Brick quality monitoring method and system based on data analysis and storage medium

Abstract

The application relates to the technical field of data analysis, and discloses a brick quality monitoring method and system based on data analysis and a storage medium. The method comprises the following steps: collecting vibration, pressure, displacement, image and air pressure data on a brick production line to obtain a data set; performing domain synchronous averaging and edge detection on the data set to obtain brick surface defect feature frequencies, brick contour feature vectors and quality anomaly indexes; performing analysis through a convolutional neural network to obtain a brick defect type judgment result and a defect severity score; performing process parameter optimization to obtain a device control strategy and transmitting the device control strategy to each execution mechanism to obtain motor control signals, servo positioning instructions, contour adjustment instructions and management layer optimization decision information. The application improves the sensitivity and accuracy of brick defect detection, enables the brick production system to continuously self-optimize, effectively responds to internal changes and external disturbances, and significantly improves the quality stability and production efficiency of brick production.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of data analysis, and particularly relates to a brick quality monitoring method and system based on data analysis and a storage medium. BACKGROUND

[0002] In the traditional brick production process, quality monitoring mainly relies on manual visual inspection and simple physical measurement. This method is not only inefficient, but also has a high misjudgment rate, and it is difficult to timely find potential problems such as polygonal defects and contour errors on the surface of the brick. On the high-speed production line, manual detection cannot meet the fine quality control requirements of modern production, resulting in a large number of unqualified products flowing into the market, increasing the risk of subsequent construction and material waste.

[0003] Although the existing automatic monitoring technology has realized the mechanization of part of the detection link, it still has obvious defects. These systems often only focus on the optimization of mechanical parameters of the conveying process, and lack data collection, analysis and feedback control of the whole production process. At the same time, the existing equipment cannot effectively handle the problem of scattering of stacked materials, resulting in uneven stress on the conveying belt, which gradually deforms after long-term use, affecting the transmission efficiency. More seriously, materials are prone to reverse and backflow during upward transportation, not only causing material waste, but also making it difficult to ensure the consistency of the quality of the final product. SUMMARY

[0004] The present application provides a brick quality monitoring method and system based on data analysis and a storage medium, which improves the sensitivity and accuracy of brick defect detection, enables the brick production system to continuously self-optimize, effectively deals with internal changes and external disturbances, and significantly improves the quality stability and production efficiency of brick production.

[0005] In a first aspect, the present application provides a brick quality monitoring method based on data analysis, which comprises:

[0006] Collecting vibration, pressure, displacement, image and air pressure data on the brick production line to obtain a data set;

[0007] Performing domain synchronous averaging and edge detection on the data set to obtain brick surface defect feature frequency, brick contour feature vector and quality anomaly index;

[0008] Inputting the brick surface defect feature frequency, the brick contour feature vector and the quality anomaly index into a convolutional neural network for analysis to obtain a brick defect type determination result and a defect severity score;

[0009] Performing process parameter optimization based on the defect type determination result and the defect severity score to obtain a device control strategy;

[0010] The device control strategy is transmitted to each execution mechanism to perform closed-loop feedback to obtain motor control signals, servo positioning instructions, profile adjustment instructions, and management layer optimization decision information.

[0011] In a second aspect, the application provides a brick quality monitoring system based on data analysis, which comprises:

[0012] A collection module is configured to collect vibration, pressure, displacement, image, and air pressure data on a brick production line to obtain a data set.

[0013] A detection module is configured to perform domain synchronous averaging and edge detection on the data set to obtain brick surface defect feature frequencies, brick profile feature vectors, and quality anomaly indicators.

[0014] An analysis module is configured to input the brick surface defect feature frequencies, the brick profile feature vectors, and the quality anomaly indicators into a convolutional neural network for analysis to obtain brick defect type determination results and defect severity scores.

[0015] A parameter optimization module is configured to perform process parameter optimization based on the defect type determination results and the defect severity scores to obtain a device control strategy.

[0016] A transmission module is configured to transmit the device control strategy to each execution mechanism to perform closed-loop feedback to obtain motor control signals, servo positioning instructions, profile adjustment instructions, and management layer optimization decision information.

[0017] In a third aspect, a computer readable storage medium is provided, which stores instructions when executed on a computer, so that the computer performs the above-mentioned brick quality monitoring method based on data analysis.

[0018] In the technical scheme provided in the application, the application collects multi-dimensional data of vibration, pressure, displacement, image and air pressure by configuring a multi-sensor network on a brick production line, forms a complete production process data set, realizes comprehensive monitoring of the whole process of brick production compared with the traditional monitoring method relying on only a single parameter, and effectively avoids misjudgment caused by single parameter abnormalities. The application applies advanced algorithms such as synchronous average processing of vibration signals, Canny edge detection analysis of images and Fourier descriptor extraction of contour features, can accurately identify the frequency of brick surface defect features, the contour feature vector and the quality abnormality index, and greatly improves the sensitivity and accuracy of brick defect detection. The application inputs the extracted features into a convolutional neural network for deep analysis, through multi-layer feature extraction and classification regression, not only can accurately determine various types of brick defects, but also can give a defect severity score, providing an accurate basis for subsequent process parameter optimization. Based on the defect analysis result, the application constructs a correlation model of brick quality and process parameters, searches in a high contribution parameter subspace through a differential evolution algorithm, and designs a specific compensation strategy for different types of defects, realizing accurate dynamic adjustment of process parameters in the brick production process. The application adopts a three-level control architecture of equipment layer, process layer and production management layer, realizes hierarchical cooperation from basic control to high-level decision-making, ensures accurate execution of control instructions through iterative contour error compensation algorithm and dynamic production plan adjustment strategy, at the same time, establishes a three-level alarm mechanism of early warning, warning and emergency warning, and improves the stability and safety of the system. Through control performance index calculation and parameter dynamic adjustment, the application forms a complete closed loop of data acquisition, analysis processing, defect diagnosis, parameter optimization and control execution, enables the brick production system to continuously self-optimize, effectively deals with internal changes and external disturbances, and significantly improves the quality stability and production efficiency of brick production. BRIEF DESCRIPTION OF DRAWINGS

[0019] In order to more clearly illustrate the technical solutions of the embodiments of the application, the drawings needed in the embodiment description will be briefly introduced. Obviously, the drawings in the following description are some embodiments of the application, and other drawings can be obtained by those skilled in the art without creative labor.

[0020] Figure 1 An embodiment schematic diagram of the brick quality monitoring method based on data analysis in the embodiment of the application;

[0021] Figure 2 An embodiment schematic diagram of the brick quality monitoring system based on data analysis in the embodiment of the application. DETAILED DESCRIPTION

[0022] The embodiment of the present application provides a brick quality monitoring method and system based on data analysis and a storage medium. The terms "first", "second", "third", "fourth" and the like (if any) in the specification and claims of the present application and the above-described drawings are used to distinguish similar objects, and do not have to be used to describe a specific order or sequence. It should be understood that the data used in this way can be interchanged under appropriate circumstances, so that the embodiments described herein can be implemented in an order other than that illustrated or described herein. In addition, the term "comprising" or "having" and any variations thereof are intended to cover non-exclusive inclusion, for example, a process, method, system, product or device including a series of steps or units does not have to be limited to those steps or units clearly listed, but can include other steps or units that are not clearly listed or inherent to these processes, methods, products or devices.

[0023] For ease of understanding, the specific flow of the embodiment of the present application is described below. Please refer to Figure 1 One embodiment of the brick quality monitoring method based on data analysis in the embodiment of the present application comprises the following steps.

[0024] Step S101, collecting vibration, pressure, displacement, image and air pressure data on the brick production line to obtain a data set;

[0025] It can be understood that the execution subject of the present application can be a brick quality monitoring system based on data analysis, and can also be a terminal or a server, and the specific execution subject is not limited herein. The embodiment of the present application takes the server as the execution subject for example.

[0026] Specifically, a number of vibration sensors are evenly installed on the surface of the conveyor belt of the brick production line, which are evenly distributed along the length direction of the conveyor belt to capture the vibration signals generated during the operation of the conveyor belt. The vibration signals reflect the dynamic changes of the bricks during the conveying process, including the stress condition of the conveyor belt, the stability of the brick movement, and the interaction characteristics between the conveyor belt and the bricks. A number of pairs of pressure sensors are installed on the first and second extrusion plates on the production line to monitor the stress state of the extrusion plates in real time, obtain the extrusion force on the bricks, the uniformity of the material during the pushing process, and the pressure change of the pushing assembly, which helps to analyze the stress characteristics of the bricks during the processing. At the same time, the displacement sensor is fixed to the bottom of the vibration strip to accurately collect the displacement change data of the vibration strip at different time points. The data reflects the displacement amplitude, movement frequency and vibration state change process of the vibration strip, which helps to analyze the relationship between the surface defects of the bricks and the vibration frequency. In terms of visual data collection, two high-speed industrial cameras are installed at the end of the conveyor belt, which capture image data of the brick surface from different angles. The image data can reflect the contour features, surface defects and geometric shape deviations of the bricks. In order to obtain the dynamic pressure change inside the cavity, an air pressure sensor is installed inside the cavity, which continuously monitors the dynamic data of the air pressure change inside the cavity with time, providing information for analyzing the material flow, air pressure fluctuation and internal environment change during the production of the bricks. The vibration signals, force data, displacement data, brick surface images and air pressure data are time-synchronized to ensure accurate matching of different types of data at the same time point. By adding a uniform time stamp to each type of data during data collection and aligning and synchronizing the time during data storage, a multi-modal data set is formed.

