System and method for producing injection molded industrial brushes

Abstract

The application discloses a system and method for producing injection glue industrial brush by using an injection molding machine, relates to the technical field of industrial brush processing, generates a multi-parameter coupling index from obtained working parameters, and binds the working parameters, the multi-parameter coupling index and batch identification in a corresponding time period; after multi-angle imaging of the brush, mark information is added, preprocessing is performed, candidate areas where defects may exist are extracted, after a plurality of defect areas are detected, a morphology coordination index is constructed to quantize the overall defect degree, and defect data of each brush is obtained; after a comprehensive index structure is constructed, an injection molding process feature vector is constructed, nonlinear correlation and local sensitivity between each working parameter and the morphology coordination index are calculated by using kernel mapping and an AWCCA method, a working parameter interval is formulated, and a parameter modification suggestion is generated; in the production process, the parameters are continuously modified with small steps, the brush defect degree and the injection molding process smoothness are balanced, and more flexible regulation and control are realized.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of industrial brush processing, in particular to a system and method for producing an injection glue industrial brush by using an injection molding machine. BACKGROUND

[0002] At present, the production of industrial brushes mainly adopts injection molding process, which injects glue or thermoplastic material into the pre-arranged brush wires to realize one-time molding. In order to meet the requirements of durability and consistency of the brush in cleaning, painting and other special fields, the bonding quality of glue and brush wire in the production process is extremely high. Therefore, modern production lines introduce multi-sensor data acquisition, machine vision detection and data analysis technology to monitor the temperature, pressure and flow of the key parts of the injection molding machine in real time, and use image processing technology to detect defects of the molded products, realizing the whole process tracking and analysis of process data and product quality.

[0003] In the existing injection molding process, due to the dynamic fluctuations of temperature, pressure and flow and other parameters, the multi-parameter interaction is difficult to be captured and quantified in real time, resulting in uneven bonding between glue and brush wire, which is manifested as loose brush wire or incomplete glue coating. This problem mainly comes from the complex influence of process parameter fluctuations on the flow and solidification state of the glue, and the existing technology cannot effectively associate the sensor data with the visual detection results, so as to accurately identify and automatically control the key process parameters causing defects, thereby seriously affecting the product quality and production efficiency.

[0004] Therefore, the present application provides a system and method for producing an injection glue industrial brush by using an injection molding machine. SUMMARY

[0005] (I) Technical problems solved

[0006] In view of the deficiencies of the prior art, the present application provides a system and method for producing an injection glue industrial brush by using an injection molding machine. After multi-angle imaging of the brush, mark information is added for preprocessing, and candidate areas that may have defects are extracted. After detecting multiple defect areas, a morphological synergy index is constructed to quantify the overall defect degree and obtain defect data of each brush. After constructing a comprehensive index structure, an injection molding process feature vector is constructed, the nonlinear correlation and local sensitivity between each working parameter and the morphological synergy index are calculated by using kernel mapping and AWCCA method, the working parameter interval is determined and parameter modification suggestions are generated. In the production process, the parameters are continuously modified in small steps to balance the defect degree of the brush and the smoothness of the injection molding process, and more flexible control is realized. Thus, the technical problems described in the background art are solved.

[0007] (II) Technical solutions

[0008] In order to achieve the above purpose, the present application is realized by the following technical solutions:

[0009] The method for producing injection-molded industrial brush by injection molding machine, comprising,

[0010] The working parameters of the injection molding machine are collected by the sensor array arranged at the key positions of the injection molding machine, the multi-parameter coupling index MCI(t) is generated from the obtained working parameters, and the working parameters, the multi-parameter coupling index MCI(t) and the batch identification in the corresponding time period are bound;

[0011] After multi-angle imaging of the brush, mark information is added, preprocessing is performed, candidate areas where defects may exist are extracted, after multiple defect areas are detected, the morphology coordination index WPMSI is constructed to quantify the overall defect degree, and defect data of each brush is obtained.

[0012] After constructing the comprehensive index structure, the injection molding process feature vector X is constructed, the nonlinear correlation and local sensitivity between each working parameter and the morphology coordination index WPMSI are calculated by using kernel mapping and AWCCA method, and then the working parameter interval is determined and parameter modification suggestions are generated;

[0013] The working parameter modification list in the index structure is read, which is superimposed with the original parameter set to form a new set value, which is automatically issued to the injection molding machine through PLC, and the working parameters are monitored and self-adaptively corrected online.

[0014] Further, a sensor array is arranged at the key positions of the injection molding machine for obtaining real-time temperature T(t), pressure P(t) and flow parameter F(t), and generating a multi-parameter coupling index MCI(t), wherein,

[0015]

[0016] In the formula: x(τ) contains respectively represent the dynamic change rate of temperature, pressure and flow at time τ; the weight matrix W is a diagonal matrix, and the main diagonal elements are used to adjust the relative proportion of temperature, pressure and flow on the bonding effect; the decay kernel function K(t-τ)=e -λ(t-τ) , λ is a non-negative constant.

[0017] Further, a high-speed industrial camera is used to perform multi-angle or multi-plane high-precision imaging of the brush to generate raw image data I raw (x,y), a mark information is uniformly added to all product images of the batch, noise filtering and brightness equalization are performed on the raw image I raw (x,y) to generate a preprocessed image I proc (x,y), and on the basis of I proc (x,y), edge enhancement is adopted to extract candidate areas where defects may exist.

[0018] Further, multi-scale feature extraction is performed on the candidate region to obtain a set of multi-dimensional feature vectors Φ k , wherein k is the index of the candidate region; for each candidate region ROI k , its feature vector Φ k is extracted start is input into the trained defect judgment model, and if it is judged as a defect region, a defect record can be generated; after detecting multiple defect regions, the overall defect degree is quantified by the following morphology coordination index WPMSI.

