Intelligent control method and system for extrusion forming of aluminum profile

Abstract

The invention relates to an intelligent control method and system for extrusion forming of an aluminum profile, and the method comprises the following steps: carrying out acoustic emission sensor array measurement and preprocessing on a mold inlet region of the aluminum profile in an extrusion process, and obtaining a preprocessed acoustic emission signal sequence; performing time-frequency domain energy distribution analysis on the acoustic emission spectrum to generate an acoustic emission energy spectrogram; carrying out material flow stability analysis based on the spectrogram to obtain a flow analysis result; if the flow instability degree exceeds the target stability range, multi-constraint optimization calculation is conducted on the main cylinder pressure and the extrusion speed based on the analysis result, and optimization control parameters are obtained; and finally, closed-loop feedback control is performed on the extruder through a distributed PID controller, so that the material flow stability is kept in a target range, intelligent high-precision forming control is realized, and the problem that in the traditional extrusion process, the material flow stability is often influenced by factors such as the extrusion speed, the main cylinder pressure and the mold design is solved. And a flow instability phenomenon is easy to occur.

Description

Technical Field

[0001] This invention relates to the field of aluminum profile technology, and in particular to an intelligent control method and system for aluminum profile extrusion molding. Background Technology

[0002] Aluminum profile extrusion is a crucial processing technology widely used in aerospace, transportation, and construction, and its forming quality directly impacts the performance and reliability of the final product. In traditional extrusion processes, the stability of material flow is often affected by factors such as extrusion speed, master cylinder pressure, and die design, easily leading to flow instability, defects such as surface cracks and uneven internal structure. However, current industry methods for monitoring material flow are still relatively limited, lacking real-time, efficient feedback mechanisms to dynamically adjust process parameters, thus hindering precise control of the extrusion process. Summary of the Invention

[0003] The main objective of this invention is to provide an intelligent control method for aluminum profile extrusion molding, which solves the technical problem that the stability of material flow is often affected by factors such as extrusion speed, master cylinder pressure and mold design during traditional extrusion processes, and is prone to flow instability.

[0004] To achieve the above objectives, the present invention provides an intelligent control method for aluminum profile extrusion molding, comprising the following steps: Acoustic emission sensor array measurement and preprocessing were performed on the die inlet area of ​​aluminum profiles during the extrusion process to obtain a preprocessed acoustic emission signal sequence; Time-frequency domain energy distribution analysis was performed on the preprocessed acoustic emission signal sequence to obtain the acoustic emission energy spectrum. Based on the acoustic emission energy spectrum, the material flow stability of the aluminum profile during the extrusion process is analyzed, and the flow analysis results are obtained. If the flow analysis results show that the degree of flow instability is not within the target flow stability range, then based on the flow analysis results, multi-constraint optimization calculations are performed on the main cylinder pressure and extrusion speed of the extruder to obtain optimized control parameters. Based on the optimized control parameters, a distributed PID controller is used to perform closed-loop feedback control on the extruder to ensure that the material flow stability of the aluminum profile during the extrusion process is within the target flow stability range.

[0005] Furthermore, the measurement and preprocessing of the acoustic emission sensor array in the die inlet area of ​​the aluminum profile during the extrusion process to obtain a preprocessed acoustic emission signal sequence includes: An initial acoustic emission data stream is obtained by real-time acquisition of elastic wave signals caused by changes in internal stress of the aluminum profile during the extrusion process using an array of acoustic emission sensors arranged in the mold inlet area. The initial acoustic emission data stream is pre-amplified to improve the signal-to-noise ratio, resulting in a preliminarily enhanced acoustic emission signal. Synchronous noise suppression is performed on each channel signal in the acoustic emission signal to eliminate environmental interference and common-mode noise, resulting in a suppressed acoustic emission signal; The suppressed acoustic emission signal is subjected to time-domain synchronous averaging to obtain a preprocessed acoustic emission signal sequence.

[0006] Furthermore, the step of performing time-frequency domain energy distribution analysis on the preprocessed acoustic emission signal sequence to obtain an acoustic emission energy spectrum includes the following steps: The preprocessed acoustic emission signal sequence is subjected to a short-time Fourier transform to obtain a time-frequency domain signal matrix, and the time-frequency domain signal matrix is ​​divided into frequency bands to obtain multi-frequency band time-frequency signals. Calculate the energy value of each frequency band in the multi-band time-frequency signal, generate a frequency band energy sequence, and perform a sliding window integration on the time axis on the frequency band energy sequence to obtain the frequency band energy integral matrix; The singular values ​​in the frequency band energy integral matrix are decomposed to obtain singular value eigenvectors, and time-frequency domain energy weights are allocated based on the singular value eigenvectors to obtain a weighted time-frequency energy matrix. The weighted time-frequency energy matrix is ​​subjected to grid encryption processing using a preset two-dimensional interpolation algorithm to obtain the processed time-frequency energy matrix. The processed time-frequency energy matrix is ​​then subjected to grayscale mapping conversion to obtain the acoustic emission energy spectrum.

[0007] Furthermore, based on the acoustic emission energy spectrum, the material flow stability analysis of the aluminum profile during the extrusion process is performed to obtain the flow analysis results, including: The acoustic emission energy spectrum is subjected to spatial gradient enhancement processing to obtain a gradient-enhanced energy matrix, and the boundary of the gradient-enhanced energy matrix is ​​identified by a two-dimensional edge detection method to obtain the energy distribution boundary contour. Based on the energy distribution boundary contour, morphological opening operations are performed to obtain an optimized energy boundary curve, and geometric features in the energy boundary curve are extracted to obtain an energy boundary morphological feature curve. The energy boundary morphology characteristic curve is decomposed into multiple scales to obtain boundary fluctuation characteristic quantities, and nonlinear dynamic analysis is performed on the boundary fluctuation characteristic quantities to obtain flow state characteristic parameters. Multi-dimensional clustering analysis is performed on the flow state characteristic parameters to obtain the flow state category determination results. Based on the flow state category determination results, the stability of the aluminum profile during the extrusion process is evaluated to obtain the flow analysis results.

[0008] Furthermore, based on the flow analysis results, multi-constraint optimization calculations are performed on the main cylinder pressure and extrusion speed of the extruder to obtain optimized control parameters, including: Based on the flow analysis results, nonlinear dynamic response analysis was performed on the main cylinder pressure and extrusion speed of the extruder to obtain the pressure-velocity coupling characteristic matrix. Then, constraint condition mapping was performed on the pressure-velocity coupling characteristic matrix to obtain the process parameter boundary constraints. The process parameter boundary constraints are decoupled into multiple objectives by Lagrange relaxation decomposition to obtain a sequence of parameter optimization sub-problems. Based on the sequence of parameter optimization sub-problems, a hierarchical iterative solution is performed to obtain a set of candidate solutions for parameter adjustment. Based on the parameter adjustment candidate solution set, the stability criterion of the extruder is verified to obtain the parameter feasible solution sequence; Gradient projection optimization is performed based on the feasible solution sequence of the parameters to obtain the optimized control parameters.

