Polypropylene film production control method and equipment based on temperature and thickness and medium

Abstract

The embodiment of the invention provides a polypropylene film production control method and equipment based on temperature and thickness and a medium. The method comprises the following steps: acquiring a temperature value of a polypropylene film through a temperature sensor; obtaining the thickness value of the polypropylene film through a thickness sensor; inputting the temperature value and the thickness value into a control model based on a bidirectional long short-term memory network and a multi-head attention mechanism to obtain control parameters of a PI D temperature control system, an SMC longitudinal drawing system and an SMC transverse drawing system of the polypropylene film production line; the PI D temperature control system, the SMC longitudinal drawing system and the SMC transverse drawing system are controlled to operate according to the control parameters; the temperature and the thickness of the polypropylene film are measured in real time without deviation, and factors influencing the production of the polypropylene film are detected in real time; for factor correlation parameter prediction, historical influence factors and current factors are connected in series by using a bidirectional long-short-term memory network, and a multi-head attention mechanism is combined to enable the model to pay attention to the parameter with the maximum influence, so that the calculation complexity is reduced, and the prediction accuracy and robustness are improved.

Description

Technical Field

[0001] The embodiments of the present invention relate to, but are not limited to, the field of data measurement, and in particular to a method, device, and medium for controlling the production of polypropylene films based on temperature and thickness. Background Art

[0002] During the production of polypropylene film, the thickness uniformity of the film directly impacts product quality and performance. Traditional detection technologies rely heavily on imported sensors and control systems, which present challenges such as high cost, poor resistance to environmental interference, and insufficient measurement accuracy. Existing ultrasonic thickness measurement methods are susceptible to fluctuations in sound velocity caused by temperature changes, requiring frequent manual calibration and resulting in low efficiency. Sensor signals are also susceptible to interference from circuit noise, making it difficult to achieve unbiased measurement. Regarding data analysis, existing models often utilize a single algorithm, which inadequately explores the spatiotemporal correlation between temperature and thickness data, making it impossible to accurately predict the risk of film rupture. Summary of the Invention

[0003] The following is a summary of the subject matter described in detail herein. This summary is not intended to limit the scope of the claims.

[0004] The embodiments of the present application provide a method, device, and medium for controlling the production of polypropylene film based on temperature and thickness, which enables the model to pay more attention to the parameters that have the greatest impact on the current temperature and thickness, thereby improving prediction accuracy and robustness.

[0005] An embodiment of the first aspect of the present application is a method for controlling polypropylene film production based on temperature and thickness, comprising:

[0006] The temperature value of the polypropylene film is obtained by a temperature sensor;

[0007] Obtain the thickness value of the polypropylene film through the thickness sensor;

[0008] Inputting the temperature value and the thickness value into a control model to obtain control parameters of a PID temperature control system, an SMC longitudinal stretching system, and an SMC transverse stretching system of a polypropylene film production line;

[0009] Control the operation of the PID temperature control system, the SMC longitudinal pulling system and the SMC transverse pulling system according to the control parameters;

[0010] Among them, the control model is based on a bidirectional long short-term memory network and a multi-head attention mechanism.

[0011] According to certain embodiments of the first aspect of the present application, the temperature sensor is a 4H-SiC vertical double diffused metal oxide semiconductor enhancement mode field effect transistor with a lateral Schottky barrier diode integrated into a class AB dual power supply complementary symmetrical power amplifier circuit in a p-type well region.

[0012] According to certain embodiments of the first aspect of the present application, when the temperature sensor is connected to a power supply, the Class AB dual-power complementary symmetrical power amplifier circuit controls the voltage across itself by regulating its own resistance, providing a current bias between the push-pull tube; when the temperature changes, the thermal motion energy of electrons changes, exciting electrons to cross the potential barrier of the lateral Schottky barrier diode, generating a current increment in the temperature sensor; obtaining the temperature value of the polypropylene film through the temperature sensor includes: determining the current value of the temperature sensor based on the current increment; obtaining the voltage value of the temperature sensor based on the current value of the temperature sensor; and obtaining the temperature value of the polypropylene film based on the voltage value of the temperature sensor.

[0013] According to certain embodiments of the first aspect of the present application, the current increment is expressed as: The voltage value of the temperature sensor is determined according to the following formula: Where ΔI is the current increment, Δx is the length increment, and W S is the Schottky contact width, A * is the Richardson constant, T is the detection temperature, q is the charge, φ B is the barrier height, η is the ideality factor of the Schottky contact, is the Boltzmann constant, V S is the potential of the lateral Schottky barrier diode at the Schottky contact, V(x) is the potential in the x direction, I T is the current value of the temperature sensor, B is the proportional coefficient, V TS is the voltage value of the temperature sensor, V R is the interface voltage drop.

[0014] According to certain embodiments of the first aspect of the present application, the thickness sensor includes a control circuit, a transmitting circuit, a signal processing circuit, a receiving circuit, a Fourier transform circuit, a phase detection circuit, an ultrasonic generating probe and an ultrasonic receiving probe; the output end of the control circuit is connected to the input end of the transmitting circuit, the output end of the transmitting circuit is connected to the ultrasonic generating probe, the ultrasonic receiving probe is connected to the input end of the receiving circuit, one output end of the receiving circuit is connected to the input end of the signal processing circuit, the output end of the signal processing circuit is connected to one input end of the control circuit, the other output end of the receiving circuit is connected to the input end of the phase detection circuit, the output end of the phase detection circuit is connected to the input end of the Fourier transform circuit, and the output end of the Fourier transform circuit is connected to the other input end of the control circuit.

