A metal plate grinding process optimization method and system
By monitoring and predicting the wear amount and temperature of high-hardness metal plates in real time, and conducting intelligent process control, the grinding burns and cracks caused by high temperature are solved, the grinding quality and efficiency are improved, and the surface quality and accuracy of the metal plates are ensured.
Patent Information
- Application Number
- CN202411001517.2
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- Filing Date
- 2024-07-25
- Publication Date
- 2024-11-12
- Estimated Expiration
- 2044-07-25
AI Technical Summary
High-hard metal plates are prone to grinding burns and cracks due to high temperatures during grinding, and lack intelligent control methods, which affects the quality and efficiency of grinding.
The wear amount and temperature are monitored in real time by predetermined sensing equipment, and a wear amount and temperature prediction model is constructed. Based on these models, process regulation instructions are issued, process optimization plans are activated, grinding process parameters are optimized, and surface quality and accuracy of metal plates are ensured.
Real-time monitoring and intelligent process control during the grinding process of high-hardness metal plates are realized, the grinding quality and efficiency are improved, the occurrence of grinding burns and cracks is reduced, and the surface quality and accuracy of the metal plates are ensured.
Smart Images

Figure CN118528084B_ABST
Abstract
Description
Technical Field
[0001] The present application relates to data processing related fields, and specifically to a metal plate grinding process optimization method and system. Background Art
[0002] The grinding process of metal plates is a key link in ensuring the surface quality and precision of metal plates, especially for high-hardness metal plates such as tool steel, hardened steel, etc. The grinding process faces great technical challenges. During the grinding process, it is not only difficult to remove surface unevenness and defects of such high-hardness metal plates, but also easy to cause grinding burns and cracks due to the high temperature of grinding, thereby affecting the performance and service life of the metal plates. The traditional grinding process mainly relies on experience and manual control, and it is difficult to monitor and accurately control the amount of wear and temperature changes in real time. This leads to unstable grinding quality and cannot guarantee the efficiency and high precision of the grinding process. In addition, the existing grinding process lacks intelligent control means, and it is difficult to optimize the process based on real-time data, which further limits the effect and efficiency of the grinding process.
[0003] Therefore, in the existing technology, grinding burns and cracks caused by high grinding temperatures are difficult to solve, and there is a lack of intelligent control methods to optimize grinding process parameters, which affects the technical problems of grinding quality and efficiency. Summary of the Invention
[0004] This application provides a method and system for optimizing the metal plate grinding process, solving the technical problems in the prior art of grinding burns and cracks caused by high grinding temperatures, as well as the lack of intelligent control methods to optimize grinding process parameters, which affects grinding quality and efficiency. The system achieves real-time monitoring of wear and temperature during the grinding of high-hardness metal plates, constructs intelligent wear prediction models and temperature prediction models, comprehensively evaluates the stability of the grinding process, and completes intelligent process control and optimization, significantly improving grinding quality and efficiency and ensuring the surface quality and precision of the metal plates.
[0005] The present application provides a method for optimizing a metal plate grinding process, the method comprising: dynamically monitoring the grinding of a target metal plate through a predetermined sensing device to obtain target grinding monitoring information. Analyzing the first wear information of the first mold in the target grinding monitoring information to obtain a first predicted wear amount, the first mold corresponding to a first predetermined wear threshold. Analyzing the first temperature information of the first point in the target grinding monitoring information to obtain a first predicted temperature. When the first predicted wear amount reaches the first predetermined wear threshold or the first predicted temperature reaches the predetermined temperature threshold, a first process control instruction is issued. Based on the first process control instruction, the target process optimization plan is activated to optimize the grinding process of the target metal plate to obtain a target optimal grinding process. The process optimization and control of the target metal plate is performed according to the target optimal grinding process.
[0006] In an implementation, the predetermined sensing device includes at least a wear sensor and a temperature sensor.
[0007] In an implementation, a predetermined grinding process is obtained. The predetermined grinding process is a process for grinding the target metal plate. The predetermined grinding process includes multiple grinding processes, each of which corresponds to a mold of a specific grit size. A first process of the target metal plate is matched to the predetermined grinding process, the first process corresponding to the first mold, and the first predetermined wear threshold is determined based on the specific grit size of the first mold.
[0008] In the implementation method, a predetermined wear feature collection strategy is read. Based on the predetermined wear feature collection strategy, multi-dimensional feature collection is performed on the first wear amount time series in the first wear information to obtain a first wear feature parameter. The first wear feature parameter is used as input information of the wear amount prediction model to obtain output information, and the output information includes the first predicted wear amount. The training process of the wear amount prediction model is as follows: the historical wear amount time series record of the same type of metal plate as the target metal plate is obtained, and the first historical record in the historical wear amount time series record is extracted. The first historical record refers to the first historical wear feature parameter before the first historical time and the first historical wear amount at the first historical time. A first data group is formed based on the first historical wear feature parameter and the first historical wear amount, and the first data group is supervised trained and learned based on the principle of neural network to obtain the wear amount prediction model.
