Intelligent programming method and system of CNC system

Abstract

The invention relates to the technical field of machining control, in particular to an intelligent programming method and system of a CNC system, and the intelligent programming method of the CNC system monitors a plurality of specified items in the CNC system in real time, carries out contact analysis on monitoring data of each specified item to generate task characteristics of the CNC system, and sends the task characteristics to the CNC system. And matching the task characteristics with a locally deployed strategy library to obtain a standard control program of the CNC system in the current task, and calling a performance monitoring curve of the CNC system to compensate the standard control program to obtain an optimized control program.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of process control, and particularly relates to an intelligent programming method and system of a CNC system. BACKGROUND

[0002] In a fierce market competition environment, enterprises need to continuously improve production efficiency to reduce costs and shorten delivery time. The traditional CNC programming process is usually cumbersome, and a large number of parameter settings and path planning need to be manually performed, and it is difficult to optimize in real time according to actual processing conditions.

[0003] Application Content Therefore, it is necessary to provide an intelligent programming method and system of a CNC system, which can quickly program the CNC system.

[0004] In a first aspect, the present application provides an intelligent programming method of a CNC system, and the method comprises the following steps: Live monitoring of a plurality of specified items in the CNC system, and contact analysis of monitoring data of each specified item to generate task characteristics of the CNC system; Matching the task characteristics with a locally deployed strategy library to obtain a standard control program of the CNC system for the current task; Retrieving a performance monitoring curve of the CNC system to compensate the standard control program and obtain an optimized control program.

[0005] In a second aspect, the present application further provides an intelligent programming system of a CNC system, which is used to implement the intelligent programming method of the CNC system in any one of the first aspect, and comprises: A task monitoring module is configured to live monitor a plurality of specified items in the CNC system, and to contact analyze monitoring data of each specified item to generate task characteristics of the CNC system; A program retrieval module is configured to match the task characteristics with a locally deployed strategy library to obtain a standard control program of the CNC system for the current task; A program optimization module is configured to retrieve a performance monitoring curve of the CNC system to compensate the standard control program and obtain an optimized control program.

[0006] The intelligent programming method of the CNC system improves machining precision and quality, adjusts programs according to real-time states of the system, reduces errors, improves production efficiency, quickly matches strategies to generate standard programs, can also be optimized in real time, reduces machining and idle time, enhances flexibility, can match strategies according to different task characteristics, adapts to diversified production, reduces costs, reduces dependence on manual programming, can discover equipment problems in time to avoid losses, and enhances enterprise competitiveness. BRIEF DESCRIPTION OF DRAWINGS

[0007] Figure 1 FIG. 1 is a schematic diagram of steps of an intelligent programming method of a CNC system in an embodiment; Figure 2 FIG. 2 is a schematic diagram of a structure of an intelligent programming system of a CNC system in an embodiment. DETAILED DESCRIPTION

[0008] In order to make the purposes, technical solutions and advantages of the present application clearer, the present application is further described in detail below with reference to the drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present application and not to limit the present application.

[0009] The intelligent programming method of the CNC system provided in the embodiments of the present application can be applied to an application environment as shown in FIG. 1. Figure 1 As shown in FIG. 1, the application environment includes a CNC system, a strategy library, and a performance monitoring curve. S1: live monitoring of a plurality of specified items in the CNC system, and contact analysis of monitoring data of each specified item to generate a task characteristic of the CNC system; S2: matching the task characteristic with a locally deployed strategy library to obtain a standard control program of the CNC system for a current task; S3: calling a performance monitoring curve of the CNC system to perform compensation processing on the standard control program to obtain an optimized control program.

[0010] Specifically, in step S1 of the embodiments provided in the present application, live monitoring obtains monitoring data of each specified item of the CNC system. The drive system is a key component of the CNC system, which provides power for the movement of the machine tool. Drive performance information (such as motor speed, torque, acceleration, etc.) directly affects the movement accuracy and efficiency of the machine tool. For example, unstable motor speed may cause errors in the machining process. By monitoring the drive performance information in real time, abnormalities in the drive system can be discovered in time, providing accurate data basis for subsequent programming and control.

[0011] More specifically, the tool condition information includes the wear, breakage and other conditions of the tool, which directly affects the machining quality and efficiency. Different tool wear levels will cause changes in cutting force, cutting temperature and other parameters, thereby affecting the surface quality and dimensional accuracy of the workpiece. Real-time monitoring of tool condition information (such as tool wear, cutting edge sharpness, etc.) can timely adjust the machining strategy when the tool has problems, avoiding the generation of waste products due to tool problems.

[0012] More specifically, the workpiece specification information is used to describe the specification data to be machined. The specification of the workpiece (such as size, shape, material, etc.) is an important basis for CNC system programming. Different specifications of the workpiece require different machining processes and parameters. For example, for materials with high hardness, appropriate tools and cutting parameters need to be selected; for complex-shaped workpieces, special machining paths need to be used. By monitoring the workpiece specification information, it can be ensured that the CNC system is programmed and machined reasonably according to the actual workpiece conditions.

[0013] More specifically, the finished product target information (such as machining accuracy, surface roughness, production efficiency, etc.) clearly defines the final requirements of the machining. The programming and control of the CNC system need to be around these targets. For example, if the finished product requires high machining accuracy, a more accurate motion control algorithm and smaller feed rate need to be selected during programming; if the production efficiency needs to be improved, the machining path and cutting parameters need to be optimized. Real-time acquisition of finished product target information can enable the CNC system to dynamically adjust the machining strategy according to the target.

[0014] More specifically, the monitoring data of each specified item is respectively encoded to generate item monitoring features corresponding to the monitoring data of each specified item. The original monitoring data is usually large and complex, containing various noise and redundant information. Feature encoding is the process of converting original data into more representative and distinguishable feature vectors. By encoding the monitoring data of each specified item, key information in the data can be extracted, reducing the dimensionality of the data and improving the processing efficiency of the data. At the same time, the item monitoring features after feature encoding are more convenient for subsequent analysis and comparison, and can better reflect the essential characteristics of each specified item.

[0015] More specifically, by combining the monitoring features of each item, the task features of the CNC system are generated. The machining task of the CNC system is a holistic process, and the various specified items are interconnected and influence each other. Individual item monitoring features can only reflect a certain aspect of the situation. However, by combining the monitoring features of each item, multiple factors such as drive performance, tool condition, workpiece specifications, and finished product targets can be comprehensively considered to form a comprehensive and complete task feature. This task feature can more accurately describe the current machining task of the CNC system, providing a more reliable basis for subsequent matching with the strategy library and generating standard control programs.

