancement treatment method of the calcium silicate board in the embodiment of the present application is shown;

[0047] Figure 3 A flowchart of sub-step S540 of the surface enhancement treatment method of the calcium silicate board in the embodiment of the present application is shown;

[0048] Figure 4 It shows a schematic structural diagram of a calcium silicate board surface enhancement treatment system in an embodiment of the present application;

[0049] Figure 5 A diagram showing an application scenario of the surface enhancement treatment method of a calcium silicate board in an embodiment of the present application is shown. DETAILED DESCRIPTION

[0050] The technical solutions in the embodiments of the present application will be described clearly and completely below in conjunction with the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present application, not all embodiments. Based on the embodiments of the present application, all other embodiments obtained by ordinary technicians in this field without creative work also fall within the scope of protection of the present application.

[0051] The terms used in this specification are those common terms currently widely used in the art in consideration of the functions of the present application, but these terms may vary according to the intentions of those skilled in the art, precedents, or new technologies in the art. In addition, specific terms may be selected, and in this case, their detailed meanings will be described in the detailed description of the present application. Therefore, the terms used in the specification should not be understood as simple names, but rather as a general description based on the meaning of the terms and the present application.

[0052] Although the present application makes various references to certain modules in the system according to the embodiments of the present application, any number of different modules can be used and run on the user terminal and / or server. The modules are only illustrative, and different aspects of the system and method can use different modules.

[0053] Flowcharts are used in the present application to illustrate the operations performed by the system according to the embodiments of the present application. It should be understood that the preceding or following operations are not necessarily performed accurately in order. On the contrary, various steps may be processed in reverse order or simultaneously as required. At the same time, other operations may also be added to these processes, or a certain step or several steps of operations may be removed from these processes.

[0054] In addition, the application architecture diagram in the embodiment of the present application is intended to more clearly illustrate the technical solution in the embodiment of the present application, and does not constitute a limitation on the technical solution provided in the embodiment of the present application. Of course, for other application architectures and business applications, the technical solution provided in the embodiment of the present application is also applicable to similar problems.

[0055] The following is a non-limiting description of the surface enhancement treatment method of the calcium silicate board provided according to at least one embodiment of the present application through several examples or embodiments. As described below, different features in these specific examples or embodiments can be combined with each other without conflicting with each other to obtain new examples or embodiments, and these new examples or embodiments also fall within the scope of protection of the present application.

[0056] Patent CN112457053A discloses a surface treatment method for a modified calcium silicate board, which comprises: immersing the calcium silicate board in a silicon dioxide solution having a solid content of more than 95%, completely absorbing it to saturation, and then drying the immersed calcium silicate board to obtain a modified calcium silicate board; applying a first gluing process on any surface of the modified calcium silicate board, wherein the first gluing process comprises one gluing process, one leveling penetration process, one drying process, one cooling process and one polishing process; applying a second gluing process on the surface of the modified calcium silicate board after the first gluing process to obtain a solidified calcium silicate board surface, wherein the second gluing process comprises a second gluing process, a second drying process and a second polishing process.

[0057] Specifically, during the secondary drying process, the drying temperature has a close influence on the feed speed. The higher the drying temperature, the faster the adhesive dries, which means that the moisture or solvent in the adhesive can be removed more effectively per unit time. Correspondingly, when the drying temperature is low, a longer time is required to ensure that the adhesive is fully dried. Therefore, there is a mutual dependence between the drying temperature and the feed speed, and the two need to be coordinated and controlled to achieve the best drying effect. However, in the actual production process, the actual temperature in the drying tunnel may deviate from the set value due to various factors such as ambient temperature fluctuations and equipment aging, showing a certain degree of volatility. Traditional drying control methods are often unable to adjust the feed speed in real time to adapt to temperature fluctuations, resulting in instability in the drying effect.

[0058] In response to the above technical problems, the technical concept of this application is to use artificial intelligence technology based on deep learning to perform real-time monitoring and data analysis on the oven temperature and feed speed, and to mine the dynamic change pattern of the oven temperature and feed speed in the global time domain through local time series coding and feature forward propagation aggregation processing, and to capture the complex correlation pattern between the oven temperature and feed speed through fine-grained time series interactive response analysis of the two, so as to perform adaptive control of the feed speed. In this way, reasonable control of the drying process can be achieved, ensuring the dynamic matching between the feed speed and the drying temperature, and effectively improving the intelligence level of the surface enhancement treatment of the calcium silicate board.

[0059] Accordingly, the present application provides a surface enhancement treatment method for a calcium silicate board, comprising: immersing the calcium silicate board in a silicon dioxide solution having a solid content of more than 95%, completely absorbing it until saturated, and then drying the immersed calcium silicate board to obtain a modified calcium silicate board; applying a first coat of glue on any surface of the modified calcium silicate board, wherein the first coat of glue application includes one coat of glue, one leveling penetration, one drying, one cooling, and one polishing; applying a second coat of glue on the surface of the modified calcium silicate board after the first coat of glue application to obtain a solidified calcium silicate board surface, wherein the second coat of glue application includes a second coat of glue application, a second drying, and a second polishing. Figure 1 As shown, secondary drying is performed on the surface of the modified calcium silicate board after the first gluing, including: S510, obtaining a time queue of real-time oven temperature collected by a temperature sensor and a time queue of real-time feed speed collected by a speed sensor; S520, performing temporal coding on the time queue of the real-time oven temperature and the time queue of the real-time feed speed based on a local time scale to obtain a sequence of local temporal correlation feature vectors of the real-time oven temperature and a sequence of local temporal correlation feature vectors of the real-time feed speed; S530, encoding the local temporal correlation feature vectors of the real-time oven temperature to the real-time oven temperature; The sequence of quantities and the sequence of local timing-related feature vectors of the real-time feed speed are input into a feature forward propagation model based on node energy decay perception to obtain a drying oven temperature timing-significant aggregation representation vector and a feed speed timing-significant aggregation representation vector; S540, feature interaction response analysis is performed on the drying oven temperature timing-significant aggregation representation vector and the feed speed timing-significant aggregation representation vector to obtain a drying oven temperature-feed speed timing-decoupling enhancement interaction representation vector; S550, based on the drying oven temperature-feed speed timing-decoupling enhancement interaction representation vector, a control instruction for the feed speed is determined.

