Urea-formaldehyde resin toilet lid hot press forming intelligent control method based on AI algorithm

Abstract

The application discloses an AI algorithm-based intelligent control method for urea-formaldehyde resin toilet cover hot press forming, relates to the technical field of intelligent hot press forming control, and comprises the transmission and preprocessing of collected sensor data; after the preprocessing of the collected sensor data, a man-machine interaction interface is constructed in a matched mode.The application improves the control precision, stability and intelligent level of the urea-formaldehyde resin toilet cover hot press forming process, reduces the dependence on manual intervention, improves the production efficiency and product quality stability, solves the problems of response lag, insufficient precision and difficulty in generalization of the traditional control method, and provides an efficient, reliable and applicable intelligent manufacturing solution for the hot press forming process.

Description

Technical Field

[0001] This invention relates to the field of intelligent hot pressing molding control technology, and in particular to an intelligent control method for hot pressing molding of urea-formaldehyde resin toilet seats based on AI algorithms. Background Technology

[0002] With the development of intelligent manufacturing and artificial intelligence technologies, the precision of parameter control and the level of intelligent modeling in the hot pressing process have become key factors affecting product quality and production efficiency. In the hot pressing process of urea-formaldehyde resin toilet seats, physical quantities such as temperature, pressure, and displacement change complexly and have strong time dependence. Traditional control methods suffer from low precision, slow response, and difficulty in generalization when handling multivariate nonlinear time-series data. Therefore, intelligent control methods for hot pressing based on AI algorithms have attracted widespread attention, especially the superior performance of deep learning models in modeling complex dynamic processes, making data-driven process prediction and optimization possible.

[0003] However, existing AI models still face many challenges when processing multivariate time series data, such as information confusion caused by the coupling of physical quantity features, the difficulty of traditional convolutional structures in capturing long-range dependencies, the lack of adaptability in local pattern extraction, and insufficient multi-scale information fusion. Summary of the Invention

[0004] In view of the problems existing in the intelligent control method for hot pressing molding of urea-formaldehyde resin toilet seats based on AI algorithms, this invention is proposed.

[0005] Therefore, the problem to be solved by this invention is: how to overcome the shortcomings of existing AI models in processing multivariate nonlinear time series data during the hot pressing process of urea-formaldehyde resin toilet seat, such as feature coupling interference, weak long-range dependency modeling ability, lack of adaptability in local pattern extraction, and insufficient multi-scale information fusion, so as to achieve intelligent closed-loop control with high precision, strong generalization, and real-time response.

[0006] To solve the above-mentioned technical problems, the present invention provides the following technical solution:

[0007] In a first aspect, embodiments of the present invention provide an intelligent control method for the hot pressing molding of urea-formaldehyde resin toilet seats based on AI algorithms, which includes: collecting sensor data; transmitting and preprocessing the collected sensor data; using the preprocessed sensor data as the original input basis for constructing an AI deep learning network model; implementing closed-loop intelligent control of the hot pressing molding process of urea-formaldehyde resin toilet seats based on the weighted fusion prediction results, and constructing a human-machine interaction interface accordingly.

[0008] As a preferred embodiment of the intelligent control method for hot pressing molding of urea-formaldehyde resin toilet seat based on AI algorithm described in this invention, the acquisition of sensor data includes temperature data acquisition, pressure data acquisition and displacement data acquisition.

[0009] The temperature data acquisition includes arranging thermocouple temperature sensors at the hot-pressing part of the urea-formaldehyde resin toilet seat. The thermocouple temperature sensors are used to convert the measured temperature data into temperature electrical signals, which are transmitted to the data acquisition card through a shielded cable. The frequency of the temperature data acquisition is set according to the hot-pressing process requirements.

[0010] The pressure data acquisition includes installing a pressure sensor at the hot-pressing part of the urea-formaldehyde resin toilet seat. The pressure sensor is used to convert the measured pressure data into a pressure electrical signal, which is then transmitted to the data acquisition card via a shielded cable.

[0011] The displacement data acquisition includes using a linear displacement sensor to measure the displacement of the mold during the hot pressing process of the urea-formaldehyde resin toilet seat, and monitoring the compression of the material. The linear displacement sensor is used to convert the output pulse signal into actual displacement data after processing by the data acquisition card.

[0012] As a preferred embodiment of the intelligent control method for hot pressing molding of urea-formaldehyde resin toilet seat based on AI algorithm described in this invention, the transmission of the collected sensor data includes using a data acquisition card to transmit the collected temperature data, pressure data, and displacement data to the AI ​​computing unit via industrial Ethernet.

[0013] The preprocessing includes normalizing the collected temperature, pressure, and displacement data, mapping different data uniformly, and dividing them according to time order to generate structured multivariate time series data.

[0014] As a preferred embodiment of the intelligent control method for hot pressing molding of urea-formaldehyde resin toilet seat based on AI algorithm described in this invention, the construction of the AI ​​deep learning network model includes constructing two complementary deep learning models, namely a temporal convolutional network model and a multi-scale representation convolutional network model.

[0015] The temporal convolutional network model includes a patched variable independent embedding method to independently encode multivariate time series data, extract the temporal features of the independently encoded data, and predict the output results.

[0016] The patched variable independent embedding method includes processing the original input using the Patchify+ variable independent embedding method, as shown in the formula:

[0017]

[0018] in, This represents the original input time series. This indicates an embedded mapping function. This indicates the embedded patch;

[0019] The independent encoding of multivariate time series data includes inputting the embedded patch into a backbone network composed of stacked modern convolutional blocks, and designing modern convolutional blocks;

[0020] The modern convolutional blocks include depthwise separable convolutions, convolutional feedforward neural networks, convolutional feedforward neural networks, and those employing residual connections.

[0021] The modern convolutional blocks are propagated to calculate the global feature representation of the time series data, extract the independently encoded temporal features, and output the extracted temporal features based on the extracted temporal features.