[0015] According to certain embodiments of the first aspect of the present application, the transmitting circuit includes a pulse signal generating circuit and a switch control circuit; the receiving circuit includes a pre-processing circuit and a secondary proportional amplification circuit; the signal processing circuit includes a full-wave rectifier circuit, an envelope detection circuit and an over-peak judgment circuit.

[0016] According to certain embodiments of the first aspect of the present application, when the thickness sensor is powered on, the switch control circuit enables the pulse signal generating circuit to generate a square wave pulse, and the ultrasonic generating probe generates an ultrasonic wave according to the square wave pulse; the ultrasonic receiving probe receives the ultrasonic wave that penetrates the film to be measured, the pre-processing circuit filters the ultrasonic wave received by the ultrasonic receiving probe, and the secondary proportional amplifier circuit amplifies the filtered signal; the full-wave rectifier circuit converts the amplified signal into a unipolar signal, retains the complete energy distribution and outputs a positive envelope; the envelope detection circuit filters out the high-frequency component of the carrier from the positive envelope and extracts the amplitude attenuation data; the over-peak judgment circuit detects the envelope peak moment, generates a falling edge pulse to trigger the control circuit to interrupt, marks the signal arrival time, and calculates the thickness value by the time difference between the signal arrival time and the square wave pulse generation time.

[0017] According to certain embodiments of the first aspect of the present application, the control circuit corrects the thickness value according to a correction formula, and the correction formula is expressed as: Where d is the thickness value, Δφ is the phase difference, v(T) is the ultrasonic velocity at time T, f is the ultrasonic frequency, k2 is the absorption compensation factor used to correct the thickness fluctuation, Z1 is the ultrasonic impedance, ω is the ultrasonic angular frequency, V ref is the reference voltage of the film being tested, V mat is the material equivalent voltage of the film being tested.

[0018] An embodiment of the second aspect of the present application is an electronic device, comprising: a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein when the processor executes the computer program, the temperature and thickness-based polypropylene film production control method as described in the embodiment of the first aspect of the present application is implemented.

[0019] An embodiment of the third aspect of the present application is a computer storage medium storing computer executable instructions, wherein the computer executable instructions are used to execute the temperature and thickness-based polypropylene film production control method as described in the embodiment of the first aspect of the present application.

[0020] The above scheme has at least the following beneficial effects: obtaining the temperature value of the polypropylene film through a temperature sensor; obtaining the thickness value of the polypropylene film through a thickness sensor; inputting the temperature value and the thickness value into a control model based on a bidirectional long short-term memory network and a multi-head attention mechanism to obtain the control parameters of the PID temperature control system, SMC longitudinal stretching system and SMC transverse stretching system of the polypropylene film production line; controlling the operation of the PID temperature control system, SMC longitudinal stretching system and SMC transverse stretching system according to the control parameters; measuring the temperature and thickness of the polypropylene film in real time without deviation, and detecting the factors affecting the production of the polypropylene film in real time; for the prediction of factor-related parameters, using a bidirectional long short-term memory network to connect historical influencing factors and current factors in series, and then combining the multi-head attention mechanism to make the model focus on the parameters with the greatest influence, thereby reducing the computational complexity and improving the prediction accuracy and robustness.

[0021] Additional aspects and advantages of the present invention will be set forth in part in the description which follows and, in part, will be obvious from the description which follows, or may be learned by practice of the present invention. BRIEF DESCRIPTION OF THE DRAWINGS

[0022] The accompanying drawings are used to provide a further understanding of the technical solution of the present invention and constitute a part of the specification. Together with the embodiments of the present invention, they are used to explain the technical solution of the present invention and do not constitute a limitation to the technical solution of the present invention.

[0023] Figure 1 It is a step diagram of the polypropylene film production control method;

[0024] Figure 2 It is a schematic diagram of the operating system environment of a polypropylene film production line;

[0025] Figure 3 It is the structural diagram of the temperature sensor;

[0026] Figure 4 It is the structural diagram of the thickness sensor;

[0027] Figure 5 is a schematic diagram of the control model;

[0028] Figure 6 This is the structural diagram of the LSTM layer;

[0029] Figure 7 This is the schematic diagram of the PID control algorithm. DETAILED DESCRIPTION

[0030] In order to make the purpose, technical solutions and advantages of the present invention more clearly understood, the present invention will be further described in detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present invention and are not intended to limit the present invention.

[0031] It should be noted that although the device schematics illustrate functional module divisions and the flowcharts illustrate logical sequences, in certain circumstances, the steps shown or described may be performed in a sequence that differs from the module divisions in the device or the sequence in the flowcharts. The terms "first," "second," and the like in the specification, claims, or accompanying drawings are used to distinguish similar items and are not necessarily used to describe a specific sequence or precedence.

[0032] The invention provides a polypropylene film production control method based on temperature and thickness.