[0016] Specifically, in step S2 of the embodiment provided in this application, task features are matched with a locally deployed strategy library to retrieve several pre-processing strategies, and a task matching index is generated for each pre-processing strategy. The locally deployed strategy library stores a large number of pre-processing strategies for different processing tasks. These strategies are effective processing schemes that have been verified in practice or derived theoretically. By matching task features with the strategy library, pre-processing strategies related to the current task can be selected from many strategies. The task matching index generated for each pre-processing strategy is to quantify the degree of matching between the strategy and the current task features. The higher the matching index, the higher the fit between the strategy and the current task. Subsequently, a suitable standard control procedure can be further screened and determined based on this index.

[0017] More specifically, based on preset standards, threshold judgments are made on the task matching indices. If a task matching index exists within a first threshold range, the corresponding pre-processing strategy is converted into the standard control program for the current task. The preset first threshold is a judgment standard used to filter out pre-processing strategies that are highly matched with the current task. When the task matching index of a certain pre-processing strategy is within the first threshold range, it indicates that the strategy is very consistent with the characteristics of the current task and can well meet the needs of the current processing task. At this time, the pre-processing strategy is directly converted into the standard control program for the current task, which can save a lot of time and computing resources, quickly provide a suitable control scheme for the CNC system, and improve processing efficiency.

[0018] More specifically, if there is no task matching index within the first threshold range, but there are several task matching indices within the second threshold range, then each pre-processing strategy with a task matching index within the second threshold range is used as a reference object. A similarity and difference analysis is performed on each reference object relative to the current task. Based on the analysis results, the strategy units of each pre-processing strategy are decomposed, merged, and evaluated to obtain a standard control program. If there is no task matching index within the first threshold range, it means that there is no strategy in the strategy library that completely matches the current task, but there are task matching indices within the second threshold range. Although these strategies cannot completely meet the requirements of the current task, they have a certain similarity to the current task. Using these strategies as reference objects and performing similarity and difference analysis can identify their similarities and differences with the current task. Based on the analysis results, the strategy units of each pre-processing strategy are decomposed, merged, and evaluated to integrate the advantages of these reference strategies and compensate for their shortcomings, thereby generating a standard control program more suitable for the current task. This method can fully utilize the existing resources in the strategy library. In the absence of a completely matching strategy, a control program that meets the requirements of the current task can be generated through optimization and combination, improving the utilization rate of the strategy library and the adaptability of the CNC system.

[0019] Specifically, in step S3 of the embodiment provided in this application, records of various tasks executed by the CNC system are collected and substituted into pre-built drive module loss models and tool module loss models to analyze module loss, thereby generating theoretical performance loss curves for the drive module and tool module. The drive module and tool module are key components of the CNC system, and their performance deteriorates with usage time and different tasks performed. The pre-built drive module loss models and tool module loss models are established based on a large amount of experimental data and theoretical analysis, and can reflect the changing patterns of module performance with task execution. By substituting the records of various tasks executed by the CNC system into these models, the theoretical performance loss of the drive module and tool module can be preliminarily estimated, generating theoretical performance loss curves, providing a basis for subsequent performance monitoring and compensation.

[0020] More specifically, by observing the performance of the CNC system during task execution through a pre-deployed observation system, task observation data can be obtained. Theoretical models have certain limitations and cannot fully and accurately reflect the performance changes of the CNC system in actual operation. By observing the performance of the CNC system during task execution through a pre-deployed observation system, actual task observation data can be obtained. This data contains various information about the CNC system in actual operation, such as machining accuracy, machining speed, and actual module wear, which can truly reflect the performance status of the system.

[0021] More specifically, based on the task observation data, the pre-built digital model of the CNC system is simulated to verify and adjust the theoretical performance loss curves of the drive module and the tool module, thereby generating performance monitoring curves for the drive module and the tool module. Simulating the operation of the pre-built digital model of the CNC system based on the task observation data allows for comparison between the theoretical model and the actual operating conditions. This comparison reveals the differences between the theoretical performance loss curves and the actual situation, allowing for the verification and adjustment of the theoretical curves. This ensures that the generated performance monitoring curves for the drive module and the tool module more accurately reflect the actual performance changes of the CNC system, providing a reliable basis for subsequent compensation processing.

[0022] More specifically, the performance monitoring curves of the CNC system are retrieved to analyze the performance characteristics of the drive module and the tool module at the current time. The performance monitoring curves can intuitively reflect the performance changes of the drive module and the tool module at different times. By retrieving and analyzing the performance monitoring curves, the performance characteristics of the drive module and the tool module of the CNC system at the current time can be obtained. These characteristics reflect the actual performance of the system at the current time, and there may be deviations from the standard performance characteristics corresponding to the standard control program. Understanding these deviations is a prerequisite for compensation processing.

[0023] More specifically, based on the performance characteristics of the drive module and the tool module, the standard performance characteristics corresponding to the standard control program are analyzed in terms of deviation form, and adaptive control parameter compensation is performed on the standard control program in response to the deviation form to obtain an optimized control program.

[0024] More specifically, the standard control program and the deviation form are each subjected to independent feature recognition. Simultaneously, interactive correlation feature recognition is performed on the independently identified features of the standard control program and the deviation form to obtain a label to be compensated. Both the standard control program and the deviation form have their own characteristics. Performing independent feature recognition on them allows for the extraction of their respective key information. At the same time, interactive correlation feature recognition identifies the inherent connection between the standard control program and the deviation form, determining the specific location and content requiring compensation, thus obtaining a label to be compensated. This label clearly identifies the specific part of the standard control program that needs adjustment, providing a precise target for subsequent parameter compensation.

[0025] More specifically, the labels to be compensated are input into a pre-trained adaptive compensation algorithm to compensate the control parameters of the standard control program, thereby obtaining an optimized control program. The pre-trained adaptive compensation algorithm is based on a large amount of experimental data and optimization algorithm training. It can reasonably adjust the control parameters of the standard control program according to the information provided by the labels to be compensated. Through the adaptive compensation algorithm, the standard control program can be automatically and accurately compensated according to the actual performance deviation of the CNC system, so that the optimized control program can better adapt to the current performance state of the CNC system and improve machining accuracy and efficiency.

[0026] This application provides an intelligent programming method for CNC systems, which has the following advantages: The intelligent programming method of the aforementioned CNC system improves machining accuracy and quality. Through real-time monitoring and compensation, it adjusts the program according to the real-time status of the system, reduces errors, improves production efficiency, quickly matches strategies to generate standard programs, and can also optimize in real time, reduce machining and idle time, enhance flexibility, match strategies according to different task characteristics, adapt to diversified production, reduce costs, reduce reliance on manual programming, and can promptly detect equipment problems to avoid losses, thereby enhancing enterprise competitiveness.

[0027] In one embodiment, the steps of real-time monitoring of several specified items in a CNC system and performing correlation analysis on the monitoring data of each specified item to generate task characteristics of the CNC system include: Real-time monitoring obtains monitoring data for various specified items of the CNC system; among which, the specified items include drive performance information, tool condition information, workpiece specification information, and finished product target information; The monitoring data of each specified project are feature-encoded to generate project monitoring features corresponding to the monitoring data of each specified project. The monitoring features of each project are combined to generate the task features of the CNC system.