[0022] The calculation formula for propagating the modern convolutional block is as follows:

[0023]

[0024] in, Indicates the first Input and output representations of a modern convolutional block Represents modern convolutional block functions;

[0025] For the initial block of a modern convolutional block, the formula is:

[0026]

[0027] in, This represents the input and output representations of the first modern convolutional block;

[0028] Based on the initial block of the modern convolutional block, subsequent blocks are calculated recursively, using the following formula:

[0029]

[0030] The independent encoding includes independently processing the calculated global feature representation of the time series data. The calculated global feature representation of the time series is given by the following formula:

[0031]

[0032] in, This represents the global feature representation of the time series. This represents a backbone network composed of K modern convolutional blocks stacked together;

[0033] The predicted output includes mapping the result to a target prediction after independent encoding, which is then used to fit the process parameters during the hot pressing process of urea-formaldehyde resin toilet seat. The formula is as follows:

[0034] Linear Flatten

[0035] in, Flatten represents the predicted output. Linear represents a one-dimensional vector. This represents a fully connected linear mapping layer.

[0036] As a preferred embodiment of the intelligent control method for hot pressing molding of urea-formaldehyde resin toilet seats based on AI algorithm described in this invention, the multi-scale representation convolutional network model includes an adaptive local pattern extraction mechanism, dynamic patch adjustment, multi-stage fully convolutional blocks with different kernel sizes, combined with reparameterization mechanism and learnable residual connections to achieve multi-scale representation dependency modeling, and weighted fusion of the prediction results of the temporal convolutional network model and the multi-scale representation convolutional network model.

[0037] The dynamic patch adjustment involves adaptive segmentation of the global feature representation of the time series, dynamically adjusting the position and scale of the patch through a learning method, preserving the semantic structure. The formula for adjusting the position and scale of the patch is as follows:

[0038]

[0039] in, Indicates the original center position. Indicates the location of the new patch center. Indicates the new patch length;

[0040] The patch boundary is calculated based on the adjusted position and scale of the patch, using the following formula:

[0041]

[0042] in, Indicates the left boundary of the patch. Indicates the right boundary of the patch;

[0043] The multi-scale representation dependency modeling includes introducing a reparameterization mechanism and introducing learnable residual connections to complete the multi-scale representation dependency modeling.

[0044] The formula for the reparameterization mechanism is:

[0045]

[0046] in, Indicates the convolution kernel weights. Indicates the first Large kernel branches in a fully convolutional block Indicates the first Small kernel branches in a fully convolutional block Indicates the first The output of each convolutional block Indicates the first The output of the depthwise convolutional layer in a fully convolutional block Indicates the activation function;

[0047] The formula for the learnable residual connection is:

[0048]

[0049] in, Indicates learnable weights, Indicates the first The output of depthwise convolution in each building block;

[0050] The weighted fusion of the prediction results from the temporal convolutional network model and the multi-scale representation convolutional network model includes: after multi-scale representation dependency modeling, mapping the results to predicted values ​​of the target hot-pressing process parameters, then introducing a linear layer to obtain the prediction results from the temporal convolutional network model and the multi-scale representation convolutional network model, and then performing weighted fusion.

[0051] The formula for introducing the linear layer is:

[0052]

[0053] in, This means that the final convolution output flattens out to a one-dimensional vector. This indicates the last fully convolutional stage. The weight matrix represents the linear transformation. Indicates bias. The output represents the prediction result;

[0054] The formula for the weighted fusion is:

[0055]

[0056] in, Indicates the weighting coefficient. This represents the prediction result after weighted fusion.

[0057] As a preferred embodiment of the intelligent control method for hot pressing molding of urea-formaldehyde resin toilet seat based on AI algorithm described in this invention, the method for implementing closed-loop intelligent control of the hot pressing molding process of urea-formaldehyde resin toilet seat includes designing three types of key control strategies.

[0058] The design employs three key control strategies: temperature control, pressure control, and pressure holding time control.

[0059] As a preferred embodiment of the intelligent control method for hot pressing molding of urea-formaldehyde resin toilet seat based on AI algorithm described in this invention, the supporting human-machine interface includes an operation interface, a monitoring interface, an alarm interface, and a historical data query and analysis interface.

[0060] The alarm interface includes prompts for the operation of the urea-formaldehyde resin toilet seat hot pressing process, and the alarm interface includes three alarm levels: high, medium and low.

[0061] Secondly, embodiments of the present invention provide an intelligent control system for the hot pressing molding of urea-formaldehyde resin toilet seats based on AI algorithms, which includes: a sensor data acquisition and preprocessing module, which acquires sensor data and transmits and preprocesses the acquired sensor data respectively.

[0062] The AI ​​deep learning network model building module preprocesses the collected sensor data and uses it as the raw input for building the AI ​​deep learning network model.

[0063] The intelligent control and human-machine interaction module for hot pressing molding implements closed-loop intelligent control of the hot pressing molding process of urea-formaldehyde resin toilet seats based on the weighted fusion prediction results, and also builds a human-machine interaction interface.

[0064] The sensor data acquisition and preprocessing module acquires sensor data and performs transmission and preprocessing on the acquired sensor data.

[0065] The AI ​​deep learning network model building module preprocesses the collected sensor data and uses it as the raw input for building the AI ​​deep learning network model.

[0066] The intelligent control and human-machine interaction module for hot pressing molding implements closed-loop intelligent control of the hot pressing molding process of urea-formaldehyde resin toilet seats based on the weighted fusion prediction results, and also builds a human-machine interaction interface.

[0067] Thirdly, embodiments of the present invention provide a computer device, including a memory and a processor, wherein the memory stores a computer program, and the processor executes the computer program to implement any step of the above-described intelligent control method for hot pressing molding of urea-formaldehyde resin toilet seats based on AI algorithms.

[0068] Fourthly, embodiments of the present invention provide a computer-readable storage medium having a computer program stored thereon, wherein: when the computer program is executed by a processor, it implements any step of the above-described intelligent control method for hot pressing molding of urea-formaldehyde resin toilet seats based on AI algorithms.