[0033] Reference Figure 1 , a polypropylene film production control method comprises the following steps:

[0034] Step S100, obtaining the temperature value of the polypropylene film through a temperature sensor;

[0035] Step S200, obtaining the thickness value of the polypropylene film through a thickness sensor;

[0036] Step S300, inputting the temperature value and the thickness value into a control model to obtain control parameters of the PID temperature control system, the SMC longitudinal stretching system, and the SMC transverse stretching system of the polypropylene film production line;

[0037] Step S400, controlling the operation of the PID temperature control system, the SMC longitudinal pulling system, and the SMC transverse pulling system according to the control parameters;

[0038] Among them, the control model is based on a bidirectional long short-term memory network and a multi-head attention mechanism.

[0039] Reference Figure 2 The operating system environment of the polypropylene film production line uses Huawei 910 graphics cards, Kunpeng 920 CPUs, and Yangtze Memory's Xtacking, and enables necessary acceleration features in the BIOS / UEFI settings. A docker image with a pre-installed software stack provided by Cambricon is used to create a docker container. Cambricon Pytorch, CNToolkit, MagicMind, and MagicMind Python are installed in this environment to ensure compatibility between hardware and ecology. Docker is then deployed on Hongmeng OS, and the training environment is migrated. Next, the deep learning framework Baidu PaddlePaddle is installed, and finally the data processing method TDengine is installed and the parameters are configured.

[0040] The microcontroller is responsible for initialization, booting, control, and calling. The operating system, HarmonyOS, transmits instructions to the graphics card and assigns tasks to the CPU. The CPU informs the memory of data storage requirements, and the memory responds with data read information. The memory then sends this displayed data read information to the graphics card. The CPU submits data coordination requests, task status feedback, and calculation results to the operating system. The data processing TDengine feeds data back to the operating system. The deep learning framework implements control information exchange, parameter aggregation processing requests, and computing task feedback, and the operating system provides data support.

[0041] The temperature of the longitudinal and transverse stretching systems on the production line is measured using a bias-free sensor, which is then converted into an electrical signal through the sensor's circuit. The electrical signal generated by the environment and the machine is filtered out through a filter circuit. The actual temperature of the BOPP film is then converted using a temperature and electrical signal conversion table. The ultrasonic thickness measurement module then drives dual probes through the film in a non-contact orthogonal manner using high voltage. The signal undergoes debiasing, amplification, full-wave rectification, and envelope detection to extract features. The AD8302 phase-locked technology is combined with FFT frequency domain analysis, the MCU integrates the SHT35 to dynamically correct the sound velocity, and Kalman filtering is used to suppress noise, outputting the dual parameters of thickness and surface density.

[0042] Reference Figure 3 As for the temperature sensor, the temperature sensor is a 4H-SiC vertical double diffused metal oxide semiconductor enhancement mode field effect transistor with a lateral Schottky barrier diode integrated into a class AB dual power supply complementary symmetrical power amplifier circuit in a p-type well region.

[0043] The temperature sensor circuit is equipped with a power-type 4H-SiC vertical double-diffused metal oxide semiconductor enhanced field-effect transistor. Its lower on-resistance can reduce power loss and improve working efficiency in high current conditions. The field-effect transistor requires a specific gate voltage to turn on, which makes it more controllable during the switching process; the switching characteristics of the field-effect transistor are more precise and can provide more stable performance. In the p-well area of ​​the sensor, a lateral SBD is used to integrate a Class A and B dual-power complementary symmetrical power amplifier circuit, effectively utilizing the p-well area to provide shielding protection for the sensor, suppressing the electrothermal crosstalk during the field-effect transistor switching process, and by integrating this module in the p-well area, the lateral distribution effect is utilized to allow further miniaturization of the sensor, and the diode's turn-on voltage can be eliminated through the power amplifier circuit. It should be noted that electrothermal crosstalk is the phenomenon of transient fluctuations in the output value of the sensor when the switch is turned on or off. In order to solve the problem of quantitatively analyzing the influence of the lateral current distribution effect and interface state on carrier migration, a lateral distribution Schottky model is proposed. By introducing lateral distribution parameters, the nonlinearity caused by excessive edge current density is reduced and the linear R is improved. 2; Establish an interface state resistance model to eliminate the temperature drift of the barrier height and improve sensitivity stability.

[0044] A lateral Schottky barrier diode is designed within the P-type well region of a power-type 4H-SiC vertical double-diffused metal oxide semiconductor enhancement mode field-effect transistor (EFMT) to integrate a Class AB dual-supply complementary symmetrical power amplifier circuit. Traditional PN junction diodes have high reverse recovery charge, leading to significant switching losses and electromagnetic interference. When powered, lateral Schottky barrier diodes exhibit no minority carrier storage effect and virtually no reverse recovery charge, significantly reducing reverse recovery time and increasing switching speed, thereby improving FET efficiency. Integrating a lateral Schottky barrier diode within the P-type well optimizes the electric field distribution and reduces local electric field peaks, thereby enhancing the device's withstand voltage and long-term reliability. A weak voltage exists across the Class AB dual-supply complementary symmetrical power amplifier circuit, and the voltage across it can be controlled by regulating its resistance. By providing a low current bias between the push-pull transistors, crossover distortion is reduced, signal fidelity is improved, and the Class AB dual-supply complementary symmetrical power amplifier circuit is suitable for diodes requiring a high turn-on voltage. Because the voltage across the diodes is directly connected to the lateral Schottky barrier diode, the Class AB dual-supply complementary symmetrical power amplifier circuit can respond quickly even at high or low temperatures.