[0019] Further, the same batch of sampling records and brush defect information are extracted, matched, and then a comprehensive index structure is constructed; the brush forming period [t end ] is searched, WPMSI is associated with T(t), P(t), F(t), MCI(t) in the interval [t start , t end ]; for each aligned product, the T(t), P(t), F(t), MCI(t) sequence in the time interval [t start , t end ] is selected, an injection molding process feature vector X is constructed, and the corresponding defect index WPMSI is regarded as an output scalar Y and introduced into the kernel function κ(·,).

[0020] Further, adaptive weighted canonical correlation analysis is introduced to quantify the degree of nonlinear correlation between X and Y, wherein the correlation R c is the optimal correlation measure of the two in the kernel space, and is defined as follows:

[0021]

[0022] wherein i represents the i-th product, X i is the injection molding process feature vector corresponding to the i-th product, Y i is the defect index WPMSI of the product, k(·,·) is the kernel function, ω i is the adaptive weighting coefficient, and u and v are direction vectors found in the kernel space.

[0023] Further, based on the correlation R c and local sensitivity obtained by analyzing each batch, the parameter range that has the greatest impact on defects is marked; if MCI(t) continuously exceeds the corresponding threshold value in a certain period of time and at the same time corresponds to a higher WPMSI defect index than expected, the corresponding high fluctuation interval is regarded as the key control object;

[0024] Based on the screened key parameters and historical data, a defect index model between WPMSI and process parameters is established by using artificial neural network ANN; the defect index model is taken as a target function, a parameter combination making WPMSI minimum is solved, specific parameter suggestions are output, and the suggested injection molding parameters are obtained.

[0025] Further, the suggested injection molding parameters output for the current batch are extracted, including the intervals that T(t), P(t) and F(t) need to be kept or limited, and the corresponding injection speed adjustment scheme, which are compared with the default settings of the original injection molding machine, and a parameter modification list is generated; the heating area, injection pressure unit, flow valve and other executive components of the injection molding machine are initially adjusted by using the programmable logic controller or the upper computer system, and a new initial adjustment parameter set is generated.

[0026] Further, T(t), P(t), F(t) and MCI(t) are reacquired, a dynamic tracking function is introduced to compare the key fluctuations during injection at a higher frequency, and if the following formula is true at any time τ:

[0027] |MCI(τ)-MCI ref |>Ω MC

[0028] Then it is judged that there is a process fluctuation overrun phenomenon, which needs to be adaptively corrected; wherein MCI ref is a reference coupling index benchmark, and Ω MC is an allowed fluctuation interval threshold; when the process fluctuation overrun or defect trend abnormality is detected, the adaptive control equation is automatically triggered and secondary correction is performed.

[0029] The system for producing industrial brush by injection molding machine, comprising,

[0030] The data acquisition unit acquires the working parameters of the injection molding machine by the sensor array arranged at the key positions of the injection molding machine, generates the multi-parameter coupling index MCI(t) from the acquired working parameters, and binds the working parameters and the multi-parameter coupling index MCI(t) in the corresponding time period with the batch identifier;

[0031] The defect detection unit adds mark information after multi-angle imaging of the brush, pre-processes the mark information, extracts the candidate area where defects may exist, constructs the morphological synergy index WPMSI to quantify the overall defect degree after detecting multiple defect areas, and obtains the defect data of each brush;

[0032] The suggestion output unit constructs the injection process feature vector X after constructing the comprehensive index structure, calculates the non-linear correlation and local sensitivity between each working parameter and the morphological synergy index WPMSI by using kernel mapping and AWCCA method, and then formulates the working parameter interval and generates the parameter modification suggestion;

[0033] A parameter correction unit reads a working parameter modification list in the index structure, superimposes it with the original parameter set to form a new setting value, automatically issues it to the injection molding machine through the PLC, and monitors and adaptively corrects the working parameters online.

[0034] (III) Beneficial effects

[0035] The present application provides a system and method for producing an injection glue industrial brush with the following beneficial effects:

[0036] 1. A multi-parameter coupling index MCI(t) is generated by continuous sampling and high-order operation to quantify the joint fluctuations of parameters such as temperature, pressure and flow, providing intuitive indicators for process state evaluation. When MCI(t) is large, it indicates that there are significant fluctuations in the process, which may cause unstable bonding of brush filaments and glue, providing early warning basis for subsequent defect detection and process control.

[0037] 2. Multi-scale feature extraction is used to effectively capture subtle changes in the contact area between the brush filaments and the glue, enhancing the detection capability of local glue leakage, loose brush filaments and other defects. The morphological synergy index WPMSI is used to quantitatively evaluate each candidate area, combining local wave packet energy and morphological complexity to accurately quantify defect severity. By finding the optimal projection in the kernel space, complex nonlinear mapping relationships can be captured while introducing adaptive weighting, which assigns higher weights to key products or products with larger fluctuations, improving the sensitivity to severe defects.

[0038] 3. Kernel mapping and adaptive weighted canonical correlation analysis are used to combine multi-dimensional process parameters into feature vectors, revealing the nonlinear relationship between temperature, pressure, flow, MCI(t) and defect index WPMSI. According to the correlation analysis and local sensitivity data, the parameter interval that has the greatest impact on defects is determined. When MCI(t) continuously exceeds the predetermined threshold in a certain time period and is accompanied by high WPMSI, this interval is marked as the key control object. The output control recommendations can be automatically executed by the parameter optimization control module, or can provide scientific basis for manual intervention, thereby reducing the defect rate and improving production consistency.