[0069] The beneficial effects of this invention are as follows: By constructing an integrated technology system encompassing sensor data acquisition and preprocessing, AI deep learning network model construction, intelligent control of hot pressing molding, and human-machine interaction, this invention significantly improves the control accuracy, stability, and intelligence level of the hot pressing molding process for urea-formaldehyde resin toilet seats. Specifically, the invention employs a patched variable independent embedding method to effectively decouple multiple physical quantity features such as temperature, pressure, and displacement, avoiding information confusion. It combines a structurally improved temporal convolutional network with an adaptive multi-scale representation convolutional network, and introduces a reparameterization mechanism, learnable residual connections, and a dual-model weighted fusion strategy, significantly enhancing the modeling ability and generalization performance for complex nonlinear temporal dynamics. Based on high-precision prediction results, it implements closed-loop intelligent control of temperature, pressure, and holding time, ensuring full resin curing, dense product structure, and consistent appearance. The accompanying human-machine interface enables flexible configuration of process parameters, real-time monitoring of process status, graded alarms for abnormal events, and traceable analysis of historical data, greatly reducing reliance on manual intervention and improving production efficiency and product quality stability. The overall solution solves the problems of slow response, insufficient accuracy, and difficulty in generalization of traditional control methods, and provides an efficient, reliable, and practical intelligent manufacturing solution for hot pressing molding processes. Attached Figure Description

[0070] To more clearly illustrate the technical solutions of the embodiments of the present invention, the accompanying drawings used in the description of the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort. Wherein:

[0071] Figure 1 A flowchart illustrating the intelligent control method for hot pressing molding of urea-formaldehyde resin toilet seats based on AI algorithms, provided in an embodiment of the present invention.

[0072] Figure 2 This is a schematic diagram of a system for an AI-based intelligent control method for the hot pressing molding of urea-formaldehyde resin toilet seats, provided in an embodiment of the present invention.

[0073] Figure 3 This is a schematic diagram of the structure of a medium for the intelligent control method of hot pressing molding of urea-formaldehyde resin toilet seats based on AI algorithms provided in an embodiment of the present invention.

[0074] Figure 4This is a schematic diagram of a computing device for an intelligent control method of urea-formaldehyde resin toilet seat hot pressing molding based on AI algorithm, provided in an embodiment of the present invention. Detailed Implementation

[0075] To make the above-mentioned objects, features, and advantages of the present invention more apparent and understandable, specific embodiments of the present invention will be described in detail below with reference to the accompanying drawings. Obviously, the described embodiments are only a part of the embodiments of the present invention, and not all of them. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort should fall within the protection scope of the present invention.

[0076] Many specific details are set forth in the following description in order to provide a full understanding of the invention. However, the invention may also be practiced in other ways different from those described herein, and those skilled in the art can make similar extensions without departing from the spirit of the invention. Therefore, the invention is not limited to the specific embodiments disclosed below.

[0077] Secondly, the term "one embodiment" or "embodiment" as used herein refers to a specific feature, structure, or characteristic that may be included in at least one implementation of the present invention. The phrase "in one embodiment" appearing in different places in this specification does not necessarily refer to the same embodiment, nor is it a single or selective embodiment that is mutually exclusive with other embodiments.

[0078] This invention is described in detail with reference to the schematic diagrams. When detailing the embodiments of this invention, for ease of explanation, the cross-sectional views illustrating the device structure may be partially enlarged, not adhering to the usual scale. Furthermore, the schematic diagrams are merely examples and should not be construed as limiting the scope of protection of this invention. In actual fabrication, the three-dimensional spatial dimensions of length, width, and depth should be included.

[0079] Furthermore, in the description of this invention, it should be noted that the terms "upper," "lower," "inner," and "outer," etc., indicate the orientation or positional relationship based on the orientation or positional relationship shown in the accompanying drawings. These terms are used solely for the convenience of describing the invention and for simplifying the description, and do not indicate or imply that the device or element referred to must have a specific orientation, or be constructed and operated in a specific orientation. Therefore, they should not be construed as limitations on the invention. In addition, the terms "first," "second," or "third" are used for descriptive purposes only and should not be construed as indicating or implying relative importance.

[0080] Unless otherwise explicitly specified and limited, the terms "installation," "connection," and "joining" in this invention should be interpreted broadly. For example, they can refer to fixed connections, detachable connections, or integral connections; similarly, they can refer to mechanical connections, electrical connections, or direct connections, or indirect connections through an intermediate medium, or internal connections between two components. Those skilled in the art can understand the specific meaning of the above terms in this invention based on the specific circumstances.

[0081] Example

[0082] Reference Figure 1 and Figure 2 This is the first embodiment of the present invention, which provides an intelligent control method for the hot pressing molding of urea-formaldehyde resin toilet seats based on AI algorithms, including:

[0083] S1: Collect sensor data, and transmit and preprocess the collected sensor data respectively;

[0084] The acquisition of sensor data includes temperature data acquisition, pressure data acquisition, and displacement data acquisition.

[0085] Temperature data acquisition includes placing thermocouple temperature sensors at the hot-pressing part of the urea-formaldehyde resin toilet seat. The thermocouple temperature sensors are used to convert the measured temperature data into temperature electrical signals, which are transmitted to the data acquisition card through shielded cables. The frequency of temperature data acquisition is set according to the hot-pressing process requirements.

[0086] Pressure data acquisition includes installing pressure sensors at the hot-pressing part of the urea-formaldehyde resin toilet seat. The pressure sensors are used to convert the measured pressure data into pressure electrical signals, which are then transmitted to the data acquisition card via a shielded cable.

[0087] Displacement data acquisition includes using a linear displacement sensor to measure the displacement of the mold during the hot pressing process of the urea-formaldehyde resin toilet seat, and monitoring the compression of the material. The linear displacement sensor is used to convert the output pulse signal into actual displacement data after processing by the data acquisition card.

[0088] S1.1: Transmitting the collected sensor data includes using a data acquisition card to transmit the collected temperature, pressure, and displacement data to the AI ​​computing unit via industrial Ethernet;

[0089] Preprocessing includes normalizing the collected temperature, pressure, and displacement data, mapping different data uniformly, and dividing them according to time order to generate structured multivariate time series data.

[0090] Further, temperature data acquisition: High-precision thermocouple temperature sensors are installed in key parts of the hot pressing mold, such as the mold cavity surface and the inside of the heating plate. These sensors measure the temperature at each point in real time, convert the temperature signal into an electrical signal, and transmit it to the data acquisition card through a shielded cable. The data acquisition frequency can be set according to the hot pressing process requirements, generally 1-10 times per second.