[0009] Furthermore, the process parameter boundary constraints are decoupled through Lagrange relaxation decomposition to obtain a sequence of parameter optimization sub-problems, including: Lagrange multiplier allocation is performed on the boundary constraints of the process parameters to obtain the multiplier allocation matrix, and dual gap evaluation is performed on the multiplier allocation matrix to obtain the dual convergence index. Based on the dual convergence index, the multiplier allocation matrix is ​​updated iteratively to obtain the updated multiplier vector. Based on the updated multiplier vector, the boundary constraints of the process parameters are relaxed to obtain the relaxed constraint conditions. The relaxed constraints are divided into sub-problem boundaries to obtain a sub-problem boundary division matrix. Based on the sub-problem boundary division matrix, the coupling variables are decoupled to obtain a set of decoupled variables. Based on the set of decoupled variables, sub-problems are constructed to obtain a sequence of parameter optimization sub-problems.

[0010] Furthermore, the gradient projection optimization calculation based on the feasible solution sequence of the parameters to obtain the optimized control parameters includes: The gradient direction is initialized on the feasible solution sequence of the parameters to obtain the gradient direction initialization vector; The gradient step size in the feasible solution sequence of the parameters is calculated based on the gradient direction initialization vector to obtain the gradient step size sequence, and line search processing is performed based on the gradient step size sequence to obtain the set of line search termination points. The set of line search termination points is subjected to projection constraint processing to obtain a set of projection constraint points, and convergence discrimination is performed based on the set of projection constraint points to obtain a convergence discrimination index. The optimal solution is selected from the set of projection constraint points based on the convergence discrimination index to obtain the optimized control parameters.

[0011] The present invention also provides an intelligent control system for aluminum profile extrusion molding, comprising: The measurement module is used to perform acoustic emission sensor array measurement and preprocessing on the die inlet area of ​​aluminum profiles during the extrusion process to obtain a preprocessed acoustic emission signal sequence; The first analysis module is used to perform time-frequency domain energy distribution analysis on the preprocessed acoustic emission signal sequence to obtain an acoustic emission energy spectrum. The second analysis module is used to perform material flow stability analysis on the aluminum profile during the extrusion process based on the acoustic emission energy spectrum, and obtain flow analysis results; The judgment module is used to perform multi-constraint optimization calculations on the main cylinder pressure and extrusion speed of the extruder based on the flow analysis results when the degree of flow instability is not within the target flow stability range, so as to obtain optimized control parameters. The control module is used to perform closed-loop feedback control of the extruder through a distributed PID controller based on the optimized control parameters, so as to ensure that the material flow stability of the aluminum profile during the extrusion process is within the target flow stability range.

[0012] The present invention also provides 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 the steps of any of the methods described above.

[0013] The present invention also provides a computer-readable storage medium having a computer program stored thereon, wherein the computer program, when executed by a processor, implements the steps of any of the methods described above.

[0014] This invention provides an intelligent control method for aluminum profile extrusion molding, comprising the following steps: measuring and preprocessing the acoustic emission signal sequence in the die inlet area of ​​the aluminum profile during the extrusion process using an array of acoustic emission sensors; performing time-frequency domain energy distribution analysis on the preprocessed acoustic emission signal sequence to obtain an acoustic emission energy spectrum; performing material flow stability analysis on the aluminum profile during the extrusion process based on the acoustic emission energy spectrum to obtain flow analysis results; when the flow analysis results indicate that the degree of flow instability is not within the target flow stability range, performing multi-constraint optimization calculations on the main cylinder pressure and extrusion speed of the extruder based on the flow analysis results to obtain optimized control parameters; and performing closed-loop feedback control on the extruder using a distributed PID controller based on the optimized control parameters to achieve material flow stability of the aluminum profile during the extrusion process within the target flow stability range. This solves the technical problem that in traditional extrusion processes, the stability of material flow is often affected by factors such as extrusion speed, main cylinder pressure, and die design, easily leading to flow instability. It achieves the conversion of the preprocessed acoustic emission signal into an energy spectrum using time-frequency domain energy distribution analysis technology, allowing researchers to gain a deeper understanding of the material's behavior characteristics during the extrusion process. This helps identify key factors affecting the stability of material flow, enabling targeted measures to optimize technical effects. Attached Figure Description

[0015] Figure 1 This is a schematic diagram of the steps of an intelligent control method for aluminum profile extrusion molding in one embodiment of the present invention; Figure 2 This is a structural block diagram of an intelligent control system for aluminum profile extrusion molding according to an embodiment of the present invention; Figure 3 This is a schematic block diagram of the structure of a computer device according to an embodiment of the present invention.

[0016] The objectives, features, and advantages of this invention will be further explained in conjunction with the embodiments and with reference to the accompanying drawings. Detailed Implementation

[0017] To make the objectives, technical solutions, and advantages of this invention clearer, the 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 merely illustrative and not intended to limit the invention.

[0018] like Figure 1 As shown, Figure 1 This invention provides an intelligent control method for aluminum profile extrusion molding, comprising the following steps: Step S1: The acoustic emission sensor array is used to measure and preprocess the aluminum profile in the die entrance area during the extrusion process to obtain a preprocessed acoustic emission signal sequence.

[0019] Specifically, in the aluminum profile extrusion process, to achieve real-time monitoring and control of material flow, it is first necessary to perform acoustic emission sensor array measurements and preprocessing on the die inlet area of ​​the aluminum profile during extrusion to obtain a preprocessed acoustic emission signal sequence that can be used for subsequent analysis. This step involves arranging an array of multiple acoustic emission sensors in the die inlet area to collect elastic wave signals caused by material deformation, crack initiation, or friction during extrusion, thereby obtaining raw information on the material's flow behavior under high temperature and high pressure. Subsequently, the collected raw acoustic emission signals undergo preprocessing operations such as filtering, noise reduction, amplification, and digitization to remove environmental noise interference and enhance effective signal characteristics, forming a preprocessed acoustic emission signal sequence that can be used for further analysis. For example, in practical applications, when the aluminum profile is extruded into the die under high temperature conditions, local microcracks or plastic deformation may occur inside the material due to stress concentration. These microscopic changes will excite acoustic emission signals. The sensor array can capture these signals and transmit them to the signal processing system for preprocessing, providing a high-quality data foundation for subsequent time-frequency domain analysis.

[0020] Step S2: Perform time-frequency domain energy distribution analysis on the preprocessed acoustic emission signal sequence to obtain an acoustic emission energy spectrum.

[0021] Specifically, after obtaining the preprocessed acoustic emission signal sequence, time-frequency domain energy distribution analysis needs to be performed on the preprocessed acoustic emission signal sequence to extract characteristic information reflecting the material flow state and finally obtain an acoustic emission energy spectrum. This step uses time-frequency analysis methods (such as short-time Fourier transform, wavelet transform, or Hilbert-Huang transform) to jointly analyze the preprocessed acoustic emission signal in both time and frequency dimensions, thereby revealing the energy distribution law of different frequency components in the signal over time. By extracting and visualizing these energy distribution characteristics, an intuitive acoustic emission energy spectrum can be formed to characterize the dynamic behavior of aluminum profile material flow during extrusion. For example, when the aluminum profile enters the mold inlet area and is subjected to high temperature and high pressure, its interior may release acoustic emission signals of different frequency ranges due to intensified plastic deformation. After time-frequency domain analysis, these signals will show specific energy concentration areas on the energy spectrum, thus providing a key basis for subsequent judgment on whether the material flow is stable.

[0022] Step S3: Based on the acoustic emission energy spectrum, perform material flow stability analysis on the aluminum profile during the extrusion process to obtain flow analysis results.