[0009] In an implementation, the predetermined wear feature collection strategy includes a wear time domain feature strategy, a wear frequency domain feature strategy, and a wear time-frequency domain feature strategy. Based on the wear time domain feature strategy, the time domain features of the first wear amount time series are collected to obtain a first wear time domain feature parameter. Based on the wear frequency domain feature strategy, the frequency domain features of the first wear amount spectrum after Fourier transform of the first wear amount time series are collected to obtain a first wear frequency domain feature parameter. Based on the wear time-frequency domain feature strategy, the time-frequency domain features of the first target modal component after modal decomposition of the first wear amount time series are collected to obtain a first wear time-frequency domain feature parameter. The first wear time domain feature parameter, the first wear frequency domain feature parameter, and the first wear time-frequency domain feature parameter are collectively used as the first wear feature parameter.
[0010] In an implementation, a polynomial regression fitting is performed on the first temperature time series obtained by analyzing the first temperature information based on a spline curve principle to obtain a first fitting formula, and the first predicted temperature is obtained based on the first fitting formula.
[0011] In an implementation, when the first predicted wear amount does not reach the first predetermined wear threshold and the first predicted temperature does not reach the predetermined temperature threshold, a process evaluation instruction is issued. Based on the process evaluation instruction, a predetermined process fitness evaluation mechanism is introduced to evaluate and analyze the first wear amount time series and the first temperature time series to obtain a target process fitness. A determination is made as to whether the target process fitness meets the fitness limit. If not, a second process control instruction is issued, and based on the second process control instruction, the grinding process of the target metal plate is optimized to obtain the target optimal grinding process.
[0012] In an implementation, according to the predetermined process fitness assessment mechanism, the first wear frequency domain characteristic parameter of the first wear amount time series and the first temperature frequency domain characteristic parameter of the first temperature time series are sequentially acquired. The first wear frequency domain characteristic parameter is weighted and normalized to obtain a first wear stability index. The first temperature frequency domain characteristic parameter is weighted and normalized to obtain a first temperature stability index. Predetermined correlation coefficients are read, including a predetermined wear correlation coefficient and a predetermined temperature correlation coefficient. The product of the weighted predetermined wear correlation coefficient and the first wear stability index, and the product of the weighted predetermined temperature correlation coefficient and the first temperature stability index are weighted to obtain the target process fitness.
[0013] In an implementation manner, the target process optimization plan refers to a plan for intelligently optimizing predetermined grinding indicators, and the predetermined grinding indicators at least include grinding speed, grinding pressure, feed speed and grinding tool rotation speed.
[0014] The present application also provides a metal plate grinding process optimization system, comprising:
[0015] The monitoring module is used to dynamically monitor the grinding of the target metal plate through a predetermined sensing device to obtain target grinding monitoring information.
[0016] The wear amount prediction module is used to analyze the first wear information of the first mold in the target grinding monitoring information to obtain a first predicted wear amount, and the first mold corresponds to a first predetermined wear threshold.
[0017] The temperature prediction module is used to analyze the first temperature information of the first point in the target polishing monitoring information to obtain a first predicted temperature.
[0018] The control instruction acquisition module is configured to issue a first process control instruction when the first predicted wear amount reaches the first predetermined wear threshold or the first predicted temperature reaches a predetermined temperature threshold.
[0019] The process optimization module is used to activate the target process optimization plan based on the first process control instruction to optimize the grinding process of the target metal plate and obtain the target optimal grinding process.
[0020] The optimization and control module is used to optimize and control the process of the target metal plate according to the target optimal grinding process.
[0021] The present application proposes a method and system for optimizing the grinding process of a metal plate, which is used to dynamically monitor the grinding of a target metal plate through a predetermined sensing device to obtain target grinding monitoring information. The first wear information of the first mold in the target grinding monitoring information is analyzed to obtain a first predicted wear amount, and the first mold corresponds to a first predetermined wear threshold. The first temperature information of the first point in the target grinding monitoring information is analyzed to obtain a first predicted temperature. When the first predicted wear amount reaches the first predetermined wear threshold or the first predicted temperature reaches the predetermined temperature threshold, a first process control instruction is issued. Based on the first process control instruction, the target process optimization plan is activated to optimize the grinding process of the target metal plate to obtain the target optimal grinding process. The process of the target metal plate is optimized and controlled according to the target optimal grinding process. This solves the technical problems in the prior art that grinding burns and cracks caused by high grinding temperatures are difficult to solve, and there is a lack of intelligent control means to optimize the grinding process parameters, which affects the grinding quality and efficiency. Real-time monitoring of wear and temperature during the grinding process of high-hardness metal plates is achieved, intelligent wear prediction models and temperature prediction models are constructed, the stability of the grinding process is comprehensively evaluated, and intelligent process control and optimization are completed, which greatly improves the grinding quality and efficiency and ensures the technical effect of the surface quality and precision of the metal plates. BRIEF DESCRIPTION OF THE DRAWINGS
[0022] To more clearly illustrate the technical solutions of the embodiments of the present disclosure, the accompanying drawings of the embodiments of the present disclosure are briefly introduced below. Flowcharts are used in this application to illustrate the operations performed by the systems according to the embodiments of the present application. It should be understood that the preceding or following operations are not necessarily performed in precise order. Instead, various steps may be processed in reverse order or simultaneously as needed. Furthermore, other operations may be added to these processes, or one or more operations may be removed from these processes.