[0027] Step S102, domain synchronous averaging and edge detection are performed on the data set to obtain the brick surface defect feature frequency, the brick contour feature vector and the quality abnormality index.

[0028] Specifically, the vibration signals collected in the dataset are segmented according to the running period of the conveying belt. The domain synchronous average signal of the periodic defects on the brick surface is obtained by averaging the vibration signals of each period. The domain synchronous average signal can eliminate random noise and non-periodic interference, thereby extracting the characteristic signal related to the periodic defects on the brick surface. The residual signal is extracted from the synchronous average signal, and the envelope spectrum of the residual signal is calculated. The characteristic frequency of the envelope spectrum can accurately represent the characteristics of the defects on the brick surface. In terms of image processing, the brick surface images in the dataset are preprocessed, including gray scale conversion, histogram equalization, and image enhancement operations to improve the contrast and detail features of the images. After image preprocessing, the Canny edge detection algorithm is used to detect the edges of the target surface image, and the edge contour information of the brick is extracted. The Fourier descriptor is used to analyze the features of the extracted brick contour, and the shape features from the 1st to the 20th order are calculated. These shape features can reflect the geometric shape changes of the brick. The Hausdorff distance between the actual contour and the standard contour is calculated, which can quantify the error between the brick contour and the standard contour, forming the brick contour feature vector. Feature extraction is performed on the air pressure data and force data. For air pressure data, the first and second derivatives are calculated to identify the inflection points, extreme points, and stable intervals of the pressure changes inside the cavity. These features can reflect important information such as material flow, air pressure fluctuations, and air cavity stability during the production process. Meanwhile, for force data, torque analysis is performed to analyze the force at different time points of the pushing assembly, and the force imbalance degree is calculated. The force imbalance degree reflects the balance of the force between the first and second extrusion plates during the pushing process, which is directly related to the uniformity of the brick material distribution. The brick surface defect characteristic frequency, brick contour feature vector, and brick material distribution features are normalized, and the different source data are integrated into the target feature vector through feature fusion. The feature fusion process ensures consistency of various data in the same feature space and eliminates the scale difference caused by different data types. The isolation forest algorithm is used to set the anomaly score threshold to detect the target feature vector. The isolation forest algorithm constructs multiple randomly divided decision trees to calculate the anomaly score of each sample, and sets the threshold according to the anomaly score to distinguish normal samples and abnormal samples. By detecting the target feature vector, various quality abnormal conditions of the brick during the production process are identified, forming the quality abnormality index.

[0029] For air pressure data, noise suppression and smoothing processing are performed, and random noise in the signal is eliminated by introducing a sliding mean filter or Gaussian filter method to obtain a smoothed air pressure data sequence. The five-point central difference method is used to calculate the derivative of the air pressure data sequence, and the first and second derivatives are calculated by the weighted difference of five adjacent points to obtain a derivative matrix representing the air pressure change rate and acceleration in the cavity. The first derivative reflects the speed of air pressure change over time, while the second derivative reveals the acceleration characteristics of air pressure change. Feature point detection is performed on the derivative matrix to identify key air pressure change feature points, including inflection points, extreme points, and stable intervals. The inflection points are identified at the intersection of the first and second derivatives, and the extreme points are determined according to the sign change of the second derivative, while the stable region is identified by analyzing the interval where the first derivative approaches zero, and a set of air pressure change feature points is constructed. These feature points reflect the key stages of air pressure change during the brick production process. At the same time, for the force data in the data set, spectral analysis is performed to extract the force amplitude of the first and second extrusion plates, and Fourier transform is used to convert the force data from the time domain to the frequency domain to identify the main frequency components and amplitude information of the extrusion plate force signal. This spectral analysis process can reveal the periodic changes and energy distribution at different frequencies in the force signal, and the extracted extrusion plate force amplitude is used to construct the force time series features. Based on the force time series features, the force imbalance degree is calculated, the force imbalance degree is calculated by comparing the force of the first and second extrusion plates to evaluate the balance of the forces on both sides during the pushing process, and a force imbalance degree curve is drawn to obtain a force distribution feature map reflecting the uniformity of material distribution and the dynamic change of extrusion force during the pushing process. The air pressure change feature point set and the force distribution feature map are time-synchronized and superimposed for analysis, and the two sets of data are aligned based on the timestamp and matched on the same time axis. The air pressure features and force imbalance features are quantified by a weighted fusion algorithm, which dynamically adjusts the contribution of each feature point based on its impact weight on material distribution uniformity, and calculates the material distribution uniformity index. The closer the material distribution uniformity index is to the ideal value, the more uniform the material distribution, and vice versa, indicating that the material distribution is abnormal or uneven.

[0030] Step S103, input the brick surface defect feature frequency, brick profile feature vector and quality abnormality index into the convolutional neural network for analysis to obtain the brick defect type determination result and defect severity score;

[0031] Specifically, the brick surface defect feature frequency, brick profile feature vector and quality anomaly index are combined into a unified feature matrix. The feature matrix is normalized and standardized. The data is converted to the same numerical range by mean-variance normalization or minimum-maximum scaling of each column feature, eliminating the influence of scale difference between different features on model training, to obtain normalized network input data for neural network analysis. The normalized network input data is input into the convolutional neural network for deep feature extraction. The structure of the convolutional neural network consists of three convolutional layers, and the convolution kernel size of each convolutional layer is 3x3, 5x5 and 7x7 respectively. Different scale convolution kernels can capture different scale local patterns in the brick feature matrix, thereby extracting more representative deep mapping features. The convolutional layer convolves the input data layer by layer through a sliding window and uses a nonlinear activation function for feature mapping to obtain a deep representation of the brick surface defects, profile features and anomaly indicators in a high-dimensional feature space. Through convolution operation, the local correlation of the feature matrix is effectively extracted, and the expression ability of the model for complex features is improved. The deep mapping features are input into two max-pooling layers of the convolutional neural network for dimension reduction. The pooling window size of the max-pooling layer is 2x2, and the information in the feature space is reduced and compressed through the pooling operation, thereby retaining key features and removing redundant information, maximizing the reduction of feature dimension while retaining the expression ability of the brick defect features, to obtain a brick feature representation vector. The brick feature representation vector is input into the defect classifier of the convolutional neural network, which classifies the pre-defined F types of brick defects. The output layer of the classifier uses a softmax activation function to calculate the probability distribution of each defect type, and a cross-entropy loss function is used to optimize and adjust the classification error of the model. The softmax activation function can convert multiple class scores output by the network into a probability distribution, so that the predicted probability of each class is equal to 1, thereby determining the defect type of the brick according to the maximum probability. At the same time, the difference between the model prediction result and the true label is evaluated by calculating the cross-entropy loss function, and the model is updated by the gradient through the back propagation algorithm, thereby continuously optimizing the performance of the classification model and improving the recognition accuracy of different brick defect types. In order to realize the scoring of defect severity, the brick feature representation vector is input into the modified output layer of the convolutional neural network. The modified output layer is different from the defect classifier and uses a sigmoid activation function regression network. The network output is limited to between 0 and 1 through the sigmoid activation function, and a continuous score representing the severity of the defect is obtained. The regression network optimizes the error of the defect severity score through the root mean square error loss function, adjusts the network parameters by comparing the mean square error between the predicted score output by the model and the true defect severity label, and improves the accuracy of the defect severity prediction.The defect type determination result and the defect severity score of the brick are obtained simultaneously through joint training of the convolutional neural network.

[0032] In step S104, process parameter optimization is performed based on the defect type determination result and the defect severity score to obtain a device control strategy.