[0023] Specifically, based on the acoustic emission energy spectrum, material flow stability analysis is performed on the aluminum profile during the extrusion process to obtain flow analysis results. This process involves in-depth interpretation of the acoustic emission energy spectrum generated in the previous step and its correlation with the material flow state. Specifically, a standard or reference model needs to be established first. This model includes typical acoustic emission energy distribution characteristics corresponding to different flow stability states. For example, under ideal flow conditions, the energy distribution may be concentrated within a specific frequency range and exhibit a regular pattern over time; while during flow instability, abnormal energy peaks or spectral expansion phenomena may be observed. By comparing the acoustic emission energy spectrum obtained during the actual extrusion process with these predefined standards, potential instability factors in the current material flow state can be identified. For example, in a specific aluminum profile extrusion application scenario, when the operator notices unusual high-frequency energy fluctuations on the energy spectrum, this may indicate localized stress concentration in the die inlet area or the generation of microcracks within the material, which are early signs of flow instability. Through precise analysis and evaluation of these characteristics, the stability of the material flow can be accurately determined, providing a scientific basis for subsequent adjustments to the extruder's operating parameters. This analysis method based on acoustic emission energy spectrum not only improves the understanding of complex extrusion processes, but also provides solid data support for achieving intelligent and precise control.

[0024] Step S4: When the flow analysis result is that the degree of flow instability is not within the target flow stability range, then based on the flow analysis result, multi-constraint optimization calculations are performed on the main cylinder pressure and extrusion speed of the extruder to obtain optimized control parameters.

[0025] Specifically, when the flow analysis results based on the acoustic emission energy spectrum show that the flow instability of the aluminum profile during extrusion is outside the set target flow stability range, it is necessary to perform multi-constraint optimization calculations on the main cylinder pressure and extrusion speed of the extruder based on the flow analysis results to obtain optimized control parameters that can improve the material flow state. Specifically, this step quantifies the current degree of flow instability as the basis for adjusting the control input variables. Considering constraints such as equipment operating limitations, process window boundaries, and material deformation capabilities, a nonlinear optimization model is constructed to solve for an optimal combination of main cylinder pressure and extrusion speed, making the material flow more stable. For example, in actual production, if the system detects abnormal fluctuations in the energy spectrum of the aluminum profile in the die inlet area due to uneven flow at a certain moment, indicating that the material is on the verge of flow instability, the control system will immediately activate the multi-constraint optimization module. By integrating historical data and real-time feedback information, it calculates new main cylinder pressure and extrusion speed values ​​and uses them as control parameters for the next stage, ensuring that the material flow stability returns to the target range during subsequent extrusion processes, thereby achieving dynamic closed-loop control of the entire forming process.

[0026] Step S5: Based on the optimized control parameters, the extruder is subjected to closed-loop feedback control through a distributed PID controller to ensure that the material flow stability of the aluminum profile during the extrusion process is within the target flow stability range.

[0027] Specifically, based on the optimized control parameters, a distributed PID controller performs closed-loop feedback control of the extruder. The aim is to precisely regulate the material flow stability of the aluminum profile during the extrusion process to within the target flow stability range. This step involves inputting optimized control parameters, such as the main cylinder pressure and extrusion speed, output from multi-constraint optimization calculations, as setpoints into distributed PID controllers located in key execution units of the extruder. Each controller adjusts physical quantities such as hydraulic system pressure and servo motor speed in real time based on the deviation between the current actual operating state and the setpoints, thereby achieving dynamic and precise control of the extrusion process. For example, when the aluminum profile enters the die inlet area and exhibits a tendency towards local flow instability, the system identifies the anomaly through acoustic emission signal analysis and, after optimization calculation, determines that the main cylinder pressure needs to be reduced and the extrusion speed slightly increased to restore flow balance. At this time, the distributed PID controller performs closed-loop adjustment of the main cylinder hydraulic circuit and the drive motor respectively, ensuring that all parameters quickly converge to the optimized values, causing the material flow state to stabilize again. This effectively prevents product defects caused by process fluctuations and ensures consistent molding quality during continuous production.

[0028] In a specific embodiment, the step of measuring and preprocessing the acoustic emission signal sequence in the die inlet area of ​​the aluminum profile during the extrusion process using an array of acoustic emission sensors to obtain the preprocessed acoustic emission signal sequence includes: An initial acoustic emission data stream is obtained by real-time acquisition of elastic wave signals caused by changes in internal stress of the aluminum profile during the extrusion process using an array of acoustic emission sensors arranged in the mold inlet area. The initial acoustic emission data stream is pre-amplified to improve the signal-to-noise ratio, resulting in a preliminarily enhanced acoustic emission signal. Synchronous noise suppression is performed on each channel signal in the acoustic emission signal to eliminate environmental interference and common-mode noise, resulting in a suppressed acoustic emission signal; The suppressed acoustic emission signal is subjected to time-domain synchronous averaging to obtain a preprocessed acoustic emission signal sequence.

[0029] Specifically, in the aluminum profile extrusion process, to achieve real-time monitoring and intelligent control of material flow, it is first necessary to perform acoustic emission sensor array measurement and preprocessing on the die inlet area of ​​the aluminum profile during extrusion to obtain a preprocessed acoustic emission signal sequence that can be used for subsequent analysis. This step specifically includes: real-time acquisition of elastic wave signals caused by internal stress changes in the aluminum profile during extrusion using an acoustic emission sensor array arranged in the die inlet area, thereby obtaining an initial acoustic emission data stream. These elastic wave signals originate from microscopic behaviors such as plastic deformation, microcrack generation, or grain boundary slip under high temperature and pressure, effectively reflecting the dynamic characteristics of material flow. Since the originally acquired signals are usually weak and contain a large amount of environmental noise, a series of signal enhancement and noise reduction processes are required to improve the accuracy of subsequent feature extraction and analysis. Subsequently, the system pre-amplifies the initial acoustic emission data stream to improve the signal-to-noise ratio, thereby obtaining a preliminarily enhanced acoustic emission signal. This step is accomplished by a high-gain, low-noise preamplifier, aiming to amplify the weak original signal to a voltage level that can be stably recognized by subsequent processing modules. Next, a multi-channel adaptive filter is used to synchronously suppress noise in each channel of the acoustic emission signal to eliminate environmental interference and common-mode noise, thus obtaining a suppressed acoustic emission signal. This filter can dynamically adjust the filtering parameters according to noise characteristics, ensuring that effective material deformation signals can still be separated even in complex industrial environments. Finally, the system performs time-domain synchronous averaging on the suppressed acoustic emission signal to further remove random noise and enhance characteristic signals with strong repeatability, ultimately obtaining a preprocessed acoustic emission signal sequence for subsequent time-frequency analysis. For example, in practical applications, when aluminum profiles are extruded into a mold under high-temperature conditions, local microcracks or severe plastic deformation may occur on their surface or inside due to stress concentration. These changes will excite transient elastic waves. The acoustic emission sensor array can capture these signals and sequentially perform pre-amplification, adaptive filtering, and synchronous averaging to form a high-quality preprocessed acoustic emission signal sequence, providing a reliable data foundation for subsequent time-frequency energy distribution analysis, thereby supporting the operation and optimization decisions of the entire closed-loop control system.

