On-line monitoring and adjusting method, device and system for impact hammer of sand making machine and medium

Abstract

The invention discloses a sand making machine impact hammer online monitoring and adjusting method, device and system and a medium. The method comprises the following steps: acquiring images of an impact hammer and a ring lining plate area when the sand making machine is in a shutdown state; determining initial wear data of the impact hammer based on the image; obtaining a corresponding wear prediction model according to the working condition parameters in the next sand making task of the sand making machine; the working condition parameters and the initial abrasion data in the next sand making task are input into the abrasion prediction model together, and predicted abrasion data of the impact hammer after the next sand making task is completed are obtained; a wear early warning threshold value is obtained, and under the condition that the predicted wear data exceed the wear early warning threshold value, a propulsion amount is generated according to the wear early warning threshold value and the initial wear data; a physical gap between the impact hammer and the ring liner is adjusted based on the amount of propulsion. According to the method, machine vision, intelligent prediction and automatic control technologies are fused, and it is ensured that the sand making machine operates at the optimal efficiency in the whole production task.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of intelligent adjustment, in particular to a sand making machine impact hammer online monitoring and adjustment method, a sand making machine impact hammer online monitoring and adjustment device, a sand making machine impact hammer online monitoring and adjustment system and a machine readable storage medium. BACKGROUND

[0002] In the production process of super fine sand, the sand making machine is the key equipment to complete the sand making operation. In the normal use process, the impact hammer in the impeller structure inside the sand making machine will be continuously worn, and the sand making machine produces by the impact of the stone on the impact hammer. The wear of the impact hammer will increase the gap between the impact hammer and the ring liner, which will reduce the quality of the finished sand. Therefore, it is necessary to regularly stop the machine to adjust the position of the impact hammer, so as to ensure that the gap between the impact hammer and the ring liner is always within a reasonable range. In the prior art, maintenance personnel often enter the cavity to perform manual operation, which is complicated and in a poor operating environment, and the production efficiency is low. SUMMARY

[0003] Therefore, it is necessary to provide a sand making machine impact hammer online monitoring and adjustment method, a sand making machine impact hammer online monitoring and adjustment device, a sand making machine impact hammer online monitoring and adjustment system and a machine readable storage medium to solve the above technical problems.

[0004] In a first aspect, the present application provides a sand making machine impact hammer online monitoring and adjustment method, comprising: acquiring an image of the impact hammer and the ring liner area of the sand making machine in the stopped state; determining the initial wear data of the impact hammer based on the image; acquiring the corresponding wear prediction model according to the working condition parameters in the next sand making task of the sand making machine; the wear prediction model is constructed based on each working condition parameter and the corresponding impact hammer wear data; inputting the working condition parameters in the next sand making task and the initial wear data into the wear prediction model to obtain the predicted wear data of the impact hammer after the completion of the next sand making task; acquiring a wear warning threshold, and generating a propulsion amount according to the wear warning threshold and the initial wear data when the predicted wear data exceeds the wear warning threshold; adjusting the physical gap between the impact hammer and the ring liner based on the propulsion amount.

[0005] In one embodiment, the steps for constructing the wear prediction model include: acquiring multiple historical sand making samples, each sample including operating parameters and measured values ​​of impact hammer wear, wherein the operating parameters include single-batch processing volume, material hardness value, and material moisture value; establishing a linear relationship model between the wear prediction value and the single-batch processing volume, material hardness, and material moisture; aiming to minimize the error between the wear prediction value and the measured wear value of the impact hammer, solving for the values ​​of each influence coefficient of the linear relationship model using a linear regression algorithm and historical sand making samples; and substituting the values ​​of each influence coefficient into the linear relationship model to obtain the wear prediction model.

[0006] In one embodiment, the linear relationship satisfies the following formula: ; In the formula, Let be the predicted wear rate value for the i-th sample; This represents the processing volume for the i-th sample in a single run. Let be the material hardness of the i-th sample; Let be the material humidity of the i-th sample; The intercept; The influence coefficient of the single processing volume; The coefficient representing the influence of the material's hardness; The influence coefficient of the material's humidity; ; ; In the formula, SSE is the sum of squared errors between the predicted wear rate and the actual wear rate of the impact hammer after a single sand task; Let be the wear rate value of the i-th sand making task. Let be the initial wear rate before the i-th sand making task is executed. The initial wear rate is the actual wear rate after the i-th sand making task is completed; both the initial wear rate and the actual wear rate are determined based on image recognition results.

[0007] In one embodiment, the step of obtaining the corresponding wear prediction model based on the operating parameters of the next sand making task of the sand making machine includes: querying the operating condition model relationship table based on the operating parameters of the next sand making task of the sand making machine; the operating condition model relationship table stores the mapping relationship between each operating condition parameter and the corresponding wear prediction model.

[0008] In one embodiment, the step of obtaining the wear warning threshold includes: generating a wear prediction curve based on the predicted wear data of the impact hammer after the completion of the next sand making task, wherein the wear prediction curve fluctuates with the task time of the next sand making task; obtaining a sand formation fluctuation curve, and when the sand formation fluctuation curve exceeds the standard line, statistically analyzing the values ​​in the wear prediction curve corresponding to the same time mapping point to generate a wear warning set; wherein the sand formation fluctuation curve is generated from the real-time operation data of historical sand making tasks under the corresponding working conditions; and calculating the wear warning threshold based on the wear warning set.

[0009] In one embodiment, the wear warning set includes a wear warning upper limit set and a wear warning lower limit set; the wear warning threshold includes a warning upper limit threshold and a warning lower limit threshold; the warning upper limit threshold is calculated based on the wear warning upper limit set, and the warning lower limit threshold is calculated based on the wear warning lower limit set; the calculation steps of the warning upper limit threshold and the warning lower limit threshold satisfy the following formula: ; In the formula, The upper limit threshold for the warning is... The task statistics for the wear warning upper limit set. to The predicted wear rate of abnormal sand making tasks in the wear warning upper limit set; ; In the formula, The lower threshold value for the early warning is... The task statistics for the wear warning lower limit set are as follows: to The wear rate is the predicted wear rate of abnormal sand making tasks in the wear warning lower limit set.

[0010] In one embodiment, the step of generating propulsion amount based on the wear warning threshold and the initial wear data includes: calculating the propulsion amount according to the following formula when the predicted wear data exceeds the warning lower limit threshold; ; In the formula, e is the propulsion amount, r is the cross-sectional area of ​​the impact hammer, s is the volume of the impact hammer, and is a fixed value. The initial wear data, The upper limit threshold for the warning is defined as follows.