[0028] Specifically, sensors are used to monitor parameters such as current, voltage, speed, and torque of the drive motor in real time, collecting operating status data of the drive system. For example, current and voltage sensors are installed in the motor circuit, and speed encoders are installed on the motor shaft to obtain corresponding data. Tools such as tool wear monitoring devices and vibration sensors are used to monitor the wear degree, damage condition, and vibration during the cutting process. For instance, optical sensors detect the wear of the tool edge, and vibration sensors sense the vibration frequency and amplitude of the tool during cutting. The drive system is the power source of the CNC system, and its performance directly affects the motion accuracy and stability of the machine tool. Real-time monitoring of drive performance information can promptly detect drive system faults and anomalies, providing accurate power status data for subsequent programming and control, ensuring smooth machining. Tool wear and damage directly affect machining quality and efficiency. By monitoring tool condition information, tool parameters can be replaced or adjusted in a timely manner when problems occur, avoiding machining errors and scrap caused by tool problems, thus improving machining quality and production efficiency.

[0029] More specifically, measuring equipment (such as calipers, micrometers, 3D scanners, etc.) is used to obtain information such as the size, shape, and material properties of the workpiece. For workpieces with complex shapes, a 3D scanner can be used for comprehensive shape scanning and size measurement. Workpieces of different specifications require different processing techniques and parameters. Accurately obtaining workpiece specification information allows the CNC system to program and process reasonably according to the actual workpiece conditions, ensuring processing accuracy and quality.

[0030] More specifically, information such as the machining accuracy requirements, surface roughness requirements, and production efficiency targets of finished products can be obtained from the production planning or control system. This information is usually stored in the production management system in the form of numbers or text and can be read directly through the interface. The finished product target information clarifies the final requirements of machining and is the basis for CNC system programming and control. Real-time monitoring of the finished product target information can enable the system to dynamically adjust the machining strategy according to the target in order to meet the quality and efficiency requirements of production.

[0031] More specifically, corresponding coding rules are formulated based on the characteristics and data types of different specified items. For example, for the speed data in the drive performance information, it can be coded according to a certain speed range, with each range corresponding to a specific coding value. The monitoring data of each specified item is converted according to the coding rules, and the original data is converted into representative feature codes. For example, if the speed is in the range of 0-1000 r / min, it is coded as "001"; in the range of 1001-2000 r / min, it is coded as "002", etc.

[0032] More specifically, raw monitoring data is usually complex and multidimensional, and may contain different data types and units. Feature encoding can transform these data into a unified and simplified feature representation, which facilitates subsequent processing and analysis. After feature encoding, the data dimensionality is reduced, the information is more compact, which can reduce the complexity of data processing, improve computational efficiency, and speed up the generation of task features. Feature encoding gives different monitoring data a unified representation, which is convenient for comparison and matching, and helps to perform accurate matching operations with the policy library in the future.

[0033] More specifically, a simple splicing method can be used to connect the monitoring features of each item in a certain order to form a complete task feature vector. For example, splicing can be done in the order of drive performance information, tool condition information, workpiece specification information, and finished product target information. The monitoring features of each item after feature encoding can be combined to obtain the final task feature of the CNC system. This task feature can comprehensively reflect the characteristics and requirements of the current machining task. The monitoring feature of a single item can only reflect a certain aspect of the situation. However, by combining the monitoring features of each item, multiple factors such as drive performance, tool condition, workpiece specification, and finished product target can be comprehensively considered to form a comprehensive and complete task feature, which more accurately describes the overall characteristics and requirements of the current machining task. The combined task feature can be matched as a whole with the locally deployed strategy library to find the standard control program most suitable for the current task, thereby improving the accuracy and effectiveness of the matching.

[0034] In one embodiment, the step of matching the task characteristics with a locally deployed policy library to obtain the standard control program of the CNC system for the current task includes: The task characteristics are matched with the locally deployed policy library to retrieve several pre-processing policies, and a task matching index is generated for each pre-processing policy. Based on preset standards, threshold judgment is performed on each task matching index. If there is a task matching index within the first threshold range, the corresponding pre-processing strategy is converted into the standard control program of the current task. If there is no task matching index within the first threshold range, but there are several task matching indices within the second threshold range, then each pre-processing strategy with a task matching index within the second threshold range is taken as a reference object, and a similarity and difference analysis is performed on each reference object relative to the current task. Based on the analysis results, the strategy units of each pre-processing strategy are decomposed, merged, and evaluated to obtain a standard control procedure.

[0035] Specifically, the generated task features are compared with the feature descriptions corresponding to each strategy in the local strategy library. Each pre-processing strategy in the strategy library has its own specific feature identifier, which is determined based on previous successful processing cases or theoretical analysis. Similarity calculation methods (such as Euclidean distance, cosine similarity, etc.) are used to measure the similarity between the task features and the features of each strategy in the strategy library. For example, if cosine similarity is used, the cosine value of the task feature vector and the feature vector of each strategy in the strategy library is calculated. The closer the cosine value is to 1, the higher the similarity.

[0036] More specifically, based on the similarity calculation results, a task matching index is generated for each retrieved pre-processing strategy. The matching index can be directly obtained from the similarity calculation results, or obtained through a certain linear transformation. For example, if the cosine similarity calculation result is 0.8, it can be directly used as the task matching index, or it can be converted to 80 through a linear transformation (assuming the matching index range is 0-100). The pre-processing strategies stored in the local strategy library are effective processing schemes that have been verified in practice or optimized theoretically. Through matching, strategies that may be related to the current task can be quickly screened from the strategy library, avoiding the time and cost consumption caused by redesigning processing strategies. The task matching index can intuitively reflect the degree of matching between each pre-processing strategy and the current task, providing a quantitative basis for subsequent screening and decision-making, and facilitating the sorting and selection of strategies based on the degree of matching.

[0037] More specifically, a first threshold range is preset. This range is determined based on actual production experience and system performance requirements. For example, setting the first threshold range to 90-100 indicates a very high matching degree. The task matching index of each pre-processing strategy is checked to see if it falls within the first threshold range. If there is a pre-processing strategy with a task matching index within the first threshold range, it means that the strategy is highly matched with the current task. It is then directly converted into the standard control program of the current task. The conversion process includes extracting information such as processing parameters and motion trajectories from the strategy and organizing and formatting them according to the programming specifications of the CNC system.

[0038] More specifically, when a strategy with a task matching index within the first threshold range exists, it indicates that the strategy is highly compatible with the current task. Directly converting it into a standard control program can save a lot of time and computing resources, quickly provide a suitable control scheme for the CNC system, and improve production efficiency. Strategies within the first threshold range have been rigorously screened and are very close to the characteristics of the current task, which can well meet the requirements of the current processing task, thereby ensuring processing quality.