[0030] In a specific embodiment, the step of performing time-frequency domain energy distribution analysis on the preprocessed acoustic emission signal sequence to obtain an acoustic emission energy spectrum includes the following steps: The preprocessed acoustic emission signal sequence is subjected to a short-time Fourier transform to obtain a time-frequency domain signal matrix, and the time-frequency domain signal matrix is ​​divided into frequency bands to obtain multi-frequency band time-frequency signals. Calculate the energy value of each frequency band in the multi-band time-frequency signal, generate a frequency band energy sequence, and perform a sliding window integration on the time axis on the frequency band energy sequence to obtain the frequency band energy integral matrix; The singular values ​​in the frequency band energy integral matrix are decomposed to obtain singular value eigenvectors, and time-frequency domain energy weights are allocated based on the singular value eigenvectors to obtain a weighted time-frequency energy matrix. The weighted time-frequency energy matrix is ​​subjected to grid encryption processing using a preset two-dimensional interpolation algorithm to obtain the processed time-frequency energy matrix. The processed time-frequency energy matrix is ​​then subjected to grayscale mapping conversion to obtain the acoustic emission energy spectrum.

[0031] Specifically, performing time-frequency domain energy distribution analysis on the preprocessed acoustic emission signal sequence to obtain an acoustic emission energy spectrum is a key technical step in realizing intelligent sensing and dynamic control of material flow state during aluminum profile extrusion molding. This step first involves performing a short-time Fourier transform (STFT) on the preprocessed acoustic emission signal sequence, converting the signal, originally represented only in the time domain, into a time-frequency domain signal matrix that characterizes both time and frequency dimensions. This allows for clear observation of the energy changes of different frequency components at different times. Subsequently, the system divides the time-frequency domain signal matrix into multiple physically meaningful frequency bands based on engineering experience or prior knowledge, forming a multi-band time-frequency signal. This division helps distinguish the acoustic emission characteristics excited by different types of material deformation behavior; for example, low-frequency bands may reflect the overall plastic deformation trend, while high-frequency bands tend to indicate the generation of local microcracks. Next, the system calculates the energy values ​​of each frequency band in the multi-band time-frequency signal and generates a frequency band energy sequence accordingly. Further, it accumulates the frequency band energy using a sliding window integration method on the time axis to enhance the temporal continuity of stable features in the signal and suppress transient noise interference, ultimately obtaining a frequency band energy integral matrix. This matrix not only preserves the trend information of energy evolution over time for each frequency band but also improves the robustness and interpretability of the data, providing a high-quality data foundation for subsequent singular value decomposition (SVD). Based on this, the system performs Singular Value Decomposition (SVD) on the frequency band energy integral matrix, extracting the main singular value eigenvectors. These eigenvectors reflect the dominant patterns and structural characteristics in the energy distribution. Then, based on the singular value eigenvectors, the system assigns time-frequency domain energy weights to the original frequency band energy, generating a weighted time-frequency energy matrix. This assigns higher weights to energy components that better represent changes in the material flow state, thereby improving the discriminative power of the entire energy spectrum. To improve image resolution and facilitate visualization analysis, the system further performs grid encryption processing on the weighted time-frequency energy matrix using a preset two-dimensional interpolation algorithm, transforming the originally sparse discrete energy points into a densely arranged time-frequency energy matrix.Two-dimensional interpolation is a mathematical method for estimating the value of any point on a two-dimensional plane using known discrete data points. To improve the time-frequency resolution of the weighted time-frequency energy matrix, the original matrix is ​​first treated as a discrete two-dimensional image, where row coordinates correspond to time, column coordinates correspond to frequency, and element values ​​are the weighted energy at that time-frequency point. The two-dimensional interpolation algorithm borrows the idea of ​​image super-resolution, performing encrypted operations on this "image." Specifically, it uses the original grid as anchor points and constructs a continuous surface in both row and column directions simultaneously, so that any new position between old grid nodes can be assigned a reliable energy value. Therefore, when it is necessary to insert three or four times the number of subdivision points between every two existing grids, the algorithm only needs to call a bilinear kernel, first linearly stretching each dimension, and then multiplying the stretching results in both directions to obtain a weighted energy value. A smooth-transitioning dense grid is used, while for sections with significant energy abrupt changes, a bicubic or cubic convolution kernel is used. Utilizing the values ​​of the surrounding sixteen original points and their first derivatives, a more rounded surface is generated, suppressing mosaic artifacts while maintaining the sharpness of abrupt edges. If the original sampling points are not uniformly distributed, Delaunay triangulation is used to divide the time-frequency plane into non-overlapping triangles, and then linear or cubic interpolation is performed within the triangles, similarly upscaling scattered data into a dense grid. Regardless of the kernel function used, the interpolation process outputs a larger time-frequency energy matrix with a significantly increased number of rows and columns. The total energy remains consistent with the original matrix, but the energy is redistributed to finer time-frequency units, resulting in a high-resolution time-frequency energy matrix after grid refinement. Finally, this matrix, after grayscale mapping, generates an intuitively visualized acoustic emission energy spectrum for subsequent material flow stability analysis. For example, in an actual aluminum profile extrusion production line, when the aluminum material experiences local flow instability due to mold wear or temperature fluctuations, abnormal acoustic emission signals within a specific frequency range will be generated inside. After the above-mentioned process, these abnormal signals will show obvious energy concentration areas on the energy spectrum, thereby providing clear feedback basis for the control system, supporting the next step of optimization control decision-making, ensuring that the material flow is always within the target stable range, and achieving high-precision, high-quality intelligent forming processing.

[0032] In a specific embodiment, the step of performing material flow stability analysis on the aluminum profile during the extrusion process based on the acoustic emission energy spectrum to obtain flow analysis results includes: The acoustic emission energy spectrum is subjected to spatial gradient enhancement processing to obtain a gradient-enhanced energy matrix, and the boundary of the gradient-enhanced energy matrix is ​​identified by a two-dimensional edge detection method to obtain the energy distribution boundary contour. Based on the energy distribution boundary contour, morphological opening operations are performed to obtain an optimized energy boundary curve, and geometric features in the energy boundary curve are extracted to obtain an energy boundary morphological feature curve. The energy boundary morphology characteristic curve is decomposed into multiple scales to obtain boundary fluctuation characteristic quantities, and nonlinear dynamic analysis is performed on the boundary fluctuation characteristic quantities to obtain flow state characteristic parameters. Multi-dimensional clustering analysis is performed on the flow state characteristic parameters to obtain the flow state category determination results. Based on the flow state category determination results, the stability of the aluminum profile during the extrusion process is evaluated to obtain the flow analysis results.