[0023] Figure 1 A schematic flow chart of a metal plate grinding process optimization method provided in an embodiment of the present application;
[0024] Figure 2 A schematic diagram of the structure of a metal plate grinding process optimization system provided in an embodiment of the present application.
[0025] Explanation of the accompanying symbols: monitoring module 11, wear amount prediction module 12, temperature prediction module 13, control instruction acquisition module 14, process optimization module 15, optimization control module 16. DETAILED DESCRIPTION
[0026] The above description is only an overview of the technical solution of the present application. In order to more clearly understand the technical means of the present application, it can be implemented in accordance with the contents of the specification. In order to make the above and other purposes, features and advantages of the present application more obvious and easy to understand, the specific implementation methods of the present application are listed below.
[0027] In order to make the purpose, technical solutions and advantages of this application clearer, the application will be further described in detail below with reference to the accompanying drawings. The described embodiments should not be regarded as limiting this application. All other embodiments obtained by ordinary technicians in this field without making creative work are within the scope of protection of this application.
[0028] In the following description, reference is made to “some embodiments”, which describes a subset of all possible embodiments, but it will be understood that “some embodiments” may be the same subset or different subsets of all possible embodiments and may be combined with each other without conflict, and the terms “first\second” involved are merely used to distinguish similar objects and do not represent a specific ordering of the objects. The terms “including” and “having” and any variations are intended to cover non-exclusive inclusions. For example, a process, method, system, product, or server that includes a series of steps or units is not necessarily limited to those steps or units that are clearly listed, but may include other steps or modules that are not clearly listed or that are inherent to these processes, methods, products, or devices. Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by those skilled in the art to which this application belongs. The terms used herein are for the purpose of describing the embodiments of this application only.
[0029] The present application provides a method and system for optimizing a metal plate grinding process. Figure 1 As shown, the method includes:
[0030] Dynamic grinding monitoring of the target metal plate is performed through predetermined sensing equipment to obtain target grinding monitoring information.
[0031] First wear information of a first mold in the target polishing monitoring information is analyzed to obtain a first predicted wear amount, and the first mold corresponds to a first predetermined wear threshold.
[0032] The first temperature information of the first point in the target polishing monitoring information is analyzed to obtain a first predicted temperature.
[0033] Dynamic grinding monitoring of the target metal plate is performed by a predetermined sensing device, and the wear data and temperature data of the metal plate during the grinding process are collected in real time to obtain target grinding monitoring information. The target grinding monitoring information is a time series parameter. The predetermined sensing device includes at least a wear sensor and a temperature sensor. The target metal plate is a high-hardness metal plate such as tool steel or hardened steel. High-hardness metal plates are difficult to grind and will generate high temperatures during the grinding process, which can easily lead to grinding burns and cracks. The wear sensor is an eddy current sensor or a laser ranging sensor to ensure high-precision measurement of the wear condition. The temperature sensor is a thermocouple or an infrared temperature sensor. Subsequently, the first wear information of the first mold in the target grinding monitoring information is analyzed and predicted by a wear amount prediction model to obtain the first wear time domain characteristic parameter, and the first predicted wear amount is obtained based on the wear amount prediction model. The first mold corresponds to a first predetermined wear threshold, wherein the first mold is a grinding head corresponding to a random process in a multi-process grinding process, i.e., a mold. Furthermore, first temperature information of a first point in the target polishing monitoring information is analyzed and predicted to obtain a first predicted temperature, where the first predicted temperature is predicted temperature data for the first point. The first point is a surface location of any metal plate, and multiple temperature sensors monitor temperatures at different locations on the polished surface of the metal plate. The first temperature information is temperature data collected at the first point over a historical time period.
[0034] The method provided in the embodiment of the present application also includes:
[0035] A predetermined grinding process is obtained, where the predetermined grinding process refers to a process for grinding the target metal plate. The predetermined grinding process includes multiple grinding processes, and each of the multiple grinding processes corresponds to a mold with a specific particle size.
[0036] A first process of the target metal plate is matched in the predetermined grinding process, the first process corresponds to the first die, and the first predetermined wear threshold is determined according to a specific grain size of the first die.
[0037] Obtaining a first predetermined wear threshold includes: the predetermined grinding process is a process for grinding the target metal plate, the predetermined grinding process includes multiple grinding processes, each grinding process uses a mold of a different particle size for grinding. A mold with a larger particle size is used for coarse grinding, and a mold with a smaller particle size is used for fine grinding. Subsequently, matching the first process of the target metal plate in the predetermined grinding process, the first process being a random process in the predetermined grinding process of the target metal plate, the first process corresponding to the first mold, and determining the first predetermined wear threshold based on the specific particle size of the first mold, the first predetermined wear threshold being a maximum wear threshold of the first mold under experimental conditions.