[0060] Based on this, in the technical solution of the present application, first, the time queue of the real-time oven temperature collected by the temperature sensor and the time queue of the real-time feed speed collected by the speed sensor are obtained. Then, considering that the processing of calcium silicate board in the oven is a dynamic process, in order to more finely extract the dynamic change trend of the oven temperature and the feed speed over time, the present application first divides the time queue of the real-time oven temperature and the time queue of the real-time feed speed at a predetermined time scale to refine the granularity of data analysis. Then, the sequence encoder based on the 1D-CNN model is used to perform time series encoding on each local time series data after segmentation, and a sliding convolution operation is performed on the time dimension of the data using a one-dimensional convolution kernel, so as to effectively extract the time series correlation features in the local time series data, and capture more subtle time series change patterns of the oven temperature and the feed speed, so as to obtain a sequence of local time series correlation feature vectors of the real-time oven temperature and a sequence of local time series correlation feature vectors of the real-time feed speed.

[0061] Correspondingly, in step S520, the time queue of the real-time drying oven temperature and the time queue of the real-time feeding speed are subjected to time series encoding based on the local time scale to obtain a sequence of local time series associated feature vectors of the real-time drying oven temperature and a sequence of local time series associated feature vectors of the real-time feeding speed, including: dividing the time queue of the real-time drying oven temperature and the time queue of the real-time feeding speed according to a predetermined time scale and inputting the split signals into a sequence encoder based on the 1D-CNN model respectively to obtain a sequence of local time series associated feature vectors of the real-time drying oven temperature and a sequence of local time series associated feature vectors of the real-time feeding speed.

[0062] Furthermore, in order to capture the dynamic change pattern of the oven temperature and feed speed over a longer time range, it is necessary to further perform feature forward propagation aggregation on the local time domain information of the oven temperature and feed speed to form global time domain information. In particular, considering that the time series change information of the oven temperature and feed speed in different time periods may have different importance, and in time series data, the influence of the information of the historical time period on the current time period tends to gradually weaken over time. Therefore, the present application introduces a feature forward propagation model based on node energy decay perception to aggregate the sequence of the local time series associated feature vectors of the real-time oven temperature and the sequence of the local time series associated feature vectors of the real-time feed speed, respectively, and simulates the propagation process of information in the time dimension by calculating the decay factor, so that the newer information has a higher weight, and the older information gradually decays, thereby achieving effective aggregation of the global time domain information of the oven temperature and feed speed. Specifically, the model first calculates the energy-significant descriptor of each local time series correlation feature vector in the input sequence based on the mean and variance, quantifies its feature importance, and assigns time information to the local time series correlation feature vector of each time node. It uses the time series decay mechanism to generate the corresponding energy time decay factor based on the time span between the local time series correlation feature vector of each historical time node and the current (i.e., the end of the time series) local time series correlation feature vector. In this way, in the process of feature forward propagation, information can be aggregated based on the feature significance of each time node and the time series decay characteristics of information, thereby generating the time series significant aggregation representation vector of the drying oven temperature and the time series significant aggregation representation vector of the feed speed.

[0063] Accordingly, in step S530, if Figure 2As shown, the sequence of the real-time local time series association feature vectors of the drying oven temperature and the sequence of the real-time local time series association feature vectors of the feed speed are input into the feature forward propagation model based on node energy decay perception to obtain the drying oven temperature time series significant aggregation representation vector and the feed speed time series significant aggregation representation vector, including: S531, based on the mean and variance of each real-time local time series association feature vector of the drying oven temperature in the sequence of the real-time local time series association feature vectors, the energy significant descriptor of each real-time local time series association feature vector of the drying oven temperature is calculated to obtain a sequence of energy significant descriptors; S532, extracting the timestamps of each real-time local time series association feature vector of the drying oven temperature in the sequence of the real-time local time series association feature vector to obtain a sequence of timestamps; S533, taking the real-time local time series association feature vector of the drying oven temperature corresponding to the current time point in the sequence of the real-time local time series association feature vectors as the current real-time local time series association feature vector of the drying oven temperature, and taking the real-time local time series association feature vectors corresponding to other time points as the real-time local time series association feature vector of the drying oven temperature. The local time series correlation feature vector of the current bake oven temperature is used as the local time series correlation feature vector of the historical real-time bake oven temperature to obtain a sequence of the current real-time bake oven temperature local time series correlation feature vector and the historical real-time bake oven temperature local time series correlation feature vector; S534, calculating the energy time decay factor of each historical real-time bake oven temperature local time series correlation feature vector in the sequence of the historical real-time bake oven temperature local time series correlation feature vector relative to the current real-time bake oven temperature local time series correlation feature vector to obtain a sequence of energy time decay factors; S535, using the sequence of the energy significant descriptors as a positive adjustment factor and the sequence of the energy time decay factors as a negative adjustment factor, performing time series propagation aggregation on the sequence of the historical real-time bake oven temperature local time series correlation feature vectors to obtain a historical node real-time bake oven temperature time series aggregation feature vector; S536, fusing the historical node real-time bake oven temperature time series aggregation feature vector and the current real-time bake oven temperature local time series association feature vector to obtain the bake oven temperature time series significant aggregation representation vector.