[0039] 4. Through the adaptive control equation, the parameters can be continuously corrected in small steps during the production process, avoiding large-scale adjustment that may cause process instability and loss of function C. WPMSI, MCI(t) and other indicators can be combined to balance the degree of brush defects and the smoothness of the injection molding process, achieving more flexible control.

[0040] 5. The control suggestions are applied to the production site through a parameter modification mechanism, and process fluctuations are fine-tuned through real-time monitoring and adaptive control equations to form a closed-loop feedback system. Ultimately, through multiple rounds of iterations, the brush defect rate is continuously reduced and the process stability is significantly improved, thereby enhancing product quality consistency and production efficiency. BRIEF DESCRIPTION OF THE DRAWINGS

[0041] Figure 1 This is a schematic flow chart of the method for producing a glue injection industrial brush according to the present invention;

[0042] Figure 2 This is a schematic diagram of the system structure for producing glue injection industrial brushes according to the present invention. DETAILED DESCRIPTION

[0043] The following will clearly and completely describe the technical solutions in the embodiments of the present invention in conjunction with the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative efforts are within the scope of protection of the present invention.

[0044] See also Figure 1 The present invention provides a method for producing a glue-injected industrial brush using an injection molding machine, comprising:

[0045] Step 1: The operating parameters of the injection molding machine are collected by a sensor array arranged at key positions of the injection molding machine, and a multi-parameter coupling index MCI(t) is generated from the acquired operating parameters. The operating parameters and the multi-parameter coupling index MCI(t) within the corresponding time period are then bound to the batch identifier.

[0046] Step 101: Temperature sensors, pressure sensors, and flow sensors are placed at key locations of the injection molding machine, such as the injection port, mold cavity, and injection channel, to construct a sensor array for acquiring real-time temperature T(t), pressure P(t), and flow parameter F(t), where t represents a time series. A unified PROCESS_DATA data structure is established for all collected temperature, pressure, and flow data, with fields containing {timestamp, T(t), P(t), F(t), batch identifier}.

[0047] When in use, temperature, pressure and flow sensors are deployed at the injection port, mold cavity and runner inlet of the injection molding machine to ensure real-time data acquisition at key locations, covering the entire process; Batch differentiation: The data of each production cycle is identified and segmented through batch identification (Batch), which facilitates traceability, comparison and fault analysis.

[0048] Step 102, continuously collect real-time values of temperature T(t), pressure P(t) and flow parameter F(t) during the operation of the injection molding machine according to the predetermined sampling frequency by the sensor array; and generate a multi-parameter coupling index MCI(t) to measure the multi-parameter interaction intensity from the starting time 0 to the current time t, which is defined as follows: let K(t-τ) = e -λ(t-τ) ; wherein: T(τ), P(τ), F(τ) represent the temperature, pressure and flow at time τ, respectively;

[0049] represents the instantaneous rate of change of the above parameters with respect to time, α T , α P , α F is the weight factor of temperature, pressure and flow, and x(τ) is a three-dimensional vector recording the instantaneous rate of change at time τ; W is a diagonal matrix corresponding to the relative weight of the three parameters in coupling analysis, K(t-τ) = e -λ(t-τ) represents that different changes at different times are given different attenuation weights, and λ is the attenuation coefficient,

[0050] Therefore, the form of the multi-parameter coupling index MCI(t) is as follows:

[0051]

[0052] Here,

[0053] Parameter meaning: x(τ): contains respectively represent the dynamic change rate of temperature, pressure and flow at time τ, which is used to capture the fluctuation amplitude and frequency of key parameters in the injection molding process; the weight matrix W is a diagonal matrix, and the main diagonal elements are used to adjust the relative proportion of temperature, pressure and flow on the bonding effect, so as to adapt to the needs of different material glue or brush silk characteristics; the attenuation kernel function K(t-τ) = e -λ(t-τ) , λ is a non-negative constant,

[0054] When used, the multi-parameter coupling index MCI(t) is generated by continuous sampling and high-order operation, which quantifies the joint fluctuation of temperature, pressure and flow and other parameters, and provides an intuitive index for process state evaluation; when MCI(t) is large, it indicates that there is significant fluctuation in the process, which may cause unstable bonding of brush silk and glue, providing a warning basis for subsequent defect detection and process control.

[0055] Step 103, write the obtained multi-parameter coupling index MCI(t) into the PROCESS_DATA data structure at the same time to form a record {timestamp T(t), P(t), F(t), MCI(t), batch identification}; after the end of each production batch, bind all {timestamp T(t), P(t), F(t), MCI(t)} records in the corresponding time period with the batch identification, and write the batch data into the PROCESS_DATA database;

[0056] If continuous production spans multiple batches, segmented processing is performed according to batch identification, and the records of each batch are independently stored to form a data collection of multiple batches in the same database. When the forming process of a specific brush product needs to be traced back, the corresponding temperature, pressure flow and MCI(t) records can be quickly retrieved according to the batch identification, and are associated with the defect detection results generated in the subsequent steps;

[0057] It should be noted that by integrating the weighted norm of the parameter change rate over the time interval [0, t] and combining the kernel function for attenuation, an index is obtained that comprehensively reflects the coupling strength of temperature, pressure and flow in the injection molding process. The larger the MCI(t) value, the higher the degree of fluctuation in that time period, which may lead to an increased risk of instability in the bonding of brush filaments and glue, which is highlighted for evaluation in subsequent processes such as machine vision defect detection.