[0045] Obtaining the temperature value of the polypropylene film through the temperature sensor includes the following steps: determining the current value of the temperature sensor according to the current increment; obtaining the voltage value of the temperature sensor according to the current value of the temperature sensor; and obtaining the temperature value of the polypropylene film according to the voltage value of the temperature sensor.

[0046] When the temperature changes, the energy of electron thermal motion changes, which stimulates more electrons to cross the potential barrier because the electrons need to overcome the potential barrier height. Only then can we move from semiconductors to metals.

[0047] The current of the temperature sensor increases. Due to the influence of the interface density, the carrier I T When passing through the interface, it is subjected to Coulomb scattering, which produces a resistance effect. According to the temperature measurement equation of the sensor, I T 、V TS , the relationship between T.

[0048] The voltage value of the temperature sensor is determined by the following formula:

[0049] Where ΔI is the current increment, Δx is the length increment, and W S is the Schottky contact width, A * is the Richardson constant, T is the detection temperature, q is the charge, φ B is the barrier height, η is the ideality factor of the Schottky contact, is the Boltzmann constant, VS is the potential of the lateral Schottky barrier diode at the Schottky contact, V(x) is the potential in the x direction, I T is the current value of the temperature sensor, B is the proportional coefficient, V TS is the voltage value of the temperature sensor, V R is the interface voltage drop.

[0050] The inferred V TS A noisy electrical signal is input into the filter circuit. The electrical signal passes through a frequency-selective network composed of components such as resistors, capacitors, inductors, and operational amplifiers. This network processes the input signal in the frequency domain based on a preset transfer function, utilizing the passband and stopband division of its amplitude-frequency characteristics to selectively attenuate high-frequency noise or low-frequency interference (electrical signals generated by ambient temperature) in the signal that exceeds the useful frequency band. Simultaneously, in the time domain, the impulse response of the circuit system is convolved with the input signal to suppress the transient interference component of the time-varying noise. At this time, the linear phase characteristic can reduce signal distortion, and ultimately, a target signal with a significantly improved signal-to-noise ratio is obtained at the output of the impedance matching network.

[0051] Finally, the temperature value detected by the temperature sensor is calculated according to the following formula:

[0052] Reference Figure 4 , for the thickness sensor, the thickness sensor includes a control circuit, a transmitting circuit, a signal processing circuit, a receiving circuit, a Fourier transform circuit, a phase detection circuit, an ultrasonic generating probe and an ultrasonic receiving probe; the output end of the control circuit is connected to the input end of the transmitting circuit, the output end of the transmitting circuit is connected to the ultrasonic generating probe, the ultrasonic receiving probe is connected to the input end of the receiving circuit, one output end of the receiving circuit is connected to the input end of the signal processing circuit, the output end of the signal processing circuit is connected to one input end of the control circuit, the other output end of the receiving circuit is connected to the input end of the phase detection circuit, the output end of the phase detection circuit is connected to the input end of the Fourier transform circuit, and the output end of the Fourier transform circuit is connected to the other input end of the control circuit.

[0053] The thickness sensor's circuitry utilizes a multi-band ultrasonic transducer, enabling dynamic adaptive matching. Combined with a non-contact, vertical dual-probe layout and mechanically adjustable spacing, this ensures orthogonal sound wave penetration through the polypropylene film, ultimately minimizing signal scattering loss to less than 5%. Furthermore, the core circuit module employs a dual-channel parallel architecture for synchronous signal processing. Channel 1 utilizes full-wave rectification and envelope detection to extract amplitude attenuation data, while channel 2 utilizes the ADuC7026 chip for FFT frequency-domain analysis. This, combined with a wideband data fusion algorithm, enhances thin material inspection accuracy. To account for the impact of temperature on thickness measurement, an SHT35 temperature and humidity sensor is integrated into the control circuit MCU. The sensor dynamically corrects sound velocity using the mass law transfer function and real-time environmental parameters (temperature, humidity) to eliminate temperature drift errors. This modular design enables calibration-free and synchronized output of thickness and surface density. Combined with Kalman filtering to suppress circuit noise, this achieves a high defect detection rate and reduces overall costs in high-speed production lines, eliminating the reliance of traditional thickness measurement technologies on sound velocity calibration and manual intervention.

[0054] The transmitting circuit includes a pulse signal generating circuit and a switch control circuit; the receiving circuit includes a pre-processing circuit and a secondary proportional amplification circuit; the signal processing circuit includes a full-wave rectification circuit, an envelope detection circuit and an over-peak judgment circuit.

[0055] When the thickness sensor is powered on, the switch control circuit causes the pulse signal generating circuit to generate a square wave pulse, and the ultrasonic generating probe generates an ultrasonic wave according to the square wave pulse; the ultrasonic receiving probe receives the ultrasonic wave that penetrates the film to be measured, the pre-processing circuit filters the ultrasonic wave received by the ultrasonic receiving probe, and the secondary proportional amplifier circuit amplifies the filtered signal; the full-wave rectifier circuit converts the amplified signal into a unipolar signal, retains the complete energy distribution and outputs a positive envelope; the envelope detection circuit filters out the high-frequency component of the carrier from the positive envelope and extracts the amplitude attenuation data; the over-peak judgment circuit detects the envelope peak moment, generates a falling edge pulse to trigger the control circuit interrupt, marks the signal arrival time, and calculates the thickness value by the time difference between the signal arrival time and the square wave pulse generation time.