[0064] Among them, in step S531, based on the mean and variance of each real-time drying oven temperature local time series associated feature vector in the sequence of the real-time drying oven temperature local time series associated feature vector, the energy significant descriptor of each real-time drying oven temperature local time series associated feature vector is calculated to obtain a sequence of energy significant descriptors, including: calculating the expected value of the fourth power of the difference between each eigenvalue in the real-time drying oven temperature local time series associated feature vector and its eigenmean, and dividing the expected value by the square of the eigenvariance of the real-time drying oven temperature local time series associated feature vector to obtain the energy significant descriptor.

[0065] Among them, in step S534, the energy time decay factor of each historical real-time drying oven temperature local time series associated feature vector in the sequence of the historical real-time drying oven temperature local time series associated feature vector is calculated relative to the current real-time drying oven temperature local time series associated feature vector to obtain a sequence of energy time decay factors, including: calculating the timestamp difference between the historical real-time drying oven temperature local time series associated feature vector and the current real-time drying oven temperature local time series associated feature vector; rounding down the timestamp difference, dividing it by a preset time decay period and then performing a square operation, and multiplying the square result by a preset decay rate to obtain an initial energy decay coefficient; calculating an exponential function value with a natural constant as the base and the initial energy decay coefficient as the exponent to obtain the energy time decay factor.

[0066] In a specific example, the sequence of the local time series correlation feature vectors of the real-time drying oven temperature and the sequence of the local time series correlation feature vectors of the real-time feed speed are input into the feature forward propagation model based on node energy decay perception to obtain the drying oven temperature time series significant aggregation representation vector and the feed speed time series significant aggregation representation vector, including: processing the sequence of the local time series correlation feature vectors of the real-time drying oven temperature with the following time series propagation fusion formula to obtain the drying oven temperature time series significant aggregation representation vector, wherein the time series propagation fusion formula is:

[0067] ;

[0068] in, represents a sequence of local time series correlation feature vectors of the real-time drying tunnel temperature, The value of is the number of feature vectors in the sequence of the local time series associated feature vectors of the real-time drying tunnel temperature, and The first, second, and third , and the local time series correlation feature vector of the current real-time drying oven temperature, Indicates the The local time series correlation feature vector of the real-time drying oven temperature eigenvalues, The value of The length of the local time series correlation feature vector of the real-time drying tunnel temperature, For the said The characteristic mean of the local time series correlation characteristic vector of the real-time drying tunnel temperature, For the said The characteristic variance of the local time series correlation feature vector of the real-time drying tunnel temperature, Indicates the The fourth-order central moment of the local time series correlation eigenvector of the real-time drying tunnel temperature, that is, The expected value of the fourth power of the difference between each eigenvalue and its eigenmean in the local time series correlation eigenvector of the real-time drying tunnel temperature, is the energy significant descriptor of the local time series correlation feature vector of the real-time drying tunnel temperature, is the energy significant descriptor of the local time series correlation feature vector of the current real-time drying oven temperature, The timestamp representing the local time series associated feature vector of the current real-time drying oven temperature, Indicates the The timestamp of the local time series correlation feature vector of the real-time drying oven temperature, Indicates rounding down. Indicates the preset decay rate, Indicates the preset time decay period, represents the exponential function operation, is the weight hyperparameter, A vector representing a significant aggregation of the oven temperature time series is shown.

[0069] Then, in order to deeply analyze the complex relationship between the drying oven temperature and the feed speed to achieve precise control of the drying process, this application introduces a feature response aggregation module based on fine-grained feature decoupling to perform fine-grained interactive response analysis on the drying oven temperature time series significant aggregation representation vector and the feed speed time series significant aggregation representation vector. Specifically, the module first decouples the drying oven temperature time series significant aggregation representation vector and the feed speed time series significant aggregation representation vector respectively, decomposing the two into two sets of local representation vector sequences to enhance the sensitivity to local detail features. Then, the local semantic association modeling is performed on the sequence of the two sets of local representation vectors after decoupling to generate a local semantic association interaction vector sequence, and on this basis, the feature interaction response strength is further quantified to generate an attention weight sequence, so that the model can focus on the most important feature interaction, thereby improving the quality of feature representation and prediction accuracy. At the same time, in order to prevent overfitting, the generated attention weight sequence is further subjected to random inactivation processing to enhance the generalization ability of the model. Finally, the attention weight sequence after random inactivation is used as the weight to perform weighted summation on the local semantic association interaction vector sequence, so as to integrate the local interaction information of the oven temperature time series significant aggregation representation vector and the feed speed time series significant aggregation representation vector, and obtain the global oven temperature-feed speed time series decoupling enhanced interaction representation vector, so as to fully explore the interaction and influence relationship between the oven temperature and the feed speed, and provide a more accurate basis for the adaptive control of the feed speed.

[0070] Correspondingly, in step S540, feature interaction response analysis is performed on the drying oven temperature timing significant aggregation representation vector and the feed speed timing significant aggregation representation vector to obtain a drying oven temperature-feed speed timing decoupling enhancement interaction representation vector, including: inputting the drying oven temperature timing significant aggregation representation vector and the feed speed timing significant aggregation representation vector into a feature response aggregation module based on feature fine-grained decoupling to obtain the drying oven temperature-feed speed timing decoupling enhancement interaction representation vector.