[0091] Pressure data acquisition: A pressure sensor is installed at the pressure application point of the hot press equipment. The pressure sensor converts the pressure signal into a standard electrical signal, which is then transmitted to the data acquisition card. To ensure the accuracy of the pressure data, the pressure sensor needs to be calibrated regularly, typically once a month.

[0092] Displacement data acquisition: A linear displacement sensor is used to measure the displacement of the mold during hot pressing to monitor the compression of the material. The pulse signal output by the displacement sensor is processed by the data acquisition card and converted into actual displacement data.

[0093] The data acquisition card transmits the collected temperature, pressure, displacement and other data to the AI ​​computing unit via industrial Ethernet. The transmission protocol uses Modbus TCP to ensure the stability and real-time performance of data transmission.

[0094] The collected raw data is normalized to map the data of different physical quantities to the [0,1] interval, thereby eliminating the influence of data units and improving the efficiency of model training. At the same time, the data is divided into training set, validation set and test set according to time order, with a ratio of 70%, 15% and 15% respectively.

[0095] S2: The collected sensor data is preprocessed and used as the raw input for building AI deep learning network models.

[0096] Among them, the construction of AI deep learning network models includes building two complementary deep learning models, namely, a temporal convolutional network model and a multi-scale representation convolutional network model.

[0097] The temporal convolutional network model employs a patched variable independent embedding method to independently encode multivariate time series data, extract the temporal features of the independently encoded data, and predict the output results.

[0098] The patched variable-independent embedding method involves processing the original input using Patchify+ variable-independent embedding, as shown in the formula:

[0099]

[0100] in, This represents the original input time series. This indicates an embedded mapping function. This indicates the embedded patch;

[0101] Independent encoding of multivariate time series data involves feeding embedded patches into a backbone network composed of stacked modern convolutional blocks, and designing modern convolutional blocks.

[0102] Modern convolutional blocks include depthwise separable convolutions, convolutional feedforward neural networks, convolutional feedforward neural networks, and those employing residual connections;

[0103] The modern convolutional blocks are propagated to calculate the global feature representation of the time series data, extract the independently encoded temporal features, and output the extracted temporal features based on the extracted temporal features.

[0104] The formula for propagating modern convolutional blocks is:

[0105]

[0106] in, Indicates the first Input and output representations of a modern convolutional block Represents modern convolutional block functions;

[0107] For the initial block of a modern convolutional block, the formula is:

[0108]

[0109] in, This represents the input and output representations of the first modern convolutional block;

[0110] Based on the initial block of the modern convolutional block, subsequent blocks are calculated recursively, using the following formula:

[0111]

[0112] Independent encoding involves independently processing the calculated global feature representation of the time series data. The calculated global feature representation of the time series is given by the following formula:

[0113]

[0114] in, This represents the global feature representation of the time series. This represents a backbone network composed of K modern convolutional blocks stacked together;

[0115] The predicted output includes mapping the independently encoded data to the target predicted result, which is then used to fit the process parameters during the hot-pressing molding of urea-formaldehyde resin toilet seats. The formula is as follows:

[0116] Linear Flatten

[0117] in, Flatten represents the predicted output. Linear represents a one-dimensional vector. This represents a fully connected linear mapping layer.

[0118] Furthermore, depthwise separable convolutions are responsible for extracting the temporal features within each variable; convolutional feedforward neural networks (ConvFFN1) are used to extract the local features of each variable; convolutional feedforward neural networks (ConvFFN2) learn the cross-channel dependencies between different variables; and residual connections are used to enhance gradient propagation and improve training stability.

[0119] Adaptive local pattern extraction: This module adaptively segments the input time series data, dynamically adjusting the position and scale of patches through learning to effectively preserve semantic structure. The input time series is first processed into... A fixed-size patch data set, where C is the number of variables (i.e., the number of channels) and T is the time step. This is the initial patch length. The center offset is calculated using the predictor, with the sliding step size as the prediction step. With scale change :

[0120]

[0121] in, Let g(x) represent the input patch, and g(x) represent the feature extraction function that extracts the feature map of the patch. This indicates a lightweight predictor (typically a convolutional projection). Indicates the center offset of the patch. This indicates the amount of change in the patch scale.

[0122] S2.1: The multi-scale representation convolutional network model includes the introduction of an adaptive local pattern extraction mechanism, dynamic patch adjustment, multi-stage fully convolutional blocks with different kernel sizes, combined with reparameterization mechanism and learnable residual connections to achieve multi-scale representation dependency modeling, and weighted fusion of the prediction results of the temporal convolutional network model and the multi-scale representation convolutional network model.

[0123] Dynamic patch adjustment involves adaptive segmentation of the global feature representation of the time series, dynamically adjusting the position and scale of the patches through learning, while preserving semantic structure. The formula for adjusting the position and scale of the patches is:

[0124]

[0125] in, Indicates the original center position. Indicates the location of the new patch center. Indicates the new patch length;

[0126] The patch boundary is calculated based on the adjusted position and scale of the patch, using the following formula:

[0127]

[0128] in, Indicates the left boundary of the patch. Indicates the right boundary of the patch;

[0129] Multi-scale representation dependency modeling includes introducing a reparameterization mechanism and introducing learnable residual connections to complete multi-scale representation dependency modeling;

[0130] The formula for the reparameterization mechanism is:

[0131]

[0132] in, Indicates the convolution kernel weights. Indicates the first Large kernel branches in a fully convolutional block Indicates the first Small kernel branches in a fully convolutional block Indicates the first The output of each convolutional block Indicates the first The output of the depthwise convolutional layer in a fully convolutional block Indicates the activation function;

[0133] The formula for learnable residual connectivity is:

[0134]

[0135] in, Indicates learnable weights, Indicates the first The output of depthwise convolution in each building block;

[0136] The prediction results of the temporal convolutional network model and the multi-scale representation convolutional network model are weighted and fused. After multi-scale representation dependency modeling, the results are mapped to the predicted values ​​of the target hot pressing process parameters. Then, a linear layer is introduced to obtain the prediction results of the temporal convolutional network model and the multi-scale representation convolutional network model. At this point, weighted fusion is performed.