[0038] The method provided in the embodiment of the present application also includes:
[0039] Read the predetermined wear signature collection strategy.
[0040] Based on the predetermined wear feature collection strategy, multi-dimensional features of the first wear amount time series in the first wear information are collected to obtain first wear feature parameters.
[0041] The first wear characteristic parameter is used as input information of a wear amount prediction model to obtain output information, where the output information includes the first predicted wear amount.
[0042] The training process of the wear amount prediction model is as follows:
[0043] Obtain a time series record of historical wear of similar metal plates of the target metal plate, and extract a first historical record in the time series record of historical wear, wherein the first historical record refers to a first historical wear characteristic parameter before a first historical time and a first historical wear at the first historical time.
[0044] A first data group is formed based on the first historical wear characteristic parameter and the first historical wear amount, and supervised training learning is performed on the first data group based on the principle of neural network to obtain the wear amount prediction model.
[0045] Read the predetermined wear feature collection strategy, which includes three main strategies: wear time domain feature strategy, wear frequency domain feature strategy and wear time-frequency domain feature strategy. The predetermined wear feature collection strategy can comprehensively collect multi-dimensional features of the first wear amount time series data in the first wear information to obtain the first wear feature parameters. Furthermore, the first wear feature parameters are used as input information of the wear amount prediction model to obtain output information, which includes the first predicted wear amount. The wear amount prediction model is constructed based on a neural network model. The training process of the wear amount prediction model is as follows: obtain the historical wear amount time series record of the same metal plate as the target metal plate, and the historical wear amount time series record contains the historical wear feature parameters of the same metal plate and the corresponding historical wear amount. Extract the first historical record from the historical wear time series record, where the first historical record refers to the first historical wear characteristic parameter before the first historical time and the first historical wear at the first historical time. The first historical record is a random group of data in the historical wear time series record. Form a first data group based on the first historical wear characteristic parameter and the first historical wear. Repeat the extraction of the first historical record and the formation of the first data group. Perform supervised training and learning on each first data group based on the principle of neural network until the accuracy of the model output meets the preset accuracy rate, thereby obtaining the wear prediction model.
[0046] The method provided in the embodiment of the present application also includes:
[0047] The predetermined wear feature collection strategy includes a wear time domain feature strategy, a wear frequency domain feature strategy and a wear time-frequency domain feature strategy.
[0048] Based on the wear time domain feature strategy, time domain features of the first wear amount time series are collected to obtain first wear time domain feature parameters.
[0049] Based on the wear frequency domain feature strategy, frequency domain features are collected on the first wear amount spectrum after Fourier transform of the first wear amount time series to obtain first wear frequency domain feature parameters.
[0050] Based on the wear time-frequency domain feature strategy, time-frequency domain features are collected on the first target modal component after modal decomposition of the first wear amount time series to obtain first wear time-frequency domain feature parameters.
[0051] The first wear time-domain characteristic parameter, the first wear frequency-domain characteristic parameter and the first wear time-frequency-domain characteristic parameter are collectively used as the first wear characteristic parameter.
[0052] Obtaining the first wear characteristic parameter includes collecting time domain characteristics of the first wear amount time series based on the wear time domain characteristic strategy to obtain the first wear time domain characteristic parameter, wherein the first wear amount time series is a relationship curve between the time series and the wear amount, and the first wear time domain characteristic parameter is a time domain characteristic such as a peak characteristic and an average change rate extracted from the first wear amount time series based on the wear time domain characteristic strategy. Collecting frequency domain characteristics of the first wear amount spectrum after Fourier transforming the first wear amount time series based on the wear frequency domain characteristic strategy to obtain the first wear frequency domain characteristic parameter, wherein the first wear frequency domain characteristic parameter includes the bandwidth after Fourier transforming the first wear amount time series, i.e., the width of the spectrum, reflecting the frequency distribution range of the wear amount, the main frequency, i.e., the frequency with the largest amplitude in the spectrum, reflecting the main periodicity of the wear amount change, and other frequency domain characteristics. Collecting time and frequency domain characteristics of the first target modal component after modal decomposition of the first wear amount time series based on the wear time and frequency domain characteristic strategy to obtain the first wear time and frequency domain characteristic parameter. Specifically, empirical mode decomposition is used to perform modal decomposition of the first wear time series, obtaining multiple components, each of which contains corresponding energy contribution data. This energy contribution data is the energy weight. The component with the largest energy weight is obtained as the first target modal component. The time and frequency domain characteristics of the first target modal component are used to represent the time and frequency domain characteristics of the wear time series, obtaining the first wear time-frequency domain characteristic parameter. Finally, the first wear time-domain characteristic parameter, the first wear frequency-domain characteristic parameter, and the first wear time-frequency domain characteristic parameter are collectively used as the first wear characteristic parameter.