[0058] When in use, all collected parameter data and calculated MCI(t) are stored in a structured manner according to batch (Batch) to form complete historical data records. The unified data format ensures the matching and joint analysis of subsequent machine vision detection data, and realizes the accurate comparison between process data and finished product quality.

[0059] Step two, after multi-angle imaging of the brush, add mark information, pre-process, extract candidate areas that may have defects, after detecting multiple defect areas, construct a morphology synergy index WPMSI to quantify the overall defect degree, and obtain defect data of each brush;

[0060] Step 201, after the brush product is demolded, a high-speed industrial camera is used to perform multi-angle or multi-plane high-precision imaging of the brush to generate raw image data I raw (x,y), to ensure that the image data is accurately matched with the corresponding PROCESS_DATA, batch identification (BatchID) and product serial number (ProductID) are uniformly added to all product images of the batch, and the image data is temporarily stored in the VISUAL_DATA structure, and the fields are shown as follows: {timestamp, BatchID, ProductID, I raw(x, y); by attaching BatchID and ProductID to each image, the subsequent defect analysis results can be seamlessly compared with the same batch of temperature T(t), pressure P(t), flow F(t) and other data, and multi-angle imaging can also improve the observation accuracy of brush wire position and glue wrapping state, avoiding the blind area caused by single angle;

[0061] In use, high-speed industrial cameras are used for multi-angle shooting to ensure that high-quality original images of each key part of the brush are obtained, and each finished product is attached with batch (Batch) and product number (Item) to ensure accurate association with PROCESS_DATA data, laying a foundation for subsequent defect quantitative analysis.

[0062] Step 202, preprocessing the original image I raw (x, y) to generate a preprocessed image I proc (x, y) on the basis of I proc (x, y), edge enhancement and morphological operation methods are used to extract candidate regions (ROI) that may have defects;

[0063] To accurately represent the defect situation between the brush wire and the glue (such as loose brush wire or incomplete glue wrapping), high-order methods such as discrete wavelet transform or adjustable scale convolution kernel are introduced to perform multi-scale feature extraction on the candidate region (ROI), obtaining a group of multi-dimensional feature vectors Φ k , where k is the index of the candidate region; through image preprocessing, interference factors in imaging are eliminated, and the distinguishability of the glue wrapping boundary and the brush wire contour is enhanced. Compared with traditional threshold segmentation, multi-scale feature extraction can better capture the adhesion differences between glue and brush wire in subtle places, and is especially suitable for detecting local glue leakage or gap areas.

[0064] Step 203, for each candidate region ROI k , its feature vector φ k is input into the trained defect judgment model (such as a classifier based on machine learning or deep learning), if it is judged as a defect region, a defect record can be generated, for example, {BatchID, ProductID, k, DefectType}; after detecting multiple defect regions, the overall defect degree is quantified by the following morphological synergy index WPMSI, where,

[0065] Let Ω k represent the coordinate domain corresponding to the kth defect region in the image plane, φ k is the multi-dimensional feature vector extracted for the defect region (obtained by wavelet transform, morphological operation or other high-order feature extraction methods), W(φ k , ψ s,θ), which represents the projection of the feature vector Φ k onto a set of wavelet bases ψ s,θ The resulting wavelet packet coefficient vector is denoted as W(Φ

[0066] In addition, a morphological metric function is introduced to characterize the shape complexity or contour irregularity of the defect region Ω k , which can be defined according to the specific requirements of industrial brush defects, for example:

[0067] Perimeter-area ratio: Convexity coefficient: measures the contour complexity based on the difference between Ω k and its minimum bounding rectangle (or minimum convex hull); Fractal dimension: quantifies the morphological richness of the defect by calculating the filling property of the region at multiple scales;

[0068] The morphological synergy index WPMSI is defined as follows:

[0069]

[0070] where m is the total number of current brush defect regions, ||W(Φ k , ψ s,θ )||2 represents the L2 norm calculation of the wavelet packet coefficients of the kth defect region to measure the energy accumulation of the feature in different wavelet bases, γ is a nonlinear amplification factor for the wavelet packet coefficient energy, which emphasizes the defect mode with higher intensity, The morphological complexity of the defect region is embedded in the energy measurement in exponential form, μ is the morphological weight coefficient; β is the coefficient for the quadratic nonlinear amplification of the overall integral result, which can further enhance the discrimination of serious defects, dA represents the spatial coordinate integration of the defect region Ω k ;

[0071] W(Φ k , ψ s,θ ) represents the high-frequency or fine structure information of the defect region, as well as the texture characteristics of the brush filament and the glue contact surface. If the energy value of this vector is large, it means that there are abnormal fluctuations in the multi-scale or multi-direction frequency band of this region, which usually corresponds to more serious or more Quantifying the morphological complexity of the defect, if the defect contour is irregular or diffuse, the value will increase accordingly, indicating that the defect is more difficult to handle or more likely to cause functional damage (such as a larger range of loose filaments);

[0072] Further, the influence of the morphological complexity degree in the exponential function can give higher weight to defect areas with complex morphology (such as large-area glue leakage or uneven wrapping), and γ and β are nonlinear amplification coefficients used to highlight serious defects in the accumulation process. By adjusting these two parameters, the system can maintain flexibility when dealing with a large number of small defects or a small number of large defects, The use of an exponential form amplifies the influence of morphological complexity on energy value, and the more complex the morphology, the larger the exponential product, which represents a potentially greater negative impact of the defect on product performance.

[0073] The detected defect data information of each brush includes {BatchID, ProductID, WPMSI, DefectMap}, as well as the type, coordinate range, and severity level of each defect area, which is written into the VISUAL_DATA structure to form a set of visual inspection results corresponding to the PROCESS_DATA.