[0011] In one embodiment, the step of determining the initial wear data of the impact hammer based on the image includes: preprocessing the image; the preprocessing includes grayscale conversion, filtering and noise reduction, and contrast enhancement; positioning and edge detection of the preprocessed image according to the encoder on the impact hammer to obtain the coordinates of a first region; extracting the features of the ring liner region from the first region coordinates and determining the reference surface to obtain the coordinates of a second region; and performing pixel coordinate transformation on the first region coordinates and the second region coordinates using a standard gap value to generate the initial wear data.

[0012] In one embodiment, after the step of adjusting the physical gap between the impact hammer and the ring liner based on the advance amount, the method further includes: acquiring an image of the area between the impact hammer and the ring liner after adjustment; determining the step amount of the impact hammer based on the image; generating an adjustment result based on the advance amount and the step amount; and feeding back fault information if the adjustment result does not meet the requirements.

[0013] Secondly, this application also provides an online monitoring and adjustment device for the impact hammer of a sand making machine, comprising: an image acquisition module for acquiring an image of the area between the impact hammer and the ring liner plate of the sand making machine in a stopped state; a feature extraction module for determining the initial wear data of the impact hammer based on the image; a model acquisition module for acquiring a corresponding wear prediction model based on the working parameters of the next sand making task of the sand making machine; the wear prediction model is constructed based on each working parameter and the corresponding impact hammer wear data; a wear prediction module for inputting the working parameters of the next sand making task and the initial wear data into the wear prediction model to obtain the predicted wear data of the impact hammer after the completion of the next sand making task; and a wear adjustment module for acquiring a wear warning threshold, and generating a propulsion amount based on the wear warning threshold and the initial wear data when the predicted wear data exceeds the wear warning threshold; and adjusting the physical gap between the impact hammer and the ring liner plate based on the propulsion amount.

[0014] Thirdly, this application also provides an online monitoring and adjustment system for the impact hammer of a sand making machine, comprising: a state perception subsystem, an intelligent analysis subsystem communicatively connected to the state perception subsystem, and a gap execution subsystem communicatively connected to the intelligent analysis subsystem; the state perception subsystem is used to acquire images of the impact hammer and the ring liner area of ​​the sand making machine in a stopped state; the intelligent analysis subsystem is used to determine the initial wear data of the impact hammer based on the images; obtain a corresponding wear prediction model according to the working parameters in the next sand making task of the sand making machine; the wear prediction model is constructed based on each working parameter and the corresponding impact hammer wear data; input the working parameters in the next sand making task and the initial wear data into the wear prediction model to obtain the predicted wear data of the impact hammer after the completion of the next sand making task; obtain a wear warning threshold, and generate a propulsion amount according to the wear warning threshold and the initial wear data when the predicted wear data exceeds the wear warning threshold; the gap execution subsystem is used to adjust the physical gap between the impact hammer and the ring liner based on the propulsion amount.

[0015] In one embodiment, the state perception subsystem includes multiple shielding protection modules and multiple data acquisition modules, which are communicatively connected to the intelligent analysis subsystem. The multiple data acquisition modules are arranged circumferentially at the top of the feed chamber of the sand making machine to collect images of the area between the impact hammer and the ring liner. The shielding protection modules correspond one-to-one with the data acquisition modules to prevent the corresponding data acquisition modules from being damaged during the sand making process.

[0016] Fourthly, embodiments of this application provide a machine-readable storage medium storing instructions that cause a machine to implement the method.

[0017] One of the above technical solutions has the following advantages or beneficial effects: It acquires an image of the impact hammer and ring liner area of ​​the sand making machine in a stopped state; determines the initial wear data of the impact hammer based on the image; obtains the corresponding wear prediction model based on the operating parameters of the next sand making task; the wear prediction model is constructed based on each operating parameter and the corresponding impact hammer wear data; inputs the operating parameters and initial wear data of the next sand making task into the wear prediction model to obtain the predicted wear data of the impact hammer after the next sand making task is completed; obtains a wear warning threshold; when the predicted wear data exceeds the wear warning threshold, generates a propulsion amount based on the wear warning threshold and the initial wear data; and adjusts the physical gap between the impact hammer and the ring liner based on the propulsion amount. This method integrates machine vision, intelligent prediction, and automatic control technologies to achieve accurate calculation of the wear state of the impact hammer of the sand making machine. It avoids the problems of low operating efficiency and high working risk of traditional methods, ensuring that the sand making machine operates at optimal efficiency throughout the entire production task without interruption, completing the entire process automatically, efficiently and safely.

[0018] Other features and advantages of the embodiments of this application will be described in detail in the following detailed description section. Attached Figure Description

[0019] The accompanying drawings are provided to further illustrate the embodiments of this application and form part of the specification. They are used together with the following detailed description to explain the embodiments of this application, but do not constitute a limitation on the embodiments of this application. In the drawings: Figure 1 The schematic diagram illustrates a process flow diagram of an online monitoring and adjustment method for a sand making machine impact hammer according to an embodiment of this application; Figure 2 The schematic diagram illustrates a process flow chart of an online monitoring and adjustment method for a sand making machine impact hammer according to yet another embodiment of this application; Figure 3 The schematic diagram illustrates a process flow diagram of an online monitoring and adjustment method for a sand making machine impact hammer according to another embodiment of this application; Figure 4 This schematic diagram illustrates the structure of a device for online monitoring and adjustment of the impact hammer of a sand making machine according to an embodiment of this application; Figure 5 This schematic diagram illustrates the component structure of an online monitoring and adjustment system for a sand making machine impact hammer according to an embodiment of this application. Figure 6 A schematic plan view of a data acquisition module according to one embodiment of this application is shown. Figure 7 A schematic plan view of a shielding protection module according to an embodiment of this application is shown. Figure 8 The diagram illustrates a top view of a sand making machine impact hammer according to one embodiment of the present application. Figure 9 for Figure 8 Sectional view of AA; Figure 10 for Figure 9 A magnified view of a section at point I; Figure 11 The diagram schematically illustrates the internal structure of a computer device in one embodiment.

[0020] Explanation of reference numerals in the attached figures Detailed Implementation

[0021] To make the objectives, technical solutions, and advantages of the embodiments of this application clearer, the technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. It should be understood that the specific embodiments described herein are only for illustration and explanation of the embodiments of this application and are not intended to limit the embodiments of this application. All other embodiments obtained by those skilled in the art based on the embodiments of this application without creative effort are within the scope of protection of this application.

[0022] It should be noted that the acquisition, transmission, storage, use, and processing of data in the technical solution of this application all comply with relevant laws and regulations. In the embodiments of this application, certain existing industry solutions such as software, components, and models may be mentioned. These should be considered exemplary, intended only to illustrate the feasibility of implementing the technical solution of this application, and do not imply that the applicant has already used or necessarily used such solutions.