[0033] Specifically, based on the acoustic emission energy spectrum, material flow stability analysis is performed on the aluminum profile during the extrusion process to obtain flow analysis results. This is the core step in the entire intelligent control method to achieve state recognition and decision-making. This step uses a series of image processing, feature extraction, and data analysis methods to transform the originally abstract time-frequency energy distribution information into a quantifiable description of the material flow state, providing a scientific basis for subsequent control strategy adjustments. Specifically, the acoustic emission energy spectrum is first subjected to spatial gradient enhancement processing to highlight the local variation trend of energy distribution, thereby obtaining a gradient-enhanced energy matrix. This operation helps to improve the contrast of the energy boundary region, making subsequent edge detection more accurate. Subsequently, the system uses a two-dimensional edge detection method to identify the boundaries of the gradient-enhanced energy matrix, extracting the energy distribution boundary contours. These contours reflect the dynamic evolution trajectory of different frequency components in the time dimension. On this basis, to further eliminate noise interference and improve the smoothness of the boundary curve, the system performs morphological opening operations on the energy distribution boundary contours to remove small burrs and isolated points, thereby obtaining an optimized energy boundary curve. Next, the system extracts geometric features from the optimized boundary curve, including curve curvature, piecewise slope, and inflection point distribution, forming an energy boundary morphology feature curve. These geometric features can effectively characterize the spatiotemporal evolution of energy distribution. For example, when local stress concentration or uneven flow occurs in the mold inlet area of ​​the aluminum profile, the energy boundary curve will show obvious inflection point shifts or slope abrupt changes, which are precursors to material flow instability. To further explore the nonlinear dynamic characteristics contained in these boundary morphology features, the system further decomposes the energy boundary morphology feature curve into multiple scales, extracting its fluctuation characteristics at different scales. This multi-scale analysis method can reveal the coupling relationship between short-term fluctuations and long-term trends in material flow, thereby capturing subtle changes that are difficult to detect by traditional single-scale analysis. Following this, the system performs nonlinear dynamic analysis on these boundary fluctuation characteristics, constructs phase space trajectories, and calculates dynamic parameters such as the Lyapunov exponent and correlation dimension, ultimately obtaining a set of flow state characteristic parameters reflecting the complexity of the material flow state. These parameters not only contain the intensity information of energy fluctuations but also integrate the structural information of their temporal evolution path, possessing high physical significance and engineering applicability. Finally, the system inputs the flow state characteristic parameters into the multi-dimensional clustering analysis module, and classifies them using K-means, DBSCAN, or other unsupervised learning algorithms to identify the current state category of the material flow, such as stable flow, critical instability, or significant instability, thereby obtaining the flow state category determination result. Based on this determination result, the system further evaluates the material flow stability of the aluminum profile during the extrusion process and outputs the final flow analysis result.For example, in practical applications, when uneven pressure distribution occurs in the inlet area due to die wear during an aluminum profile extrusion process, the acoustic emission signal collected by the system will show an abnormal energy accumulation area on the energy spectrum. After the above-mentioned process analysis, the flow state characteristic parameters will deviate from the normal range. Cluster analysis will identify this state as belonging to the "critical instability" category. Based on this, the control system determines that the current material flow stability is no longer within the target range, thereby triggering the next step of multi-constraint optimization control process to ensure the consistency of product quality and the stability of the process. In summary, this step, through progressive image processing and feature modeling of the acoustic emission energy spectrum, achieves high-precision identification and quantitative evaluation of the material flow state of aluminum profiles during extrusion, providing solid data support and decision-making basis for closed-loop feedback control.

[0034] In a specific embodiment, based on the flow analysis results, multi-constraint optimization calculations are performed on the main cylinder pressure and extrusion speed of the extruder to obtain optimized control parameters, including: Based on the flow analysis results, nonlinear dynamic response analysis was performed on the main cylinder pressure and extrusion speed of the extruder to obtain the pressure-velocity coupling characteristic matrix. Then, constraint condition mapping was performed on the pressure-velocity coupling characteristic matrix to obtain the process parameter boundary constraints. The process parameter boundary constraints are decoupled into multiple objectives by Lagrange relaxation decomposition to obtain a sequence of parameter optimization sub-problems. Based on the sequence of parameter optimization sub-problems, a hierarchical iterative solution is performed to obtain a set of candidate solutions for parameter adjustment. Based on the parameter adjustment candidate solution set, the stability criterion of the extruder is verified to obtain the parameter feasible solution sequence; Gradient projection optimization is performed based on the feasible solution sequence of the parameters to obtain the optimized control parameters.

[0035] Specifically, based on the flow analysis results, multi-constraint optimization calculations are performed on the main cylinder pressure and extrusion speed of the extruder to obtain optimized control parameters. This is a key technical step in achieving closed-loop control of material flow stability during the extrusion process of aluminum profiles. This step establishes a nonlinear dynamic relationship model between process parameters and material flow state, and combines this with the physical and process constraints in actual production to ultimately solve for a set of control parameter combinations that meet performance requirements and are engineering-feasible. Specifically, the system first performs nonlinear dynamic response analysis on the main cylinder pressure and extrusion speed of the extruder based on the flow analysis results, identifies the influence of these two parameters on the material flow stability under different values, and thus constructs a pressure-velocity coupling characteristic matrix. This matrix reflects the direction and magnitude of the material flow instability trend when the main cylinder pressure changes by 1 MPa or the extrusion speed increases by 0.1 mm / s under the current operating conditions. Subsequently, the system maps the feature matrix to the set process parameter boundary constraints, including the upper and lower limits of the main cylinder pressure (e.g., 30~200 MPa), the range of extrusion speed variation (e.g., 5~60 mm / s), and the material deformation resistance threshold (e.g., not exceeding 350 MPa), ensuring that subsequent optimization results do not exceed the safe operating range of the equipment or cause material failure. Based on this, the system uses the Lagrange relaxation decomposition method to perform multi-objective decoupling processing on the process parameter boundary constraints, breaking down the originally complex coupled problem into a sequence of multiple interrelated but independently solvable parameter optimization sub-problems. Each sub-problem focuses on the optimal adjustment strategy under a specific parameter combination, such as: how to minimize speed fluctuations while keeping the main cylinder pressure constant, or how to compensate for pressure changes to maintain overall stress balance. Subsequently, the system performs hierarchical iterative solutions based on the parameter optimization sub-problem sequence, gradually approaching the global optimum, and finally forming a candidate solution set for parameter adjustment. This set includes multiple possible main cylinder pressure adjustment values ​​(e.g., +5 MPa, -3 MPa), speed compensation increments (e.g., +2 mm / s, -1.5 mm / s), and process parameter correction coefficients (e.g., 0.98~1.02), which are used for subsequent screening and verification. To ensure that the selected parameter combination has good dynamic response characteristics and stability in practical applications, the system further verifies the stability based on the candidate solution set for parameter adjustment. That is, it evaluates whether the system response of each set of parameters is stable in actual operation and whether there is a risk of overshoot or oscillation through simulation or empirical models, thereby screening out feasible solution sequences of parameters that meet the control requirements, including pressure adjustment timing (e.g., adjusted every 3 seconds), speed adjustment step size (e.g., adjusted ±1 mm / s each time), and control parameter update cycle (e.g., updated parameters every 5 seconds), to ensure that the control system has good transient performance and robustness during dynamic adjustment.Finally, the system performs gradient projection optimization calculations based on the feasible solution sequence of the parameters. By searching for the optimal path along the gradient descent direction within the feasible region, it gradually converges to a set of optimal control parameter configurations, which serve as the final output optimized control parameters. This set of parameters typically includes specific master cylinder pressure settings (e.g., 150 MPa), extrusion speed settings (e.g., 40 mm / s), and parameter adjustment time intervals (e.g., fine-tuning every 7 seconds), effectively guiding material flow back to the target stable range. For example, during a continuous extrusion of aluminum profiles, local flow instability occurred due to temperature fluctuations in the die inlet area. After identifying the anomaly through acoustic emission energy spectrum, the system triggered a multi-constraint optimization process. After the above analysis and calculations, it was finally determined to increase the master cylinder pressure from 140 MPa to 148 MPa and reduce the extrusion speed from 42 mm / s to 39 mm / s, thereby successfully suppressing material flow instability and ensuring product quality and production continuity. In summary, this step, through nonlinear dynamic modeling of the main cylinder pressure and extrusion speed, multi-objective decoupling, stability verification, and gradient optimization, achieves intelligent optimization and precise adjustment of control parameters under complex process conditions, providing core support for the efficient operation of the entire closed-loop feedback control system.