[0039] More specifically, a second threshold range is pre-defined, which is usually lower than the first threshold range, for example, set to 70-89, indicating a certain degree of matching but not a complete match. It checks whether there are no task matching indices within the first threshold range, while there are several task matching indices within the second threshold range. For the preparatory machining strategies with task matching indices within the second threshold range, their similarity and differences with the current task are analyzed from multiple aspects (such as machining process, tool path, cutting parameters, etc.). For example, it analyzes whether the machining process of each strategy is suitable for the workpiece material and shape of the current task, and whether the tool path can be referenced, etc.

[0040] More specifically, each reference strategy is decomposed into multiple strategy units. Each strategy unit can be a specific machining step, a specific tool action, or a set of cutting parameters. Based on the results of similarity and difference analysis, the advantageous strategy units from different reference strategies are combined and fused. For example, cutting parameters suitable for the current workpiece material in one strategy are combined with a more reasonable tool path in another strategy. The fused strategy is evaluated, and the evaluation indicators can include machining accuracy, machining efficiency, cost, etc. By simulating the machining process or using an evaluation model, the performance of the fused strategy on these indicators is calculated, and the optimal fused strategy is selected as the standard control program.

[0041] More specifically, in actual production, there may not be a strategy that perfectly matches the current task, but there are some strategies that are similar to it. By analyzing and optimizing these strategies, we can make full use of the existing resources in the strategy library to generate standard control programs that meet the requirements of the current task. Through similarity and difference analysis, as well as the decomposition, fusion and evaluation of strategy units, we can integrate the advantages of different strategies, make up for the shortcomings of individual strategies, generate standard control programs that are more suitable for the current task, improve processing accuracy and efficiency, and reduce costs.

[0042] In one embodiment, the performance monitoring curves of the CNC system include a drive module performance monitoring curve and a tool module performance monitoring curve, and the steps for generating the performance monitoring curves include: Collect records of all tasks executed by the CNC system, and substitute them into the pre-built drive module loss model and tool module loss model to analyze the module loss, so as to generate theoretical performance loss curves for the drive module and tool module. The performance of the CNC system during task execution is observed through a pre-deployed observation system to obtain task observation data; Based on the task observation data, the pre-built digital model of the CNC system is simulated to verify and adjust the theoretical performance loss curves of the drive module and the tool module, so as to generate the performance monitoring curves of the drive module and the tool module.

[0043] Specifically, detailed records of all tasks executed by the CNC system are collected from the system's log files, databases, or monitoring equipment. These records include the start and end times of the tasks, machining parameters (such as cutting speed, feed rate, and depth of cut), workpiece information (such as material and dimensions), and the operating status of the modules (such as the current and speed of the drive motor, and the usage time of the tool). The collected task records are then substituted into pre-built drive module loss models and tool module loss models. The drive module loss model is based on the physical processes of motor heating, wear, and fatigue, while the tool module loss model considers the wear mechanism of the tool (such as abrasive wear and adhesive wear).

[0044] More specifically, through model calculations, the wear and tear of the drive module and tool module during different task execution processes is analyzed, thereby generating theoretical performance wear curves for the drive module and tool module. These curves reflect the theoretical trend of module performance changes with factors such as the number of task executions and time. The pre-built wear models for the drive module and tool module are based on physical principles and a large amount of experimental data, and can theoretically describe the changes in module performance with task execution. By collecting historical task records and substituting them into the model for analysis, the performance wear and tear of the drive module and tool module can be preliminarily predicted, providing a theoretical basis for subsequent monitoring and optimization. The theoretical performance wear curves can reflect the long-term trend of module performance changes, helping operators to understand the wear and tear of the modules in advance, formulate reasonable maintenance plans and replacement strategies, and avoid production interruptions and quality problems caused by module failures.

[0045] More specifically, various sensors and monitoring devices, such as current sensors, temperature sensors, vibration sensors, and displacement sensors, are installed in the CNC system to monitor the operating status of the drive module and tool module in real time. At the same time, image recognition, laser measurement and other technologies are used to observe the workpiece quality and tool wear during the machining process.

[0046] More specifically, during the execution of tasks by the CNC system, the observation system collects various data in real time, including the current, temperature, and vibration of the drive module, the wear and cutting force of the tool, and the machining accuracy and surface roughness of the workpiece. The collected data is recorded and organized to form task observation data, which reflects the actual performance of the CNC system during actual operation.

[0047] More specifically, theoretical models have certain limitations and cannot fully and accurately reflect the performance changes of a CNC system during actual operation. Real-time observation through a pre-deployed monitoring system allows for the acquisition of actual data on the CNC system during machining. This data includes the influence of various practical factors, such as the machining environment and operating habits, and can accurately reflect the system's performance status. Task monitoring data can help operators promptly identify potential problems during CNC system operation, such as abnormal vibrations of the drive module or excessive tool wear. By addressing these issues in a timely manner, machining quality and production efficiency can be improved, and production costs reduced.

[0048] More specifically, the task observation data is input into a pre-built digital model of the CNC system. This digital model is obtained by mathematically modeling the structure, dynamic characteristics, and machining process of the CNC system. By simulating the operation of the CNC system during the actual task execution process, the performance change data of the drive module and tool module during the simulation process are obtained.

[0049] More specifically, the simulated performance change data is compared with the theoretical performance loss curve to verify the accuracy of the theoretical curve. If a significant difference is found, it indicates that the theoretical model may contain errors or overlooked factors. Based on the comparison results, the theoretical performance loss curve is adjusted. Adjustment methods may include modifying model parameters, introducing new influencing factors, etc., to make the theoretical curve closer to reality. The verified and adjusted theoretical performance loss curves are the drive module performance monitoring curves and tool module performance monitoring curves. These curves can more accurately reflect the actual performance changes of the drive module and tool module in the CNC system.

[0050] More specifically, inputting task observation data into a digital model for simulation allows for comparison between the theoretical model and actual operating conditions. This comparison verifies the accuracy of the theoretical model, identifies problems and shortcomings, and allows for adjustments to the theoretical performance degradation curve based on the comparison results. This enables the curve to more accurately reflect the actual performance changes of the CNC system. The generated performance monitoring curves for the drive module and tool module provide a more reliable basis for intelligent programming and control of the CNC system, achieving real-time monitoring and optimization of system performance.

[0051] In one embodiment, the performance monitoring curves and task observation data of multiple CNC systems are transmitted to a designated cloud platform via a wireless network. The cloud platform then compares the performance monitoring curves and task observation data of the multiple CNC systems to obtain the correlation characteristics between the performance monitoring curves and task observation data of each CNC system. By performing consistency analysis on the correlation characteristics of each CNC system, a confidence parameter is assigned to the performance monitoring curve of each CNC system. When the confidence parameter of a CNC system is lower than a predetermined value standard, manual inspection and processing are arranged for the CNC system.