[0074] In use, image clarity is improved and environmental interference is reduced through noise filtering, brightness equalization, edge enhancement, and other preprocessing methods; candidate region precise positioning: multi-scale feature extraction (such as discrete wavelet transform or adjustable scale convolution kernel) is used to effectively capture subtle changes in the contact area between the bristles and the glue, enhancing the detection capability for local glue leakage, loose bristles, and other defects. The morphological synergy index WPMSI is used to quantitatively evaluate each candidate region, combining local wave packet energy and morphological complexity to accurately quantify defect severity; the calculated WPMSI is associated with Batch and Item and stored in VISUAL_DATA, ensuring effective comparison with the process parameters in PROCESS_DATA for subsequent accurate defect cause analysis.

[0075] Step three, after constructing the comprehensive index structure, construct the injection molding process feature vector X, calculate the nonlinear correlation and local sensitivity between each working parameter and the morphological synergy index WPMSI using kernel mapping and AWCCA method, and then determine the working parameter range and generate parameter modification suggestions;

[0076] Step 301, extract PROCESS_DATA under the same batch identifier BatchID, which contains T(t), P(t), F(t), and MCI(t) sampling records over time, and obtain brush defect information such as WPMSI and defect area mapping from VISUAL_DATA under the corresponding BatchID;

[0077] The two are matched according to batch identification BatchID and product serial number ProductID, and a comprehensive index structure ANALYSIS_DATA is constructed, which is {BatchID, ProductID, T(t), P(t), F(t), MCI(t), WPMSI, …};

[0078] To match the defect index WPMSI of a specific brush with its corresponding forming process, the forming period [t start , t end ] of the brush needs to be retrieved in PROCESS_DATA, and WPMSI is associated with T(t), P(t), F(t), MCI(t) in the interval [t start , t end ], which can be analyzed subsequently to ensure that the analysis results are based on accurate timing correspondence;

[0079] When constructing VISUAL_DATA, it should be ensured that all records are indexed using the same batch identification and product serial number, so that all parameters and defect information of a specific brush can be quickly retrieved in subsequent steps, and the accuracy of the time window mapping directly affects the subsequent inference of the causal relationship between temperature, pressure, flow fluctuation and defect generation;

[0080] When used, it can be ensured that the forming data of each brush product corresponds to the visual detection results one by one, providing a data basis for establishing the causal relationship between defects and process parameters subsequently.

[0081] Step 302, for each aligned product, select T(t), P(t), F(t), MCI(t) sequence in the time interval [t start , t end ], construct injection molding process feature vector X, and introduce corresponding defect index WPMSI as output scalar Y into kernel function κ(·,·), map X to high-dimensional feature space to capture the potential nonlinear relationship between parameters and defect index;

[0082] Introduce adaptive weighted canonical correlation analysis to quantify the nonlinear correlation between X and Y, where the correlation R c is the optimal correlation measure of the two in the kernel space, and the higher the value, the more significant the correlation between the injection molding process features and the defect index, which is defined as follows:

[0083]

[0084] Where: i represents the ith product, X i is the injection molding process feature vector corresponding to the ith product, and Y iwhere WPMSI is the defect index of the product, κ(·,·) is a kernel function (such as RBF kernel or polynomial kernel) used to realize nonlinear mapping, ω is an adaptive weighting coefficient, u and v are direction vectors found in the kernel space, and used to maximize the projection correlation between X and Y. i where WPMSI is the defect index of the product, κ(·,·) is a kernel function (such as RBF kernel or polynomial kernel) used to realize nonlinear mapping, ω is an adaptive weighting coefficient, u and v are direction vectors found in the kernel space, and used to maximize the projection correlation between X and Y.

[0085] R c , PartialSensitivity, …} ; PartialSensitivity can be further subdivided into the influence weights of temperature, pressure, flow rate, and coupling index MCI(t), and used to identify which process parameters are strongly correlated with defects. c , PartialSensitivity, …} ; PartialSensitivity can be further subdivided into the influence weights of temperature, pressure, flow rate, and coupling index MCI(t), and used to identify which process parameters are strongly correlated with defects.

[0086] By finding the optimal projection in the kernel space, complex nonlinear mapping relationships can be captured, and adaptive weighting is introduced, so that key products or products with greater fluctuations are assigned higher weights, and the sensitivity to serious defects is improved. By writing the result data uniformly into ANALYSIS_DATA, it can be directly connected with subsequent steps, so that parameter name conflicts or data inconsistencies will not occur.

[0087] In use, kernel mapping and adaptive weighting canonical correlation analysis (AWCCA) are used to combine multi-dimensional process parameters into feature vectors, revealing the nonlinear relationship between temperature, pressure, flow rate, MCI(t), and defect index WPMSI. The local sensitivity of each parameter to defects is calculated, and its comprehensive correlation degree is quantified, providing a basis for identifying key influencing factors and abnormal intervals.

[0088] Step 303, based on the correlation R c and local sensitivity obtained by analyzing each batch (or a certain time window), the parameter range that has the greatest impact on defects (such as temperature T within a certain interval, pressure P near a certain critical value) is marked. If MCI(t) continuously exceeds the corresponding threshold value for a certain period of time, and at the same time corresponds to a WPMSI defect index higher than expected, the high fluctuation interval is taken as the key control object.