[0071] Specifically, Figure 3 As shown, the oven temperature time series significant aggregation representation vector and the feed speed time series significant aggregation representation vector are input into a feature response aggregation module based on feature fine-grained decoupling to obtain the oven temperature-feed speed time series decoupling enhanced interaction representation vector, including: S541, feature decoupling the oven temperature time series significant aggregation representation vector and the feed speed time series significant aggregation representation vector respectively to obtain a sequence of oven temperature time series feature local representation vectors and a sequence of feed speed time series feature local representation vectors; S542, calculating the oven temperature time series feature local representation vector and the feed speed time series feature local representation vector of each group corresponding to each group in the sequence of the oven temperature time series feature local representation vector and the sequence of the feed speed time series feature local representation vector. The local semantic association interaction vectors between the quantities are calculated to obtain a sequence of oven temperature-feed speed local semantic association interaction vectors; S543, the fine-grained attention weights of each oven temperature-feed speed local semantic association interaction vector in the sequence of oven temperature-feed speed local semantic association interaction vectors are calculated to obtain a sequence of fine-grained attention weights; S544, the sequence of fine-grained attention weights is input into a random deactivation module to obtain a sequence of fine-grained attention weights after deactivation processing; S545, the sequence of fine-grained attention weights after deactivation processing is used as a weight distribution, and the weighted sum of the sequence of oven temperature-feed speed local semantic association interaction vectors is calculated to obtain the oven temperature-feed speed temporal decoupling reinforcement interaction representation vector.

[0072] Among them, in step S542, the local semantic association interaction vector between each group of corresponding local representation vectors of drying tunnel temperature timing features and local representation vectors of feed speed timing features in the sequence of local representation vectors of drying tunnel temperature timing features and the sequence of local representation vectors of feed speed timing features is calculated to obtain a sequence of local semantic association interaction vectors of drying tunnel temperature-feed speed, including: calculating the position point addition, position point subtraction and position point multiplication between the local representation vectors of drying tunnel temperature timing features and the corresponding local representation vectors of feed speed timing features to obtain a first fusion result, a second fusion result and a third fusion result; after cascading the first fusion result, the second fusion result and the third fusion result into a multi-scale fusion feature vector of drying tunnel temperature-feed speed, one-dimensional convolution processing and maximum value pooling processing are performed on them to obtain the local semantic association interaction vector of drying tunnel temperature-feed speed.

[0073] Among them, in step S543, the fine-grained attention weights of each drying oven temperature-feed speed local semantic association interaction vector in the sequence of the drying oven temperature-feed speed local semantic association interaction vectors are calculated to obtain a sequence of fine-grained attention weights, including: multiplying the drying oven temperature-feed speed local semantic association interaction vector by a weight coefficient matrix, adding the weight to the bias vector and then passing the result through a tanh function to obtain a drying oven temperature-feed speed local semantic association interaction encoding feature vector; using an attention score reference vector to multiply the drying oven temperature-feed speed local semantic association interaction encoding feature vector to obtain the fine-grained attention weight.

[0074] In a specific example, the drying tunnel temperature time series significant aggregation representation vector and the feed speed time series significant aggregation representation vector are input into a feature response aggregation module based on feature fine-grained decoupling to obtain the drying tunnel temperature-feed speed time series decoupling enhanced interaction representation vector, including: processing the drying tunnel temperature time series significant aggregation representation vector and the feed speed time series significant aggregation representation vector with the following interaction response analysis formula to obtain the drying tunnel temperature-feed speed time series decoupling enhanced interaction representation vector, wherein the interaction response analysis formula is:

[0075] ;

[0076] in, The vector representing the significant aggregation of the drying tunnel temperature time series is: The vector representing the significant aggregation of the feed speed time series is, Indicates decoupling processing, and They represent the first, second, and third vectors of the local representation vector of the drying oven temperature time series feature. and The local representation vector of the time series characteristics of the drying tunnel temperature, The value of is the number of feature vectors in the sequence of local representation vectors of the drying oven temperature time series feature, and respectively represent the first, second, and third vectors in the sequence of the local representation vector of the feed speed time series feature. and A local representation vector of the feed speed time series characteristics, Indicates cascade operation, represents the maximum pooling operation, represents a one-dimensional convolution operation, represents dot product, Indicates point addition, Indicates point reduction, The first in the sequence of the local semantic association interaction vector of the drying oven temperature and the feeding speed is represented The local semantic association interaction vector of oven temperature and feed speed, is the weight coefficient matrix, is the bias vector, is the hyperbolic tangent function, is the attention score reference vector, Indicates fine-grained attention weights, represents random dropout processing, Indicates The fine-grained attention weights after dropout processing, A vector representing the oven temperature-feeding speed timing decoupling enhancement interaction is shown.

[0077] Next, the drying oven temperature-feed speed timing decoupling enhanced interaction representation vector is input into the classifier-based feed speed controller to obtain a control instruction, indicating whether the feed speed at the next time point should increase, decrease, or remain unchanged, so that the feed speed can dynamically adapt to the change of the drying oven temperature, thereby realizing real-time optimization of the drying process.