[0023] It should be noted that if the embodiments of this application involve directional indicators (such as up, down, left, right, front, back, etc.), the directional indicators are only used to explain the relative positional relationship and movement of each component in a certain specific posture (as shown in the figure). If the specific posture changes, the directional indicators will also change accordingly.

[0024] In one embodiment, such as Figure 1 As shown in the figure, this application provides an online monitoring and adjustment method for the impact hammer of a sand making machine, including: S102, acquire an image of the impact hammer and ring liner area of ​​the sand making machine when it is stopped.

[0025] Each impact hammer is fixed to a corresponding bottom liner plate. The impact hammer has a U-shaped structure. After the end of the U-shaped opening of the impact hammer wears down, the gap between it and the ring liner plate will increase, reducing the sand quality of the sand making machine. Therefore, images of the corresponding area can be collected when the sand making machine is stopped, and the actual physical gap between the impact hammer and the ring liner plate can be accurately measured in a non-contact manner.

[0026] S104, based on this image, determine the initial wear data of the impact hammer.

[0027] The initial wear data refers to the current wear rate of the impact hammer calculated through image analysis. By performing a series of steps such as preprocessing, feature region segmentation, etc., the pixel distance extracted from the image is converted into real wear data.

[0028] S106, Obtain the corresponding wear prediction model based on the working condition parameters of the sand making machine in the next sand making task; The wear prediction model is constructed based on each working condition parameter and the corresponding impact hammer wear data.

[0029] In this embodiment, the operating parameters for the next sand-making task refer to the production conditions set by the user for the upcoming production task. These operating parameters directly affect the wear rate and may include: planned throughput, material hardness, material moisture content, estimated running time, and material particle size, etc. It should be understood that the operating parameters can be manually input by the user or pre-configured by the system. In addition, material hardness and material moisture content are determined by the characteristics of the raw materials themselves. This method can also collect images of the materials provided by the user, identify the type of raw materials based on the images, and then query the material characteristic table embedded in the system to obtain specific material parameters.

[0030] The wear prediction model is based on a mathematical model trained on historical sand making tasks. It is used to establish the relationship between operating parameters and wear data. Multiple linear regression models, machine learning models, or neural networks can be used. Multiple historical sand making samples can be collected and grouped according to factors such as material type and power grid stability to establish multiple wear prediction models. This makes the wear prediction model obtained based on the operating parameters of the next sand making task more consistent with the actual situation of the next sand making task.

[0031] S108: Input the working parameters and initial wear data of the next sand making task into the wear prediction model to obtain the predicted wear data of the impact hammer after the next sand making task is completed.

[0032] Predicted wear data refers to the predicted total wear rate of the impact hammer after successfully completing the next sand making task. The system can alert the user based on the predicted wear data and automatically adjust the clearance.

[0033] S110: Obtain the wear warning threshold. If the predicted wear data exceeds the wear warning threshold, generate the propulsion amount based on the wear warning threshold and the initial wear data.

[0034] Among them, the wear warning threshold refers to the safety boundary value. When the predicted wear data exceeds this boundary, the system believes that the equipment will face risks or a decrease in production efficiency if it continues to operate. The wear warning threshold can be a fixed value set in advance or a dynamic value calculated based on historical sand making samples.

[0035] The push amount is to compensate for wear and restore the optimal working clearance; it is the physical displacement that the impact hammer needs to be pushed.

[0036] S112, based on the propulsion amount, adjusts the physical gap between the impact hammer and the ring liner.

[0037] The feed rate can be used to adjust the physical clearance between the impact hammer and the ring liner. The feed rate can be adjusted for a single impact hammer based on its code, or for multiple impact hammers depending on the actual situation. This process achieves automated and intelligent adjustment, eliminating the need for frequent manual intervention and ensuring that the clearance between the impact hammer and the ring liner remains within a reasonable range. Furthermore, the aforementioned calculation method allows for more accurate clearance control.

[0038] This method effectively avoids equipment downtime due to excessive wear of the impact hammer by predicting its wear condition in advance and adjusting it accordingly. Maintenance can be completed automatically before sand-making problems occur, achieving predictive maintenance and unmanned operation. Furthermore, by monitoring the wear status of components over a long period, the lifespan of vulnerable parts can be predicted, allowing for advance planning of spare parts and maintenance schedules. This transforms scheduled maintenance into on-demand maintenance, ensuring that the impact hammer operates at its optimal clearance during operation, significantly stabilizing and improving the quality of the finished sand.

[0039] In one embodiment, the steps for constructing the wear prediction model include: acquiring multiple historical sand making samples, each sample including operating parameters and measured values ​​of impact hammer wear, the operating parameters including single-batch throughput, material hardness, and material moisture content; establishing a linear relationship model between the wear prediction value and the single-batch throughput, material hardness, and material moisture content; aiming to minimize the error between the wear prediction value and the measured wear value of the impact hammer, solving for the values ​​of each influence coefficient of the linear relationship model using a linear regression algorithm and historical sand making samples; and substituting the values ​​of each influence coefficient into the linear relationship model to obtain the wear prediction model.

[0040] In this embodiment, the historical sand making samples are data records of complete sand making tasks under various working conditions. Each sample includes working condition parameters, measured values ​​of impact hammer wear, and sand formation fluctuation curves in the sand making task, which are used to establish wear prediction models for each working condition. The single processing capacity refers to the total weight of material processed in one sand making task, which indirectly includes running time information. Single processing capacity = processing capacity × running time.

[0041] The linear relationship model is a mathematical model that represents the weighted sum relationship between the wear prediction value and various operating parameters. The contribution of each parameter to wear is determined by the corresponding influence coefficient. Among them, the influence coefficient is the core parameter of the linear relationship model, which quantifies the degree of influence of each operating parameter on wear. It can be calculated by fitting historical sample data through a linear regression algorithm, such as the least squares method, with the goal of minimizing the error between the wear prediction value and the actual wear value of the impact hammer.

[0042] In one embodiment, the linear relationship satisfies the following formula: ; In the formula, Let be the predicted wear rate value for the i-th sample; This represents the processing volume for the i-th sample in a single run. Let be the material hardness of the i-th sample; Let be the material humidity of the i-th sample; The intercept; The influence coefficient for a single processing volume; The coefficient representing the influence of material hardness; Let be the influence coefficient of material moisture; the objective function used to solve for the values ​​of each influence coefficient in the linear relationship model satisfies the following equation: ; ; In the formula, SSE is the sum of squared errors between the predicted wear rate and the actual wear rate of the impact hammer after a single sand task; Let be the wear rate value of the i-th sand making task. Let be the initial wear rate before the i-th sand making task is executed. The actual wear rate is the value after the i-th sand making task is completed; both the initial wear rate and the actual wear rate are determined based on the image recognition results.