[0089] Based on the screened key parameters and historical data, an artificial neural network ANN is used to establish a defect index model between WPMSI and process parameters; the defect index model is taken as a target function to solve the parameter combination that minimizes WPMSI, wherein T(t), P(t), and F(t) need to meet the physical constraints of the process (such as the upper and lower limits of temperature, pressure, etc.); at this time, specific parameter suggestions such as keeping the temperature curve in [T min ,T max ], and the pressure peak does not exceed a certain critical value are output, and the recommended injection molding parameters are obtained;

[0090] The recommendations are written into the recommendation field in the index structure ANALYSIS_DATA, and by comparing the significant correlation between different parameter intervals and the defect index, the optimal or near-optimal combination that reduces the defect rate can be sought in the multi-dimensional parameter space, and the output control recommendations can be automatically executed in subsequent steps or provided for artificial intervention decision-making reference.

[0091] According to the correlation analysis and local sensitivity data, the parameter interval that has the greatest impact on defects (such as temperature within a certain range, pressure near a critical value) is determined; at the same time, when MCI(t) continuously exceeds a predetermined threshold for a certain period of time and is accompanied by high WPMSI, the interval is marked as a key control object; the output control recommendations can be automatically executed by the parameter optimization control module or can provide scientific basis for artificial intervention, thereby reducing the defect rate and improving production consistency.

[0092] Step four, read the working parameter modification list in the index structure, superimpose it with the original parameter set to form new set values, and automatically issue them to the injection molding machine through PLC for online monitoring and adaptive correction of working parameters;

[0093] Step 401, extract the recommended injection molding parameters output for the current batch, including the intervals that T(t), P(t), and F(t) may need to maintain or limit, and the corresponding injection speed adjustment scheme, compare them with the original injection molding machine default settings, and generate a parameter modification list, which can be defined as: {BatchID, ΔT, ΔP, ΔF, SpeedAdj, …}, wherein ΔT, ΔP, and ΔF are the adjustment amounts of temperature, pressure, and flow relative to the original set values, SpeedAdj is the correction coefficient of injection speed, etc.

[0094] The heating zone, injection pressure unit, flow valve, and other execution components of the injection molding machine are initially adjusted using a programmable logic controller or an upper computer system, a new initial adjustment parameter set is generated, and it is registered and associated with the corresponding BatchID for subsequent comparison and tracing. The difference between the old set value and the new set value is recorded for subsequent verification of adjustment effect and continuous optimization.

[0095] Read the parameter modification list from ANALYSIS_DATA, compare with the existing original injection molding parameter set, calculate the new target set value, use PLC or host computer system to issue the new set value to the temperature control unit, pressure unit, flow valve and injection speed adjustment mechanism of the injection molding machine, realize the initial process adjustment; Write the successfully effective adjustment value in the initial adjustment parameter set format to the data management device, and associate with the corresponding Batch, provide basis for subsequent comparison and traceability.

[0096] Step 402, after the start of a new round of production, repeat the sensor collection in step one: real-time acquisition of T(t), P(t), F(t), MCI(t), form new process data, introduce dynamic tracking function to compare the key fluctuations during injection at higher frequency, if at any time τ the following formula is true:

[0097] |MCI(τ)-MCI ref |>Ω MC

[0098] Then judge that there is a process fluctuation overrun phenomenon, which needs to be adaptively corrected; Wherein MCI ref is the reference coupling index benchmark, Ω MC is the allowable fluctuation interval threshold;

[0099] When detecting process fluctuation overrun or defect trend abnormality, automatically trigger the following adaptive control equation for secondary correction of ΔT, ΔP, ΔF, etc., wherein,

[0100] Wherein: ΔT old , ΔP old , ΔF old are the parameter adjustment amounts in the current execution, ΔT new , ΔP new , ΔF new are the new adjustment amounts after correction η T , η P , η F are the step coefficients of each parameter, which determine the speed of correction action; Indicates the partial derivative of a certain comprehensive loss function C with respect to temperature, pressure and flow, which is used to quantify the comprehensive influence on brush defect indicators (such as WPMSI) and injection process fluctuations (MCI(t));

[0101] Update the actual execution parameter record of the current batch at any time τ in the production process with the new parameter value after correction, form {BatchID, τ, T exec , P exec , F exec,MCI(τ)}, so that subsequent defect detection and associated data analysis can trace back to the corresponding real process status according to the specific correction moment;

[0102] Through the adaptive control equation, the parameters can be corrected in small steps and continuously during the production process to avoid process instability caused by one-time large-scale adjustments. Indicators such as WPMSI and MCI(t) can be combined in the loss function C to balance the degree of brush defects and the smoothness of the injection molding process, achieving more flexible control.

[0103] After the new settings take effect, the sensor continues to collect T(t), P(t), F(t), and MCI(t) in real time and writes them to PROCESS_DATA, forming a continuous process data stream. A dynamic tracking function is used to monitor if MCI(t) exceeds the preset threshold at any time. The adaptive control equation is automatically triggered to make small step corrections to ΔT, ΔP, and ΔF to ensure that the production process always remains within the target parameter range. The parameters and corresponding timestamps after each correction are recorded in the data management device, providing detailed data for real-time feedback and subsequent reanalysis.

[0104] Step 403: After completing this production run, repeatedly obtain new defect detection indicators (such as WPMSI), associate the new detection indicators with T(t), P(t), F(t), MCI(t) and all corrected parameters corresponding to this production run, and write them into ANALYSIS_DATA.

[0105] If the test results still indicate a high proportion of uneven brush bonding, step three is further called to output optimization suggestions again based on the new collected data and defect indicators, and then return to this step (step four is used for the next round of adjustment and feedback). Through multiple iterations, a relative balance and optimal area are reached between the fluctuation of the injection molding process and the defect rate of the finished product, so as to achieve a stable improvement in the bonding quality of the brush filament and the glue; this process actually forms a closed loop of data collection, defect detection, correlation analysis, parameter optimization, re-collection and re-analysis. When the system has been verified through multiple iterations, the defect rate is significantly reduced and tends to be stable, and the optimal injection molding parameters of this round can be solidified in stages for reference in subsequent production.