[0052] Specifically, in each CNC system, the previously generated performance monitoring curves (including drive module performance monitoring curves and tool module performance monitoring curves) and task observation data (such as various sensor data during machining, task execution records, etc.) are organized and packaged. Using wireless network technologies (such as Wi-Fi, 4G / 5G, etc.), the data from each CNC system is sent to a designated cloud platform. During transmission, the data needs to be encrypted to ensure data security and integrity. At the same time, the stability of the network connection must be ensured to avoid data loss or transmission errors.

[0053] More specifically, centralizing the data of multiple CNC systems on a cloud platform for management facilitates unified storage, analysis, and processing of the data. Compared to performing data analysis on each CNC system individually, the cloud platform has more powerful computing capabilities and storage capacity, enabling it to handle large-scale data. By transmitting data via wireless network, remote monitoring and analysis of multiple CNC systems are achieved. Technicians can access the cloud platform from anywhere with a network connection to obtain performance information of each CNC system, promptly identify system anomalies, and improve the efficiency of production management.

[0054] More specifically, after receiving data from each CNC system, the cloud platform stores it in a database and integrates the data to ensure consistency and standardization of data formats, facilitating subsequent comparative analysis. Using data analysis techniques (such as statistical analysis and machine learning algorithms), key features are extracted from performance monitoring curves and task observation data, such as curve shape characteristics and statistical features of the data (mean, variance, etc.). Then, these features are compared pairwise across multiple CNC systems to identify similarities and differences. For example, the changing trends of tool wear curves under the same machining task in different CNC systems can be compared, or the current fluctuation characteristics of the drive module at different time periods can be compared. Based on the comparison results, correlation features between the performance monitoring curves and task observation data of each CNC system are generated. These correlation features can reflect the performance correlation and mutual influence between different CNC systems. For example, the performance changes of some CNC systems under specific machining conditions may be similar, or the performance changes of a certain CNC system may have a specific causal relationship with other systems.

[0055] More specifically, by comparing data from multiple CNC systems, potential problems that are difficult to detect with a single system can be discovered. For example, if the performance monitoring curve of a certain CNC system shows an anomaly, but observing the data of this system alone cannot determine the root cause of the problem, comparing the data with that of other systems can identify whether there are common problems or special anomalies in this system. The analysis of correlation features can help enterprises understand the performance correlation between different CNC systems, thereby optimizing the production process. For example, if it is found that the performance of certain CNC systems is similar under a specific machining task, the parameters of these systems can be uniformly adjusted or maintained to improve production efficiency and product quality.

[0056] More specifically, select appropriate consistency analysis methods, such as cluster analysis and correlation analysis. Cluster analysis can group CNC systems with similar correlation characteristics into one category, while correlation analysis can calculate the correlation coefficient between the correlation characteristics of different CNC systems. Based on the results of the consistency analysis, assign a confidence parameter to the performance monitoring curve of each CNC system. If the correlation characteristics of a CNC system have high consistency with most other systems, it indicates that its performance monitoring curve is relatively reliable, and a higher confidence parameter is assigned; conversely, if its correlation characteristics differ significantly from other systems, there may be data anomalies or system failures, and a lower confidence parameter is assigned.

[0057] More specifically, reliability parameters can intuitively reflect the reliability of the performance monitoring curve of each CNC system. When conducting data analysis and decision-making, technicians can judge the availability of data based on reliability parameters to avoid making wrong decisions due to unreliable data. Through consistency analysis and the allocation of reliability parameters, CNC systems with abnormal data can be identified in a timely manner. These abnormalities are caused by system failures, sensor damage, or data transmission errors. Timely detection and handling of these abnormalities can prevent production accidents and ensure the continuity and stability of production.

[0058] More specifically, a predetermined value standard for a reliability parameter is set in advance. This standard can be adjusted based on actual production experience and system requirements. For example, the threshold for the reliability parameter can be set to 0.7. When the reliability parameter of a CNC system falls below 0.7, the system's data is considered abnormal. When the reliability parameter of a CNC system is found to be below the predetermined value standard, the cloud platform automatically triggers an alarm mechanism and notifies relevant technical personnel. Based on the alarm information, the technical personnel manually inspect the CNC system, including the operating status of the system hardware, the working condition of the sensors, and whether the data acquisition and transmission processes are normal, to determine if there is a fault in the system and to perform corresponding repairs.

[0059] More specifically, when the reliability parameter of a CNC system is lower than the predetermined value standard, it indicates that the system's data has significant uncertainty and potential faults. Arranging manual inspection can promptly identify and resolve problems in the system, ensuring the normal operation of the CNC system and reducing production losses caused by system failures. Manual inspection can conduct a comprehensive check of the system, ensuring the accuracy and stability of the data acquisition and transmission process. By repairing system faults and adjusting sensor parameters, the quality of data can be improved, making the performance monitoring curve more accurately reflect the actual performance of the system.

[0060] In one embodiment, the step of retrieving the performance monitoring curve of the CNC system to compensate the standard control program and obtain an optimized control program includes: The performance monitoring curves of the CNC system are retrieved to analyze the performance characteristics of the drive module and the tool module of the CNC system at the current time. Based on the performance characteristics of the drive module and the tool module, the standard performance characteristics corresponding to the standard control program are analyzed for deviation forms, and adaptive control parameter compensation is performed on the standard control program for the deviation forms to obtain an optimized control program.

[0061] Specifically, based on the identification information of the current CNC system, the corresponding drive module performance monitoring curve and tool module performance monitoring curve are retrieved from the database or file system of storage performance monitoring curves. These curves were obtained through long-term monitoring and analysis of the CNC system and record the performance changes of the drive module and tool module at different times.

[0062] More specifically, the retrieved performance monitoring curves are analyzed to extract key performance characteristics at the current time point. For the drive module, performance characteristics include the real-time speed, torque, and power consumption of the motor; for the tool module, performance characteristics include the wear degree of the tool, the magnitude of the cutting force, and the cutting edge temperature. Signal processing techniques (such as filtering and interpolation) and data analysis methods (such as statistical analysis and curve fitting) can be used to accurately extract these characteristics.

[0063] More specifically, the performance monitoring curves can reflect the performance changes of the drive module and the tool module in real time. By retrieving the curves and analyzing the performance characteristics at the current time, we can accurately understand the actual operating status of the CNC system at the current moment, providing basic data for subsequent deviation analysis and compensation processing. Analyzing the performance characteristics can promptly identify potential problems in the drive module and the tool module. For example, if the cutting force of the tool module suddenly increases, it means that the tool has worn or broken, and timely measures need to be taken to deal with it to avoid affecting the machining quality.

[0064] More specifically, the standard performance characteristics of the drive module and tool module corresponding to the standard control program are clearly defined. These standard characteristics are performance indicators pre-set under ideal machining conditions to achieve specific machining tasks, such as standard motor speed and standard tool wear rate. The performance characteristics of the drive module and tool module obtained from the current analysis are compared with the corresponding standard performance characteristics, and the deviation between the two is calculated. The deviation can be expressed in numerical form, such as absolute deviation (the difference between the actual value and the standard value) or relative deviation (the ratio of the absolute deviation to the standard value). For example, if the standard motor speed is 1000 r / min and the current actual speed is 980 r / min, then the absolute deviation is -20 r / min and the relative deviation is -2%. The calculated deviation is further analyzed to determine the type, magnitude, and trend of the deviation. For example, it is determined whether the deviation is a long-term systematic deviation or a short-term random deviation; whether the deviation gradually increases or decreases over time, etc.