[0043] In this embodiment, the objective function can be solved in matrix form. Let matrix (X) be defined, which is an m×4 matrix where m is the number of samples, n is the number of features, and the first column is typically all 1s. The intercept term is represented by the design matrix as follows. The target vector (y) is an m×1 column vector containing the true wear rate values ​​of all samples, represented as... Coefficient vector ( () is a 4×1 column vector containing all the parameters to be calculated. Therefore, the predicted value vector of the model is represented as Sum of squared errors To further minimize the SSE value, for SSE with respect to the vector Differentiate, and thus obtain the normal equation. Solving the normal equation yields the following results. By obtaining the influence coefficients of various parameters, the wear prediction model can be solved based on these coefficients to obtain the predicted wear data of the impact hammer. This method makes full use of historical data and considers the influence of different working conditions on the wear of the impact hammer. It not only predicts wear but also analyzes the main causes of wear aggravation, providing a reliable basis for subsequent wear prediction.

[0044] In one embodiment, the step of obtaining the corresponding wear prediction model based on the operating parameters of the next sand making task of the sand making machine includes: querying the operating condition model relationship table based on the operating parameters of the next sand making task of the sand making machine; the operating condition model relationship table stores the mapping relationship between each operating condition parameter and the corresponding wear prediction model.

[0045] In this embodiment, the working condition model relationship table stores the mapping relationship between different working conditions and wear prediction models specifically trained for those conditions. Since different customers have different usage scenarios, there may be differences in raw materials, power grid stability, and other aspects, resulting in varying wear conditions of the impact hammer. Therefore, it is necessary to establish separate wear prediction models for different usage scenarios and store the model files in the wear database. Based on the working condition parameters of the next sand making task of the sand making machine, for example, material hardness = 8, material moisture = 2%, material type = granite, the system automatically searches the table to find working condition ranges that cover the currently input parameters and matches the corresponding wear prediction model. This method solves the problem that a single linear model cannot accurately describe the complex laws under all working conditions, and it has strong scalability and wide applicability.

[0046] In one embodiment, the step of obtaining the wear warning threshold includes: generating a wear prediction curve based on the predicted wear data of the impact hammer after the completion of the next sand making task, wherein the wear prediction curve fluctuates with the task time of the next sand making task; obtaining the sand formation fluctuation curve, and when the sand formation fluctuation curve exceeds the standard line, statistically analyzing the values ​​in the wear prediction curve corresponding to the same time mapping point to generate a wear warning set; the sand formation fluctuation curve is generated from the real-time operation data of the historical sand making task corresponding to the working condition; and calculating the wear warning threshold based on the wear warning set.

[0047] In this embodiment, the wear prediction curve is a curve showing the wear rate change over time during the predicted next sand making task, representing the theoretical expectation of the future equipment wear state. The sand formation fluctuation curve is a curve extracted from the real-time operation database of historical sand making tasks, reflecting the fluctuation of the sand formation rate over time. The sand formation fluctuation curve generally fluctuates around the optimal sand formation rate control line, which is provided by the manufacturer, with a control difference line of ±p%. When the historical sand formation rate begins to deteriorate, the corresponding wear state is a threshold that needs to be monitored. Therefore, by aligning the time base of the two curves, all time points when the sand formation rate values ​​exceed the standard line control zone are captured. A wear warning set is generated based on the values ​​in the wear prediction curve corresponding to these abnormal time points. The wear warning threshold is obtained by statistically calculating this set.

[0048] In one embodiment, the wear warning set includes an upper limit set and a lower limit set; the wear warning threshold includes an upper limit threshold and a lower limit threshold; the upper limit threshold is calculated based on the upper limit set, and the lower limit threshold is calculated based on the lower limit set; the calculation steps for the upper and lower limit thresholds satisfy the following formula: ; In the formula, This is the upper limit threshold for early warning. The task statistics for the upper limit set of wear and tear warnings. to The predicted wear rate for abnormal sand making tasks in the wear warning upper limit set; ; In the formula, This is the lower threshold for early warning. This is a statistical measure for the task of setting the lower limit of wear warning. to This represents the predicted wear rate for abnormal sand-making tasks within the wear warning lower limit set.

[0049] Specifically, if the sand formation rate is higher than the upper limit of the control zone, it is stored in the upper limit set of wear warnings; if it is lower than the lower limit, it is stored in the lower limit set of wear warnings. The RMS (Root Mean Square) values ​​of h groups of data in each set are calculated and used as the upper and lower limit thresholds for the warnings. The number of tasks in the upper and lower limit sets of wear warnings can be the same or different. This method ensures that the wear warning thresholds are dynamic, intelligent thresholds determined based on actual production results and historical data.

[0050] In one embodiment, the step of generating propulsion amount based on wear warning threshold and initial wear data includes: calculating propulsion amount according to the following formula when the predicted wear data exceeds the warning lower limit threshold: ; In the formula, e is the propulsion amount, r is the cross-sectional area of ​​the impact hammer, and s is the volume of the impact hammer, which is a fixed value. This is the initial wear data. This is the upper limit threshold for early warning.

[0051] The advance amount is the distance the impact hammer needs to move outward, and control commands can be output based on the advance amount. The warning upper limit threshold is the maximum allowable wear depth, indicating that when the wear reaches this value, the equipment will enter a high-risk or inefficient operating state. The wear rate difference calculates the volumetric wear rate that needs to be compensated. s is the volume of the impact hammer in its unworn state, which is a fixed value. The wear rate difference and the impact hammer volume are used to calculate the wear volume that needs to be compensated. When wear occurs at the end of the impact hammer, the increased clearance after wear is reflected across the entire cross-sectional area. Therefore, the propulsion amount is calculated by dividing by the cross-sectional area of ​​the impact hammer to achieve propulsion amount calculation based on volume compensation.

[0052] In one embodiment, such as Figure 2 As shown, the steps for determining the initial wear data of the impact hammer based on the image include: preprocessing the image; the preprocessing includes grayscale conversion, filtering and noise reduction, and contrast enhancement; positioning and edge detection of the preprocessed image according to the encoder on the impact hammer to obtain the coordinates of the first region; extracting the features of the ring liner region from the coordinates of the first region and determining the reference surface to obtain the coordinates of the second region; and performing pixel coordinate transformation on the coordinates of the first region and the coordinates of the second region using a standard gap value to generate the initial wear data.

[0053] To improve image quality, image preprocessing is necessary. Preprocessing includes grayscale conversion, noise reduction filtering, and contrast enhancement, specifically as follows: First, the color image is converted to grayscale to reduce subsequent computation, using a weighted method: Gray (grayscale value) = 0.299 × R + 0.587 × G + 0.114 × B. Second, median filtering or Gaussian filtering can be used to remove noise caused by dust and high-speed motion. Gaussian filtering uses a Gaussian kernel convolved with the image, and the Gaussian function... Where (x, y) is the offset distance between the point and the kernel center, and σ is the standard deviation used to control the smoothing degree. Finally, the CLAHE method (limited contrast adaptive histogram equalization) is used for contrast enhancement. This method divides the image into small blocks, performs histogram equalization on each block, and limits the magnification factor, thereby enhancing edge contrast while suppressing background noise.