[0033] Specifically, according to the defect type determination result and the defect severity score, a correlation analysis between the brick quality and the process parameters is performed. In the analysis process, the defect characteristics are matched with the key process parameters to obtain a process parameter vector containing multiple key process parameters, including the conveyor belt speed, the initial positions of the first and second extrusion plates, the spring pressure, the baffle height, the vibration bar height, and the opening and closing time of the electronic stop valve. These process parameters directly affect the material distribution state, forming accuracy, and surface defect generation during the brick production process. Based on the process parameter vector, a mathematical model for process parameter optimization is constructed. By establishing a mapping relationship between the brick quality objective function and the process parameters, the model quantifies the influence of different parameters on the quality objective, and through sensitivity analysis, the contribution degree of each process parameter to the quality objective is calculated to generate a parameter contribution degree matrix. This matrix reflects the comprehensive influence of different process parameters on the brick quality, production efficiency, and material utilization rate. Through analysis of the contribution degree matrix, the process parameters are divided into a high contribution degree group and a low contribution degree group. High contribution degree parameters have a stronger influence on the quality objective, so in the parameter optimization process, a differential evolution algorithm is executed in the subspace composed of high contribution degree parameters to search for the optimal solution. The differential evolution algorithm updates individuals in the population through continuous iteration, combining mutation, crossover, and selection operations to find the optimal solution in the high-dimensional parameter space and obtain the optimal process parameter configuration. The optimization process calculates the objective function value after each iteration and dynamically adjusts the search range based on the improvement degree of the quality objective to ensure that the final parameter combination meets the quality objective requirements. Based on the optimal process parameter configuration, corresponding compensation strategies are designed for different types of brick defects to allow for differentiated control during the production process according to the specific defect type. For polygonal defects, the conveyor belt speed and vibration bar height are adjusted to change the brick motion state, thereby compensating for shape abnormalities caused by material accumulation or uneven conveying; for contour errors, the initial positions of the first and second extrusion plates and the baffle position are adjusted to accurately control the contour features of the brick during forming, thereby eliminating contour deviations caused by uneven extrusion; and for material mixing unevenness, the opening and closing time of the electronic stop valve is adjusted to control the uniform distribution of materials in the cavity, preventing internal defects caused by material accumulation or uneven mixing. Through defect-specific compensation schemes, the finished product rate and quality consistency of the bricks are effectively improved. The defect-specific compensation schemes are converted into equipment control strategies to ensure accurate implementation of the compensation strategies at each equipment level.The device control strategy includes various specific control instructions, wherein the speed control instruction adjusts the motor output torque through a PID controller to ensure that the conveying belt speed reaches the expected optimal value; the position control instruction controls the positions of the first and second extrusion plates through the precise positioning signal of the servo motor to achieve contour error compensation; the spring pressure control instruction adjusts the spring pressure through a pressure adjusting mechanism to maintain the force balance of the pushing process; the baffle height control instruction dynamically adjusts the baffle height through the control of the air pressure system to ensure that the optimal pressure state is reached during the material conveying process; and the timing control instruction of the electronic stop valve controls the material flow by accurately setting the switching time to achieve the optimization of material distribution uniformity.

[0034] In step S105, the device control strategy is transmitted to each execution mechanism to execute closed-loop feedback, and motor control signals, servo positioning instructions, contour adjustment instructions and management layer optimization decision information are obtained.

[0035] Specifically, the device control strategy is decomposed into three levels, including device layer control, process layer control, and production management layer control. Among them, the device layer control is responsible for directly controlling the actuators on the brick production line, ensuring that the production equipment operates according to the set process parameters; the process layer control is responsible for quality monitoring and dynamic compensation of the production process based on real-time data to ensure that the brick quality meets the expected standards; the production management layer control focuses on the optimization and adjustment of the overall production plan, and optimizes parameters, adjusts plans, and warns of quality based on long-term production data trends. In the device layer control, the control strategy is decomposed into instructions, which are converted into specific control commands and transmitted to each actuator through a field bus (such as PROFINET or EtherCAT). This process establishes a device-based control network that networks motor, extrusion plate, vibration components, electronic stop valve, and baffle actuators to achieve real-time transmission and feedback of control instructions. In terms of motor control, vector control algorithm is applied to accurately control the speed and torque of the motor. The vector control algorithm converts the three-phase current of the AC motor into a DC component for decoupling control, thereby achieving accurate adjustment of motor speed and torque. This control signal dynamically adjusts the motor output through a closed-loop feedback mechanism to ensure that the conveyor belt speed reaches the optimal value. At the same time, for extrusion plate position control, servo positioning control is implemented to achieve high-precision pushing positioning by tracking the position of the servo motor of the pushing component. The servo positioning instruction is based on the optimized initial position of the extrusion plate and combined with real-time displacement feedback for position correction, ensuring that the extrusion plate maintains the best position during the pushing process. In the process layer control, the iterative contour error compensation algorithm is used to dynamically adjust the contour error of the brick surface. The motion synchronization point of the pushing component and the vibration component is calculated based on the motion synchronization control principle to ensure that the pushing action and the vibration action are consistent in time and space. Based on the motion synchronization point, a brick contour error compensation function is designed. This compensation function analyzes the variation law of the contour error to generate a compensation signal, which is superimposed on the original control signal to realize real-time compensation of the contour error. This process generates a contour adjustment instruction, which is transmitted to the pushing component and the vibration component through the servo control system to achieve accurate control of the brick contour shape, thereby reducing contour deviation and improving the contour accuracy of finished bricks. In the production management layer control, a dynamic production plan adjustment strategy is used to set up a condition response mechanism such as quality trend prediction, parameter self-optimization program, and emergency adjustment program. When consecutive B-block bricks have the same type of defects, the parameter self-optimization program is triggered to recalculate the optimal process parameter configuration and dynamically adjust the device control parameters, thereby achieving adaptive optimization of the production process. When serious defects are detected, the emergency adjustment program is triggered, which quickly corrects abnormal situations in the production process by pausing production, checking device status, and resetting key parameters.According to the results of quality trend prediction, the production plan is dynamically adjusted, and potential quality risks are early warned in the production process, so as to realize preventive control. The dynamic production plan adjustment mechanism identifies potential quality fluctuations through trend analysis and prediction of historical production data, and adjusts key process parameters in advance to prevent the occurrence of quality abnormalities. Through the multi-level control architecture of the equipment layer, the process layer and the production management layer, closed-loop feedback control of the brick production process is realized. The motor control signal adjusts the motor speed in real time through the vector control algorithm, the servo positioning instruction ensures the accurate position control of the extrusion plate and the pushing assembly, the contour adjustment instruction realizes the dynamic compensation of contour error based on the motion synchronization point, and the management layer optimization decision information ensures the stability and efficiency of the production process through the mechanisms of quality trend prediction, parameter self-optimization and emergency adjustment.

[0036] In the embodiments of the present application, a multi-element sensor network is configured on the brick production line to collect multi-dimensional data such as vibration, pressure, displacement, image and air pressure, forming a complete production process data set. Compared with the traditional monitoring method relying on only a single parameter, the present application realizes comprehensive monitoring of the whole process of brick production, effectively avoiding misjudgment caused by single parameter abnormalities. The present application applies advanced algorithms such as synchronous average processing of vibration signals, Canny edge detection analysis of images and Fourier descriptor extraction of contour features, which can accurately identify the frequency of brick surface defect features, contour feature vectors and quality abnormal indicators, greatly improving the sensitivity and accuracy of brick defect detection. The present application inputs the extracted features into a convolutional neural network for deep analysis, through multi-layer feature extraction and classification regression, not only can accurately determine various types of brick defects, but also can give a defect severity score, providing an accurate basis for subsequent process parameter optimization. Based on the defect analysis results, the present application constructs a correlation model between brick quality and process parameters, searches in the high-contribution parameter subspace through the differential evolution algorithm, and designs specific compensation strategies for different types of defects, realizing accurate dynamic adjustment of process parameters in the brick production process. The present application adopts a three-level control architecture of equipment layer, process layer and production management layer, realizes hierarchical collaboration from basic control to high-level decision-making, through iterative contour error compensation algorithm and dynamic production plan adjustment strategy, ensures the accurate execution of control instructions, and at the same time establishes a three-level alarm mechanism of early warning, warning and emergency warning, improves the stability and safety of the system. Through control performance index calculation and parameter dynamic adjustment, the present application forms a complete closed loop of data acquisition, analysis processing, defect diagnosis, parameter optimization and control execution, so that the brick production system can continuously self-optimize, effectively cope with internal changes and external disturbances, and significantly improve the quality stability and production efficiency of brick production.

[0037] In a specific embodiment, the process of step S101 can specifically include the following steps:

[0038] n vibration sensors are evenly installed on the surface of the conveying belt on the brick production line, and vibration signals representing the dynamic characteristics of the conveying belt are collected through the vibration sensors;

[0039] m pairs of pressure sensors are installed on the first and second extrusion plates on the brick production line, and force data representing the force state of the extrusion plates are collected through the pressure sensors;

[0040] The displacement sensor is fixed to the bottom of the vibration strip, and displacement data representing the movement state of the vibration strip are collected through the displacement sensor;

[0041] Two high-speed industrial cameras installed at the end of the conveying belt are used to collect multi-angle brick surface images, and an air pressure sensor installed inside the cavity is used to collect air pressure data representing the dynamic changes of the internal pressure of the cavity;

[0042] The vibration signals, force data, displacement data, brick surface images and air pressure data are time-synchronized and stored to obtain the data set.