[0137] The formula for introducing a linear layer is:

[0138]

[0139] in, This means that the final convolution output flattens out to a one-dimensional vector. This indicates the last fully convolutional stage. The weight matrix represents the linear transformation. Indicates bias. The output represents the prediction result;

[0140] The formula for weighted fusion is:

[0141]

[0142] in, Indicates the weighting coefficient. This represents the prediction result after weighted fusion.

[0143] S3: Based on the weighted fusion prediction results, a closed-loop intelligent control is implemented for the hot pressing molding process of urea-formaldehyde resin toilet seats, and a human-machine interface is built accordingly.

[0144] Among them, implementing closed-loop intelligent control for the hot pressing molding process of urea-formaldehyde resin toilet seats includes designing three key control strategies.

[0145] The design incorporates three key control strategies: temperature control, pressure control, and pressure holding time control.

[0146] Furthermore, regarding temperature control strategies, the AI ​​algorithm analyzes real-time temperature data collected at key locations of the mold and compares it with a preset hot-pressing temperature control curve to predict the temperature change trend at the next time step. Based on the time-series prediction capabilities of the deep learning network model, the system can accurately judge dynamic temperature fluctuations and respond in advance.

[0147] When the predicted temperature is lower than the lower edge of the target temperature curve, the AI ​​computing unit sends control commands to the heating system of the hot pressing equipment in real time to automatically increase the heating power.

[0148] When the predicted temperature is higher than the upper edge of the target temperature curve, the heating power should be reduced appropriately to avoid overheating.

[0149] The entire process achieves adaptive closed-loop control, which keeps the temperature stable within the target range. This control method can accurately maintain the hot pressing temperature within the set range, with a control accuracy of ±2℃, effectively improving the consistency of the resin curing reaction and the appearance quality of the finished product.

[0150] The pressure control strategy utilizes AI algorithms to model and predict pressure data collected in real time during the hot pressing process, and dynamically compares it with a preset pressure-time process curve. If the prediction results indicate a deviation trend in pressure, the system immediately takes adjustment measures.

[0151] In the initial stage of hot pressing, AI will dynamically adjust the pressure rise rate according to the material's density, thickness and flowability characteristics to ensure that the material is compacted slowly and evenly, preventing cracks or bubbles from forming.

[0152] During the constant pressure phase, if the pressure is found to be slightly higher or lower than the target value, the system will quickly correct it by adjusting the output pressure of the hydraulic pump, so that the pressure always remains within the stable range.

[0153] This strategy ensures uniform stress distribution during the hot pressing process, avoids product deformation and local underpressure, and achieves a control accuracy of ±0.5MPa, significantly improving product consistency and structural strength.

[0154] The pressure holding time control strategy differs from traditional hot pressing control, where the pressure holding time is typically set by operators based on experience. This solution utilizes an AI model to comprehensively analyze multiple real-time process parameters such as temperature, pressure, and displacement, and combines the curing characteristics of different materials with the quality indicators of the target finished product to intelligently determine the optimal pressure holding time.

[0155] When the system detects that key indicators such as temperature, pressure, and displacement have stabilized and reached the target range set by the process, and have been maintained for a set time, the AI ​​calculation unit triggers a signal to end the pressure holding phase.

[0156] After the pressure holding process is completed, the system automatically switches to the cooling process to ensure product dimensional stability and prevent stress cracking during subsequent cooling.

[0157] This intelligent pressure holding control method can be customized according to different raw materials and mold structures, with a control accuracy of ±5 seconds. It avoids excessive pressure holding and wastes energy, while ensuring sufficient curing reaction, ultimately improving product qualification rate and energy efficiency.

[0158] S3.1: The human-computer interaction interface is built to include the operation interface, monitoring interface, alarm interface, and historical data query and analysis interface;

[0159] The alarm interface includes prompts for the operation of the urea-formaldehyde resin toilet seat hot-pressing process, and includes three alarm levels: high, medium, and low.

[0160] Furthermore, an operation interface is constructed, serving as the system's main control entry point. This interface allows operators to flexibly configure hot-pressing process parameters according to the production requirements of different products. The interface employs a graphical design with a rational overall layout and clear module divisions, providing a superior human-computer interaction experience. Users can quickly set core process parameters such as hot-pressing temperature, applied pressure, holding time, and cooling rate through drop-down menus, slider adjustments, and numerical input boxes. A preset parameter template function is provided; for different batches or models of urea-formaldehyde resin toilet seat products, the system can automatically load historically optimal process schemes, improving setup efficiency. A configuration confirmation function allows operators to perform secondary confirmation before starting the hot-pressing process, preventing accidental operation. The interface is compatible with touch screens and mouse operation, offering high usability and flexibility.

[0161] A monitoring interface is constructed to display the real-time operating status and process parameter fluctuations of the hot pressing equipment, creating an intuitive and transparent process visualization window. It employs various graphical components such as line charts, bar charts, and dashboards to dynamically display real-time change curves of key process parameters such as current temperature, pressure, and displacement; displays the current status of the hot pressing equipment (e.g., heating, holding pressure, cooling, standby), and can track the execution flow of operation commands; supports multi-point curve comparison, facilitating operators to observe the coupling relationship between multiple variables, such as the coordinated change trend of temperature and pressure; and features zoom and drag functions, allowing for fine-grained observation and analysis of parameter fluctuations over any time period. This module provides operators with comprehensive control over the hot pressing process, helping to promptly detect anomalies and adjust strategies.

[0162] To ensure the safe and stable operation of the hot pressing process, an intelligent alarm interface is designed, capable of real-time sensing and proactive alerts. When the system detects excessively high or low temperatures, abnormal pressure fluctuations, sensor malfunctions, communication interruptions, or abnormal equipment shutdowns, an alarm window will pop up, accompanied by a flashing red indicator and a buzzer sound, alerting operators to take immediate action. Alarm information includes the alarm type, trigger time, affecting parameters, and suggested handling measures, presented clearly and concisely. All alarm events are automatically recorded in the background database, generating a complete alarm log for subsequent troubleshooting, maintenance management, and experience summarization. The system supports a tiered alarm mechanism, allowing for the setting of high, medium, and low alarm levels, corresponding to different severity levels and response strategies. This module significantly enhances the system's proactive defense capabilities and operational safety.