[0053] The method provided in the embodiment of the present application also includes:
[0054] Based on the spline curve principle, a polynomial regression fitting is performed on the first temperature time series obtained by analyzing the first temperature information to obtain a first fitting formula.
[0055] The first predicted temperature is obtained based on the first fitting formula.
[0056] A first temperature time series is constructed for the first temperature information collected based on the time series at the first point. The first temperature time series is the correspondence between time series and temperature. A polynomial is fitted to the first temperature time series using the least squares method. The polynomial can be a cubic polynomial, a quadratic polynomial, a linear polynomial, etc. The specific polynomial can be selected based on the data characteristics of the first temperature time series, and a first fitting formula is obtained. Furthermore, a predicted temperature for the future time series is obtained based on the first fitting formula, thereby obtaining the first predicted temperature.
[0057] When the first predicted wear amount reaches the first predetermined wear threshold or the first predicted temperature reaches a predetermined temperature threshold, a first process control instruction is issued.
[0058] Based on the first process control instruction, the target process optimization plan is activated to optimize the grinding process of the target metal plate to obtain the target optimal grinding process.
[0059] The process of the target metal plate is optimized and regulated according to the target optimal grinding process.
[0060] When the first predicted wear amount reaches or is greater than or equal to the first predetermined wear threshold or the first predicted temperature reaches or is greater than or equal to the predetermined temperature threshold, that is, when either the predicted temperature or the predicted wear amount reaches the threshold, then the corresponding metal plate grinding process is abnormal and needs to be adjusted in time, and a first process control instruction is issued. The predetermined temperature threshold is the pre-set maximum temperature data for metal plate grinding. Based on the first process control instruction, the target process optimization plan is activated to optimize the grinding process of the target metal plate to obtain the target optimal grinding process. The target process optimization plan is a predetermined grinding index and a corresponding parameter control range based on big data collection. When obtaining the target optimal grinding process, the control range corresponding to the predetermined grinding index is obtained based on the target process optimization plan. The predetermined grinding index includes at least grinding speed, grinding pressure, feed speed and mold speed. Based on the control range corresponding to the predetermined grinding index, the control parameter corresponding to the grinding index is randomly extracted to obtain the first control parameter. Based on the first control parameter, the corresponding temperature time series record data and wear time series record data are extracted from the historical grinding process record data set. The temperature time series record data and the wear time series record data are input into the predetermined process fitness evaluation mechanism for fitness evaluation, and the fitness evaluation result is obtained. The traversal is continued until the required fitness threshold is met or the preset number of traversals is reached, and the grinding process with the highest fitness evaluation result is obtained as the target optimal grinding process. The process optimization and control of the target metal plate are performed according to the target optimal grinding process. It solves the technical problems in the prior art that grinding burns and cracks caused by high grinding temperature are difficult to solve, and there is a lack of intelligent control means to optimize the grinding process parameters, which affects the grinding quality and efficiency. Real-time monitoring of the wear amount and temperature during the grinding process of high-hardness metal plates is realized, an intelligent wear amount prediction model and a temperature prediction model are constructed, the stability of the grinding process is comprehensively evaluated, and intelligent process control and optimization are completed, which greatly improves the grinding quality and efficiency and ensures the technical effect of the surface quality and precision of the metal plates.
[0061] The method provided in the embodiment of the present application also includes:
[0062] When the first predicted wear amount does not reach the first predetermined wear threshold and the first predicted temperature does not reach the predetermined temperature threshold, a process evaluation instruction is issued.
[0063] Based on the process evaluation instruction, a predetermined process fitness evaluation mechanism is introduced to evaluate and analyze the first wear amount time series and the first temperature time series to obtain a target process fitness.
[0064] Determine whether the target process fitness meets the fitness limit.
[0065] If it does not meet the requirements, a second process control instruction is issued, and the grinding process of the target metal plate is optimized based on the second process control instruction to obtain the target optimal grinding process.
[0066] When the first predicted wear amount does not reach or is less than the first predetermined wear threshold, and the first predicted temperature does not reach or is less than the predetermined temperature threshold, then there is no abnormality in the predicted wear and the predicted temperature, and a process evaluation instruction is issued. Based on the process evaluation instruction, a predetermined process fitness evaluation mechanism is introduced to evaluate and analyze the first wear amount time series and the first temperature time series to obtain the target process fitness. The predetermined process fitness evaluation mechanism is constructed using a fitness function, and the calculation formula is:
[0067]
[0068] in, is the process adaptability, is the predetermined wear correlation coefficient, is the wear stability index, is the predetermined temperature dependence coefficient, is the temperature stability index. Furthermore, it is determined whether the target process fitness meets, i.e., is greater than or equal to, a fitness limit, where the fitness limit is a pre-set minimum value for the fitness to meet the requirements. The calculated target process fitness is compared with the fitness limit. If it does not meet, i.e., is less than the fitness limit, it is necessary to adjust the process parameters, optimize the grinding process, issue a second process control instruction, and optimize the grinding process of the target metal plate based on the second process control instruction to obtain the target optimal grinding process. If it meets, the current process parameters are considered appropriate and no adjustment is required.