[0043] Specifically, according to the length of the conveyor belt and the actual situation of the brick conveying path, the number n of vibration sensors and their uniform distribution positions are determined. The vibration sensors are installed at different positions on the surface of the conveyor belt to ensure real-time monitoring of the dynamic characteristics of the conveyor belt during the transportation of the bricks, including the vibration amplitude, frequency change, and abnormal vibration state of the conveyor belt. By uniformly distributing the vibration sensors along the length of the conveyor belt, it is ensured that the data covers all key stages of the conveying process, preventing inaccurate monitoring results due to local data loss. The vibration signals collected by the vibration sensors reflect the stress conditions of the conveyor belt, the friction changes between the bricks and the conveyor belt during the conveying process, and the uniformity of the material distribution. At the same time, pressure sensors are installed at the key stress positions of the first and second extrusion plates of the pushing assembly, and their positions are ensured to cover the stress area. The installation positions of the pressure sensors are accurately calculated according to the stress distribution of the extrusion plates to accurately monitor the forces acting on the extrusion plates during the pushing process. Each pair of pressure sensors is installed symmetrically on the first and second extrusion plates, respectively, to measure the stress on the extrusion plates during the pushing process, analyze whether the stress on the extrusion plates is uniform, and monitor the unbalanced stress problems occurring during the pushing process. The force data collected by the pressure sensors reflect the magnitude of the force acting on the material during the extrusion process, the pushing pressure change, and the fluctuation of the material density and hardness. The displacement sensor is fixed to the bottom of the vibration strip, and the displacement data representing the motion state of the vibration strip are collected by the displacement sensor. The displacement sensor is installed at the center position of the bottom of the vibration strip to capture the displacement change information of the vibration strip at different time points in real time. By monitoring the motion state of the vibration strip through the displacement sensor, the displacement amplitude, motion period, and transient change of the vibration strip are obtained, which helps to analyze the relationship between the running state of the vibration strip and the surface defects of the bricks. At the same time, the displacement data reflect the vibration abnormalities of the vibration strip during the pushing process. Two high-speed industrial cameras are installed on both sides of the end of the conveyor belt to image the surface of the bricks from different angles. The high-speed industrial cameras have high frame rate and resolution to ensure that the details of the brick surface are captured during high-speed motion and to avoid image quality degradation due to motion blur. The installation angle of the cameras is accurately adjusted according to the geometry of the brick surface to cover different areas of the brick surface, ensuring that complete brick surface images are obtained. Through multi-angle image acquisition, the contour information, edge features, and surface defect information of the brick surface are extracted. At the same time, in order to monitor the pressure change inside the cavity, a pressure sensor installed inside the cavity is used to collect pressure data representing the dynamic change of the pressure inside the cavity, reflecting the pressure fluctuation inside the cavity, the density change of the material, and the expansion and compression of the gas during the pushing process of the material. These data help to analyze the uniformity of the material distribution and the stability of the brick forming. The vibration signals, force data, displacement data, brick surface images, and pressure data are time-synchronized and stored to obtain the data set.A high-precision time synchronization mechanism is used to add uniform timestamps to all data acquisition devices and align different data types in time. The time synchronization mechanism uses high-precision network clock protocols (such as PTP) or GPS time synchronization technology to uniformly time-stamp the data from each sensor, ensuring that data from different sources is stored under the same time reference. The vibration signals, force data, displacement data, brick surface images, and air pressure data collected after time synchronization are transmitted to the central data acquisition controller through a high-speed communication bus and stored in the data management system. The data storage format uses a block storage method, with each sampling period's data being block labeled and stored as a complete data set.

[0044] In a specific embodiment, the process of performing step S102 can specifically include the following steps:

[0045] The vibration signals in the data set are segmented according to the running period of the conveyor belt and the synchronous average signal of N periods is calculated to obtain the domain synchronous average signal representing the periodic defects on the brick surface;

[0046] The residual signal is extracted from the domain synchronous average signal and its envelope spectrum is calculated to obtain the defect feature frequency of the brick surface;

[0047] The brick surface images in the data set are subjected to grayscale conversion, histogram equalization, and image enhancement to obtain target surface images, and the target surface images are subjected to Canny edge detection to obtain the brick contour;

[0048] The Fourier descriptor is used to calculate the shape features of the first to twentieth order of the brick contour and calculate the Hausdorff distance between the actual contour and the standard contour to obtain the brick contour feature vector;

[0049] The first and second derivatives are calculated based on the air pressure data in the data set and the inflection points, extreme points, and stable intervals are extracted, and the moment analysis is performed based on the force data in the data set and the force imbalance degree is calculated to obtain the brick material distribution feature;

[0050] The brick surface defect feature frequency, brick contour feature vector, and brick material distribution feature are normalized and fused to obtain the target feature vector, and the isolation forest algorithm is used to set an abnormal score threshold to detect the target feature vector to obtain the quality anomaly index.

[0051] Specifically, the vibration signals collected in the data set are preprocessed, and the vibration signals are segmented based on the running period T of the conveyor belt. The running period of the conveyor belt is calculated by monitoring the speed signal of the conveyor belt or according to the time interval of the brick passing through a specific sensor position. The vibration signals are segmented according to the period T, and each segment contains a fixed number of sampling points. The segmented signals are represented as VbThe segmented sequence of (t). To eliminate random noise and periodic interference, after acquiring the segmented signal of N periods, the signals of all periods are calculated by synchronous averaging, and the signals of each period are accumulated and averaged to obtain the domain synchronous average signal. This signal can effectively retain the characteristics related to the periodic defects of the brick surface, while suppressing non-periodic noise interference. The residual signal is extracted from the domain synchronous average signal and its envelope spectrum is calculated to obtain the defect characteristic frequency of the brick surface. The calculation of the residual signal is by difference operation between the original vibration signal and the domain synchronous average signal, to obtain the residual signal irrelevant to the periodic characteristics. These residual signals contain high-frequency characteristic information of the defects on the brick surface. By envelope spectrum analysis of the residual signal, the signal is converted to the frequency domain and the characteristic frequency related to the defects on the brick surface is extracted. These characteristic frequencies reflect the periodic changes of the polygonal defects, profile errors and micro defects on the brick surface. At the same time, the brick surface images in the data set are preprocessed, including gray scale conversion, histogram equalization and image enhancement, to obtain the target surface image. The original brick image is converted from RGB color space to grayscale image to eliminate the interference of color information, so as to focus more on the edge features of the image. The histogram equalization method is used to adjust the gray scale distribution of the image to enhance the contrast and detail features of the image, so that the defects and profile of the brick surface are more prominent. The image enhancement technology is applied, and the image is processed by using adaptive filtering or edge enhancement algorithm, so that the boundary of the brick surface is clearer. After the preprocessing operation, the target surface image is input into the Canny edge detection algorithm, the gradient of the image is calculated and double threshold screening is performed to extract the profile edge of the brick surface and generate a binary image reflecting the profile characteristics of the brick. After obtaining the profile information of the brick, Fourier descriptor is used to extract the shape features of the brick profile, and the first to 20th order shape features are calculated. Fourier descriptor is a mathematical tool for converting profile shape information to frequency domain for analysis. The extracted profile boundary curve is converted into a set of Fourier coefficients, and the first 20 Fourier coefficients are analyzed to describe the geometric shape characteristics of the brick. Fourier descriptor captures the overall shape information of the profile and the change of local features. By calculating the Hausdorff distance between the actual profile and the standard profile, the size of the brick profile error is quantified. Hausdorff distance is a measure for calculating the maximum and minimum distance between two point sets, which measures the similarity between the actual profile and the standard profile, and generates the profile feature vector of the brick. At the same time, the first and second derivatives of the air pressure data in the data set are calculated, and the inflection points, extreme points and stable intervals are extracted to represent the dynamic change characteristics of the air pressure in the cavity. After smoothing the air pressure data, the central difference method is used to calculate the derivative of the air pressure data, and the first derivative reflects the air pressure change rate and the second derivative reflects the air pressure acceleration.The extraction of the inflection point is based on the sign change of the first derivative and the second derivative, the identification of the extreme point is determined by judging the point where the derivative is zero, and the identification of the stable interval is extracted by analyzing the time period where the air pressure change rate is close to zero. At the same time, the force data in the data set is subjected to moment analysis, and the force imbalance degree is calculated to represent the characteristics of the brick material distribution. The moment analysis is realized by calculating the force change of the first and second extrusion plates during the pushing process, and the force imbalance degree is calculated by comparing the force difference of the two extrusion plates, effectively reflecting the uniformity of the material distribution during the pushing process, thereby providing a basis for analyzing the internal material state of the brick. The surface defect feature frequency of the brick, the brick contour feature vector and the brick material distribution feature are normalized, and are integrated into a target feature vector through feature fusion. The normalization process maps data of different dimensions into the same numerical range by linear scaling of the feature data, thereby eliminating the differences between different feature scales and preventing high-dimensional features from deviating the results in the calculation process. Feature fusion is to combine different types of normalized features, integrate the information of different features into a unified feature vector through feature splicing or weighted fusion, and form a brick feature representation. The target feature vector is input into the isolation forest algorithm for anomaly detection, and the isolation forest algorithm calculates the anomaly score of each sample by constructing multiple randomly divided decision trees, and judges whether the sample is abnormal according to the set anomaly score threshold. Through the isolation forest algorithm for detecting the target feature vector, the quality abnormality in the brick production process is identified, and the quality abnormality index is obtained.