[0163] The system integrates powerful historical data management and analysis functions to achieve production process traceability and optimization. Users can filter hot-pressing historical data by time period, product batch, or operator, including process data such as temperature, pressure, and displacement, equipment operating status, process parameter settings, and corresponding product quality inspection results. It provides data visualization functions, such as trend charts, fluctuation analysis charts, and anomaly annotation charts, facilitating operators to identify potential patterns or optimization points. Data can be exported to Excel, CSV, and other formats, and daily, weekly, or batch reports can be automatically generated to help management understand production efficiency and process consistency. This module provides strong support for process optimization, product quality improvement, and enterprise digital transformation.

[0164] In a preferred embodiment, an AI-based intelligent control system for the hot-pressing molding of urea-formaldehyde resin toilet seats includes a sensor data acquisition and preprocessing module, which acquires sensor data and transmits and preprocesses the acquired sensor data; an AI deep learning network model construction module, which preprocesses the acquired sensor data and uses it as the original input for constructing the AI ​​deep learning network model; and a hot-pressing intelligent control and human-machine interaction module, which implements closed-loop intelligent control of the hot-pressing molding process of urea-formaldehyde resin toilet seats based on the weighted fusion prediction results and constructs a corresponding human-machine interaction interface.

[0165] The above-mentioned unit modules can be embedded in the processor of the computer device in hardware form or independent of it, or they can be stored in the memory of the computer device in software form, so that the processor can call and execute the corresponding operations of the above modules.

[0166] In one embodiment, a computer device is provided, which may be a terminal. The computer device includes a processor, memory, a communication interface, a display screen, and an input device connected via a system bus. The processor of the computer device provides computing and control capabilities. The memory of the computer device includes a non-volatile storage medium and internal memory. The non-volatile storage medium stores an operating system and computer programs. The internal memory provides an environment for the operation of the operating system and computer programs stored in the non-volatile storage medium. The communication interface of the computer device is used for wired or wireless communication with external terminals. Wireless communication can be achieved through Wi-Fi, carrier networks, NFC (Near Field Communication), or other technologies. The display screen of the computer device may be an LCD screen or an e-ink display screen. The input device of the computer device may be a touch layer covering the display screen, or buttons, a trackball, or a touchpad located on the casing of the computer device, or an external keyboard, touchpad, or mouse, etc.

[0167] In summary, this invention significantly improves the control accuracy, stability, and intelligence level of the urea-formaldehyde resin toilet seat hot-pressing process by constructing an integrated technology system encompassing sensor data acquisition and preprocessing, AI deep learning network model construction, intelligent control of hot-pressing molding, and human-machine interaction. Specifically, the patched variable independent embedding method effectively decouples multiple physical quantities such as temperature, pressure, and displacement, avoiding information confusion. Combining a structurally improved temporal convolutional network and an adaptive multi-scale representation convolutional network, and introducing a reparameterization mechanism, learnable residual connections, and a dual-model weighted fusion strategy, it significantly enhances the modeling ability and generalization performance for complex nonlinear temporal dynamics. Based on high-precision prediction results, closed-loop intelligent control of temperature, pressure, and holding time is implemented to ensure full resin curing, dense product structure, and consistent appearance. The accompanying human-machine interface enables flexible configuration of process parameters, real-time monitoring of process status, hierarchical alarms for abnormal events, and traceable analysis of historical data, greatly reducing reliance on manual intervention and improving production efficiency and product quality stability. The overall solution solves the problems of lag, insufficient accuracy, and difficulty in generalization inherent in traditional control methods, providing an efficient, reliable, and practical intelligent manufacturing solution for hot-pressing molding processes.

[0168] Reference Figure 3 and Figure 4 After introducing the method and system of exemplary embodiments of the present invention, the following references are made. Figure 3 A computer-readable storage medium according to exemplary embodiments of the present invention will be described, please refer to... Figure 3 The computer-readable storage medium shown is an optical disc 30, on which a computer program (i.e., a program product) is stored. When the computer program is run by a processor, it implements the steps described in the above method implementation, such as: collecting sensor data; transmitting and preprocessing the collected sensor data; using the preprocessed sensor data as the raw input for building an AI deep learning network model; and implementing closed-loop intelligent control of the urea-formaldehyde resin toilet seat hot-pressing molding process based on the weighted fusion prediction results, and constructing a corresponding human-machine interface. The specific implementation methods of each step will not be repeated here.

[0169] It should be noted that examples of computer-readable storage media may also include, but are not limited to, phase-change memory (PRAM), static random access memory (SRAM), dynamic random access memory (DRAM), other types of random access memory (RAM), read-only memory (ROM), electrically erasable programmable read-only memory (EEPROM), flash memory or other optical and magnetic storage media, which will not be elaborated here.

[0170] After introducing the methods and media of exemplary embodiments of the present invention, the following references are made. Figure 4A computational device for adaptive recovery of low-voltage power grid self-healing control according to an exemplary embodiment of the present invention.

[0171] Figure 4 A block diagram is shown of an exemplary computing device 40 suitable for implementing embodiments of the present invention. The computing device 40 may be a computer system or a server. Figure 4 The computing device 40 shown is merely an example and should not impose any limitation on the functionality and scope of use of the embodiments of the present invention.

[0172] like Figure 4 As shown, the components of computing device 40 may include, but are not limited to: one or more processors or processing units 401, system memory 402, and bus 403 connecting different system components (including system memory 402 and processing unit 401).

[0173] The computing device 40 typically includes a variety of computer system readable media. These media can be any available media that can be accessed by the computing device 40, including volatile and non-volatile media, and removable and non-removable media.

[0174] System memory 402 may include computer system readable media in the form of volatile memory, such as random access memory (RAM) 4021 and / or cache memory 4022. Computing device 40 may further include other removable / non-removable, volatile / non-volatile computer system storage media. By way of example only, ROM 4023 may be used to read and write non-removable, non-volatile magnetic media (…). Figure 4 (Not shown in the image, usually referred to as "hard drive"). Although not shown in... Figure 4 The diagram illustrates that disk drives for reading and writing to removable non-volatile disks (e.g., "floppy disks") and optical disc drives for reading and writing to removable non-volatile optical discs (e.g., CD-ROMs, DVD-ROMs, or other optical media) can be provided. In these cases, each drive can be connected to bus 403 via one or more data media interfaces. System memory 402 may include at least one program product having a set (e.g., at least one) of program modules configured to perform the functions of the embodiments of the present invention.