[0078] Accordingly, in step S550, based on the drying tunnel temperature-feed speed timing decoupling enhanced interaction representation vector, a control instruction for the feed speed is determined, including: inputting the drying tunnel temperature-feed speed timing decoupling enhanced interaction representation vector into a classifier-based feed speed controller to obtain the control instruction, wherein the control instruction is used to indicate that the feed speed at the next time point should increase, decrease, or remain unchanged.

[0079] It should be understood that the role of the classifier is to use the given categories and known training data to learn classification rules and classifiers, and then classify (or predict) unknown data. Logistic regression, SVM, etc. are often used to solve binary classification problems. For multi-class classification problems, logistic regression or SVM can also be used, but multiple binary classifications are required to form a multi-classification, but this is prone to errors and inefficient. Commonly used multi-classification methods include the Softmax classification function.

[0080] In a preferred example, the drying oven temperature-feed speed timing decoupling and enhanced interaction representation vector is input into a classifier-based feed speed controller to obtain the control instruction, including: optimizing the drying oven temperature-feed speed timing decoupling and enhanced interaction representation vector to obtain an optimized drying oven temperature-feed speed timing decoupling and enhanced interaction representation vector; inputting the optimized drying oven temperature-feed speed timing decoupling and enhanced interaction representation vector into the classifier-based feed speed controller to obtain the control instruction.

[0081] The method comprises: optimizing the drying oven temperature-feed speed timing decoupling and strengthening interaction representation vector to obtain an optimized drying oven temperature-feed speed timing decoupling and strengthening interaction representation vector, comprising: clustering all eigenvalues ​​of the drying oven temperature-feed speed timing decoupling and strengthening interaction representation vector based on the L2 distance between the eigenvalues, and arranging the clustering features into a drying oven temperature-feed speed timing decoupling and strengthening interaction clustering vector; determining the number of eigenvalues ​​of the drying oven temperature-feed speed timing decoupling and strengthening interaction clustering vector and the ... The cluster ratio value of the number of eigenvalues ​​of the interaction representation vector is strengthened; the second norm of the drying oven temperature-feeding speed timing decoupling strengthening interaction clustering vector is divided by the second norm of the drying oven temperature-feeding speed timing decoupling strengthening interaction representation vector to obtain the rolling speed-adding amount decoupling strengthening timing interaction conflict representation value; the first power value of the one-norm of the drying oven temperature-feeding speed timing decoupling strengthening interaction clustering vector with the cluster ratio value as the exponent is divided by the one-norm of the drying oven temperature-feeding speed timing decoupling strengthening interaction representation vector with the cluster ratio value as the exponent The second power value of the number is used to obtain the rolling speed-additive amount decoupling reinforcement timing interaction confrontation representation value; for each eigenvalue of the oven temperature-feeding speed timing decoupling reinforcement interaction clustering vector, it is multiplied by the reciprocal of the difference between the rolling speed-additive amount decoupling reinforcement timing interaction conflict representation value and the rolling speed-additive amount decoupling reinforcement timing interaction confrontation representation value to obtain the optimized eigenvalue of the oven temperature-feeding speed timing decoupling reinforcement interaction clustering vector; for each eigenvalue outside the cluster in the oven temperature-feeding speed timing decoupling reinforcement interaction representation vector The optimized out-of-class eigenvalue of the drying oven temperature-feeding speed timing decoupling and strengthening interaction representation vector is formed by the optimized eigenvalue of the drying oven temperature-feeding speed timing decoupling and strengthening interaction clustering vector and the optimized out-of-class eigenvalue of the drying oven temperature-feeding speed timing decoupling and strengthening interaction representation vector.

[0082] In one example, optimizing the drying tunnel temperature-feed speed timing decoupling and strengthening interaction representation vector to obtain an optimized drying tunnel temperature-feed speed timing decoupling and strengthening interaction representation vector includes: optimizing the drying tunnel temperature-feed speed timing decoupling and strengthening interaction representation vector using the following optimization formula to obtain the optimized drying tunnel temperature-feed speed timing decoupling and strengthening interaction representation vector; wherein the optimization formula is:

[0083] ;

[0084] in, is the vector representing the oven temperature-feed speed timing decoupling enhancement interaction, is the number of eigenvalues ​​of the vector representing the oven temperature-feed speed timing decoupling enhancement interaction, is the drying tunnel temperature-feeding speed timing decoupling enhanced interaction clustering vector, is the number of eigenvalues ​​of the oven temperature-feed speed timing decoupling enhanced interaction clustering vector, represents the clustering ratio value, represents the clustering feature set corresponding to the drying oven temperature-feeding speed time series decoupling enhanced interaction clustering vector, and Represents the binary norm and the uninorm of the vector respectively. Power, represents the characteristic value of each position in the drying oven temperature-feeding speed timing decoupling enhanced interaction representation vector, The eigenvalues ​​of each position in the optimized drying tunnel temperature-feeding speed timing decoupling enhanced interaction representation vector are represented.

[0085] That is, considering that the oven temperature time series significant aggregation representation vector and the feed speed time series significant aggregation representation vector respectively represent the local time series correlation characteristics of the oven temperature and the feed speed based on the forward propagation aggregation characteristics of the significant attenuation of the time series node energy, when further performing the local fine-grained sequence interaction response aggregation based on feature decoupling, the local fine-grained sequence interaction response alignment conflict will be caused due to the forward propagation aggregation difference of the time series feature distribution, thereby causing the loss of aggregation key information, affecting the expression effect of the oven temperature-feed speed time series decoupling enhanced interaction representation vector.