[0052] In a specific embodiment, the execution step calculates the first derivative and the second derivative from the air pressure data in the data set and extracts the inflection point, the extreme point and the stable interval, and at the same time, performs moment analysis on the force data in the data set and calculates the force imbalance degree to obtain the brick material distribution characteristics, which can specifically include the following steps:

[0053] The air pressure data in the data set is subjected to noise suppression and smoothing processing to obtain a preprocessed air pressure data sequence;

[0054] The first derivative and the second derivative are calculated from the preprocessed air pressure data sequence by five-point central difference method to obtain a derivative matrix representing the air pressure change rate and acceleration in the cavity;

[0055] The derivative matrix is subjected to feature point detection algorithm to identify the inflection point, the extreme point and the stable interval to obtain an air pressure change feature point set;

[0056] The force data in the data set is subjected to frequency spectrum analysis and the force amplitudes of the first and second extrusion plates are extracted to obtain extrusion plate force time sequence features;

[0057] Based on the stress timing characteristics of the extruded plate, the stress imbalance degree is calculated and the stress imbalance degree curve with time is drawn to obtain the stress distribution characteristic map;

[0058] The air pressure change feature point set is time-synchronized and superimposed with the stress distribution characteristic map for analysis, and a material distribution uniformity index is quantified by a weighted fusion algorithm to obtain the brick material distribution characteristics.

[0059] Specifically, the collected air pressure data is denoised and filtered. Due to the influence of factors such as equipment vibration, environmental fluctuations and material flow during the brick production process, the original data is accompanied by high-frequency noise and abnormal fluctuations. In the data preprocessing stage, the sliding mean filter or Gaussian filter method is used to smooth the air pressure data. The mean value of the data is calculated through the sliding window or the adjacent data is weighted and averaged by using the Gaussian weight, so as to remove the high-frequency noise in the short period and retain the main trend of the air pressure change. At the same time, in order to avoid data distortion, a reasonable filter window length is set, so that the smoothed air pressure data truly reflects the change process of the air pressure in the cavity, and the smoothed air pressure data sequence after noise suppression is obtained. The first derivative and the second derivative of the preprocessed air pressure data sequence are calculated by the five-point central difference method, and the derivative matrix representing the air pressure change rate and acceleration in the cavity is obtained. The five-point central difference method is a numerical differentiation method, which calculates the derivative of the data through five adjacent time points, thereby improving the accuracy of the differential calculation. For the calculation of the first derivative, the five-point central difference formula is used to process the air pressure data sequence, which accurately reflects the change rate of the air pressure in the cavity with time, while the calculation of the second derivative describes the acceleration characteristics of the air pressure change. The first derivative reflects the speed of the air pressure change in the cavity, that is, the air pressure change rate, which reveals the transient fluctuations generated by the material pushing, vibration and material flow, while the second derivative reflects the acceleration of the air pressure change, which can identify the inflection points, extreme points and stable intervals in the air pressure change process, thereby constructing the derivative matrix. The derivative matrix is executed to detect the feature points, identify the inflection points, extreme points and stable intervals, and obtain the air pressure change feature point set. The core of the feature point detection lies in analyzing the change trend of the first derivative and the second derivative. By detecting the point where the first derivative is zero and the sign of the second derivative changes, the extreme points in the air pressure change process are identified, which reflect the local maximum and minimum values of the pressure in the cavity during the material pushing and forming process. By analyzing the zero crossing point of the second derivative, the inflection point is identified, which corresponds to the critical point of the air pressure change acceleration or deceleration. These points are the turning points of the pressure change in the cavity. By identifying the time period when the first derivative approaches zero, the stable interval of the air pressure change is detected, which corresponds to the time period when the pressure in the cavity reaches the equilibrium state, reflecting the process of material distribution tending to be stable. Through the feature point detection of the derivative matrix, the air pressure change feature point set is obtained. At the same time, the force data in the data set is executed to perform spectrum analysis and extract the force amplitude of the first extrusion plate and the second extrusion plate, and the extrusion plate force time sequence characteristics are obtained. In the extrusion process, the force state of the first extrusion plate and the second extrusion plate directly affects the material distribution and the uniformity of the brick forming, and by performing spectrum analysis on the force data, the main frequency component of the extrusion plate force signal is identified, and the corresponding amplitude feature is extracted.The spectrum analysis converts the time domain signal to the frequency domain through Fourier transform, reveals the periodic characteristics hidden in the force signal and the force change amplitude corresponding to different frequency components, and identifies the characteristic frequencies related to the pushing material cycle, material fluctuation and vibration signal. By extracting the force amplitude of the first and second extrusion plates and combining the time series information, the extrusion plate force time sequence feature is generated. Based on the extracted force time sequence feature, the force imbalance degree is calculated and the force imbalance degree curve with time is drawn to obtain the force distribution feature map. The force imbalance degree is calculated by comparing the force difference between the first and second extrusion plates during the pushing process, and the force imbalance degree reflects the unevenness of the material force during the pushing process and the distribution state of the pushing pressure. By analyzing the force difference at different time points and drawing the force imbalance degree curve with time, the force change trend of the material at different pushing stages is displayed, so as to construct the force distribution feature map. The air pressure change feature point set and the force distribution feature map are time-synchronized and superimposed, the air pressure change feature point and the force distribution feature map are aligned on the time axis, a unified time stamp is added to different data types, and the matching is ensured under the same time reference, so that the time synchronization of the feature point set and the force feature is realized. After time synchronization, the air pressure feature point and the force feature are quantified by using a weighted fusion algorithm, the weighted fusion algorithm dynamically adjusts the contribution degree of the feature point and the feature map according to the influence weight of different features on the material distribution uniformity, and the material distribution uniformity index is calculated by the feature weighted summation. The material distribution uniformity index is a quantitative index reflecting the uniform distribution of the material in the cavity during the brick production process. The closer the index is to the ideal value, the more uniform the material distribution is, and vice versa. Through the above steps, the brick material distribution feature is finally obtained.

[0060] In a specific embodiment, the process of performing step S103 can specifically include the following steps:

[0061] The brick surface defect feature frequency, the brick profile feature vector and the quality abnormality index are combined into a unified feature matrix and normalized and standardized to obtain network input data;

[0062] The network input data is input into the three-layer convolution structure of the convolutional neural network for feature extraction to obtain deep mapping features;

[0063] The deep mapping features are input into the two maximum pooling layers of the convolutional neural network for dimension reduction to obtain a brick feature representation vector;

[0064] The brick feature representation vector is input into a convolutional neural network defect classifier, and a defect classification result is obtained by the defect classifier corresponding to a predefined F brick defect types and using a softmax activation function and a cross-entropy loss function.

[0065] The brick feature representation vector is input into a convolutional neural network modification output layer, and a defect severity score is obtained by executing a root mean square error loss calculation on a regression network with a sigmoid activation function in the modification output layer.

[0066] Specifically, the features obtained from different data sources are fused to integrate the brick surface defect feature frequency, the brick profile feature vector, and the quality abnormality index into a unified feature matrix. The brick surface defect feature frequency is a frequency domain feature calculated by domain synchronous averaging of the vibration signal and residual signal envelope spectrum. These features reflect the periodic characteristics of the brick surface polygonal defect and profile error. The brick profile feature vector is a shape feature obtained by Canny edge detection, Fourier descriptor calculation, and Hausdorff distance measurement on the brick surface image. These features describe the edge profile changes of the brick and the deviation between the actual profile and the standard profile. Meanwhile, the quality abnormality index is an index obtained by weighted fusion of the brick material distribution feature, force imbalance degree, and air pressure change feature point. It reflects the quality change trend in the brick production process. These features are uniformly normalized and standardized. By scaling the data range of different features to the same numerical interval, the differences in feature value scales are eliminated. Through standardization, the feature data is converted to a standard distribution with a mean of zero and a variance of one. Thus, the consistency of different features in the same feature space is ensured, and the network input data required by the convolutional neural network is obtained. The network input data is input into the three-layer convolutional structure of the convolutional neural network for feature extraction. The three-layer convolutional structure of the convolutional neural network uses different size convolution kernels, including 3x3, 5x5, and 7x7 convolution kernels, to perform convolution operations on the input feature matrix, capture local features at different scales, and extract multi-dimensional feature information of brick surface defects, profile shapes, and quality abnormalities. Convolution operation performs point-by-point convolution on input data through a sliding window, element-wise multiplication of the local region of input data and the convolution kernel, and summation to generate a feature mapping matrix. After each convolution layer, a ReLU activation function is introduced for non-linear transformation to enhance the expression ability of the network. Batch normalization layers are added to standardize the convolution results, thereby accelerating network training and preventing gradient vanishing. After three layers of convolution processing, the feature mapping matrix obtained contains deep mapping features of brick surface defects, profile features, and quality abnormalities. The deep mapping features are input into two max-pooling layers of the convolutional neural network for dimension reduction to obtain a brick feature representation vector. Max-pooling is a dimension reduction operation that takes the maximum value in the window by sliding a fixed-size window over the feature mapping matrix, thereby retaining the main information of the features and removing redundant data. Pooling operation can effectively reduce the feature dimension while maintaining the sparsity of the feature space and enhancing the generalization ability of the network. In the convolutional neural network, two max-pooling layers use 2x2 and 3x3 pooling windows to perform dimension reduction on the deep mapping features, converting high-dimensional feature mapping matrices into low-dimensional feature vectors, and obtaining brick feature representation vectors with fixed dimensions. The brick feature representation vector is input into the defect classifier of the convolutional neural network to classify the pre-defined F types of brick defects.The defect classifier consists of two fully connected layers, which perform linear transformation by multiplying the feature vector with a weight matrix and adding a bias vector to generate a probability distribution for defect classification. The output layer uses a softmax activation function to normalize the classification results, converting the original classification scores into a probability distribution, and calculates the classification error of the model through a cross-entropy loss function. The softmax activation function can normalize the output results into probability values, so that the predicted probability of each defect type sums to 1, so that the defect type of the brick is determined according to the maximum probability. The cross-entropy loss function measures the performance of the classification model by comparing the difference between the model's predicted probability distribution and the true label, and adjusts the network weights through the backpropagation algorithm to continuously optimize the classification accuracy. Through the training and optimization of the defect classifier, the determination result of the brick defect type is obtained. The brick feature representation vector is input into the modified output layer of the convolutional neural network, and the defect severity is scored through the modified output layer. The modified output layer adopts a regression network structure, which outputs a continuous severity score through a single neuron with a sigmoid activation function. The sigmoid activation function limits the output result to between 0 and 1, so that the defect severity score is mapped to a fixed numerical interval, thereby facilitating the quantitative comparison of the severity of different defects. The training target of the regression network is to minimize the root mean square error loss function, which optimizes the network parameters by calculating the mean square error between the predicted severity score and the true severity label, thereby improving the prediction accuracy of the model for defect severity. The root mean square error can effectively measure the prediction error of the regression model, and the error is propagated to each layer of the neural network through backpropagation for gradient update, ultimately forming a regression model that quantifies defect severity. After completing the model training, the defect severity score is automatically generated based on the input feature data.