[0175] A program / utility 4025 having a set (at least one) of program modules 4024 may be stored, for example, in system memory 402, and such program modules 4024 include, but are not limited to, an operating system, one or more application programs, other program modules, and program data, each or some combination of these examples may include an implementation of a network environment. Program modules 4024 typically perform the functions and / or methods described in the embodiments of the present invention.

[0176] The computing device 40 can also communicate with one or more external devices 404 (such as a keyboard, pointing device, display, etc.). This communication can be performed via the input / output (I / O) interface 405. Furthermore, the computing device 40 can also communicate with one or more networks (e.g., local area networks (LANs), wide area networks (WANs), and / or public networks, such as the Internet) via a network adapter 406. Figure 4 As shown, network adapter 406 communicates with other modules of computing device 40 (such as processing unit 401) via bus 403. It should be understood that, although... Figure 4 As not shown, it can be used in conjunction with computing device 40 with other hardware and / or software modules.

[0177] The processing unit 401 executes various functional applications and data processing by running programs stored in the system memory 402. For example, it collects sensor data, transmits and preprocesses the collected sensor data; after preprocessing the collected sensor data, it uses it as the original input basis for building an AI deep learning network model; based on the prediction results after weighted fusion, it implements closed-loop intelligent control of the hot pressing molding process of urea-formaldehyde resin toilet seat, and builds a human-machine interaction interface accordingly.

[0178] Those skilled in the art will understand that, for the sake of convenience and brevity, the specific working processes of the systems, devices, and units described above can be referred to the corresponding processes in the foregoing method embodiments, and will not be repeated here.

[0179] In the several embodiments provided by this invention, it should be understood that the disclosed systems, apparatuses, and methods can be implemented in other ways. The apparatus embodiments described above are merely illustrative. For example, the division of units is only a logical functional division, and in actual implementation, there may be other division methods. Furthermore, multiple units or components may be combined or integrated into another system, or some features may be ignored or not executed. Additionally, the coupling or direct coupling or communication connection shown or discussed may be through some communication interface; the indirect coupling or communication connection between apparatuses or units may be electrical, mechanical, or other forms.

[0180] The units described as separate components may or may not be physically separate. The components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the units can be selected to achieve the purpose of this embodiment according to actual needs.

[0181] In addition, the functional units in the various embodiments of the present invention can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit.

[0182] If the functionality is implemented as a software functional unit and sold or used as an independent product, it can be stored in a processor-executable, non-volatile, computer-readable storage medium. Based on this understanding, the technical solution of this invention, or the part that contributes to the prior art, or a portion of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods of the various embodiments of this invention. The aforementioned storage medium includes various media capable of storing program code, such as USB flash drives, portable hard drives, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks.

[0183] Finally, it should be noted that the above embodiments are merely specific implementations of the present invention, used to illustrate the technical solutions of the present invention, and not to limit it. The scope of protection of the present invention is not limited thereto. Although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that any person skilled in the art can still modify or easily conceive of changes to the technical solutions described in the foregoing embodiments within the technical scope disclosed in the present invention, or make equivalent substitutions for some of the technical features; and these modifications, changes, or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of the present invention, and should all be covered within the scope of protection of the present invention. Therefore, the scope of protection of the present invention should be determined by the scope of the claims.

[0184] Furthermore, although the operations of the method of the present invention are described in a specific order in the accompanying drawings, this does not require or imply that these operations must be performed in that specific order, or that all the operations shown must be performed to achieve the desired result. Additionally or alternatively, certain steps may be omitted, multiple steps may be combined into one step, and / or one step may be broken down into multiple steps.

[0185] It should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and are not intended to limit it. Although the present invention has been described in detail with reference to preferred embodiments, those skilled in the art should understand that modifications or equivalent substitutions can be made to the technical solutions of the present invention without departing from the spirit and scope of the technical solutions of the present invention, and all such modifications or substitutions should be covered within the scope of the claims of the present invention.

Claims

1. An intelligent control method for thermoforming urea-formaldehyde resin toilet seats based on AI algorithms, characterized in that: include, Sensor data is collected, and the collected sensor data is transmitted and preprocessed. The collected sensor data is preprocessed and then used as the raw input for building AI deep learning network models. Based on the weighted fusion prediction results, a closed-loop intelligent control is implemented for the hot pressing molding process of urea-formaldehyde resin toilet seats, and a human-machine interface is constructed accordingly.

2. The intelligent control method for hot pressing molding of urea-formaldehyde resin toilet seats based on AI algorithms as described in claim 1, characterized in that: The acquisition of sensor data includes temperature data acquisition, pressure data acquisition, and displacement data acquisition; The temperature data acquisition includes arranging thermocouple temperature sensors at the hot-pressing part of the urea-formaldehyde resin toilet seat. The thermocouple temperature sensors are used to convert the measured temperature data into temperature electrical signals, which are transmitted to the data acquisition card through a shielded cable. The frequency of the temperature data acquisition is set according to the hot-pressing process requirements. The pressure data acquisition includes installing a pressure sensor at the hot-pressing part of the urea-formaldehyde resin toilet seat. The pressure sensor is used to convert the measured pressure data into a pressure electrical signal, which is then transmitted to the data acquisition card via a shielded cable. The displacement data acquisition includes using a linear displacement sensor to measure the displacement of the mold during the hot pressing process of the urea-formaldehyde resin toilet seat, and monitoring the compression of the material. The linear displacement sensor is used to convert the output pulse signal into actual displacement data after processing by the data acquisition card.

3. The intelligent control method for hot pressing molding of urea-formaldehyde resin toilet seats based on AI algorithms as described in claim 2, characterized in that: The transmission of the collected sensor data includes using a data acquisition card to transmit the collected temperature data, pressure data, and displacement data to the AI ​​computing unit via an industrial Ethernet network. The preprocessing includes normalizing the collected temperature, pressure, and displacement data, mapping different data uniformly, and dividing them according to time order to generate structured multivariate time series data.