[0086] Therefore, in order to avoid the loss of key suffix semantic information relative to the original feature set as a whole due to aggregation conflict caused by the drying oven temperature-feed speed timing decoupling and strengthening interaction representation vector based on the aggregation feature, the adversarial decision of the set absolute representation of the drying oven temperature-feed speed timing decoupling and strengthening interaction clustering vector and the one-norm of the drying oven temperature-feed speed timing decoupling and strengthening interaction representation vector is performed by taking the clustering ratio of the number of eigenvalues ​​of the drying oven temperature-feed speed timing decoupling and strengthening interaction clustering vector relative to the number of eigenvalues ​​of the drying oven temperature-feed speed timing decoupling and strengthening interaction representation vector as the decision function, and respectively comparing the absolute representation of the set of the drying oven temperature-feed speed timing decoupling and strengthening interaction clustering vector with the drying oven temperature-feed speed timing decoupling and strengthening interaction clustering vector. Positive and negative interactions are performed with the clustered intrinsic conflict representation of the two norm of the drying oven temperature-feed speed timing decoupling and enhanced interaction representation vector to construct a solid alignment guardrail of the optimized drying oven temperature-feed speed timing decoupling and enhanced interaction representation vector based on the aggregated features and the original feature set as a whole, thereby achieving the mitigation of the intention of harmful information loss based on the aggregated risk transferability of the optimized drying oven temperature-feed speed timing decoupling and enhanced interaction representation vector, improving the expression effect of the optimized drying oven temperature-feed speed timing decoupling and enhanced interaction representation vector, thereby improving the accuracy of the control instructions obtained by the drying oven temperature-feed speed timing decoupling and enhanced interaction representation vector input into the classifier-based feed speed controller.

[0087] Based on the above embodiments, see Figure 4 The figure is a schematic diagram of the structure of a calcium silicate board surface enhancement treatment system 800 in an embodiment of the present application. The calcium silicate board surface enhancement treatment system 800 includes: a data acquisition module 810, which is used to obtain the time queue of the real-time drying oven temperature collected by the temperature sensor, and the time queue of the real-time feeding speed collected by the speed sensor; a time series encoding module 820, which is used to perform time series encoding on the time queue of the real-time drying oven temperature and the time queue of the real-time feeding speed based on the local time scale to obtain a sequence of local time series correlation feature vectors of the real-time drying oven temperature and a sequence of local time series correlation feature vectors of the real-time feeding speed; an aggregation module 830, which is used to aggregate the sequence of the local time series correlation feature vectors of the real-time drying oven temperature and the real-time feeding speed. The sequence input of the local timing-related feature vector of the real-time feed speed is based on the feature forward propagation model of node energy decay perception to obtain the oven temperature timing significant aggregation representation vector and the feed speed timing significant aggregation representation vector; the feature interaction response analysis module 840 is used to perform feature interaction response analysis on the oven temperature timing significant aggregation representation vector and the feed speed timing significant aggregation representation vector to obtain the oven temperature-feed speed timing decoupling and enhanced interaction representation vector; the feed speed control module 850 is used to determine the control instruction of the feed speed based on the oven temperature-feed speed timing decoupling and enhanced interaction representation vector.

[0088] Here, those skilled in the art will appreciate that the specific functions and operations of each module in the calcium silicate board surface enhancement treatment system 800 have been described in detail above. Figures 1 to 3 The surface enhancement treatment method of the calcium silicate board has been described in detail, and therefore, its repeated description will be omitted.

[0089] Figure 5 FIG. 1 is an application scenario diagram of the surface enhancement treatment method of calcium silicate board according to an embodiment of the present application. Figure 5 As shown, in this application scenario, first, the time queue of the real-time oven temperature collected by the temperature sensor is obtained (for example, Figure 5 D1 shown in the figure) and the time queue of the real-time feed speed collected by the speed sensor (for example, Figure 5 Then, the time queue of the real-time oven temperature and the time queue of the real-time feed speed are input to a server (for example, Figure 5 In S), the server can use the calcium silicate board surface enhancement processing algorithm to process the time queue of the real-time oven temperature and the time queue of the real-time feed speed to obtain a control instruction indicating that the feed speed at the next time point should be increased, decreased or kept unchanged.

[0090] Based on the above embodiment, another exemplary embodiment of an electronic device is also provided in the embodiment of the present application. In some possible implementations, the electronic device in the embodiment of the present application may include a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the steps of the surface enhancement treatment method of the calcium silicate board in the above embodiment can be implemented when the processor executes the program.

[0091] The embodiment of the present application also provides a computer-readable storage medium, on which computer-executable instructions are stored. When the computer-executable instructions are executed by a processor, the surface enhancement treatment method of the calcium silicate board according to the embodiment of the present application described with reference to the above figures can be executed. The computer-readable storage medium includes, but is not limited to, for example, volatile memory and / or non-volatile memory. The volatile memory may include, for example, a random access memory (RAM) and / or a cache memory (cache), etc. The non-volatile memory may include, for example, a read-only memory (ROM), a hard disk, a flash memory, etc.

[0092] The embodiment of the present application also provides a computer program product or a computer program, which includes computer executable instructions, and the computer executable instructions are stored in a computer readable storage medium. The processor of the computer device reads the computer executable instructions from the computer readable storage medium, and the processor executes the computer executable instructions, so that the computer device performs the surface enhancement treatment method of the calcium silicate board according to the embodiment of the present application.

[0093] Those skilled in the art will appreciate that the contents disclosed in this application may be subject to various variations and improvements. For example, the various devices or components described above may be implemented by hardware, or by software, firmware, or a combination of some or all of the three.