[0056] Specifically, when the power is turned on, a high-voltage square wave pulse is generated in the transmitting circuit by a switch control circuit based on the NOT gate chip 74HC04 and the driver chip MC34151P. This is then driven by the MOS transistor IRF840 to drive the ultrasonic probe to transmit the ultrasonic wave. After the ultrasonic wave penetrates the film being measured, the receiving probe captures the attenuated signal. After capturing the signal, the pre-processing circuit eliminates the DC bias through capacitors and resistors and uses diode limiting to suppress ambient noise. The signal then enters the secondary proportional amplifier circuit, which boosts the microvolt-level signal to the volt level. The amplified signal is converted to a unipolar signal by the full-wave rectifier circuit, preserving the complete energy distribution and outputting a positive envelope. At this point, the envelope detection circuit filters out the high-frequency components of the carrier and extracts the amplitude attenuation data, which provides a key parameter for surface density calculation. Channel 1 then detects the envelope peak moment through the over-peak detection circuit, generating a falling-edge pulse that triggers an MCU interrupt, accurately marking the signal arrival time. The initial thickness value is then calculated based on the time difference. At the same time, the FFT chip of channel 2 performs FFT transformation on the time domain signal and outputs the amplitude spectrum and phase spectrum. At the same time, the AD8302 phase detection module extracts the phase difference between the transmitted and received signals in real time.

[0057] The MCU control circuit integrates the temperature and humidity sensor SHT35 to dynamically correct and update the thickness d according to the ambient temperature and ambient sound speed v(T).

[0058] The correction formula is expressed as: Where d is the thickness value, Δφ is the phase difference, v(T) is the ultrasonic velocity at time T, f is the ultrasonic frequency, k2 is the absorption compensation factor used to correct the thickness fluctuation, Z1 is the ultrasonic impedance, ω is the ultrasonic angular frequency, V ref is the reference voltage of the film being tested, V mat is the material equivalent voltage of the film being tested.

[0059] Among them, the absorption compensation factor k2 is dynamically updated through an online learning algorithm to compensate for nanoscale surface density fluctuations and micron-level thickness changes.

[0060] In order to make the signal output more accurate and eliminate the influence of noise and vibration generated during the film production process on the results, a digital filter optimization module is designed, and an adaptive filtering algorithm (such as Kalman filtering) is embedded in the MCU. The algorithm constructs the state space model state equation x k and the observation equation z k :x k =Ax k-1 +w k-1 and z k =Hx k +v k Among them, x k is the state vector of the system at time step k; Axk-1 is the state transfer matrix, which describes the dynamic relationship from the previous state to the current state; w k-1 is the process noise, representing the random disturbance in the production process. k is the observation matrix of time step k, Hx k is to map the system state to the sensor measurement space; v k It is the measurement noise, which reflects sensor error or environmental interference.

[0061] The above model suppresses circuit noise and mechanical vibration interference. By suppressing circuit noise, the system can reduce the standard deviation of thickness measurements from 2.1mm to 0.7mm under a 10°C temperature fluctuation. The output signal is ultimately sent to a microcontroller, where the thickness data is displayed on a 1602 LCD. The system supports dual outputs for density and thickness, and provides real-time production line feedback through Hongmeng OS. The system supports dual-mode switching (TOF mode and phase-amplitude mode), with a temperature drift error of <0.5µm / °C, meeting the requirements of high-speed production lines for calibration-free, anti-interference, and multi-parameter output.

[0062] The temperature value and the thickness value are input into the control model to obtain the control parameters of the PID temperature control system, SMC longitudinal stretching system and SMC transverse stretching system of the polypropylene film production line; wherein, the control model is based on a data preprocessing module, a bidirectional long short-term memory network and a multi-head attention mechanism.

[0063] Reference Figure 5 In the control model, the input features are convolved and max-pooled through the CNN network, and then feature maps of different sizes are passed through different LSTM layers respectively. The feature maps output from the LSTM layer enter the multi-head attention mechanism, undergo full-time step output, matrix multiplication, delayed step output processing, and pass through the fully connected layer. The feature maps output by the fully connected layer are processed by multiple LSTM layers, and the results are mapped by the linear activation function and output by the output layer.

[0064] During the preprocessing phase of the temperature and thickness data detected by the two sensors, TDengine is used for data preprocessing. The sensor temperature and thickness data are loaded into the database through the TDengine interface. This data is timestamped to facilitate subsequent time series analysis. The loaded data is then cleaned, and missing values ​​and outliers are identified and processed. Data conversion and normalization are then performed, converting the temperature data from Fahrenheit to Celsius and the thickness units from centimeters to microns. The temperature data is then scaled to the [0, 1] range using the min-max normalization method. The TDengine database is then used to partition the data by time range and compress the partitioned data using the TSZ compression algorithm. This achieves full-state compression from lossy to lossless floating-point data, further improving the compression rate. Finally, based on the time series, the acquired coordinates (x, y) are organized into 2x2 features. The output of the partitioned data contains timestamps.