[0065] More specifically, standard control procedures are based on ideal conditions. However, in actual machining processes, the performance of the CNC system can deviate from the standard due to various factors. By calculating and analyzing these deviations, the gap between the current system performance and the standard performance can be clearly identified, providing direction for subsequent compensation. Analyzing the form of the deviation helps to analyze its causes. Different types of deviations are caused by different factors. For example, systematic deviations are caused by equipment aging, environmental changes, etc., while random deviations are caused by accidental factors in the machining process. Understanding the causes of deviations helps to develop more effective compensation strategies.

[0066] More specifically, based on the analytical results of the deviation form, corresponding control parameter compensation strategies are formulated. For example, if the speed deviation of the drive module is a systematic deviation caused by load changes, it can be compensated by adjusting the control voltage or current of the motor. If the wear deviation of the tool module is large, cutting parameters (such as feed rate and cutting speed) can be appropriately adjusted to ensure machining quality. The formulated compensation strategies are applied to the standard control program, and the control parameters are adjusted accordingly. For example, a compensation item is added to the standard control program to dynamically adjust the motor control signal or the cutting parameters of the tool based on the deviation calculation results. After parameter compensation, an optimized control program is obtained. This program takes into account the actual performance of the current CNC system, can better adapt to system changes, and improve machining accuracy and efficiency.

[0067] More specifically, by compensating the parameters of the standard control program, the CNC system can better approximate the standard performance during actual operation, reduce machining errors caused by performance deviations, and improve machining accuracy. At the same time, reasonable parameter compensation can also optimize the machining process and improve production efficiency. The performance of the CNC system will change with time and usage conditions, and the standard control program may not always adapt to the changes in the system. Adaptive control parameter compensation can enable the control program to dynamically adjust according to the actual performance of the system, enhance the system's adaptability to different working conditions, and ensure the stability of machining quality.

[0068] In one embodiment, the step of adaptively compensating the control parameters of the standard control program for the form of deviation includes: Independent feature recognition is performed on the standard control program and the deviation form, and interactive correlation feature recognition is performed on the independently recognized features of the standard control program and the deviation form to obtain the label to be compensated; The label to be compensated is input into a pre-trained adaptive compensation algorithm to compensate the control parameters of the standard control program, thereby obtaining an optimized control program.

[0069] Specifically, key information is extracted from the standard control program. This information includes, but is not limited to, machining paths, cutting parameters (such as cutting speed, feed rate, and depth of cut), and motion commands (such as linear interpolation and circular interpolation). This information is then organized into a structured data format for subsequent analysis. Appropriate feature extraction algorithms, such as rule-based methods or machine learning algorithms (such as decision trees and support vector machines), are used to process the extracted data and identify the characteristics of the standard control program. For example, by analyzing the geometry of the machining path and the range of values ​​for the cutting parameters, the characteristic patterns of the standard control program can be determined.

[0070] More specifically, the previously calculated deviation data (such as the deviation between the performance characteristics of the drive module and the standard performance characteristics, and the deviation between the performance characteristics of the tool module and the standard performance characteristics) are organized to make different types of deviation data comparable. Similarly, feature extraction algorithms are used to extract key features from the deviation data, such as analyzing the trend of deviation changes (increase, decrease, fluctuation), the magnitude of deviation, and the periodicity of deviation.

[0071] More specifically, standard control procedures contain a wealth of information. Feature recognition can extract their core features, helping to gain a deeper understanding of the nature and characteristics of standard control procedures. This helps to determine which parameters are key parameters and which parameters need to be adjusted to adapt to deviations. Independently identifying the features of standard control procedures is the foundation for subsequent correlation analysis. Only by accurately identifying the features of standard control procedures can they be effectively correlated with the features of deviation forms, and the inherent connection between the two be found.

[0072] More specifically, deviation data contains a wealth of information. Feature recognition can extract key characteristics of the deviation, such as the pattern of deviation change and the magnitude of the deviation. This helps to accurately grasp the nature of the deviation and provide a basis for subsequent compensation decisions. The characteristics of the deviation form can indicate the cause and trend of the deviation, thereby helping to determine the direction in which the standard control procedure needs to be compensated. For example, if the deviation shows an upward trend, more aggressive compensation measures need to be taken.

[0073] More specifically, association analysis methods (such as association rule mining and correlation analysis) are used to construct an association model between standard control program features and deviation form features. This model is used to analyze the intrinsic relationship between the two, identify which standard control program features are associated with which deviation form features, and extract standard control program features related to deviation forms through association analysis. These features are called associated features. For example, if it is found that when the tool wear deviation is large, the cutting speed feature in the standard control program is strongly correlated with it, then the cutting speed feature becomes an associated feature.

[0074] More specifically, based on the analysis results of the correlation features, a compensation label is defined. The compensation label clearly indicates which parameters in the standard control program need to be compensated, as well as the direction and approximate degree of compensation. For example, if the correlation analysis shows that the tool wear deviation is related to the cutting speed and the deviation is positive (actual tool wear is greater than standard wear), then the compensation label can be defined as "reduce the cutting speed by an amount of X".

[0075] More specifically, there is an inherent correlation between the standard control program and the form of deviation. This relationship can be identified through correlation feature recognition, which helps to understand how the deviation affects the standard control program and which parameters of the standard control program should be adjusted to compensate for the deviation. The compensation label clarifies the specific content and direction of the standard control program that needs to be compensated, providing a clear input for the subsequent adaptive compensation algorithm. It transforms the complex deviation analysis and compensation problem into specific and operable labels, making it easier for the algorithm to process.

[0076] More specifically, based on the characteristics and compensation requirements of the CNC system, a suitable adaptive compensation algorithm is selected. Common algorithms include neural networks, fuzzy control algorithms, and genetic algorithms. These algorithms have strong adaptive capabilities and can automatically adjust the compensation strategy according to different compensation labels. Before using the algorithm, it needs to be trained. The algorithm is trained using historical data (including standard control programs, deviation data, and corresponding compensation adjustment results) to learn the mapping relationship between standard control program parameter adjustments and deviation compensation.

[0077] More specifically, the label to be compensated is input into a pre-trained adaptive compensation algorithm. Based on the label information and the mapping relationship it has learned, the algorithm calculates the amount of compensation needed for the standard control program parameters. For example, if the label to be compensated is "reduce cutting speed," the algorithm will calculate the specific value at which the cutting speed needs to be reduced. Based on the calculated compensation amount, the corresponding parameters in the standard control program are adjusted. The adjusted control program is then used as an optimized control program for subsequent CNC system machining operations.