[0054] The preprocessed image is positioned and edge detected by the encoder on the impact hammer to obtain the coordinates of the first region; the features of the ring liner region are extracted from the coordinates of the first region and the reference plane is determined to obtain the coordinates of the second region; the coordinates of the first region and the coordinates of the second region are transformed into pixel coordinates by using the standard gap value.

[0055] Further, the U-shaped contour of the impact hammer is located. Based on the impeller speed and the encoder trigger phase of the impact hammer, a Region of Interest (ROI) is defined to estimate the approximate area where the impact hammer will appear in the image, thus narrowing the search range. Edge detection is performed within the ROI using the Canny operator. The Canny operator first calculates the image gradient, then performs non-maximum suppression and dual-threshold hysteresis connections to obtain refined edges. Image gradient. , , and These are the horizontal and vertical gradients obtained after convolution using the Sobel operator. The Hough transform is used to detect straight line segments in the edges. The U-shaped structure consists of two distinct parallel vertical lines and a horizontal line (or arc) connecting them. An algorithm is used to find combinations of straight lines that form the U-shape, thereby locating the outer edge of the impact hammer's wear end. After edge detection and straight line / curve fitting, the set of pixel coordinates of the impact hammer's wear edge in the image is obtained, generating the coordinates of the first region.

[0056] Further, it is necessary to locate the edge of the ring liner as the measurement reference. The honeycomb structure of the ring liner is a periodic texture feature, which can be segmented using methods such as LBP (Local Binary Pattern) or GLCM (Gray-Level Co-occurrence Matrix) to separate the ring liner region from the background. Binarization and morphological closing operations are performed on the texture region to connect discontinuous edges. Then, a contour-finding algorithm is used to find the contours of honeycomb holes or honeycomb protrusions, selecting the outer edge of one or more honeycomb units closest to the impact hammer. Since the honeycomb surface may be worn, the algorithm can calculate the average position of multiple honeycomb edges at the same radial depth to improve anti-interference and accuracy. Finally, the pixel coordinates of the determined ring liner reference edge in the image are used as the coordinates of the second region.

[0057] In this embodiment, the extracted pixel distance also needs to be converted into the actual physical distance. First, the number of pixels between two pixels is calculated to obtain the minimum Euclidean distance between the two edge point sets. In practical applications, multiple positions can be taken along the radial direction of the image for calculation, and then the average value is taken to eliminate the error caused by tilt. Second, a camera calibration model is established. In the initial stage of impact hammer installation, a standard gap value is measured and the corresponding pixel distance is measured to obtain the conversion coefficient k. Then, the pixel distance between the obtained first area coordinates and second area coordinates is substituted into k for conversion to generate initial wear data.

[0058] This method uses image processing technology to transform physical information into digital information that can be processed and monitored, providing a data foundation for subsequent prediction and automatic control.

[0059] In one embodiment, such as Figure 3 As shown, after adjusting the physical gap between the impact hammer and the ring liner based on the advance amount, the method further includes: acquiring an image of the area between the impact hammer and the ring liner after adjustment; determining the step amount of the impact hammer based on the image; generating an adjustment result based on the advance amount and the step amount; and providing fault information if the adjustment result does not meet the requirements.

[0060] To verify whether the adjustment results meet the standards, the system triggers the camera again to acquire a new image of the area between the impact hammer and the ring liner. Based on this image, the system determines the step size of the impact hammer, compares the advance amount with the actual adjustment step size, sets an allowable error range, and generates the adjustment result. If the adjustment does not meet the standards, an alarm signal is issued on the control page to report the fault, allowing maintenance personnel to intervene promptly.

[0061] In addition, during normal operation of the sand making machine, the real-time sand production rate curve will fall within the range of ±p% of the optimal sand production rate control difference line. If the sand production rate curve deviates abnormally and exceeds a certain period of time, it indicates that there may be faults such as the fixed impact hammer falling off, iron entering the raw material, or abnormal wear of the material distribution chamber. At this time, the system will alarm the user, and the operator will identify the fault and carry out emergency shutdown and maintenance.

[0062] In one embodiment, such as Figure 4 As shown, an online monitoring and adjustment device for the impact hammer of a sand making machine is provided, comprising: Image acquisition module 401 is used to acquire images of the impact hammer and ring liner area of ​​the sand making machine when it is stopped.

[0063] Feature extraction module 402 is used to determine the initial wear data of the impact hammer based on the image.

[0064] The model acquisition module 403 is used to acquire the corresponding wear prediction model based on the working condition parameters of the sand making machine in the next sand making task; the wear prediction model is constructed based on each working condition parameter and the corresponding impact hammer wear data.

[0065] The wear prediction module 404 is used to input the working parameters and initial wear data of the next sand making task into the wear prediction model to obtain the predicted wear data of the impact hammer after the next sand making task is completed.

[0066] The wear adjustment module 405 is used to obtain the wear warning threshold. When the predicted wear data exceeds the wear warning threshold, it generates a propulsion amount based on the wear warning threshold and the initial wear data. Based on the propulsion amount, it adjusts the physical gap between the impact hammer and the ring liner.

[0067] In one embodiment, a prediction model construction module is further included, which is used to obtain multiple historical sand making samples, each sample including operating parameters and measured values ​​of impact hammer wear. The operating parameters include single-batch throughput, material hardness, and material moisture content. A linear relationship model between the wear prediction value and the single-batch throughput, material hardness, and material moisture content is established. With the goal of minimizing the error between the wear prediction value and the measured wear value of the impact hammer, the values ​​of each influence coefficient of the linear relationship model are solved by a linear regression algorithm and historical sand making samples. The values ​​of each influence coefficient are substituted into the linear relationship model to obtain the wear prediction model.

[0068] In one embodiment, the linear relationship satisfies the following formula: ; In the formula, Let be the predicted wear rate value for the i-th sample; This represents the processing volume for the i-th sample in a single run. Let be the material hardness of the i-th sample; Let be the material humidity of the i-th sample; The intercept; The influence coefficient for a single processing volume; The coefficient representing the influence of material hardness; Let be the influence coefficient of material moisture; the objective function used to solve for the values ​​of each influence coefficient in the linear relationship model satisfies the following equation: ; ; In the formula, SSE is the sum of squared errors between the predicted wear rate and the actual wear rate of the impact hammer after a single sand task; Let be the wear rate value of the i-th sand making task. Let be the initial wear rate before the i-th sand making task is executed. The actual wear rate is the value after the i-th sand making task is completed; both the initial wear rate and the actual wear rate are determined based on the image recognition results.