[0067] In a specific embodiment, the process of performing step S104 can specifically include the following steps:

[0068] Based on the defect type determination result and the defect severity score, perform correlation analysis of the brick quality and the process parameters to obtain a process parameter vector including the conveying belt speed, the initial position of the extrusion plate, the spring pressure, the baffle height, the vibration strip height, and the electronic stop valve switching time;

[0069] Based on the process parameter vector, construct a process parameter optimization mathematical model, and calculate the contribution degree of each process parameter to the quality target according to the process parameter optimization mathematical model to obtain a parameter contribution degree matrix;

[0070] According to the parameter contribution degree matrix, divide the process parameters into a high contribution degree group and a low contribution degree group, and perform a differential evolution algorithm in the subspace formed by the high contribution degree parameters to search for the optimal process parameter configuration;

[0071] Based on the optimal process parameter configuration, corresponding compensation strategies are designed for different types of brick defects. For polygonal defects, the speed of the conveying belt and the height of the vibrating strip are adjusted. For profile errors, the position of the extrusion plate and the position of the baffle are adjusted. For uneven material mixing, the switching time of the electronic stop valve is adjusted to obtain a defect-specific compensation scheme.

[0072] The defect-specific compensation scheme is converted into a device control strategy, which includes a speed control instruction for the PID controller to adjust the motor output torque, a position control instruction for the servo motor control signal, a spring pressure control instruction for the pressure regulating mechanism, a baffle height control instruction for the air pressure system, and a timing control instruction for the electronic stop valve.

[0073] Specifically, the data from the quality monitoring system is deeply correlated with historical production process data. Through multivariate regression analysis or machine learning-based feature selection methods, key process parameters that affect brick quality are identified, and the degree of influence of each parameter on the quality target is quantified. These parameters include conveyor belt speed, extrusion plate initial position, spring pressure, baffle height, vibration bar height, and electronic stop valve switching time. Conveyor belt speed affects the stability of brick movement and the uniformity of material distribution during transportation, while the initial position of the extrusion plate determines the distribution of forces during material forming; the size of the spring pressure directly affects the thrust of the extrusion plate and the material compaction effect, the baffle height and vibration bar height affect the speed and accumulation form of material flow, thereby affecting the shape and density of the brick, and the switching time of the electronic stop valve determines the amount and rhythm of material feeding, which has an important influence on material mixing uniformity. These key process parameters are combined into a process parameter vector. Based on the process parameter vector, a process parameter optimization mathematical model is constructed, and a parameter contribution matrix is formed by calculating the contribution of each process parameter to the quality target. The core of the process parameter optimization mathematical model is to establish a mathematical expression with the brick quality target as the optimization objective. The objective function uses a linear regression model to describe the linear relationship between parameters and quality targets, or a nonlinear model based on neural networks or support vector machines to fit the complex relationship between parameter changes and quality changes. In the solving process, by slightly perturbing the process parameters and observing their impact on the quality target, the contribution of each process parameter to the brick quality target is calculated, forming a parameter contribution matrix. Each element in the matrix represents the contribution value of a specific parameter under different quality targets, and this matrix reflects which parameters have higher influence in the quality target optimization process. According to the results of the parameter contribution matrix, the process parameters are divided into high-contribution and low-contribution groups, and the differential evolution algorithm is executed in the subspace formed by high-contribution parameters to search for the optimal process parameter configuration. The differential evolution algorithm is a global optimization algorithm based on population, suitable for solving optimization problems in complex parameter spaces. By initializing a population containing multiple process parameter configurations, the differential evolution algorithm performs mutation, crossover, and selection operations on the population in each iteration, gradually approaching the optimal solution. During optimization, the algorithm focuses on searching in the subspace formed by high-contribution parameters, and dynamically adjusts the importance of different parameters according to the parameter contribution matrix, thereby improving optimization efficiency and solution accuracy. After multiple iterations, the differential evolution algorithm finds an optimal process parameter configuration that meets the quality target. Based on the optimal process parameter configuration, corresponding compensation strategies are designed for different types of brick defects to ensure specific correction of different types of defects during the production process.For polygonal defects, the motion state during material conveying is changed by adjusting the conveyor belt speed and the vibration bar height, thereby optimizing the shape characteristics of the brick surface; for profile errors, the initial positions of the first and second extrusion plates and the position of the partition plate are adjusted to compensate for the influence of profile deviation during the material pushing process, thereby improving the profile accuracy of the brick; for the problem of uneven material mixing, the feeding rhythm of the material is changed by adjusting the opening and closing time of the electronic stop valve, thereby optimizing the uniformity of material mixing in the cavity. These compensation strategies are adjusted according to the characteristics of different defect types to form defect-specific compensation schemes. The defect-specific compensation schemes are converted into equipment control strategies to realize precise execution of various control instructions. The equipment control strategies include various specific control instructions, among which the speed control instruction is adjusted by the PID controller to control the motor output torque to accurately control the speed of the conveyor belt; the position control instruction adjusts the positions of the first and second extrusion plates through the control signal of the servo motor to realize dynamic compensation of the profile error; the spring pressure control instruction of the pressure regulating mechanism adjusts the stress of the material pushing assembly through feedback adjustment to ensure the uniformity of the stress during the material pushing process; the baffle height control instruction adjusts the air pressure inside the cavity based on the feedback signal of the air pressure system to change the height of the baffle, thereby ensuring the stability of the material conveying process; and the timing control instruction of the electronic stop valve adjusts the opening and closing state of the stop valve through an accurate time control algorithm to ensure the uniformity of material flow. Through the synergistic effect of these control strategies, closed-loop control is realized in the brick production process, which enables the production process to adaptively adjust according to real-time monitoring data, thereby improving production efficiency and quality stability.

[0074] In a specific embodiment, the process of step S105 can specifically include the following steps:

[0075] The equipment control strategy is divided into equipment layer control, process layer control and production management layer;

[0076] The control strategy in the equipment layer control is decomposed into instructions and transmitted to each actuator through the field bus to obtain an equipment basic control network;

[0077] Based on the equipment basic control network, a three-phase current control model is constructed by applying a vector control algorithm to motor control, and servo positioning is implemented for extrusion plate position control to obtain motor control signals and servo positioning instructions;

[0078] Based on the iterative profile error compensation algorithm in the process layer control, the motion synchronization points of the material pushing assembly and the vibration assembly are calculated, and a brick profile error compensation function is designed to obtain profile adjustment instructions;

[0079] Based on the dynamic production plan adjustment strategy in production management layer control, the same defect trigger parameter self-optimization program for continuous B-block bricks, the emergency adjustment program triggered by serious defects and the condition response mechanism for quality trend prediction are set to obtain management layer optimization decision information.