[0065] First, the temperature and thickness data are fed into a convolutional neural network, which outputs a processed, enhanced feature map. The enhanced feature map is then fed into the upper bidirectional LSTM layer. The core of the LSTM layer lies in its three key gating mechanisms: the forget gate, the input gate, and the output gate. These gating mechanisms control the flow of information through their built-in sigmoid activation function, enabling selective memorization and forgetting of historical temperature and thickness information.

[0066] Reference Figure 6 LSTM processes and transmits information through the forget gate, input gate, and output gate. First, the forget gate receives the previous hidden state ht and the current input xt, calculates the forgetting coefficient (between 0 and 1), and determines whether to retain or discard the temperature and thickness information in the previous cell state. The closer the forgetting coefficient is to 0, the more information is forgotten; the closer it is to 1, the more information is retained. Next, the input gate also receives ht and xt, uses sigmoid to calculate the input coefficient, determines the amount of newly measured temperature and thickness information to be included, and updates the candidate cell state C: generated by tanh. The cell state combines the outputs of the forget gate and input gate to complete the information update. Finally, the output gate determines which information in the cell state Ct should be output to the hidden state h. Ct is compressed to [-1, 1] using tanh and multiplied by the output gate result to obtain the final h. This hidden state is used for the current prediction and is passed as input to the subsequent multi-head attention module, which comprehensively represents the current input and context information.

[0067] After the LSTM layer processes the sequence data, it generates a hidden state output h containing historical information. t-1 The present invention will hide the output h t-1The data is then passed to the multi-head attention layer. This layer first splits the input feature data into multiple "heads," each of which independently processes different parts of the input data. Through linear transformations, it calculates the query matrix Q, key matrix K, and value matrix V. Each head has an independent weight matrix, allowing it to understand the data from different perspectives. Next, each head uses the dot-product attention mechanism to calculate the similarity between Q and K, and then uses the softmax function to convert this into attention weights. Subsequently, the multi-head attention layer calculates the attention output of each head in parallel and concatenates the outputs of all heads to obtain the preliminary results of the multi-head attention.

[0068] In order to map the preliminary results back to the original dimension, this layer also contains a final linear transformation layer, whose formula is: MultiHead(Q,K,V)=Concat(head1,…,head h )·W o ; Among them, head i represents the attention output of the i-th head, h is the number of heads, W o is the final output linear transformation matrix.

[0069] The multi-head attention layer outputs a two-dimensional feature map consisting of a weighted summed and linearly transformed vector. This vector incorporates information from multiple independent attention heads. Finally, the lower layer of the bidirectional LSTM layer operates in much the same way as the upper layer. It takes as input the predicted BOPP film temperature and thickness values ​​activated by a ReLU-Sigmoid combined with a nonlinear activation function. The lower layer of the bidirectional LSTM layer performs deep time series modeling on the input data, effectively capturing long-term dependencies and complex patterns in the data. Finally, the output of the linear activation function output layer is passed to the specific PID and SMC algorithms. This parameter not only contains information about the current input temperature and thickness, but also incorporates key features from historical data. This provides rich contextual information about film temperature and thickness for subsequent predictions of BOPP film rupture risk, improving the predictive performance and generalization capabilities of deep learning models when processing time series data.

[0070] A correlation algorithm derives control parameters for the PID and SMC algorithms. Once these parameters are loaded, adaptive control is achieved using PID control of the temperature control system and SMC control of the longitudinal and transverse stretching systems. The three control systems work together to achieve precise temperature control and stable stretching, optimizing BOPP production quality.

[0071] The parameters C (sliding surface parameters), K required by the SMC algorithm are converted into s (Switching Gain) and K r(Arrival law gain) is uploaded to the transverse and longitudinal pulling systems, and by designing appropriate sliding surfaces, the dynamic behavior of the system in the transverse and longitudinal directions is ensured to be stably controlled. In the transverse pulling system, the SMC controller achieves accurate tracking of the system movement by defining a sliding surface related to the lateral position or speed, and overcomes external interference and system uncertainty during the control process. In the longitudinal pulling system, the SMC controller controls the traction force and speed so that the longitudinal displacement can accurately reach the target position. By designing appropriate switching rules and control laws, it is ensured that the system can enter and remain within the sliding surface within a limited time, thereby achieving high-precision and robust control and adapting to different loads and environmental changes. When the system uses PID to control the temperature system, it detects the error between the current temperature and the set temperature, and uses the K provided by the association algorithm p (proportional gain), K i (integral gain) and K d The (integral gain) control law generates a control signal to adjust the power to the heating or cooling device to maintain a stable temperature.

[0072] Reference Figure 7 In the PID control algorithm, the deviation between the theoretical and actual values ​​is calculated through the integral, proportional, and differential stages to obtain the integral, proportional, and differential values. These values ​​are then linearly combined to form the control variable, which is then fed into the temperature control system and measured by a bias-free sensor to obtain the current actual value. The PID controller continuously calculates and adjusts the operating state of the heater or cooling unit to quickly reach the set temperature and maintain stability despite environmental changes. This reduces overshoot, oscillation, and steady-state errors, achieving precise temperature control.