[0069] In one embodiment, the feature extraction module 402 includes an image processing module for preprocessing the image; the preprocessing includes grayscale conversion, filtering and noise reduction, and contrast enhancement; the preprocessed image is positioned and edge detected according to the encoder on the impact hammer to obtain the coordinates of a first region; the features of the ring liner region are extracted from the coordinates of the first region and the reference surface is determined to obtain the coordinates of a second region; the coordinates of the first region and the second region are converted into pixel coordinates using a standard gap value to generate initial wear data.

[0070] In one embodiment, the model acquisition module 403 is specifically used to query the working condition model relationship table based on the working condition parameters in the next sand making task of the sand making machine; the working condition model relationship table stores the mapping relationship between each working condition parameter and the corresponding wear prediction model.

[0071] In one embodiment, the wear adjustment module 405 is specifically used to generate a wear prediction curve based on the predicted wear data of the impact hammer after the next sand making task is completed. The wear prediction curve fluctuates with the task time of the next sand making task. The sand forming fluctuation curve is obtained. When the sand forming fluctuation curve exceeds the standard line, the values ​​in the wear prediction curve corresponding to the same time mapping point are statistically analyzed to generate a wear warning set. The sand forming fluctuation curve is generated by the real-time operation data of the historical sand making task corresponding to the working condition. The wear warning threshold is calculated based on the wear warning set.

[0072] In one embodiment, the wear warning set includes an upper limit set and a lower limit set; the wear warning threshold includes an upper limit threshold and a lower limit threshold; the upper limit threshold is calculated based on the upper limit set, and the lower limit threshold is calculated based on the lower limit set; the calculation steps for the upper and lower limit thresholds satisfy the following formula: ; In the formula, This is the upper limit threshold for early warning. The task statistics for the upper limit set of wear and tear warnings. to The predicted wear rate for abnormal sand making tasks in the wear warning upper limit set; ; In the formula, This is the lower threshold for early warning. This is a statistical measure for the task of setting the lower limit of wear warning. to This represents the predicted wear rate for abnormal sand-making tasks within the wear warning lower limit set.

[0073] In one embodiment, the wear adjustment module 405 further includes a gap calculation module for calculating the propulsion amount according to the following formula when the predicted wear data exceeds the warning lower limit threshold. ; In the formula, e is the propulsion amount, r is the cross-sectional area of ​​the impact hammer, and s is the volume of the impact hammer, which is a fixed value. This is the initial wear data. This is the upper limit threshold for early warning.

[0074] In one embodiment, the device further includes an adjustment feedback module for acquiring an image of the area between the impact hammer and the ring liner after adjustment; determining the step size of the impact hammer based on the image; generating an adjustment result based on the advance amount and the step size; and providing fault information if the adjustment result does not meet the requirements.

[0075] In one embodiment, such as Figure 5 As shown, the area between the impact hammer and the ring liner is set as the wear zone. Since there are 12 sets of fixed impact hammers, there are 12 sets of wear zones. By detecting the blank areas of the wear zone through image recognition and calculating the ratio of the blank areas to the wear zone, the wear rate can be obtained, which is used to represent the wear condition of the impact hammer.

[0076] Please refer to the following: Figure 6 to Figure 10 This application also provides an online monitoring and adjustment system for the impact hammer of a sand making machine, including: a state perception subsystem, an intelligent analysis subsystem communicatively connected to the state perception subsystem, and an interval execution subsystem communicatively connected to the intelligent analysis subsystem; The system comprises the following subsystems: a state perception subsystem for acquiring images of the area between the impact hammer 6 and the ring liner 8 of the sand making machine when it is stopped; an intelligent analysis subsystem for determining the initial wear data of the impact hammer 6 based on these images; a wear prediction model for obtaining the corresponding wear prediction model based on the operating parameters of the next sand making task; a wear prediction model constructed based on each operating parameter and the corresponding wear data of the impact hammer 6; inputting the operating parameters and initial wear data of the next sand making task into the wear prediction model to obtain the predicted wear data of the impact hammer 6 after the next sand making task is completed; obtaining a wear warning threshold; and generating a propulsion amount based on the wear warning threshold and the initial wear data when the predicted wear data exceeds the wear warning threshold; and a gap execution subsystem for adjusting the physical gap between the impact hammer 6 and the ring liner 8 based on the propulsion amount.

[0077] The state perception subsystem is the information acquisition mechanism, responsible for collecting all direct or indirect signals related to the gap. It may include components such as an industrial camera, encoder, and lighting system.

[0078] The intelligent analysis subsystem is responsible for receiving and processing the acquired image signals to determine the wear condition of the impact hammer, predict the future, and make decisions. The intelligent analysis subsystem can consist of an edge computing gateway, a digital model, an AI fusion algorithm, and decision-making and threshold judgment functions. The edge computing gateway can preprocess the raw data returned by all state perception subsystems; the digital model can build a device-level digital model, which includes the device's physical parameters, initial geometric dimensions, and a wear prediction model trained on a large amount of historical data. The AI ​​fusion algorithm can include image processing algorithms and computer vision algorithms to fuse and analyze information, ultimately outputting a high-confidence estimate of the physical gap. The decision-making and threshold judgment function determines whether the wear condition has reached a level requiring adjustment based on preset rules and generates adjustment instructions. Correspondingly, the warning threshold can be dynamically generated according to different material characteristics.

[0079] The intermittent execution subsystem may include signal transceivers, motors, etc., and can perform actions based on signals from the intelligent analysis subsystem.

[0080] In one embodiment, such as Figure 6 , Figure 7 As shown, the state perception subsystem includes multiple shielding protection modules and multiple data acquisition modules. The data acquisition modules and shielding protection modules are respectively connected to the intelligent analysis subsystem. The multiple data acquisition modules are arranged in a circumferential direction at the top of the feed chamber 1 of the sand making machine to collect images of the impact hammer and the ring liner area. The shielding protection modules correspond one-to-one with the data acquisition modules to prevent the corresponding data acquisition modules from being damaged during the sand making process.

[0081] The primary function of the shielding protection module is to protect the data acquisition module. During normal operation of the sand making machine, the internal working environment is extremely harsh, and the crushed sand and gravel can easily damage the data acquisition module. Therefore, the data acquisition module needs to be shielded for protection during normal operation of the sand making machine. Figure 7 As shown, 10 shielding doors 3 and 10 data acquisition modules are arranged in the top circumferential direction of the feeding chamber 1. The shielding doors 3 are opened and closed by the shielding door screws 5 driven by the shielding door motor 4. The data acquisition modules can communicate with the intelligent analysis and decision-making subsystem through 4G or WIFI network and issue a forced power-off command to put the sand making machine in a forced power-off state. At this time, the shielding doors 3 are opened, and the data acquisition modules take pictures of the wear of the impact hammers through the industrial camera 2. One data acquisition module can be set to take pictures of the wear area of ​​all impact hammers, or multiple modules can be set to take pictures one by one with the impact hammers 6 and transmit the data back. The intelligent analysis subsystem processes the image information.