[0094] In addition, although the present application makes various references to certain units in the system according to embodiments of the present application, any number of different units can be used and run on the client and / or server. The units are illustrative only, and different aspects of the systems and methods can use different units.

[0095] Those skilled in the art will appreciate that all or part of the steps in the above method can be completed by instructing related hardware through a program, and the program can be stored in a computer-readable storage medium, such as a read-only memory, a disk or an optical disk. Optionally, all or part of the steps in the above embodiment can also be implemented using one or more integrated circuits. Accordingly, each module / unit in the above embodiment can be implemented in the form of hardware or in the form of software function modules. The present application is not limited to any particular form of combination of hardware and software.

[0096] Unless otherwise defined, all terms (including technical and scientific terms) used herein have the same meaning as commonly understood by those of ordinary skill in the art to which this application belongs. It should also be understood that terms such as those defined in common dictionaries should be interpreted as having a meaning consistent with their meaning in the context of the relevant technology, and should not be interpreted in an idealized or extremely formal sense, unless explicitly defined as such herein.

[0097] The above is an explanation of the present application and should not be considered as limiting thereof. Although several exemplary embodiments of the present application have been described, those skilled in the art will readily appreciate that many modifications may be made to the exemplary embodiments without departing from the novel teachings and advantages of the present application.

Claims

1. A surface enhancement treatment method for a calcium silicate board, comprising: The calcium silicate board is immersed in a silicon dioxide solution with a solid content of more than 95%, and completely absorbed until saturated, and then the immersed calcium silicate board is dried to obtain a modified calcium silicate board; a first gluing is performed on any surface of the modified calcium silicate board, and the first gluing includes one gluing, one leveling penetration, one drying, one cooling and one polishing; a second gluing is performed on the surface of the modified calcium silicate board after the first gluing to obtain a solidified calcium silicate board surface, and the second gluing includes a second gluing, a second drying and a second polishing, characterized in that the second drying is performed on the surface of the modified calcium silicate board after the first gluing, including: Acquire the time queue of the real-time drying tunnel temperature collected by the temperature sensor and the time queue of the real-time feeding speed collected by the speed sensor; Performing temporal coding based on the local time scale on the time queue of the real-time drying oven temperature and the time queue of the real-time feeding speed to obtain a sequence of local temporal correlation feature vectors of the real-time drying oven temperature and a sequence of local temporal correlation feature vectors of the real-time feeding speed; Inputting the sequence of local time series associated feature vectors of the real-time drying oven temperature and the sequence of local time series associated feature vectors of the real-time feeding speed into the feature forward propagation model based on node energy decay perception to obtain a drying oven temperature time series significant aggregation representation vector and a feeding speed time series significant aggregation representation vector; Performing feature interaction response analysis on the oven temperature time series significant aggregation representation vector and the feed speed time series significant aggregation representation vector to obtain an oven temperature-feed speed time series decoupling enhancement interaction representation vector; Determine the control instruction of the feed speed based on the drying tunnel temperature-feed speed timing decoupling enhanced interaction representation vector; Inputting the sequence of the local time series associated feature vectors of the real-time drying oven temperature and the sequence of the local time series associated feature vectors of the real-time feed speed into the feature forward propagation model based on node energy decay perception to obtain the drying oven temperature time series significant aggregation representation vector and the feed speed time series significant aggregation representation vector, including: Based on the mean and variance of each real-time drying tunnel temperature local time series correlation feature vector in the sequence of the real-time drying tunnel temperature local time series correlation feature vector, calculating the energy significant descriptor of each real-time drying tunnel temperature local time series correlation feature vector to obtain a sequence of energy significant descriptors; Extracting the timestamp of each real-time drying oven temperature local time series correlation feature vector in the sequence of the real-time drying oven temperature local time series correlation feature vector to obtain a sequence of timestamps; The real-time drying oven temperature local time series association feature vector corresponding to the current time point in the sequence of the real-time drying oven temperature local time series association feature vectors is used as the current real-time drying oven temperature local time series association feature vector, and the real-time drying oven temperature local time series association feature vectors corresponding to other time points are used as the historical real-time drying oven temperature local time series association feature vectors, so as to obtain a sequence of the current real-time drying oven temperature local time series association feature vector and the historical real-time drying oven temperature local time series association feature vector; Calculating the energy time decay factor of each historical real-time drying oven temperature local time series associated feature vector in the sequence of the historical real-time drying oven temperature local time series associated feature vector relative to the current real-time drying oven temperature local time series associated feature vector to obtain a sequence of energy time decay factors; Using the sequence of energy significant descriptors as positive adjustment factors and the sequence of energy time decay factors as negative adjustment factors, the sequence of local time series correlation feature vectors of the historical real-time oven temperature is subjected to time series propagation aggregation to obtain a time series aggregation feature vector of the historical node real-time oven temperature; Fusion of the historical node real-time oven temperature time series aggregation feature vector and the current real-time oven temperature local time series association feature vector to obtain the oven temperature time series significant aggregation representation vector; Performing feature interaction response analysis on the oven temperature time series significant aggregation representation vector and the feed speed time series significant aggregation representation vector to obtain an oven temperature-feed speed time series decoupling enhancement interaction representation vector, including: Input the oven temperature time series significant aggregation representation vector and the feed speed time series significant aggregation representation vector into a feature response aggregation module based on feature fine-grained decoupling to obtain the oven temperature-feed speed time series decoupling enhanced interaction representation vector; Inputting the oven temperature time series significant aggregation representation vector and the feed speed time series significant aggregation representation vector into a feature response aggregation module based on feature fine-grained decoupling to obtain the oven temperature-feed speed time series decoupling enhanced interaction representation vector, including: Decoupling the drying tunnel temperature time series significant aggregation representation vector and the feeding speed time series significant aggregation representation vector to obtain a sequence of drying tunnel temperature time series feature local representation vectors and a sequence of feeding speed time series feature local representation vectors; Calculating the local semantic association interaction vectors between each corresponding group of the local representation vectors of the drying oven temperature time series features and the local representation vectors of the feeding speed time series features in the sequence of the local representation vectors of the drying oven temperature time series features and the sequence of the local representation vectors of the feeding speed time series features to obtain a sequence of the local semantic association interaction vectors of the drying oven temperature and the feeding speed; Calculating the fine-grained attention weight of each oven temperature-feed speed local semantic association interaction vector in the sequence of oven temperature-feed speed local semantic association interaction vectors to obtain a sequence of fine-grained attention weights; Inputting the sequence of fine-grained attention weights into a random dropout module to obtain a sequence of fine-grained attention weights after dropout processing; The sequence of fine-grained attention weights after the deactivation process is used as the weight distribution, and the weighted sum of the sequence of the oven temperature-feeding speed local semantic association interaction vectors is calculated to obtain the oven temperature-feeding speed temporal decoupling reinforcement interaction representation vector.