[0073] An embodiment of the present application provides an electronic device comprising: a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor implements the polypropylene film production control method described above when executing the computer program.

[0074] The electronic device may be any intelligent terminal including a computer.

[0075] In general, for the hardware structure of the electronic device, the processor can be implemented using a general-purpose CPU (Central Processing Unit), a microprocessor, an application-specific integrated circuit (ASIC), or one or more integrated circuits, etc., to execute relevant programs to implement the technical solutions provided in the embodiments of the present application.

[0076] The memory can be implemented in the form of a read-only memory (ROM), a static storage device, a dynamic storage device, or a random access memory (RAM). The memory can store an operating system and other application programs. When the technical solutions provided in the embodiments of this specification are implemented through software or firmware, the relevant program code is stored in the memory and is called by the processor to execute the methods of the embodiments of this application.

[0077] The input / output interface is used to realize information input and output.

[0078] The communication interface is used to realize the communication interaction between this device and other devices. Communication can be achieved through wired means (such as USB, network cable, etc.) or wireless means (such as mobile network, WIFI, Bluetooth, etc.).

[0079] The bus transmits information between the various components of the device (such as the processor, memory, input / output interface, and communication interface). The processor, memory, input / output interface, and communication interface communicate with each other within the device through the bus.

[0080] An embodiment of the present application provides a computer storage medium storing computer-executable instructions for executing the polypropylene film production control method described above.

[0081] Those skilled in the art will appreciate that all or some of the steps and systems in the method disclosed above can be implemented as software, firmware, hardware, and appropriate combinations thereof. Some physical components or all physical components can be implemented as software executed by a processor, such as a central processing unit, a digital signal processor, or a microprocessor, or implemented as hardware, or implemented as an integrated circuit, such as an application-specific integrated circuit. Such software can be distributed on a computer-readable medium, and the computer-readable medium can include computer storage media (or non-transitory media) and communication media (or temporary media). As known to those skilled in the art, the term computer storage media is included in any method or technology for storing information (such as computer-readable instructions, data structures, program modules, or other data) and is volatile and non-volatile, removable, and non-removable. Computer storage media includes, but is not limited to, RAM, ROM, EEPROM, flash memory, or other memory technology, CD-ROM, digital versatile disks (DVD), or other optical disk storage, magnetic cassettes, magnetic tapes, disk storage, or other magnetic storage devices, or any other medium that can be used to store desired information and can be accessed by a computer. In addition, it is well known to those skilled in the art that communication media generally contain computer-readable instructions, data structures, program modules, or other data in a modulated data signal such as a carrier wave or other transmission mechanism, and may include any information delivery medium. In the above description of this specification, the reference terms "one embodiment / example", "another embodiment / example" or "certain embodiments / examples" and the like are intended to mean that the specific features, structures, materials or characteristics described in conjunction with the embodiment or example are included in at least one embodiment or example of the present application. In this specification, the schematic representation of the above terms does not necessarily refer to the same embodiment or example. Moreover, the specific features, structures, materials or characteristics described may be combined in any one or more embodiments or examples in a suitable manner.

[0082] Those skilled in the art will appreciate that all or some of the steps in the methods, systems, and functional modules / units in the devices disclosed above may be implemented as software, firmware, hardware, or appropriate combinations thereof.

[0083] The units described above as separate components may or may not be physically separate, and 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 these units may be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0084] In addition, the functional units in the various embodiments of the present application may be integrated into a single processing unit, or each unit may exist physically separately, or two or more units may be integrated into a single unit. The aforementioned integrated units may be implemented in the form of hardware or software functional units.

[0085] If the integrated unit is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present application, or the part that contributes to the prior art, or all or part of the technical solution can be embodied in the form of a software product, which is stored in a storage medium and includes multiple instructions for enabling a computer device (which can be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods of various embodiments of the present application. The aforementioned storage medium includes: various media that can store programs, such as a USB flash drive, a mobile hard disk, a read-only memory (ROM), a random access memory (RAM), a magnetic disk or an optical disk.

[0086] In the several embodiments provided in the present application, it should be understood that the disclosed devices and methods can be implemented in other ways. For example, the device embodiments described above are merely schematic. For example, the division of the above-mentioned units is only a logical function division. There may be other division methods in actual implementation, such as multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be through some interfaces, indirect coupling or communication connection of devices or units, which can be electrical, mechanical or other forms. Although the embodiments of the present application have been shown and described, those skilled in the art will understand that various changes, modifications, substitutions and variations can be made to these embodiments without departing from the principles and purpose of the present application, and the scope of the present application is defined by the claims and their equivalents.

[0087] The above is a specific description of the preferred implementation of the present invention, but the present invention is not limited to the above implementation. Those skilled in the art can also make various equivalent modifications or substitutions under the shared conditions that do not violate the spirit of the present invention. These equivalent modifications or substitutions are all included in the scope defined by the claims of the present invention.

Claims

1. A polypropylene film production control method based on temperature and thickness, characterized in that: include: The temperature value of the polypropylene film is obtained by a temperature sensor; Obtain the thickness value of the polypropylene film through the thickness sensor; Inputting the temperature value and the thickness value into a control model to obtain control parameters of a PID temperature control system, an SMC longitudinal stretching system, and an SMC transverse stretching system of a polypropylene film production line; Control the operation of the PID temperature control system, the SMC longitudinal pulling system and the SMC transverse pulling system according to the control parameters; Among them, the control model is based on a bidirectional long short-term memory network and a multi-head attention mechanism.