[0082] In one embodiment, such as Figure 8 , Figure 9 , Figure 10As shown, the gap execution subsystem includes a stepper control module, which comprises a drive motor 10, a connecting screw 9, and a nut fitted onto the connecting screw 9. The nut is connected to the bottom of the impact hammer 6. The stepper control module is communicatively connected to the intelligent analysis subsystem and is used to receive the propulsion amount to rotate the screw and drive the impact hammer 6 to move radially towards the impeller. The drive motor can be piezoelectric, shape memory alloy, or micro hydraulic, etc., and is not limited here.

[0083] The 12 impact hammers 6 mounted on the impeller structure of the sand making machine have built-in electronic codes, which are respectively coded as A1, A2, A3, A4, A5, A6, A7, A8, A9, A10, A11, and A12. The intelligent analysis subsystem can calculate the gap of each impact hammer individually, and can make the stepping amount and stepping control of each impact hammer 6 independent.

[0084] Furthermore, after the drive motor 10 in the intermittent execution subsystem completes the stepping action, it feeds back to the intelligent analysis subsystem, which in turn instructs the status perception subsystem to perform secondary sensing and re-capture images of the area between the impact hammer 6 and the ring liner plate 8. The image information is then sent back to the intelligent analysis subsystem to verify whether each impact hammer 6 has stepped into position. If it has, the status perception subsystem is powered off, the shielding door 3 is closed simultaneously, and the sand making host is allowed to be powered on. If it has not stepped into position, the stepping adjustment is performed again. If the secondary fine-tuning is still not in place, an alarm signal is issued and a fault is reported, thereby prompting maintenance personnel to intervene.

[0085] In one embodiment, the system further includes a human-machine interaction module for real-time display of data such as gap values, equipment status, trend curves, and alarm information.

[0086] In one embodiment, the system also includes a log recording module that can record all gap data, operation logs, and alarm events for production reporting and process optimization.

[0087] In one embodiment, a machine-readable storage medium is provided having a computer program stored thereon that, when executed by a processor, implements the steps in the above method embodiments.

[0088] Machine-readable media, including both permanent and non-permanent, removable and non-removable media, can store information using any method or technology. Information can be computer-readable instructions, data structures, program modules, or other data. Examples of storage media include, but are not limited to, phase-change memory (PRAM), static random access memory (SRAM), dynamic random access memory (DRAM), other types of random access memory (RAM), read-only memory (ROM), electrically erasable programmable read-only memory (EEPROM), flash memory or other memory technologies, CD-ROM, digital versatile optical disc (DVD) or other optical storage, magnetic tape, magnetic disk storage or other magnetic storage devices, or any other non-transfer medium that can be used to store information accessible by a computing device.

[0089] This invention provides a processor for running a program, wherein the program executes a method for online monitoring and adjustment of the impact hammer of a sand making machine.

[0090] In one embodiment, a computer device is provided, which may be a mobile terminal, and the internal structure diagram of the computer device may be as follows: Figure 11 As shown, the computer device includes a processor, memory, input / output interface, communication interface, display unit, and input device. The processor, memory, and input / output interface are connected via a system bus, and the communication interface, display unit, and input device are also connected to the system bus via the input / output interface. The processor provides computing and control capabilities. The memory includes non-volatile storage media and internal memory. The non-volatile storage media stores the operating system and computer programs. The internal memory provides an environment for the operation of the operating system and computer programs in the non-volatile storage media. The input / output interface is used for exchanging information between the processor and external devices. The communication interface is used for wired or wireless communication with external terminals; wireless communication can be achieved through Wi-Fi, mobile cellular networks, NFC (Near Field Communication), or other technologies. When the computer program is executed by the processor, it implements a method for online monitoring and adjustment of the impact hammer of a sand making machine. The display unit is used to form a visually visible image and can be a display screen, projection device, or virtual reality imaging device. The display screen can be an LCD screen or an e-ink screen. The input device of the computer device can be a touch layer covering the display screen, or buttons, trackballs, or touchpads set on the casing of the computer device, or external keyboards, touchpads, or mice, etc.

[0091] This application also provides a computer program product, including a computer program that, when executed by a processor, implements the steps in the above-described method embodiments.

[0092] Those skilled in the art will understand that embodiments of this application can be provided as methods, systems, or computer program products. Therefore, this application can take the form of a completely hardware embodiment, a completely software embodiment, or an embodiment combining software and hardware aspects.

[0093] This application is described with reference to flowchart illustrations and / or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of this application. It will be understood that each block of the flowchart illustrations and / or block diagrams, and combinations of blocks in the flowchart illustrations and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, special-purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, generate instructions for implementing the flowchart... Figure 1 One or more processes and / or boxes Figure 1 A device that provides the functions specified in one or more boxes.

[0094] It should also be noted that the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such process, method, article, or apparatus. Unless otherwise specified, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the process, method, article, or apparatus that includes that element.

[0095] The above are merely embodiments of this application and are not intended to limit the scope of this application. Various modifications and variations can be made to this application by those skilled in the art. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of this application should be included within the scope of the claims of this application.

Claims

1. A method for online monitoring and adjustment of the impact hammer of a sand making machine, characterized in that, include: Acquire images of the impact hammer and ring liner area of ​​the sand making machine when it is stopped; The initial wear data of the impact hammer was determined based on this image; The wear prediction model is obtained based on the operating parameters of the sand making machine in the next sand making task; the wear prediction model is constructed based on each operating parameter and the corresponding impact hammer wear data. The working parameters of the next sand making task and the initial wear data are input into the wear prediction model to obtain the predicted wear data of the impact hammer after the completion of the next sand making task. Obtain a wear warning threshold, and if the predicted wear data exceeds the wear warning threshold, generate an advance amount based on the wear warning threshold and the initial wear data; Based on the propulsion amount, adjust the physical gap between the impact hammer and the ring liner.

2. The method according to claim 1, characterized in that, The steps for constructing the wear prediction model include: Multiple historical sand making samples were obtained. Each sample included operating parameters and measured values ​​of impact hammer wear. The operating parameters included single processing volume, material hardness value, and material moisture value. A linear relationship model was established between the wear prediction value and the single processing volume, material hardness, and material moisture content. With the goal of minimizing the error between the wear prediction value and the actual wear value of the impact hammer, the values ​​of each influence coefficient of the linear relationship model were solved by linear regression algorithm and historical sand making samples. Substituting the values ​​of each influence coefficient into the linear relationship model yields the wear prediction model.