[0069] The method provided in the embodiment of the present application also includes:
[0070] According to the predetermined process fitness evaluation mechanism, the first wear frequency domain characteristic parameter of the first wear amount time series and the first temperature frequency domain characteristic parameter of the first temperature time series are obtained in sequence.
[0071] The first wear stability index is obtained by performing weighted normalization on the first wear frequency domain characteristic parameters.
[0072] The first temperature frequency domain characteristic parameter after weighted normalization is used to obtain a first temperature stability index.
[0073] Predetermined correlation coefficients are read, the predetermined correlation coefficients including a predetermined wear correlation coefficient and a predetermined temperature correlation coefficient.
[0074] The product of the predetermined wear correlation coefficient and the first wear stability index and the product of the predetermined temperature correlation coefficient and the first temperature stability index are weighted to serve as the target process adaptability.
[0075] According to the predetermined process fitness evaluation mechanism, the first wear frequency domain characteristic parameters of the first wear amount time series and the first temperature frequency domain characteristic parameters of the first temperature time series are sequentially obtained. The frequency domain characteristic parameters include bandwidth and main frequency. The first wear frequency domain characteristic parameters are normalized and weighted summed to obtain a first wear stability index. The first temperature frequency domain characteristic parameters are normalized and weighted summed to obtain a first temperature stability index. The predetermined correlation coefficient is read. The predetermined correlation coefficient includes a predetermined wear correlation coefficient and a predetermined temperature correlation coefficient. The predetermined correlation coefficient is a pre-set correlation parameter. The higher the correlation, the larger the corresponding correlation coefficient value. The sum of the predetermined wear correlation coefficient and the predetermined temperature correlation coefficient is 1. The sum of the weighted predetermined wear correlation coefficient and the first wear stability index, and the weighted predetermined temperature correlation coefficient and the first temperature stability index is taken as the target process fitness.
[0076] In the above, refer to Figure 1 A metal plate grinding process optimization method according to an embodiment of the present invention is described in detail. Figure 2 A metal plate grinding process optimization system according to an embodiment of the present invention is described.
[0077] A metal plate grinding process optimization system according to an embodiment of the present invention solves the technical problems in the prior art that grinding burns and cracks caused by high grinding temperatures are difficult to solve, and there is a lack of intelligent control means to optimize grinding process parameters, which affects the grinding quality and efficiency. It realizes real-time monitoring of wear and temperature during the grinding process of high-hardness metal plates, constructs intelligent wear prediction models and temperature prediction models, comprehensively evaluates the stability of the grinding process, and completes intelligent process control and optimization, greatly improving the grinding quality and efficiency, and ensuring the technical effect of the surface quality and precision of the metal plate. A metal plate grinding process optimization system includes: a monitoring module 11, a wear prediction module 12, a temperature prediction module 13, a control instruction acquisition module 14, a process optimization module 15, and an optimization control module 16.
[0078] The monitoring module 11 is used to perform dynamic grinding monitoring on the target metal plate through a predetermined sensing device to obtain target grinding monitoring information.
[0079] The wear amount prediction module 12 is configured to analyze first wear information of a first mold in the target grinding monitoring information to obtain a first predicted wear amount, where the first mold corresponds to a first predetermined wear threshold.
[0080] The temperature prediction module 13 is configured to analyze the first temperature information of the first point in the target polishing monitoring information to obtain a first predicted temperature.
[0081] The control instruction acquisition module 14 is configured to issue a first process control instruction when the first predicted wear amount reaches the first predetermined wear threshold or the first predicted temperature reaches a predetermined temperature threshold.
[0082] The process optimization module 15 is used to activate the target process optimization plan based on the first process control instruction to optimize the grinding process of the target metal plate to obtain the target optimal grinding process.
[0083] The optimization and control module 16 is used to optimize and control the process of the target metal plate according to the target optimal grinding process.
[0084] The specific configuration of the wear prediction module 12 will be described in detail below. The wear prediction module 12 may further include: obtaining a predetermined grinding process, where the predetermined grinding process refers to a grinding process for the target metal plate, and the predetermined grinding process may include multiple grinding processes, each of which corresponds to a mold of a specific grit size. Matching the target metal plate with a first process in the predetermined grinding process, where the first process corresponds to the first mold, and determining the first predetermined wear threshold based on the specific grit size of the first mold.
[0085] Below, the specific configuration of the wear amount prediction module 12 will be described in detail. The wear amount prediction module 12 may further include: reading a predetermined wear feature collection strategy. Based on the predetermined wear feature collection strategy, multi-dimensional feature collection is performed on the first wear amount time series in the first wear information to obtain a first wear feature parameter. The first wear feature parameter is used as input information of the wear amount prediction model to obtain output information, and the output information includes the first predicted wear amount. The training process of the wear amount prediction model is as follows: obtain the historical wear amount time series record of the same type of metal plate as the target metal plate, and extract the first historical record in the historical wear amount time series record, the first historical record refers to the first historical wear feature parameter before the first historical time and the first historical wear amount at the first historical time. A first data group is formed based on the first historical wear feature parameter and the first historical wear amount, and the first data group is supervised trained and learned based on the principle of neural network to obtain the wear amount prediction model.