[0080] Specifically, according to the different control requirements of the brick production process, the control strategy is refined and classified according to the hierarchy. The device layer control is responsible for the basic operation of the device and the real-time instruction of the actuator, including the accurate control of key components such as motors, extrusion plates, vibration components, electronic stop valves and baffles. The process layer control is responsible for real-time compensation and dynamic adjustment of quality errors that occur in the production process, and optimizes the production process through data feedback. The production management layer control focuses on the management and optimization of the overall production plan, and adjusts the process parameters according to the production data trend and quality detection results to ensure the dynamic balance of the production plan and quality target. After decomposing the device control strategy into different levels, the response speed and stability of the control system are effectively improved, and dynamic control is realized from the bottom device control to the high-level production management. In the device layer control, the control strategy is further decomposed into instructions, and the instructions are transmitted to each actuator through the field bus, thereby constructing the device basic control network. The field bus can realize high-speed data communication between different actuators, ensuring that the control instructions can be accurately issued to each device node within milliseconds. Through instruction decomposition, complex control strategies are broken down into specific control commands, and different types of control signals are generated according to the functions and control requirements of different devices, such as motor torque control signals, servo motor position control signals, pressure regulation signals, baffle height adjustment signals, and electronic stop valve switch control signals. These control signals are transmitted to the corresponding actuators through the field bus to form the device basic control network, ensuring that all devices on the production line can work cooperatively. Based on the device basic control network, a three-phase current control model is constructed by applying a vector control algorithm to motor control, and a servo positioning is implemented for the extrusion plate position control, obtaining motor control signals and servo positioning instructions. The vector control algorithm decouples the three-phase current of the motor to convert the three-phase current of the AC motor into a DC component, thereby realizing independent control of the motor torque and magnetic flux. By constructing a three-phase current control model, the speed and torque of the motor are controlled to make the conveyor belt run at an optimal speed range while avoiding speed changes caused by load fluctuations. The servo positioning control accurately adjusts the position of the extrusion plate through closed-loop feedback control. The feedback signal of the servo motor is compared with the target position to generate an error signal, which is adjusted in real time by a PID controller to ensure that the position accuracy of the extrusion plate during the pushing process reaches a control accuracy of ±0.1mm. The motor control signals and servo positioning instructions are issued in real time to the actuators through the device basic control network to ensure high-speed and stable device layer control during the production process. In the process layer control, the motion synchronization points of the pushing assembly and the vibration assembly are calculated based on the iterative contour error compensation algorithm, and a brick contour error compensation function is designed to generate contour adjustment instructions.The iterative contour error compensation algorithm calculates the motion synchronization point of the pushing assembly and the vibrating assembly at a set delay time τ by analyzing their motion data, and adjusts the compensation strategy according to the difference in motion trajectories. The calculation of the motion synchronization point is based on the kinematic model, ensuring that the pushing action and the vibrating strip motion are highly coordinated in time and space, thereby eliminating the contour error generated during the pushing process. Based on the motion synchronization point, a brick contour error compensation function is designed, which generates a set of dynamic compensation signals by analyzing the variation law of the contour error. These compensation signals are superimposed on the original control signals of the pushing assembly and the vibrating assembly, thereby realizing real-time compensation of the contour error. The contour adjustment instructions are issued to the actuators through the device basic control network, ensuring high-precision contour control during brick molding, and improving the quality and consistency of the brick surface. In the production management layer control, a continuous B-block brick appears the same defect trigger parameter self-optimization program, a serious defect trigger emergency adjustment program, and a quality trend prediction condition response mechanism are set based on the dynamic production plan adjustment strategy, thereby generating management layer optimization decision information. The dynamic production plan adjustment strategy analyzes historical production data, real-time quality detection results, and production targets, and predicts future production quality trends using machine learning algorithms, and dynamically adjusts the production plan based on the prediction results. When the same type of defect is detected in consecutive B-block bricks, the system automatically triggers the parameter self-optimization program to recalculate the optimal process parameter configuration and dynamically adjust the device control strategy to eliminate persistent defects. For serious defects detected, such as large-scale damage to the brick surface or contour error exceeding the set threshold, the system triggers the emergency adjustment program, suspends the production line, and checks the device state, while adjusting the key process parameters to prevent the production of a large number of unqualified products. By analyzing the quality trend prediction results, potential quality fluctuation trends are identified, and key process parameters are adjusted in advance to achieve preventive control of the production process. The management layer optimization decision information is issued to each actuator through the device basic control network, and combined with the dynamic adjustment strategies of the device layer control and the process layer control, a closed-loop control of the whole process is realized. The motor control signal, servo positioning instruction, contour adjustment instruction, and management layer optimization decision information work together to ensure precise control and dynamic adjustment of each link in the brick production process, thereby achieving continuous optimization of brick quality and improvement of production efficiency.

[0081] The above describes the brick quality monitoring method based on data analysis in the embodiments of the present application. The following describes the brick quality monitoring system based on data analysis in the embodiments of the present application. Please refer to Figure 2 The brick quality monitoring system based on data analysis in the embodiments of the present application includes one embodiment:

[0082] The acquisition module 201 is configured to acquire vibration, pressure, displacement, image, and air pressure data on the brick production line to obtain a data set.

[0083] The detection module 202 is configured to perform field synchronous averaging and edge detection on the data set to obtain brick surface defect feature frequency, brick contour feature vector and quality anomaly index.

[0084] The analysis module 203 is configured to input the brick surface defect feature frequency, the brick contour feature vector and the quality anomaly index into a convolutional neural network for analysis to obtain a brick defect type determination result and a defect severity score.

[0085] The parameter optimization module 204 is configured to perform process parameter optimization based on the defect type determination result and the defect severity score to obtain a device control strategy.

[0086] The transmission module 205 is configured to transmit the device control strategy to each execution mechanism to perform closed-loop feedback to obtain motor control signals, servo positioning instructions, contour adjustment instructions and management layer optimization decision information.

[0087] Through the cooperation of the above components, the present application configures a multi-sensor network on the brick production line, collects multi-dimensional data such as vibration, pressure, displacement, image and air pressure, forms a complete production process data set, and realizes comprehensive monitoring of the whole process of brick production compared with the traditional monitoring method relying on only a single parameter, effectively avoiding misjudgment caused by single parameter abnormalities. The present application applies advanced algorithms such as domain synchronous averaging processing of vibration signals, Canny edge detection analysis of images and Fourier descriptor extraction of contour features, which can accurately identify brick surface defect feature frequency, contour feature vector and quality anomaly index, greatly improving the sensitivity and accuracy of brick defect detection. The present application inputs the extracted features into a convolutional neural network for deep analysis, through multi-layer feature extraction and classification regression, not only can accurately determine various types of brick defects, but also can give a defect severity score, providing an accurate basis for subsequent process parameter optimization. Based on the defect analysis result, the present application constructs a correlation model of brick quality and process parameters, searches in the high contribution parameter subspace through differential evolution algorithm, and designs specific compensation strategies for different types of defects, realizing accurate dynamic adjustment of process parameters in the brick production process. The present application adopts a three-level control architecture of device layer, process layer and production management layer, realizes hierarchical cooperation from basic control to high-level decision-making, through iterative contour error compensation algorithm and dynamic production plan adjustment strategy, ensures the accurate execution of control instructions, and at the same time establishes a three-level alarm mechanism of early warning, warning and emergency warning, improving the stability and safety of the system. Through control performance index calculation and parameter dynamic adjustment, the present application forms a complete closed loop of data acquisition, analysis processing, defect diagnosis, parameter optimization and control execution, so that the brick production system can continuously self-optimize, effectively cope with internal changes and external disturbances, and significantly improve the quality stability and production efficiency of brick production.

[0088] The application further provides a computer readable storage medium, which can be a non-volatile computer readable storage medium or a volatile computer readable storage medium, and the computer readable storage medium stores instructions, and the instructions make a computer execute the steps of the brick quality monitoring method based on data analysis when the instructions are run on the computer.

[0089] Those skilled in the art can clearly understand that, for the convenience and brevity of description, the specific working processes of the above-described system, system and unit can refer to the corresponding processes in the foregoing method embodiments, and will not be described here.

[0090] The integrated unit, if realized in the form of a software function unit and sold or used as an independent product, can be stored in a computer readable storage medium. Based on such understanding, the technical solutions of the application or the whole or part of the technical solutions that make essential contributions to the prior art can be embodied in the form of a software product. The computer software product is stored in a storage medium and includes a plurality of instructions for making a brick quality monitoring device based on data analysis (which can be a personal computer, a server, or a network device, etc.) execute all or part of the steps of the method described in each embodiment of the application. The foregoing storage medium includes a U disk, a mobile hard disk, a read-only memory (ROM), a random access memory (RAM), a magnetic disk or an optical disk, and various media that can store program codes.

[0091] The above-described embodiments are only used to illustrate the technical solutions of the application, rather than limit the application; although the application has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that: it can still modify the technical solutions recorded in the foregoing embodiments, or make equivalent replacement for part of the technical features; and these modifications or replacements do not make the corresponding technical solutions deviate from the spirit and scope of the technical solutions of the embodiments of the application.