3. The method according to claim 2, characterized in that, The linear relationship satisfies the following formula: ； In the formula, Let be the predicted wear rate value for the i-th sample; This represents the processing volume for the i-th sample in a single run. Let be the material hardness of the i-th sample; Let be the material humidity of the i-th sample; The intercept; The influence coefficient of the single processing volume; The coefficient representing the influence of the material's hardness; The influence coefficient of the material's humidity; The objective function used to solve for the values ​​of each influence coefficient in the linear relationship model satisfies the following equation: ； ； In the formula, SSE is the sum of squared errors between the predicted wear rate and the actual wear rate of the impact hammer after a single sand task; Let be the wear rate value of the i-th sand making task. Let be the initial wear rate before the i-th sand making task is executed. The initial wear rate and the actual wear rate are both determined based on image recognition results after the i-th sand making task is completed.

4. The method according to claim 3, characterized in that, The step of obtaining the corresponding wear prediction model based on the operating parameters of the sand making machine in the next sand making task includes: Based on the operating parameters of the sand making machine in the next sand making task, query the operating condition model relationship table; The working condition model relationship table stores the mapping relationship between each working condition parameter and the corresponding wear prediction model.

5. The method according to any one of claims 1-4, characterized in that, The step of obtaining the wear warning threshold includes: Based on the predicted wear data of the impact hammer after the completion of the next sand making task, a wear prediction curve is generated, and the wear prediction curve fluctuates with the task time of the next sand making task. The sand formation fluctuation curve is obtained, which is generated by the real-time operation data of the historical sand making task corresponding to the working condition; When the sand formation fluctuation curve exceeds the standard line, the values ​​in the wear prediction curve corresponding to the same time mapping point are statistically analyzed to generate a wear warning set. The wear warning threshold is calculated based on the wear warning set.

6. The method according to claim 5, characterized in that, The wear warning set includes a wear warning upper limit set and a wear warning lower limit set; the wear warning threshold includes a warning upper limit threshold and a warning lower limit threshold. The upper warning threshold is calculated based on the wear warning upper limit set, and the lower warning threshold is calculated based on the wear warning lower limit set; The calculation steps for the upper and lower warning thresholds satisfy the following formula: ； In the formula, The upper limit threshold for the warning is... The task statistics for the wear warning upper limit set. to The predicted wear rate of abnormal sand making tasks in the wear warning upper limit set; ； In the formula, The lower threshold value for the early warning is... The task statistics for the wear warning lower limit set are as follows: to The wear rate is the predicted wear rate of abnormal sand making tasks in the wear warning lower limit set.

7. The method according to claim 6, characterized in that, The step of generating a propulsion amount based on the wear warning threshold and the initial wear data when the predicted wear data exceeds the wear warning threshold includes: If the predicted wear data exceeds the warning lower limit threshold, the propulsion amount is calculated according to the following formula; ； In the formula, e is the propulsion amount, r is the cross-sectional area of ​​the impact hammer, s is the volume of the impact hammer, and is a fixed value. The initial wear data, The upper limit threshold for the warning is defined as follows.

8. The method according to any one of claims 1-4, characterized in that, The step of determining the initial wear data of the impact hammer based on the image includes: The image is preprocessed; the preprocessing includes grayscale conversion, filtering and noise reduction, and contrast enhancement. The preprocessed image is used to perform positioning and edge detection based on the encoder on the impact hammer to obtain the coordinates of the first region. The second region coordinates are obtained by extracting the features of the ring liner region from the first region coordinates and determining the reference plane. The initial wear data is generated by converting the coordinates of the first and second regions into pixel coordinates using a standard gap value.

9. The method according to any one of claims 1-4, characterized in that, After the step of adjusting the physical gap between the impact hammer and the ring liner based on the propulsion amount, the method further includes: Acquire images of the area between the impact hammer and the ring liner after adjustment; The step size of the impact hammer is determined based on the image; An adjustment result is generated based on the propulsion amount and the step amount; If the adjustment results do not meet the requirements, feedback of fault information will be provided.

10. An online monitoring and adjustment device for the impact hammer of a sand making machine, characterized in that, include: The image acquisition module is used to acquire images of the impact hammer and ring liner area of ​​the sand making machine when it is stopped. The feature extraction module is used to determine the initial wear data of the impact hammer based on the image; The model acquisition module is used to acquire the corresponding wear prediction model based on the working condition parameters of the sand making machine in the next sand making task; the wear prediction model is constructed based on each working condition parameter and the corresponding impact hammer wear data. The wear prediction module is used to input the working condition parameters of the next sand making task and the initial wear data into the wear prediction model to obtain the predicted wear data of the impact hammer after the completion of the next sand making task. The wear adjustment module is used to obtain a wear warning threshold, and when the predicted wear data exceeds the wear warning threshold, generate a propulsion amount based on the wear warning threshold and the initial wear data; and adjust the physical gap between the impact hammer and the ring liner based on the propulsion amount.

11. An online monitoring and adjustment system for the impact hammer of a sand making machine, characterized in that, include: A state-aware subsystem, an intelligent analysis subsystem communicatively connected to the state-aware subsystem, and an intermittent execution subsystem communicatively connected to the intelligent analysis subsystem; The state perception subsystem is used to collect images of the impact hammer and ring liner area of ​​the sand making machine when it is stopped. The intelligent analysis subsystem is used to determine the initial wear data of the impact hammer based on the image; The wear prediction model is obtained based on the operating parameters of the sand making machine in the next sand making task; the wear prediction model is constructed based on each operating parameter and the corresponding impact hammer wear data. The working parameters of the next sand making task and the initial wear data are input into the wear prediction model to obtain the predicted wear data of the impact hammer after the completion of the next sand making task. Obtain a wear warning threshold, and if the predicted wear data exceeds the wear warning threshold, generate an advance amount based on the wear warning threshold and the initial wear data; The gap execution subsystem is used to adjust the physical gap between the impact hammer and the ring liner based on the propulsion amount.

12. The system according to claim 11, characterized in that, The state perception subsystem includes multiple shielding protection modules and multiple data acquisition modules, and the data acquisition modules and the shielding protection modules are respectively communicatively connected to the intelligent analysis subsystem. The multiple data acquisition modules are arranged circumferentially at the top of the feed chamber of the sand making machine to acquire images of the area between the impact hammer and the ring liner. The shielding protection module corresponds one-to-one with the data acquisition module, and is used to prevent the corresponding data acquisition module from being damaged during the sand making process.

13. A machine-readable storage medium, characterized in that, The machine-readable storage medium stores instructions for causing the machine to perform the method according to any one of claims 1-9.

Images