[0036] In a specific embodiment, the step of performing multi-objective decoupling processing on the process parameter boundary constraints through Lagrange relaxation decomposition to obtain a sequence of parameter optimization sub-problems includes: Lagrange multiplier allocation is performed on the boundary constraints of the process parameters to obtain the multiplier allocation matrix, and dual gap evaluation is performed on the multiplier allocation matrix to obtain the dual convergence index. Based on the dual convergence index, the multiplier allocation matrix is ​​updated iteratively to obtain the updated multiplier vector. Based on the updated multiplier vector, the boundary constraints of the process parameters are relaxed to obtain the relaxed constraint conditions. The relaxed constraints are divided into sub-problem boundaries to obtain a sub-problem boundary division matrix. Based on the sub-problem boundary division matrix, the coupling variables are decoupled to obtain a set of decoupled variables. Based on the set of decoupled variables, sub-problems are constructed to obtain a sequence of parameter optimization sub-problems.

[0037] Specifically, the multi-objective decoupling of the process parameter boundary constraints through Lagrange relaxation decomposition yields a sequence of parameter optimization sub-problems, which is a key mathematical modeling step for achieving coordinated optimization control of master cylinder pressure and extrusion speed during aluminum profile extrusion. This step aims to transform the originally tightly coupled and difficult-to-solve multivariate optimization problem into several relatively independent and easily solvable sub-problems, thus providing a structured foundation for subsequent hierarchical iteration and parameter optimization. Specifically, after obtaining the pressure-velocity coupling characteristic matrix driven by the flow analysis results, the system first performs Lagrange multiplier allocation on the process parameter boundary constraints corresponding to this matrix. That is, a corresponding Lagrange multiplier is introduced for each boundary constraint condition (such as the master cylinder pressure upper limit of 200 MPa, the material deformation resistance threshold of 350 MPa, etc.), forming a multiplier allocation matrix. These multipliers represent the "weight" or "penalty coefficient" of each constraint in the overall optimization objective function, used to measure the cost of violating a certain constraint. Subsequently, the system performs duality gap evaluation on the multiplier allocation matrix to determine the convergence state of the current Lagrange relaxation model, obtaining a dual convergence index, which includes the difference between the dual function value and the original function value. For example, if the current dual function value is 12.4 and the original function value is 14.6, the difference between the two is 2.2, reflecting the degree of approximation between the current solution and the optimal solution. Based on the above dual convergence index, the system further iterates the multiplier update on the multiplier allocation matrix, using the subgradient ascent method or other update strategies to gradually adjust the values ​​of each Lagrange multiplier until the difference between the dual function value and the original function value is reduced to a preset range (e.g., less than 0.5), thereby obtaining the updated multiplier vector. Based on this, the system performs constraint relaxation on the original process parameter boundary constraints using the updated multiplier vector, transforming some hard constraints into soft constraints to form relaxed constraint conditions. This makes the optimization problem more feasible and flexible. For example, the original requirement that the main cylinder pressure must be strictly maintained between 140 and 160 MPa can be relaxed to allow fluctuations within the range of 138 to 162 MPa, but at a corresponding cost. Next, the system performs sub-problem boundary partitioning on the relaxed constraint conditions, dividing the entire parameter space into several non-overlapping but logically related sub-regions, forming a sub-problem boundary partitioning matrix. Each sub-region corresponds to a specific process state or control objective, such as the main cylinder pressure adjustment range of 140 to 150 MPa and the extrusion speed adjustment range of 35 to 45 mm / s. Subsequently, the system performs coupling variable decoupling processing based on the sub-problem boundary partitioning matrix, identifying which variables can change independently in different sub-problems and which variables need to be coordinated, thereby extracting a set of decoupled variables. For example, the main cylinder pressure can be adjusted independently, while the extrusion speed needs to change in tandem with the die temperature.Finally, the system constructs sub-problems based on the decoupled variable set, generating a sequence of parameter optimization sub-problems, including a main cylinder pressure optimization sub-problem, an extrusion speed optimization sub-problem, and a constraint coordination optimization sub-problem. The main cylinder pressure optimization sub-problem focuses on minimizing pressure fluctuations while ensuring material deformation capacity; for example, while maintaining a constant extrusion speed, the main cylinder pressure should be adjusted from 145 MPa to 148 MPa. The extrusion speed optimization sub-problem focuses on improving production efficiency while avoiding material instability; for example, when good flow stability is detected, the speed can be increased from 40 mm / s to 43 mm / s. The constraint coordination optimization sub-problem is responsible for establishing consistent constraints among multiple sub-problems, ensuring that parameter adjustments do not conflict with each other or exceed the equipment's safe range. For example, in practical applications, when the aluminum profile exhibits increased flow instability due to localized stress concentration in the die inlet area, the system identifies abnormal signals through acoustic emission energy spectrum analysis and triggers a multi-objective optimization process. At this point, based on the current flow analysis results, the system initiates a Lagrange relaxation decomposition mechanism. By performing multiplier allocation and dual convergence evaluation on boundary constraints such as the upper and lower limits of the main cylinder pressure and the range of extrusion speed variation, the system gradually updates the multiplier vectors and completes constraint relaxation. Subsequently, the system divides the boundaries of sub-problems and decouples key variables such as main cylinder pressure and extrusion speed, constructing independent optimization models for each. Finally, under the premise of satisfying all constraints, the system outputs a set of optimized control parameters, for example: the main cylinder pressure is set to 152 MPa, the extrusion speed is set to 38 mm / s, and the parameter adjustment time interval is once every 6 seconds. This effectively guides the material flow back to the target stable range, ensuring product quality and process stability.

[0038] In a specific embodiment, the step of performing gradient projection optimization calculation based on the feasible solution sequence of parameters to obtain optimized control parameters includes: The gradient direction is initialized on the feasible solution sequence of the parameters to obtain the gradient direction initialization vector; The gradient step size in the feasible solution sequence of the parameters is calculated based on the gradient direction initialization vector to obtain the gradient step size sequence, and line search processing is performed based on the gradient step size sequence to obtain the set of line search termination points. The set of line search termination points is subjected to projection constraint processing to obtain a set of projection constraint points, and convergence discrimination is performed based on the set of projection constraint points to obtain a convergence discrimination index. The optimal solution is selected from the set of projection constraint points based on the convergence discrimination index to obtain the optimized control parameters.