[0078] More specifically, the adaptive compensation algorithm has strong adaptive capabilities, automatically calculating the compensation amount and adjusting parameters based on the label to be compensated. This avoids the tediousness and inaccuracy of manual adjustments, improving the efficiency and accuracy of compensation. By compensating the parameters of the standard control program, the resulting optimized control program can better adapt to the actual performance deviations of the CNC system, thereby improving machining accuracy, quality, and efficiency, and enabling the CNC system to maintain good operating conditions under different working conditions.

[0079] In one embodiment, the method further includes: obtaining the sequence characteristics of the production tasks executed by the CNC system at the current moment within the production batch to determine whether the current task of the CNC system needs to perform performance verification of the CNC system; if it needs to be performed, then analyzing the performance characteristics based on the performance characteristics of the drive module and the performance characteristics of the tool module to obtain error risk information of the performance characteristics, and evaluating the performance verification value of various parameter adjustment forms of the control parameters of the standard control program based on the error risk information, so as to determine the parameter adjustment forms of the standard control program required to perform performance verification.

[0080] Specifically, the task list for the current production batch is retrieved from the production management module of the CNC system or a relevant database. This list records the execution order and related information of each production task. Based on the current task's identifier, its position in the task list is located to determine its sequence number within the production batch. Further analysis of the task sequence number is then performed to extract sequence features. For example, the sequence number can be converted into relative positional information (such as whether it is in the beginning, middle, or end stage of the production batch), or the distance to the start and end tasks of the batch can be calculated.

[0081] More specifically, task sequence characteristics can reflect the position of the current task within the entire production batch. Different sequence positions correspond to different production stages and conditions. For example, in the early stages of a production batch, the CNC system may be in the debugging and break-in phase, and its performance may be unstable. In the later stages, due to factors such as tool wear, machining quality may be prone to problems. Understanding task sequence characteristics helps determine whether performance verification is necessary. Based on task sequence characteristics, the timing and frequency of performance verification can be reasonably arranged to avoid unnecessary verification operations and improve production efficiency. For example, performing performance verification at critical stages or stages prone to problems in a production batch can promptly identify and resolve issues, ensuring production quality.

[0082] More specifically, performance verification rules are pre-defined based on production experience and quality requirements. These rules can be based on task sequence characteristics, such as specifying that performance verification needs to be performed on the first few tasks, the last few tasks, or every certain number of tasks in a production batch. The rules can also consider other factors, such as the complexity of the production tasks and the quality requirements. For example, for complex production tasks or tasks with high quality requirements, performance verification is required regardless of their order. The obtained task sequence characteristics are compared with the pre-defined verification rules to determine whether the current task needs to perform performance verification. If it meets the rule requirements, then performance verification is required; otherwise, verification is not required.

[0083] More specifically, performance verification can promptly identify performance issues in the CNC system, ensuring that production tasks are completed as required and improving product quality. For tasks with high quality requirements or prone to problems, performance verification can effectively prevent the generation of defective products due to poor system performance. Not all tasks need performance verification. By using reasonable rules to determine the appropriate tasks, unnecessary verification can be avoided, saving time and resources and improving production efficiency.

[0084] More specifically, error risk indicators related to the performance characteristics of the drive module and the tool module are defined. For example, for the drive module, the fluctuation range of motor speed and the degree of torque deviation can be defined as risk indicators; for the tool module, the threshold of tool wear and abnormal changes in cutting force can be defined as risk indicators. Based on the current performance characteristics of the drive module and the tool module, a comparative analysis is performed with the defined risk indicators. If a certain performance characteristic exceeds the range of the risk indicator, it is considered that there is an error risk. Further analysis is conducted on the performance characteristics with error risk to assess the severity of the risk and the possible impact. For example, if the tool wear is close to the threshold, it will lead to a decrease in machining accuracy, and timely measures need to be taken.

[0085] More specifically, the drive module and tool module are key components of the CNC system. The error risk of their performance characteristics directly affects machining quality and production efficiency. By analyzing the error risk information of performance characteristics, potential problems can be identified in advance, providing a basis for subsequent parameter adjustments and performance verification. Error risk information can quantify the uncertainty of performance characteristics, enabling operators to intuitively understand the severity of the risk and its potential impact. This helps to formulate reasonable countermeasures and reduce risks.

[0086] More specifically, list the executable control parameter adjustments in the standard control program, such as adjusting cutting speed, feed rate, and depth of cut. For each parameter adjustment, describe in detail its adjustment method (e.g., increase, decrease, proportional adjustment) and possible adjustment range. For each parameter adjustment, in conjunction with error risk information, evaluate its value in reducing error risk and improving performance verification. Evaluation can be conducted using methods such as simulation analysis, historical data statistics, or expert judgment. For example, by simulating and analyzing the impact of different cutting speed adjustments on tool wear and machining accuracy, the performance verification value of this parameter adjustment can be evaluated.

[0087] More specifically, different control parameter adjustment methods have different effects on reducing error risk and improving performance verification effectiveness. By evaluating the value of performance verification, the most effective parameter adjustment method can be selected, the standard control program can be optimized, and the accuracy and effectiveness of performance verification can be improved. Evaluating the value of performance verification can avoid blindly trying various parameter adjustment methods, save time and resources, and improve the efficiency of performance verification.

[0088] More specifically, based on the evaluation results of performance verification value, various parameter adjustment methods are ranked, and the parameter adjustment method with the highest performance verification value is selected as the adjustment method required for performing performance verification. If the verification values ​​of multiple parameter adjustment methods are similar, these adjustment methods can be used in combination to achieve better performance verification results. Determining the specific parameter adjustment method provides a clear operational direction for performance verification, enabling operators to make targeted adjustments to the standard control program to ensure that the performance verification can achieve the expected results. Selecting the appropriate parameter adjustment method can minimize error risks, improve the performance of the CNC system, and thus ensure the smooth completion of production tasks and the stability of product quality.

[0089] In one embodiment, such as Figure 2 As shown, an intelligent programming system for a CNC system is provided, used to implement the intelligent programming method for a CNC system as described in any one of the first aspects, comprising: The task monitoring module is used to monitor several specified items in the CNC system in real time, and to perform correlation analysis on the monitoring data of each specified item in order to generate task characteristics of the CNC system. The program retrieval module is used to match the task characteristics with the locally deployed strategy library to obtain the standard control program of the CNC system for the current task. The program optimization module is used to retrieve the performance monitoring curves of the CNC system to compensate the standard control program and obtain an optimized control program.

[0090] In this embodiment, the specific implementation of each module in the above system embodiment is described in the above method embodiment, and will not be repeated here.

[0091] It should be noted that the user information (including but not limited to user device information, user personal information, etc.) and data (including but not limited to data used for analysis, data stored, data displayed, etc.) involved in this application are all information and data authorized by the user or fully authorized by all parties, and the collection, use and processing of the relevant data shall comply with the relevant laws, regulations and standards of the relevant countries and regions.