[0106] After the production cycle ends, machine vision detection is performed again to obtain the latest defect index WPMSI, and these detection results are associated with the current executed process parameter data and written into ANALYSIS_DATA; if the defect rate is still high, step 303 is returned to perform associated analysis again based on the latest data, output a new round of parameter modification suggestions, realize multi-round closed-loop optimization, and until the defect rate is stably reduced and reaches the expected target. The control suggestions are applied to the production site through a parameter modification issuing mechanism, and the process fluctuation is fine-tuned through real-time monitoring and a self-adaptive control equation to form a closed-loop feedback system; finally, through multiple iterations, the brush defect rate is continuously reduced and the process stability is significantly improved, and the product quality consistency and production efficiency are enhanced.

[0107] Please refer to Figure 2 The present application provides a system for producing industrial brush with injection molding machine, comprising,

[0108] The data acquisition unit acquires the working parameters of the injection molding machine through the sensor array arranged at the key positions of the injection molding machine, generates a multi-parameter coupling index MCI(t) from the acquired working parameters, and binds the working parameters and the multi-parameter coupling index MCI(t) in the corresponding time period with the batch identification;

[0109] The defect detection unit adds mark information after multi-angle imaging of the brush, pre-processes the image, extracts the candidate area where defects may exist, constructs a morphology coordination index WPMSI to quantify the overall defect degree after detecting multiple defect areas, and obtains the defect data of each brush;

[0110] The suggestion output unit constructs an injection molding process feature vector X after constructing a comprehensive index structure, calculates the non-linear correlation and local sensitivity between each working parameter and the morphology coordination index WPMSI by using kernel mapping and AWCCA method, and then formulates the working parameter interval and generates parameter modification suggestions;

[0111] The parameter correction unit reads the working parameter modification list in the index structure, superimposes it with the original parameter set to form a new set value, automatically issues it to the injection molding machine through PLC, and monitors and adaptively corrects the working parameters online.

[0112] Those skilled in the art can realize that the units and algorithm steps of the examples described in combination with the embodiments disclosed herein can be realized by electronic hardware or a combination of computer software and electronic hardware. Whether the functions are realized in hardware or software depends on the specific application and design constraints of the technical solution. Professional technicians can use different methods to realize the described functions for each specific application, but such implementation should not be considered beyond the scope of the present application.

[0113] Those skilled in the art can clearly understand the specific working process of the system, the device and the unit described above can refer to the corresponding process in the foregoing method embodiments for description convenience and brevity, and details are not described herein.

[0114] In several embodiments provided in the present application, it should be understood that the disclosed system, device and method can be implemented in other ways. For example, the device embodiments described above are merely illustrative, for example, the division of the units is only some logical function division, and actual implementation can have another division manner, for example, a plurality of units or components can be combined or integrated into another system, or some features can be ignored or not executed. In addition, the coupling or direct coupling or communication connection between the units shown or discussed can be indirect coupling or communication connection through some interfaces, devices or units, and can be electrical, mechanical or other forms.

[0115] The units described as separate components can or can not be physically separated, and the components shown as units can or can not be physical units, that is, can be located in one place, or can be distributed on a plurality of network units. Part or all of the units can be selected to achieve the purpose of the embodiment scheme according to actual needs.

[0116] The above is only a specific implementation of the present application, but the protection scope of the present application is not limited thereto, any person skilled in the art can easily think of changes or replacements within the technical range disclosed in the present application, which should be covered in the protection scope of the present application. Therefore, the protection scope of the present application should be subject to the protection scope of the claims.

Claims

1. A method for producing glue-injected industrial brushes using an injection molding machine, characterized in that: include, The operating parameters of the injection molding machine are collected by a sensor array arranged at key locations of the injection molding machine. The multi-parameter coupling index (MCI(t)) is generated from the acquired operating parameters. The operating parameters and the multi-parameter coupling index (MCI(t)) within the corresponding time period are then bound to the batch identifier. After multi-angle imaging of the brush, marking information is added and pre-processed to extract candidate areas with possible defects. After detecting multiple defective areas, the morphological synergy index (WPMSI) is constructed to quantify the overall defect degree and obtain defect data for each brush. After building a comprehensive index structure, the injection molding process characteristic vector X is constructed. The nonlinear correlation and local sensitivity between each working parameter and the morphological synergy index WPMSI are calculated using kernel mapping and AWCCA methods, and then the working parameter range is determined and parameter modification suggestions are generated. Read the working parameter modification list in the index structure, superimpose it with the original parameter set to form a new set value, and automatically send it to the injection molding machine through the PLC to monitor and adaptively correct the working parameters online; A sensor array is arranged at key locations of the injection molding machine to obtain real-time temperature T(t), pressure P(t), and flow parameters F(t), and generate a multi-parameter coupling index MCI(t), where: Where: x(τ): contains Respectively represent the dynamic change rate of temperature, pressure and flow at time τ; the weight matrix W is a diagonal matrix, and the main diagonal elements are Used to adjust the relative weight of the influence of temperature, pressure and flow on the bonding effect; attenuation kernel function K(t-τ) = e -λ(t-τ) ,λ is a non-negative constant; Use high-speed industrial cameras to perform high-precision imaging of the brush at multiple angles or planes to generate raw image data. raw (x,y), add label information to all finished product images of this batch, and add label information to the original image I raw (x,y) performs noise filtering and brightness equalization to generate the preprocessed image I proc (x,y), in I proc Based on (x,y), edge enhancement is used to extract candidate areas with defect risks; Perform multi-scale feature extraction on the candidate area to obtain a set of multi-dimensional feature vectors Φ k , where k is the index of the candidate region; for each candidate region ROI k , its eigenvector Φ k Input the trained defect determination model. If it is determined to be a defect area, a defect record can be generated. After multiple defect areas are detected, the overall defect degree is quantified by the following morphological synergy index WPMSI. The morphological synergy index WPMSI is defined as follows: Where: m is the total number of current brush defect areas, || W(Φ k ,ψ s,θ )||2 represents the L2 norm calculation of the wave packet coefficient of the k-th defect area to measure the energy accumulation of the feature under different wavelet bases. γ is the nonlinear amplification factor of the wave packet coefficient energy, which is used to emphasize the defect mode with higher intensity. The morphological complexity of the defect area is embedded into the energy measurement in an exponential form. μ is the morphological weight coefficient; β is the coefficient of the quadratic nonlinear amplification of the overall integral result, which further enhances the discrimination of severe defects. dA represents the Ω k The spatial coordinate integral of W(Φ k ,ψ s,θ ) characterizes the high-frequency or fine structural information of the defect area, as well as the texture characteristics of the contact surface between the brush and the glue. If the energy value of this vector is large, it means that there are abnormal fluctuations in the multi-scale or multi-directional frequency bands in this area.