[0086] Below, the specific configuration of the wear amount prediction module 12 will be described in detail. The wear amount prediction module 12 may further include: the predetermined wear feature collection strategy includes a wear time domain feature strategy, a wear frequency domain feature strategy and a wear time-frequency domain feature strategy. Based on the wear time domain feature strategy, the time domain feature of the first wear amount time series is collected to obtain the first wear time domain feature parameter. Based on the wear frequency domain feature strategy, the frequency domain feature of the first wear amount spectrum after Fourier transform of the first wear amount time series is collected to obtain the first wear frequency domain feature parameter. Based on the wear time-frequency domain feature strategy, the time-frequency domain feature of the first target modal component after modal decomposition of the first wear amount time series is collected to obtain the first wear time-frequency domain feature parameter. The first wear time domain feature parameter, the first wear frequency domain feature parameter and the first wear time-frequency domain feature parameter are collectively used as the first wear feature parameter.
[0087] The specific configuration of the temperature prediction module 13 will be described in detail below. The temperature prediction module 13 further includes: performing polynomial regression fitting on the first temperature time series obtained by analyzing the first temperature information based on the spline curve principle to obtain a first fitting formula. The first predicted temperature is obtained based on the first fitting formula.
[0088] The specific configuration of the control instruction acquisition module 14 will be described in detail below. The control instruction acquisition module 14 further includes: when the first predicted wear amount does not reach the first predetermined wear threshold, and the first predicted temperature does not reach the predetermined temperature threshold, issuing a process evaluation instruction. Based on the process evaluation instruction, a predetermined process fitness evaluation mechanism is introduced to evaluate and analyze the first wear amount time series and the first temperature time series to obtain the target process fitness. It is determined whether the target process fitness meets the fitness limit. If not, a second process control instruction is issued, and based on the second process control instruction, the grinding process of the target metal plate is optimized to obtain the target optimal grinding process.
[0089] Below, the specific configuration of the process optimization module 15 will be described in detail. The process optimization module 15 further includes: according to the predetermined process fitness evaluation mechanism, sequentially obtaining the first wear frequency domain characteristic parameters of the first wear amount time series and the first temperature frequency domain characteristic parameters of the first temperature time series. The first wear stability index is obtained by weighting and normalizing the first wear frequency domain characteristic parameters. The first temperature frequency domain characteristic parameters are weighted and normalized to obtain the first temperature stability index. Read the predetermined correlation coefficient, which includes a predetermined wear correlation coefficient and a predetermined temperature correlation coefficient. The product of the weighted predetermined wear correlation coefficient and the first wear stability index, and the product of the weighted temperature correlation coefficient and the first temperature stability index are used as the target process fitness.
[0090] A metal plate grinding process optimization system provided by an embodiment of the present invention can execute a metal plate grinding process optimization method provided by any embodiment of the present invention, and has corresponding functional modules and beneficial effects of the execution method.
[0091] Although this application makes various references to certain modules in the system according to the embodiments of the present application, any number of different modules may be used and run on the user terminal and / or server. The various units and modules included are only divided according to functional logic, but are not limited to the above divisions, as long as the corresponding functions can be achieved. In addition, the specific names of the functional units are only for the purpose of distinguishing them from each other and are not intended to limit the scope of protection of the present invention.
[0092] The above specific embodiments do not constitute a limitation on the scope of protection of this application. Those skilled in the art should understand that various modifications, combinations, and substitutions may be made based on design requirements and other factors. Any modifications, equivalent substitutions, and improvements made within the spirit and principles of this application shall be included within the scope of protection of this application.