[0039] Specifically, the gradient projection optimization calculation based on the feasible solution sequence of parameters to obtain optimized control parameters is the final mathematical optimization process in the intelligent control system for aluminum profile extrusion forming to achieve coordinated control of master cylinder pressure and extrusion speed. This step follows the feasible solution sequence of parameters generated in the previous step through Lagrange relaxation decomposition, multi-objective decoupling, and stability criterion verification. Its core lies in gradually approaching the globally optimal combination of control parameters along the optimal descent direction under the premise of satisfying process boundary constraints, thereby providing precise execution instructions for closed-loop feedback control. Specifically, after obtaining the feasible solution sequence of parameters consisting of master cylinder pressure adjustment (e.g., +5 MPa), speed compensation increment (e.g., -1.5 mm / s), and control parameter update cycle (e.g., adjusted every 7 seconds), the system first performs gradient direction initialization processing on it, that is, determines the initial search direction based on the performance difference between the current parameter combination and the historical optimal solution, forming a gradient direction initialization vector. For example, in a real-world operation, if the flow instability trend persists even with a current pressure of 148 MPa and a velocity of 39 mm / s, the system will identify the direction in which the velocity needs to be further reduced or the pressure increased, and initialize the gradient direction accordingly. Subsequently, the system calculates the gradient step size in the feasible solution sequence based on the initialization vector of the gradient direction, obtaining a gradient step size sequence. This step evaluates the changing trend of material flow stability under different step sizes, selecting the set of step size values ​​that causes the objective function (such as the flow instability index) to decrease the fastest as the basis for subsequent line searches. For example, the system might try increasing the main cylinder pressure from 148 MPa to 150 MPa (step size of 2 MPa) while simultaneously reducing the extrusion velocity from 39 mm / s to 37 mm / s (step size of -2 mm / s), and observe whether the corresponding energy spectrum tends to stabilize. Next, the system performs a line search based on the gradient step size sequence, searching for several key points along the gradient direction on the optimal solution path in the parameter space, ultimately obtaining a set of line search termination points. These termination points represent multiple locally optimal states achievable under the current constraints and parameter combinations. For example, when the main cylinder pressure increases to 150 MPa and the speed decreases to 37 mm / s, the system detects that the energy distribution of the acoustic emission signal tends to concentrate, indicating that the material flow has stabilized. At this point, this combination can be marked as a line search termination point. To further ensure that all candidate solutions meet the actual process constraints, the system performs projection constraint processing on the set of line search termination points. That is, solutions that exceed the upper and lower limits of the main cylinder pressure (e.g., 30~200 MPa) or the range of extrusion speed variation (e.g., 5~60 mm / s) are "projected" back into the feasible region, forming a set of projection constraint points. For example, if a candidate solution requires a main cylinder pressure of 205 MPa, the system will correct it to 200 MPa and adjust the speed accordingly to maintain overall stress balance, thereby avoiding equipment overload or material failure.Based on this, the system further performs convergence discrimination based on the set of projected constraint points to determine whether the current iteration is sufficiently close to the global optimum. This process is accomplished by calculating the change in the objective function between adjacent iterations, obtaining a convergence discrimination index. For example, if the difference in the objective function value between two consecutive iterations is less than 0.1%, the algorithm is considered to have converged, and the search can be stopped. Finally, the system uses the convergence discrimination index to screen the optimal solution from the set of projected constraint points, selecting the parameter combination that provides the best material flow stability, lowest energy consumption, and fastest response as the final output optimized control parameters. For example, in a continuous aluminum profile production process, the system screened a set of control parameters through the above process: the main cylinder pressure was set to 150 MPa, the extrusion speed was set to 37 mm / s, and the parameter adjustment time interval was once every 6 seconds. This successfully restored the material flow stability to the target range, significantly improving product quality and process consistency. In summary, this step establishes a complete gradient projection optimization mechanism by initializing the gradient direction, calculating the step size, performing line search, and applying projection constraints and convergence criteria to the sequence of feasible solutions. This enables high-precision optimization of control parameters under complex process boundary constraints, providing crucial support for the efficient operation of the entire closed-loop feedback system.

[0040] The intelligent control method for aluminum profile extrusion molding in the embodiments of the present invention has been described above. The intelligent control system for aluminum profile extrusion molding in the embodiments of the present invention is described below. Please refer to [link / reference]. Figure 2 One embodiment of the intelligent control system for aluminum profile extrusion molding in this invention includes: Measurement module 21 is used to perform acoustic emission sensor array measurement and preprocessing on the die inlet area of ​​aluminum profile during the extrusion process to obtain a preprocessed acoustic emission signal sequence; The first analysis module 22 is used to perform time-frequency domain energy distribution analysis on the preprocessed acoustic emission signal sequence to obtain an acoustic emission energy spectrum. The second analysis module 23 is used to perform material flow stability analysis on the aluminum profile during the extrusion process based on the acoustic emission energy spectrum, and obtain flow analysis results; The judgment module 24 is used to perform multi-constraint optimization calculations on the main cylinder pressure and extrusion speed of the extruder based on the flow analysis results when the flow instability degree is not within the target flow stability range, so as to obtain optimized control parameters. The control module 25 is used to perform closed-loop feedback control of the extruder through a distributed PID controller based on the optimized control parameters, so as to ensure that the material flow stability of the aluminum profile during the extrusion process is within the target flow stability range.

[0041] In this embodiment, the specific implementation of each unit in the above system embodiment is described in the above method embodiment, and will not be repeated here.

[0042] Reference Figure 3 This invention also provides a computer device whose internal structure can be as follows: Figure 3 As shown, the computer device includes a processor, memory, display screen, input device, network interface, and database connected via a system bus. The processor provides computing and control capabilities. The memory includes a non-volatile storage medium and internal memory. The non-volatile storage medium stores the operating system, computer programs, and database. The internal memory provides an environment for the operation of the operating system and computer programs in the non-volatile storage medium. The database stores the data corresponding to this embodiment. The network interface is used to communicate with external terminals via a network connection. When the computer program is executed by the processor, it implements the above-described method.

[0043] Those skilled in the art will understand that Figure 3 The structures shown are merely block diagrams of some structures related to the present invention and do not constitute a limitation on the computer devices on which the present invention is applied.

[0044] An embodiment of the present invention also provides a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the above-described method. It is understood that the computer-readable storage medium in this embodiment can be a volatile readable storage medium or a non-volatile readable storage medium.

[0045] Those skilled in the art will understand that all or part of the processes in the above embodiments can be implemented by a computer program instructing related hardware. The computer program can be stored in a non-volatile computer-readable storage medium. When executed, the computer program can include the processes of the embodiments of the above methods. Any references to memory, storage, databases, or other media used in the present invention and embodiments can include non-volatile and / or volatile memory. Non-volatile memory can include read-only memory (ROM), programmable ROM (PROM), electrically programmable ROM (EPROM), electrically erasable programmable ROM (EEPROM), or flash memory. Volatile memory can include random access memory (RAM) or external cache memory. By way of illustration and not limitation, RAM is available in various forms, such as static RAM (SRAM), dynamic RAM (DRAM), synchronous DRAM (SDRAM), dual-rate SDRAM (SSRSDRAM), enhanced SDRAM (ESDRAM), synchronous link DRAM (SLDRAM), Rambus direct RAM (RDRAM), direct memory bus dynamic RAM (DRDRAM), and memory bus dynamic RAM, etc.

[0046] It should be noted that, in this document, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, apparatus, article, or method that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such process, apparatus, article, or method. Unless otherwise specified, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the process, apparatus, article, or method that includes that element.

[0047] The above description is merely a preferred embodiment of the present invention and does not limit the patent scope of the present invention. Any equivalent structural or procedural transformations made based on the content of the present invention's specification and drawings, or direct or indirect applications in other related technical fields, are similarly included within the patent protection scope of the present invention.

Claims

1. An intelligent control method for aluminum profile extrusion molding, characterized in that, Includes the following steps: Acoustic emission sensor array measurement and preprocessing were performed on the die inlet area of ​​aluminum profiles during the extrusion process to obtain a preprocessed acoustic emission signal sequence; Time-frequency domain energy distribution analysis was performed on the preprocessed acoustic emission signal sequence to obtain the acoustic emission energy spectrum. Based on the acoustic emission energy spectrum, the material flow stability of the aluminum profile during the extrusion process is analyzed, and the flow analysis results are obtained. If the flow analysis results show that the degree of flow instability is not within the target flow stability range, then based on the flow analysis results, multi-constraint optimization calculations are performed on the main cylinder pressure and extrusion speed of the extruder to obtain optimized control parameters. Based on the optimized control parameters, a distributed PID controller is used to perform closed-loop feedback control on the extruder to ensure that the material flow stability of the aluminum profile during the extrusion process is within the target flow stability range.