2. The method for producing a glue-injected industrial brush using an injection molding machine according to claim 1, characterized in that: Extract the same batch of sampling records and brush defect information, match them and build a comprehensive index structure; retrieve the brush molding period [t start , t end ], associate WPMSI to [t start ,t end ] interval T(t), P(t), F(t), MCI(t); for each aligned product, select the time interval [t start , t end ], construct the injection molding process feature vector X, and regard the corresponding defect index WPMSI as the output scalar Y to introduce the kernel function κ(·,·) to map X to a high-dimensional feature space.

3. The method for producing a glue-injected industrial brush using an injection molding machine according to claim 2, characterized in that: Adaptive weighted canonical correlation analysis is introduced to quantify the nonlinear correlation between X and Y, where the correlation R c is the optimal correlation measure between the two in the kernel space, which is defined as follows: Where: i represents the i-th product, X i is the injection molding process feature vector corresponding to the i-th product, Y i is the defect index WPMSI of the product, κ(·,·) is the kernel function, ω i is the adaptive weighting coefficient, u and v are the direction vectors to be searched in the kernel space.

4. The method for producing a glue-injected industrial brush using an injection molding machine according to claim 3, wherein: Based on the correlation R obtained from the analysis of each batch c With local sensitivity, the parameter range with the greatest impact of defects will be marked. If MCI(t) continues to exceed the corresponding threshold for a period of time and corresponds to a higher-than-expected WPMSI defect index, the corresponding high-fluctuation range will be the focus of regulation; Based on the screened key parameters and historical data, an artificial neural network (ANN) is used to establish a defect index model between WPMSI and process parameters. The defect index model is used as the objective function to solve the parameter combination that minimizes WPMSI, output specific parameter recommendations, and obtain recommended injection molding parameters.

5. The method for producing a glue-injected industrial brush using an injection molding machine according to claim 4, characterized in that: Extract the recommended injection molding parameters output for the current batch, including the ranges that need to be maintained or limited for T(t), P(t), and F(t), as well as the corresponding injection speed adjustment plan, compare them with the original injection molding machine default settings, and generate a parameter modification list; use a programmable logic controller or host computer system to perform initial adjustments on the injection molding machine's heating zone, injection pressure unit, flow valve and other actuators to generate a new set of initial adjustment parameters.

6. The method for producing a glue-injected industrial brush using an injection molding machine according to claim 5, characterized in that: Re-acquire T(t), P(t), F(t) and MCI(t), and introduce a dynamic tracking function to compare the key fluctuations during injection molding at a higher frequency. If the following equation holds at any time τ: |MCI(t)-MCI ref |>Oh MC It is judged that there is a process fluctuation exceeding the limit at this time, and adaptive correction is required; among them, MCI ref is the reference coupling index benchmark, Ω MC is the allowable fluctuation range threshold; when it is detected that the process fluctuation exceeds the limit or the defect trend is abnormal, the adaptive control equation is automatically triggered and a secondary correction is performed.

7. A system for producing glue-injected industrial brushes using an injection molding machine, applying the method for producing glue-injected industrial brushes using an injection molding machine according to any one of claims 1 to 6, characterized in that: include, The data acquisition unit collects the working parameters of the injection molding machine using a sensor array arranged at key positions of the injection molding machine, generates a multi-parameter coupling index MCI(t) based on the acquired working parameters, and binds the working parameters and the multi-parameter coupling index MCI(t) within the corresponding time period to the batch identifier; The defect detection unit performs multi-angle imaging of the brush, adds marking information, and then performs pre-processing to extract candidate areas where defects may exist. After detecting multiple defective areas, it constructs the morphological synergy index (WPMSI) to quantify the overall defect degree and obtain defect data for each brush. The output unit is recommended. After building a comprehensive index structure, the injection molding process feature vector X is constructed. The nonlinear correlation and local sensitivity between each working parameter and the morphological synergy index WPMSI are calculated using the kernel mapping and AWCCA method, and then the working parameter range is formulated and parameter modification suggestions are generated. The parameter correction unit reads the working parameter modification list in the index structure, superimposes it with the original parameter set to form a new set value, and automatically sends it to the injection molding machine through the PLC to monitor and adaptively correct the working parameters online.

Images