Claims
1. A method for optimizing the metal plate grinding process, It is characterized in that include: Dynamically monitor the grinding of the target metal plate through a predetermined sensor device to obtain target grinding monitoring information; Analyzing first wear information of a first mold in the target grinding monitoring information to obtain a first predicted wear amount, wherein the first mold corresponds to a first predetermined wear threshold; Analyze the first temperature information of the first point in the target polishing monitoring information to obtain a first predicted temperature; When the first predicted wear amount reaches the first predetermined wear threshold or the first predicted temperature reaches a predetermined temperature threshold, issuing a first process control instruction; Activating a target process optimization plan based on the first process control instruction to optimize the grinding process of the target metal plate to obtain a target optimal grinding process; Performing process optimization and control of the target metal plate according to the target optimal grinding process; The method further comprises: Reading a predetermined wear signature collection strategy; Based on the predetermined wear feature collection strategy, multi-dimensional feature collection is performed on the first wear amount time series in the first wear information to obtain a first wear feature parameter; Using the first wear characteristic parameter as input information of a wear amount prediction model to obtain output information, wherein the output information includes the first predicted wear amount; The training process of the wear amount prediction model is as follows: Obtaining a time series record of historical wear of a metal plate of the same type as the target metal plate, and extracting a first historical record in the time series record of historical wear, wherein the first historical record refers to a first historical wear characteristic parameter before a first historical time and a first historical wear at the first historical time; A first data set is formed based on the first historical wear characteristic parameter and the first historical wear amount, and supervised training and learning is performed on the first data set based on the neural network principle to obtain the wear amount prediction model; The predetermined wear feature collection strategy includes a wear time domain feature strategy, a wear frequency domain feature strategy and a wear time-frequency domain feature strategy; Based on the wear time domain feature strategy, time domain features of the first wear amount time series are collected to obtain first wear time domain feature parameters; Based on the wear frequency domain feature strategy, frequency domain features are collected on the first wear amount spectrum after Fourier transforming the first wear amount time series to obtain first wear frequency domain feature parameters; Based on the wear time-frequency domain feature strategy, the first target modal component after modal decomposition of the first wear amount time series is collected in the time-frequency domain to obtain the first wear time-frequency domain feature parameter; The first wear time-domain characteristic parameter, the first wear frequency-domain characteristic parameter and the first wear time-frequency-domain characteristic parameter are collectively used as the first wear characteristic parameter.
2. According to the metal plate grinding process optimization method of claim 1, It is characterized in that The predetermined sensing device includes at least a wear sensor and a temperature sensor.
3. According to the metal plate grinding process optimization method of claim 1, It is characterized in that include: Obtaining a predetermined grinding process, wherein the predetermined grinding process refers to a process for grinding the target metal plate, and the predetermined grinding process includes multiple grinding processes, and each of the multiple grinding processes corresponds to a mold of a specific particle size; A first process of matching the target metal plate in the predetermined grinding process, the first process corresponds to the first die, and the first predetermined wear threshold is determined according to a specific grain size of the first die.
4. According to the metal plate grinding process optimization method of claim 1, It is characterized in that include: Performing polynomial regression fitting on the first temperature time series obtained by analyzing the first temperature information based on the spline curve principle to obtain a first fitting formula; The first predicted temperature is obtained based on the first fitting formula.
5. A metal plate grinding process optimization method according to claim 4, It is characterized in that include: When the first predicted wear amount does not reach the first predetermined wear threshold and the first predicted temperature does not reach the predetermined temperature threshold, issuing a process evaluation instruction; Based on the process evaluation instruction, a predetermined process fitness evaluation mechanism is introduced to evaluate and analyze the first wear amount time series and the first temperature time series to obtain a target process fitness; Determining whether the target process fitness meets the fitness limit; If it does not meet the requirements, a second process control instruction is issued, and the grinding process of the target metal plate is optimized based on the second process control instruction to obtain the target optimal grinding process.
6. A metal plate grinding process optimization method according to claim 5, It is characterized in that include: According to the predetermined process fitness evaluation mechanism, sequentially acquiring a first wear frequency domain characteristic parameter of the first wear amount time series and a first temperature frequency domain characteristic parameter of the first temperature time series; The first wear frequency domain characteristic parameter after weighted normalization processing is used to obtain a first wear stability index; The first temperature frequency domain characteristic parameter after weighted normalization processing obtains a first temperature stability index; Reading a predetermined correlation coefficient, wherein the predetermined correlation coefficient includes a predetermined wear correlation coefficient and a predetermined temperature correlation coefficient; The product of the predetermined wear correlation coefficient and the first wear stability index and the product of the predetermined temperature correlation coefficient and the first temperature stability index are weighted to serve as the target process fitness.
7. A metal plate grinding process optimization method according to claim 1, It is characterized in that The target process optimization plan refers to a plan for intelligently optimizing predetermined grinding indicators, and the predetermined grinding indicators at least include grinding speed, grinding pressure, feed speed and mold rotation speed.
8. A metal plate grinding process optimization system, It is characterized in that The system is used to execute the method according to any one of claims 1 to 7, and the system comprises: A monitoring module, used to dynamically monitor the grinding of a target metal plate through a predetermined sensor device to obtain target grinding monitoring information; a wear amount prediction module, configured to analyze first wear information of a first mold in the target grinding monitoring information to obtain a first predicted wear amount, wherein the first mold corresponds to a first predetermined wear threshold; A temperature prediction module, used for analyzing the first temperature information of the first point in the target polishing monitoring information to obtain a first predicted temperature; a control instruction acquisition module, configured to issue a first process control instruction when the first predicted wear amount reaches the first predetermined wear threshold or the first predicted temperature reaches a predetermined temperature threshold; A process optimization module, configured to activate a target process optimization plan based on the first process control instruction to optimize the grinding process of the target metal plate to obtain a target optimal grinding process; The optimization and control module is used to optimize and control the process of the target metal plate according to the target optimal grinding process.