2. The polypropylene film production control method based on temperature and thickness according to claim 1, characterized in that: The temperature sensor is a 4H-SiC vertical double diffused metal oxide semiconductor enhancement mode field effect transistor (EFMT) with a lateral Schottky barrier diode integrated into a class AB dual power supply complementary symmetrical power amplifier circuit in a p-type well region.

3. The polypropylene film production control method based on temperature and thickness according to claim 1, characterized in that: When the temperature sensor is connected to a power source, the Class AB dual-power complementary symmetrical power amplifier circuit controls the voltage across itself by adjusting its own resistance, providing a current bias between the push-pull transistors. When the temperature changes, the energy of electron thermal motion changes, exciting electrons to cross the potential barrier of the lateral Schottky barrier diode, generating a current increment in the temperature sensor. The method of obtaining the temperature value of the polypropylene film by using the temperature sensor includes: determining the current value of the temperature sensor according to the current increment; obtaining the voltage value of the temperature sensor according to the current value of the temperature sensor; and obtaining the temperature value of the polypropylene film according to the voltage value of the temperature sensor.

4. The polypropylene film production control method based on temperature and thickness according to claim 3, characterized in that: The current increment is expressed as: The voltage value of the temperature sensor is determined according to the following formula: Where ΔI is the current increment, Δx is the length increment, and W S is the Schottky contact width, A * is the Richardson constant, T is the detection temperature, q is the charge, φ B is the barrier height, η is the ideality factor of the Schottky contact, is the Boltzmann constant, V S is the potential of the lateral Schottky barrier diode at the Schottky contact, V(x) is the potential in the x direction, I T is the current value of the temperature sensor, B is the proportional coefficient, V TS is the voltage value of the temperature sensor, V R is the interface voltage drop.

5. The polypropylene film production control method based on temperature and thickness according to claim 1, characterized in that: The thickness sensor includes a control circuit, a transmitting circuit, a signal processing circuit, a receiving circuit, a Fourier transform circuit, a phase detection circuit, an ultrasonic generating probe and an ultrasonic receiving probe; the output end of the control circuit is connected to the input end of the transmitting circuit, the output end of the transmitting circuit is connected to the ultrasonic generating probe, the ultrasonic receiving probe is connected to the input end of the receiving circuit, one output end of the receiving circuit is connected to the input end of the signal processing circuit, the output end of the signal processing circuit is connected to one input end of the control circuit, the other output end of the receiving circuit is connected to the input end of the phase detection circuit, the output end of the phase detection circuit is connected to the input end of the Fourier transform circuit, and the output end of the Fourier transform circuit is connected to the other input end of the control circuit.

6. The polypropylene film production control method based on temperature and thickness according to claim 5, characterized in that: The transmitting circuit includes a pulse signal generating circuit and a switch control circuit; the receiving circuit includes a pre-processing circuit and a secondary proportional amplification circuit; the signal processing circuit includes a full-wave rectification circuit, an envelope detection circuit and an over-peak judgment circuit.

7. The polypropylene film production control method based on temperature and thickness according to claim 6, characterized in that: When the thickness sensor is powered on, the switch control circuit causes the pulse signal generating circuit to generate a square wave pulse, and the ultrasonic generating probe generates an ultrasonic wave according to the square wave pulse; the ultrasonic receiving probe receives the ultrasonic wave that penetrates the film to be measured, the pre-processing circuit filters the ultrasonic wave received by the ultrasonic receiving probe, and the secondary proportional amplifier circuit amplifies the filtered signal; the full-wave rectifier circuit converts the amplified signal into a unipolar signal, retains the complete energy distribution and outputs a positive envelope; the envelope detection circuit filters out the high-frequency component of the carrier from the positive envelope and extracts the amplitude attenuation data; the over-peak judgment circuit detects the envelope peak moment, generates a falling edge pulse to trigger the control circuit interrupt, marks the signal arrival time, and calculates the thickness value by the time difference between the signal arrival time and the square wave pulse generation time.

8. The polypropylene film production control method based on temperature and thickness according to claim 5, characterized in that: The control circuit corrects the thickness value according to a correction formula, which is expressed as: Where d is the thickness value, Δφ is the phase difference, v(T) is the ultrasonic velocity at time T, f is the ultrasonic frequency, k2 is the absorption compensation factor used to correct the thickness fluctuation, Z1 is the ultrasonic acoustic impedance, ω is the ultrasonic angular frequency, V ref is the reference voltage of the film being tested, V mat is the material equivalent voltage of the film being tested.

9. An electronic device, characterized in that: include: A memory, a processor, and a computer program stored in the memory and executable on the processor, wherein when the processor executes the computer program, the method for controlling the production of polypropylene film based on temperature and thickness as claimed in any one of claims 1 to 8 is implemented.

10. A computer storage medium, characterized in that Computer-executable instructions are stored, and the computer-executable instructions are used to execute the polypropylene film production control method based on temperature and thickness according to any one of claims 1 to 8.

Images