Claims

1. A method for monitoring the quality of bricks based on data analysis, characterized in that, The method comprises the following steps: Collecting vibration, pressure, displacement, image and air pressure data on the brick production line to obtain a data set; Performing field synchronous averaging and edge detection on the data set to obtain brick surface defect feature frequency, brick profile feature vector and quality anomaly index; Inputting the brick surface defect feature frequency, the brick profile feature vector and the quality anomaly index into a convolutional neural network for analysis to obtain a brick defect type determination result and a defect severity score; Performing process parameter optimization based on the defect type determination result and the defect severity score to obtain a device control strategy; Transmitting the device control strategy to each execution mechanism to perform closed-loop feedback to obtain motor control signals, servo positioning instructions, profile adjustment instructions and management layer optimization decision information; The process parameter optimization based on the defect type determination result and the defect severity score to obtain a device control strategy comprises: Performing correlation analysis of brick quality and process parameters based on the defect type determination result and the defect severity score to obtain a process parameter vector containing the conveying belt speed, the extrusion plate initial position, the spring pressure, the baffle height, the vibration strip height and the electronic stop valve switching time; Based on the process parameter vector, a process parameter optimization mathematical model is constructed, and the contribution of each process parameter to the quality target is calculated based on the process parameter optimization mathematical model to obtain a parameter contribution matrix; According to the parameter contribution matrix, the process parameters are divided into a high-contribution group and a low-contribution group, and a differential evolution algorithm is performed in the subspace formed by the high-contribution parameters to search for the optimal process parameter configuration; Based on the optimal process parameter configuration, corresponding compensation strategies are designed for different types of brick defects, the conveying belt speed and the vibration strip height are adjusted for polygonal defects, the extrusion plate position and the baffle position are adjusted for profile error, and the electronic stop valve switching time is adjusted for material mixing unevenness to obtain a defect-specific compensation scheme; The defect-specific compensation scheme is converted into a device control strategy, which includes a speed control instruction for the PID controller to adjust the motor output torque, a position control instruction for the servo motor control signal, a spring pressure control instruction for the pressure adjusting mechanism, a baffle height control instruction for the air pressure system and a timing control instruction for the electronic stop valve; The device control strategy is transmitted to each execution mechanism to perform closed-loop feedback to obtain motor control signals, servo positioning instructions, profile adjustment instructions and management layer optimization decision information, which comprises: The device control strategy is decomposed into device layer control, process layer control and production management layer; The control strategy in the device layer control is decomposed into instructions and transmitted to each execution mechanism through a field bus to obtain a device basic control network; Based on the device basic control network, a three-phase current control model is constructed by applying a vector control algorithm to motor control, and servo positioning is implemented for extrusion plate position control to obtain motor control signals and servo positioning instructions; The motion synchronization point of the pushing assembly and the vibration assembly is calculated based on an iterative profile error compensation algorithm in the process layer control, and a brick profile error compensation function is designed to obtain a profile adjustment instruction; the iterative profile error compensation algorithm calculates the motion synchronization point of the pushing assembly and the vibration assembly at a set delay time τ by analyzing the motion data of the two assemblies, and adjusts the compensation strategy according to the difference between the motion trajectories; Based on the dynamic production plan adjustment strategy in the production management layer control, a continuous B-block brick same defect trigger parameter self-optimization program, a serious defect trigger emergency adjustment program and a quality trend prediction condition response mechanism are set to obtain management layer optimization decision information.

2. The data analysis based brick quality monitoring method as claimed in claim 1 wherein, The vibration, pressure, displacement, image and air pressure data on the brick production line are collected to obtain a data set, including: n vibration sensors are uniformly installed on the surface of the conveying belt on the brick production line, and vibration signals representing the dynamic characteristics of the conveying belt are collected through the vibration sensors; m pairs of pressure sensors are installed on the first and second extrusion plates on the brick production line, and force data representing the force state of the extrusion plates are collected through the pressure sensors; a displacement sensor is fixed to the bottom of the vibration strip, and displacement data representing the motion state of the vibration strip are collected through the displacement sensor; two high-speed industrial cameras installed at the end of the conveying belt are used to collect multi-angle brick surface images, and an air pressure sensor installed inside the cavity is used to collect air pressure data representing the dynamic change of the internal pressure of the cavity; The vibration signal, the force data, the displacement data, the brick surface image and the air pressure data are time-synchronized and stored to obtain a data set.

3. The data analysis based brick quality monitoring method as claimed in claim 1 wherein, The data set is domain-synchronized and averaged, and edge detection is performed to obtain brick surface defect feature frequency, brick profile feature vector and quality anomaly index, including: The vibration signals in the data set are segmented according to the running period of the conveying belt and the synchronous average signal of N cycles is calculated to obtain the domain-synchronous average signal representing the periodic defects on the surface of the brick; The residual error signal is extracted from the domain-synchronous average signal and its envelope spectrum is calculated to obtain the brick surface defect feature frequency; The brick surface image in the data set is subjected to grayscale conversion, histogram equalization and image enhancement to obtain a target surface image, and Canny edge detection is performed on the target surface image to obtain the brick profile; Fourier descriptors are used to calculate the shape features of the first to the twentieth order of the brick profile, and the Hausdorff distance between the actual profile and the standard profile is calculated to obtain the brick profile feature vector; The first and second derivatives of the air pressure data in the data set are calculated, and the inflection points, extreme points and stable intervals are extracted, and the moment analysis is performed on the force data in the data set to calculate the force imbalance degree to obtain the brick material distribution characteristics; The brick surface defect feature frequency, the brick profile feature vector and the brick material distribution characteristics are normalized and fused to obtain a target feature vector, and the target feature vector is detected by setting an abnormal score threshold through the isolation forest algorithm to obtain a quality anomaly index.

4. The data analysis based brick quality monitoring method as claimed in claim 3 wherein, The first-order derivative and the second-order derivative are calculated according to the air pressure data in the data set, and the inflection points, extreme points and stable intervals are extracted, and at the same time, the moment analysis is performed on the force data in the data set, and the stress imbalance degree is calculated, so as to obtain the brick material distribution characteristics, including: The air pressure data in the data set is subjected to noise suppression and smoothing processing to obtain a preprocessed air pressure data sequence; The first-order derivative and the second-order derivative are calculated by five-point central difference method on the preprocessed air pressure data sequence to obtain a derivative matrix representing the air pressure change rate and acceleration in the cavity; The derivative matrix is subjected to feature point detection algorithm to identify the inflection points, extreme points and stable intervals to obtain an air pressure change feature point set; The force data in the data set is subjected to frequency spectrum analysis and the stress amplitudes of the first and second extrusion plates are extracted to obtain the extrusion plate stress timing characteristics; The stress imbalance degree is calculated based on the extrusion plate stress timing characteristics and the stress imbalance degree-time curve is drawn to obtain the stress distribution characteristic map; The air pressure change feature point set and the stress distribution characteristic map are subjected to time synchronization superposition analysis and the material distribution uniformity index is quantitatively integrated by a weighted fusion algorithm to obtain the brick material distribution characteristics.

5. The data analysis based brick quality monitoring method as claimed in claim 1 wherein, The brick surface defect feature frequency, the brick profile feature vector and the quality abnormality index are input into the convolutional neural network for analysis to obtain the brick defect type judgment result and the defect severity score, including: The brick surface defect feature frequency, the brick profile feature vector and the quality abnormality index are combined into a unified feature matrix and subjected to normalization and standardization processing to obtain network input data; The network input data is input into the three-layer convolutional structure of the convolutional neural network for feature extraction to obtain deep mapping features; The deep mapping features are input into the two maximum pooling layers of the convolutional neural network for dimension reduction to obtain a brick feature representation vector; The brick feature representation vector is input into the defect classifier of the convolutional neural network, and the defect classifier corresponds to the pre-defined F types of brick defect types, and the softmax activation function and the cross-entropy loss function are used to calculate the brick defect type judgment result; The brick feature representation vector is input into the modified output layer of the convolutional neural network, and the root mean square error loss calculation is performed by the regression network with the activation function of sigmoid in the modified output layer to obtain the defect severity score.

6. A data analysis based brick quality monitoring system characterized in that, The data analysis-based brick quality monitoring system for implementing the method according to any one of claims 1-5 comprises: A collection module for collecting vibration, pressure, displacement, image and air pressure data on a brick production line to obtain a data set; A detection module for performing domain synchronous averaging and edge detection on the data set to obtain a brick surface defect feature frequency, a brick profile feature vector and a quality abnormality index; An analysis module for inputting the brick surface defect feature frequency, the brick profile feature vector and the quality abnormality index into a convolutional neural network for analysis to obtain a brick defect type judgment result and a defect severity score. a parameter optimization module, configured to perform process parameter optimization based on the defect type determination result and the defect severity score, to obtain a device control strategy; a transmission module, configured to transmit the device control strategy to each execution mechanism to perform closed-loop feedback, to obtain a motor control signal, a servo positioning instruction, a contour adjustment instruction, and management layer optimization decision information.

7. A computer-readable storage medium having stored thereon a computer program, characterized in that, The computer program, when executed by a processor, causes the processor to perform the brick quality monitoring method based on data analysis according to any one of claims 1 to 5.

Images