2. The surface enhancement treatment method of calcium silicate board according to claim 1, characterized in that: The time queue of the real-time drying oven temperature and the time queue of the real-time feeding speed are subjected to time series encoding based on the local time scale to obtain a sequence of local time series correlation feature vectors of the real-time drying oven temperature and a sequence of local time series correlation feature vectors of the real-time feeding speed, including: The time queue of the real-time drying oven temperature and the time queue of the real-time feeding speed are divided into predetermined time scales and respectively input into a sequence encoder based on a 1D-CNN model to obtain a sequence of local temporal correlation feature vectors of the real-time drying oven temperature and a sequence of local temporal correlation feature vectors of the real-time feeding speed.

3. The surface enhancement treatment method of calcium silicate board according to claim 2, characterized in that: Based on the mean and variance of each real-time drying tunnel temperature local time series associated feature vector in the sequence of the real-time drying tunnel temperature local time series associated feature vector, calculating the energy significant descriptor of each real-time drying tunnel temperature local time series associated feature vector to obtain a sequence of energy significant descriptors, including: The expected value of the fourth power of the difference between each eigenvalue in the local time series associated eigenvalue vector of the real-time drying tunnel temperature and its eigenmean is calculated, and the expected value is divided by the square of the eigenvariance of the local time series associated eigenvalue vector of the real-time drying tunnel temperature to obtain the energy significant descriptor.

4. The surface enhancement treatment method of calcium silicate board according to claim 3, characterized in that: Calculating the energy time decay factor of each historical real-time drying oven temperature local time series correlation feature vector in the sequence of the historical real-time drying oven temperature local time series correlation feature vector relative to the current real-time drying oven temperature local time series correlation feature vector to obtain a sequence of energy time decay factors, including: Calculate the timestamp difference between the historical real-time drying oven temperature local time series associated feature vector and the current real-time drying oven temperature local time series associated feature vector; The timestamp difference is rounded down and divided by a preset time decay period and then squared, and the square result is multiplied by a preset decay rate to obtain an initial energy decay coefficient; An exponential function value with a natural constant as the base and the initial energy decay coefficient as the exponent is calculated to obtain the energy time decay factor.

5. The surface enhancement treatment method of calcium silicate board according to claim 4, characterized in that: Calculating the local semantic association interaction vector between each group of corresponding local representation vectors of drying oven temperature time series features and local representation vectors of feeding speed time series features in the sequence of local representation vectors of drying oven temperature time series features and the sequence of local representation vectors of feeding speed time series features to obtain a sequence of local semantic association interaction vectors of drying oven temperature and feeding speed, including: Calculate the position point addition, position point subtraction and position point multiplication between the local representation vector of the drying tunnel temperature time series feature and the corresponding local representation vector of the feeding speed time series feature to obtain a first fusion result, a second fusion result and a third fusion result; After the first fusion result, the second fusion result and the third fusion result are cascaded into a drying tunnel temperature-feed speed multi-scale fusion feature vector, one-dimensional convolution processing and maximum pooling processing are performed on it to obtain the drying tunnel temperature-feed speed local semantic association interaction vector.

6. The surface enhancement treatment method of calcium silicate board according to claim 5, characterized in that: Calculating the fine-grained attention weight of each oven temperature-feed speed local semantic association interaction vector in the sequence of oven temperature-feed speed local semantic association interaction vectors to obtain a sequence of fine-grained attention weights, including: The drying oven temperature-feeding speed local semantic association interaction vector is multiplied by a weight coefficient matrix, and then added to the bias vector and passed through a tanh function to obtain a drying oven temperature-feeding speed local semantic association interaction encoding feature vector; The attention score reference vector is multiplied by the oven temperature-feed speed local semantic association interaction encoding feature vector to obtain the fine-grained attention weight.

7. The surface enhancement treatment method of calcium silicate board according to claim 6, characterized in that: Based on the drying tunnel temperature-feeding speed timing decoupling enhanced interaction representation vector, a control instruction for the feeding speed is determined, including: The drying tunnel temperature-feeding speed timing decoupling enhanced interaction representation vector is input into the classifier-based feeding speed controller to obtain the control instruction, and the control instruction is used to indicate whether the feeding speed at the next time point should increase, decrease or remain unchanged.

Images