2. The intelligent control method for aluminum profile extrusion forming according to claim 1, characterized in that, The method involves measuring and preprocessing the acoustic emission signal sequence in the die inlet area during the extrusion process of the aluminum profile using an array of acoustic emission sensors, resulting in a preprocessed acoustic emission signal sequence, including: An initial acoustic emission data stream is obtained by real-time acquisition of elastic wave signals caused by changes in internal stress of the aluminum profile during the extrusion process using an array of acoustic emission sensors arranged in the mold inlet area. The initial acoustic emission data stream is pre-amplified to improve the signal-to-noise ratio, resulting in a preliminarily enhanced acoustic emission signal. Synchronous noise suppression is performed on each channel signal in the acoustic emission signal to eliminate environmental interference and common-mode noise, resulting in a suppressed acoustic emission signal; The suppressed acoustic emission signal is subjected to time-domain synchronous averaging to obtain a preprocessed acoustic emission signal sequence.

3. The intelligent control method for aluminum profile extrusion forming according to claim 1, characterized in that, The step of performing time-frequency domain energy distribution analysis on the preprocessed acoustic emission signal sequence to obtain the acoustic emission energy spectrum includes the following steps: The preprocessed acoustic emission signal sequence is subjected to a short-time Fourier transform to obtain a time-frequency domain signal matrix, and the time-frequency domain signal matrix is ​​divided into frequency bands to obtain multi-frequency band time-frequency signals. Calculate the energy value of each frequency band in the multi-band time-frequency signal, generate a frequency band energy sequence, and perform a sliding window integration on the time axis on the frequency band energy sequence to obtain the frequency band energy integral matrix; The singular values ​​in the frequency band energy integral matrix are decomposed to obtain singular value eigenvectors, and time-frequency domain energy weights are allocated based on the singular value eigenvectors to obtain a weighted time-frequency energy matrix. The weighted time-frequency energy matrix is ​​subjected to grid encryption processing using a preset two-dimensional interpolation algorithm to obtain the processed time-frequency energy matrix. The processed time-frequency energy matrix is ​​then subjected to grayscale mapping conversion to obtain the acoustic emission energy spectrum.

4. The intelligent control method for aluminum profile extrusion forming according to claim 1, characterized in that, The material flow stability analysis of the aluminum profile during the extrusion process, based on the acoustic emission energy spectrum, yields the following flow analysis results: The acoustic emission energy spectrum is subjected to spatial gradient enhancement processing to obtain a gradient-enhanced energy matrix, and the boundary of the gradient-enhanced energy matrix is ​​identified by a two-dimensional edge detection method to obtain the energy distribution boundary contour. Based on the energy distribution boundary contour, morphological opening operations are performed to obtain an optimized energy boundary curve, and geometric features in the energy boundary curve are extracted to obtain an energy boundary morphological feature curve. The energy boundary morphology characteristic curve is decomposed into multiple scales to obtain boundary fluctuation characteristic quantities, and nonlinear dynamic analysis is performed on the boundary fluctuation characteristic quantities to obtain flow state characteristic parameters. Multi-dimensional clustering analysis is performed on the flow state characteristic parameters to obtain the flow state category determination results. Based on the flow state category determination results, the stability of the aluminum profile during the extrusion process is evaluated to obtain the flow analysis results.

5. The intelligent control method for aluminum profile extrusion forming according to claim 1, characterized in that, Based on the flow analysis results, multi-constraint optimization calculations are performed on the main cylinder pressure and extrusion speed of the extruder to obtain optimized control parameters, including: Based on the flow analysis results, nonlinear dynamic response analysis was performed on the main cylinder pressure and extrusion speed of the extruder to obtain the pressure-velocity coupling characteristic matrix. Then, constraint condition mapping was performed on the pressure-velocity coupling characteristic matrix to obtain the process parameter boundary constraints. The process parameter boundary constraints are decoupled into multiple objectives by Lagrange relaxation decomposition to obtain a sequence of parameter optimization sub-problems. Based on the sequence of parameter optimization sub-problems, a hierarchical iterative solution is performed to obtain a set of candidate solutions for parameter adjustment. Based on the parameter adjustment candidate solution set, the stability criterion of the extruder is verified to obtain the parameter feasible solution sequence; Based on the feasible solution sequence of the parameters, gradient projection optimization calculation is performed to obtain optimized control parameters, wherein the optimized control parameters include the master cylinder pressure setpoint, the extrusion speed setpoint, and the parameter adjustment time interval.

6. The intelligent control method for aluminum profile extrusion forming according to claim 5, characterized in that, The process parameter boundary constraints are decoupled through Lagrange relaxation decomposition to obtain a sequence of parameter optimization sub-problems, including: Lagrange multiplier allocation is performed on the boundary constraints of the process parameters to obtain the multiplier allocation matrix, and dual gap evaluation is performed on the multiplier allocation matrix to obtain the dual convergence index. Based on the dual convergence index, the multiplier allocation matrix is ​​updated iteratively to obtain the updated multiplier vector. Based on the updated multiplier vector, the boundary constraints of the process parameters are relaxed to obtain the relaxed constraint conditions. The relaxed constraints are divided into sub-problem boundaries to obtain a sub-problem boundary division matrix. Based on the sub-problem boundary division matrix, the coupling variables are decoupled to obtain a set of decoupled variables. Based on the set of decoupled variables, sub-problems are constructed to obtain a sequence of parameter optimization sub-problems.

7. The intelligent control method for aluminum profile extrusion forming according to claim 6, characterized in that, The gradient projection optimization calculation based on the feasible solution sequence of the parameters to obtain the optimized control parameters includes: The gradient direction is initialized on the feasible solution sequence of the parameters to obtain the gradient direction initialization vector; The gradient step size in the feasible solution sequence of the parameters is calculated based on the gradient direction initialization vector to obtain the gradient step size sequence, and line search processing is performed based on the gradient step size sequence to obtain the set of line search termination points. The set of line search termination points is subjected to projection constraint processing to obtain a set of projection constraint points, and convergence discrimination is performed based on the set of projection constraint points to obtain a convergence discrimination index. The optimal solution is selected from the set of projection constraint points based on the convergence discrimination index to obtain the optimized control parameters.

8. An intelligent control system for aluminum profile extrusion molding, characterized in that, include: The measurement module is used to perform acoustic emission sensor array measurement and preprocessing on the die inlet area of ​​aluminum profiles during the extrusion process to obtain a preprocessed acoustic emission signal sequence; The first analysis module is used to perform time-frequency domain energy distribution analysis on the preprocessed acoustic emission signal sequence to obtain an acoustic emission energy spectrum. The second analysis module is used to perform material flow stability analysis on the aluminum profile during the extrusion process based on the acoustic emission energy spectrum, and obtain flow analysis results; The judgment module is used to perform multi-constraint optimization calculations on the main cylinder pressure and extrusion speed of the extruder based on the flow analysis results when the degree of flow instability is not within the target flow stability range, so as to obtain optimized control parameters. The control module is used to perform closed-loop feedback control of the extruder through a distributed PID controller based on the optimized control parameters, so as to ensure that the material flow stability of the aluminum profile during the extrusion process is within the target flow stability range.

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 method according to 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 a processor, it implements the steps of the method according to any one of claims 1 to 7.

Images