4. The intelligent control method for hot pressing molding of urea-formaldehyde resin toilet seats based on AI algorithms as described in claim 3, characterized in that: The construction of the AI ​​deep learning network model includes building two complementary deep learning models, namely a temporal convolutional network model and a multi-scale representation convolutional network model. The temporal convolutional network model includes a patched variable independent embedding method to independently encode multivariate time series data, extract the temporal features of the independently encoded data, and predict the output results. The patched variable independent embedding method includes processing the original input using the Patchify+ variable independent embedding method, as shown in the formula: in, This represents the original input time series. This indicates an embedded mapping function. This indicates the embedded patch; The independent encoding of multivariate time series data includes inputting the embedded patch into a backbone network composed of stacked modern convolutional blocks, and designing modern convolutional blocks; The modern convolutional blocks include depthwise separable convolutions, convolutional feedforward neural networks, convolutional feedforward neural networks, and those employing residual connections. The modern convolutional blocks are propagated to calculate the global feature representation of the time series data, extract the independently encoded temporal features, and output the extracted temporal features based on the extracted temporal features. The calculation formula for propagating the modern convolutional block is as follows: in, Indicates the first Input and output representations of a modern convolutional block Represents modern convolutional block functions; For the initial block of a modern convolutional block, the formula is: in, This represents the input and output representations of the first modern convolutional block; Based on the initial block of the modern convolutional block, subsequent blocks are calculated recursively, using the following formula: The independent encoding includes independently processing the calculated global feature representation of the time series data. The calculated global feature representation of the time series is given by the following formula: in, This represents the global feature representation of the time series. This represents a backbone network composed of K modern convolutional blocks stacked together; The predicted output includes mapping the result to a target prediction after independent encoding, which is then used to fit the process parameters during the hot pressing process of urea-formaldehyde resin toilet seat. The formula is as follows: Linear Flatten in, Flatten represents the predicted output. Linear represents a one-dimensional vector. This represents a fully connected linear mapping layer.

5. The intelligent control method for hot pressing molding of urea-formaldehyde resin toilet seats based on AI algorithms as described in claim 4, characterized in that: The multi-scale representation convolutional network model includes an adaptive local pattern extraction mechanism, dynamic patch adjustment, multi-stage fully convolutional blocks with different kernel sizes, and a reparameterization mechanism combined with learnable residual connections to achieve multi-scale representation dependency modeling. The prediction results of the temporal convolutional network model and the multi-scale representation convolutional network model are weighted and fused. The dynamic patch adjustment involves adaptive segmentation of the global feature representation of the time series, dynamically adjusting the position and scale of the patch through a learning method, preserving the semantic structure. The formula for adjusting the position and scale of the patch is as follows: in, Indicates the original center position. Indicates the location of the new patch center. Indicates the new patch length; The patch boundary is calculated based on the adjusted position and scale of the patch, using the following formula: in, Indicates the left boundary of the patch. Indicates the right boundary of the patch; The multi-scale representation dependency modeling includes introducing a reparameterization mechanism and introducing learnable residual connections to complete the multi-scale representation dependency modeling. The formula for the reparameterization mechanism is: in, Indicates the convolution kernel weights. Indicates the first Large kernel branches in a fully convolutional block Indicates the first Small kernel branches in a fully convolutional block Indicates the first The output of each convolutional block Indicates the first The output of the depthwise convolutional layer in a fully convolutional block Indicates the activation function; The formula for the learnable residual connection is: in, Indicates learnable weights, Indicates the first The output of depthwise convolution in each building block; The weighted fusion of the prediction results from the temporal convolutional network model and the multi-scale representation convolutional network model includes: after multi-scale representation dependency modeling, mapping the results to predicted values ​​of the target hot-pressing process parameters, then introducing a linear layer to obtain the prediction results from the temporal convolutional network model and the multi-scale representation convolutional network model, and then performing weighted fusion. The formula for introducing the linear layer is: in, This means that the final convolution output flattens out to a one-dimensional vector. This indicates the last fully convolutional stage. The weight matrix represents the linear transformation. Indicates bias. The output represents the prediction result; The formula for the weighted fusion is: in, Indicates the weighting coefficient. This represents the prediction result after weighted fusion.

6. The intelligent control method for hot pressing molding of urea-formaldehyde resin toilet seats based on AI algorithms as described in claim 5, characterized in that: The implementation of closed-loop intelligent control for the hot pressing molding process of urea-formaldehyde resin toilet seats includes the design of three key control strategies. The design employs three key control strategies: temperature control, pressure control, and pressure holding time control.

7. The intelligent control method for hot pressing molding of urea-formaldehyde resin toilet seats based on AI algorithms as described in claim 6, characterized in that: The supporting human-computer interaction interface includes an operation interface, a monitoring interface, an alarm interface, and a historical data query and analysis interface. The alarm interface includes prompts for the operation of the urea-formaldehyde resin toilet seat hot pressing process, and the alarm interface includes three alarm levels: high, medium and low.

8. An intelligent control system for the hot pressing molding of urea-formaldehyde resin toilet seats based on AI algorithms, based on the intelligent control method for the hot pressing molding of urea-formaldehyde resin toilet seats based on AI algorithms as described in any one of claims 1 to 7, characterized in that: include, The sensor data acquisition and preprocessing module acquires sensor data and performs transmission and preprocessing on the acquired sensor data. The AI ​​deep learning network model building module preprocesses the collected sensor data and uses it as the raw input for building the AI ​​deep learning network model. The intelligent control and human-machine interaction module for hot pressing molding implements closed-loop intelligent control of the hot pressing molding process of urea-formaldehyde resin toilet seats based on the weighted fusion prediction results, and also builds a human-machine interaction interface.

9. A computer device comprising a memory and a processor, wherein the memory stores a computer program, characterized in that: When the processor executes the computer program, it implements the steps of the intelligent control method for hot pressing molding of urea-formaldehyde resin toilet seats based on AI algorithms as described in any one of claims 1 to 7.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by the processor, it implements the steps of the intelligent control method for hot pressing molding of urea-formaldehyde resin toilet seats based on AI algorithms as described in any one of claims 1 to 7.

Images