[0092] Those skilled in the art will understand that all or part of the processes in the above embodiments can be implemented by a computer program instructing related hardware. The computer program can be stored in a non-volatile computer-readable storage medium, and when executed, it can include the processes of the embodiments described above. Any references to memory, databases, or other media used in the embodiments provided in this application can include at least one of non-volatile and volatile memory. Non-volatile memory can include read-only memory (ROM), magnetic tape, floppy disk, flash memory, optical memory, high-density embedded non-volatile memory, resistive random access memory (ReRAM), magnetic random access memory (MRAM), ferroelectric random access memory (FRAM), phase change memory (PCM), graphene memory, etc. Volatile memory can include random access memory (RAM) or external cache memory, etc. By way of illustration and not limitation, RAM can take many forms, such as Static Random Access Memory (SRAM) or Dynamic Random Access Memory (DRAM). The databases involved in the embodiments provided in this application may include at least one type of relational database and non-relational database. Non-relational databases may include, but are not limited to, blockchain-based distributed databases. The processors involved in the embodiments provided in this application may be general-purpose processors, central processing units, graphics processing units, digital signal processors, programmable logic devices, quantum computing-based data processing logic devices, etc., and are not limited to these.

[0093] The technical features of the above embodiments can be combined in any way. For the sake of brevity, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0094] The embodiments described above are merely illustrative of several implementation methods of this application, and while the descriptions are specific and detailed, they should not be construed as limiting the scope of this patent application. It should be noted that those skilled in the art can make various modifications and improvements without departing from the concept of this application, and these all fall within the protection scope of this application. Therefore, the protection scope of this application should be determined by the appended claims.

Claims

1. An intelligent programming method for a CNC system, characterized in that, include: The system monitors several specified items in the CNC system in real time and performs correlation analysis on the monitoring data of each specified item to generate task characteristics of the CNC system. The task characteristics are matched with the locally deployed strategy library to obtain the standard control program of the CNC system for the current task; The performance monitoring curves of the CNC system are retrieved to compensate the standard control program, thereby obtaining an optimized control program.

2. The intelligent programming method for a CNC system as described in claim 1, characterized in that, The steps for real-time monitoring of several specified items in a CNC system and correlation analysis of the monitoring data of each specified item to generate task characteristics of the CNC system include: Real-time monitoring obtains monitoring data for various specified items of the CNC system; among which, the specified items include drive performance information, tool condition information, workpiece specification information, and finished product target information; The monitoring data of each specified project are feature-encoded to generate project monitoring features corresponding to the monitoring data of each specified project. The monitoring features of each project are combined to generate the task features of the CNC system.

3. The intelligent programming method for a CNC system as described in claim 1, characterized in that, The steps for matching the task characteristics with the locally deployed strategy library to obtain the standard control program of the CNC system for the current task include: The task characteristics are matched with the locally deployed policy library to retrieve several pre-processing policies, and a task matching index is generated for each pre-processing policy. Based on preset standards, threshold judgment is performed on each task matching index. If there is a task matching index within the first threshold range, the corresponding pre-processing strategy is converted into the standard control program of the current task. If there is no task matching index within the first threshold range, but there are several task matching indices within the second threshold range, then each pre-processing strategy with a task matching index within the second threshold range is taken as a reference object, and a similarity and difference analysis is performed on each reference object relative to the current task. Based on the analysis results, the strategy units of each pre-processing strategy are decomposed, merged, and evaluated to obtain a standard control procedure.

4. The intelligent programming method for a CNC system as described in claim 1, characterized in that, The performance monitoring curves of the CNC system include the performance monitoring curves of the drive module and the tool module, and the steps for generating the performance monitoring curves include: Collect records of all tasks executed by the CNC system, and substitute them into the pre-built drive module loss model and tool module loss model to analyze the module loss, so as to generate theoretical performance loss curves for the drive module and tool module. The performance of the CNC system during task execution is observed through a pre-deployed observation system to obtain task observation data; Based on the task observation data, the pre-built digital model of the CNC system is simulated to verify and adjust the theoretical performance loss curves of the drive module and the tool module, so as to generate the performance monitoring curves of the drive module and the tool module.

5. The intelligent programming method for a CNC system as described in claim 4, characterized in that, The performance monitoring curves and task observation data of multiple CNC systems are transmitted to a designated cloud platform via a wireless network. The cloud platform then compares the performance monitoring curves and task observation data of the multiple CNC systems to obtain the correlation characteristics between the performance monitoring curves and task observation data of each CNC system. By performing consistency analysis on the correlation characteristics of each CNC system, a confidence parameter is assigned to the performance monitoring curve of each CNC system. When the confidence parameter of a CNC system is lower than a predetermined value standard, manual inspection and processing are arranged for the CNC system.

6. The intelligent programming method for a CNC system as described in claim 1, characterized in that, The steps of retrieving the performance monitoring curves of the CNC system to compensate for the standard control program and obtain the optimized control program include: The performance monitoring curves of the CNC system are retrieved to analyze the performance characteristics of the drive module and the tool module of the CNC system at the current time. Based on the performance characteristics of the drive module and the tool module, the standard performance characteristics corresponding to the standard control program are analyzed for deviation forms, and adaptive control parameter compensation is performed on the standard control program for the deviation forms to obtain an optimized control program.

7. The intelligent programming method for a CNC system as described in claim 6, characterized in that, The steps for adaptively compensating the control parameters of the standard control program for the aforementioned deviation include: Independent feature recognition is performed on the standard control program and the deviation form, and interactive correlation feature recognition is performed on the independently recognized features of the standard control program and the deviation form to obtain the label to be compensated; The label to be compensated is input into a pre-trained adaptive compensation algorithm to compensate the control parameters of the standard control program, thereby obtaining an optimized control program.

8. The intelligent programming method for a CNC system as described in claim 6, characterized in that, Also includes: The sequence characteristics of the production tasks executed by the CNC system at the current moment within the production batch are obtained to determine whether the current task of the CNC system needs to perform performance verification. If it needs to be performed, the error risk information of the performance characteristics is obtained based on the performance characteristics of the drive module and the tool module. The performance verification value of various parameter adjustment forms of the control parameters of the standard control program is evaluated based on the error risk information to determine the parameter adjustment forms of the standard control program required to perform performance verification.

9. An intelligent programming system for a CNC system, characterized in that, A method for implementing an intelligent programming system for a CNC system according to any one of claims 1-8 includes: The task monitoring module is used to monitor several specified items in the CNC system in real time, and to perform correlation analysis on the monitoring data of each specified item in order to generate task characteristics of the CNC system. The program retrieval module is used to match the task characteristics with the locally deployed strategy library to obtain the standard control program of the CNC system for the current task. The program optimization module is used to retrieve the performance monitoring curves of the CNC system to compensate the standard control program and obtain an optimized control program.

Images