Cutter adaptive control method based on artificial intelligence and intelligent agent control system

Abstract

The invention provides a cutter adaptive control method based on artificial intelligence and an intelligent agent control system, and relates to the technical field of intelligent control processing. The method comprises the following steps: calculating a cutter vibration energy growth rate; calculating the edge weight of each connecting edge directly connected with the cutter; calculating a vibration influence degree and a vibration flow direction factor of each connecting edge in combination with the edge weight; according to the vibration influence degree, the vibration flow direction factor and the tool vibration energy growth rate, the tool abrasion condition is judged; and adjusting according to the wear condition of the cutter. By implementing the scheme, the limitation of traditional single-point monitoring is broken through, the working condition characteristics of the cutter can be accurately captured, and targeted maintenance is facilitated.

Description

Technical Field

[0001] This application relates to the field of intelligent control and processing technology, and in particular to an artificial intelligence-based tool adaptive control method and intelligent agent control system. Background Technology

[0002] In the field of intelligent manufacturing, AI-based adaptive process control technology, which optimizes production processes through real-time data, has become a core means of improving machining accuracy and efficiency. Among these technologies, accurate assessment of tool wear is a crucial step in achieving adaptive process adjustments.

[0003] Currently, most manufacturers use the vibration amplitude threshold method to determine tool wear. This method collects tool vibration signals using an accelerometer, compares the vibration amplitude with a preset fixed threshold, and determines that the tool needs to be replaced when the amplitude exceeds the threshold, enabling adaptive control.

[0004] However, this single-index threshold judgment carries a high risk of misjudgment. When connecting components (such as the spindle or tool holder) malfunction and generate abnormal vibrations, the tool vibration amplitude may also exceed the standard, leading to misjudging the component malfunction as tool wear and resulting in unnecessary tool replacement. Therefore, there is an urgent need for an intelligent judgment method that can accurately distinguish vibration sources and adaptively adjust based on operating conditions. Summary of the Invention

[0005] This application provides an artificial intelligence-based tool adaptive control method and intelligent agent control system to solve the problems of low accuracy of traditional tool wear monitoring and difficulty in adapting to complex working conditions for intelligent fault diagnosis and intelligent control strategies.

[0006] In a first aspect, this application provides an artificial intelligence-based adaptive tool control method applied to an intelligent agent control system. The method includes: calculating the tool vibration energy growth rate; calculating the edge weight of each connection edge directly connected to the tool; calculating the vibration influence degree and vibration flow direction factor of each connection edge based on the edge weight; judging the tool wear condition based on the vibration influence degree, the vibration flow direction factor and the tool vibration energy growth rate; and making adjustments based on the tool wear condition.

[0007] By employing the above technical solution, the tool vibration signal is divided into frequency bands and differentiated energy weights are set, enabling precise capture of the impact of vibration energy in different frequency bands on tool wear. High-frequency vibration typically reflects microscopic damage to the tool, while low-frequency vibration is related to system stability. By dynamically adjusting the weights in conjunction with tool properties (material hardness, wear resistance) and cutting parameters (cutting speed, feed rate), the vibration energy calculation becomes more closely aligned with actual working conditions, providing an accurate data foundation for subsequent wear assessment. This multi-dimensional data fusion analysis effectively improves the sensitivity of early wear characteristic identification and avoids misjudgment based on a single parameter.

[0008] In conjunction with some embodiments of the first aspect, in some embodiments, the calculation of the tool vibration energy growth rate includes: acquiring a tool vibration signal, dividing the vibration signal into multiple frequency bands according to a preset frequency band division strategy, and calculating the energy of each frequency band; when the frequency band of the tool vibration signal is greater than a set frequency band threshold, determining a first energy weight for the corresponding frequency band, and when the frequency band of the tool vibration signal is not greater than the set frequency band threshold, determining a second energy weight for the corresponding frequency band; calculating a first product of the first energy weight and the energy of the corresponding frequency band, calculating a second product of the second energy weight and the energy of the corresponding frequency band, and calculating the total vibration energy of the tool, wherein the total vibration energy is the sum of the sum of each first product and the sum of each second product; calculating the average total vibration energy of multiple sampling points within a preset time window, wherein the average total vibration energy is the arithmetic mean of the total vibration energy of each sampling point within the time window, and the preset time window contains at least two consecutive sampling points; calculating the tool vibration energy growth rate of adjacent time windows, wherein the tool vibration energy growth rate is the ratio of the difference between the average total vibration energy of adjacent time windows to the average total vibration energy of the previous time window.

[0009] By adopting the above technical solution, the energy weight is dynamically adjusted based on tool properties, achieving personalized adaptation of the monitoring model. Tools with high material hardness are more sensitive to high-frequency vibrations, while tools with high wear resistance are more significantly affected by low-frequency energy changes. Cutting speed and feed rate directly affect the cutting force distribution, thereby altering the vibration energy propagation characteristics. Through threshold matching and coefficient correction, the energy weight is highly coupled with the actual working conditions of the tool, enhancing the physical meaning of vibration energy calculation and improving the system's compatibility and diagnostic accuracy for different tool types.

[0010] In conjunction with some embodiments of the first aspect, in some embodiments, the step of determining a first energy weight for a corresponding frequency band when the frequency band of the tool vibration signal is greater than a set frequency band threshold, and determining a second energy weight for a corresponding frequency band when the frequency band of the tool vibration signal is not greater than the set frequency band threshold, specifically includes: acquiring attribute data of the currently used tool, a first cutting speed, and a first feed rate, wherein the attribute data includes the tool's material hardness and wear resistance values; matching the corresponding material hardness threshold range in a preset tool attribute threshold table based on the material hardness to determine the corresponding first preliminary energy weight; and matching the corresponding wear resistance threshold range in the preset tool attribute threshold table based on the wear resistance value to determine the first preliminary energy weight. Determine the corresponding energy weight reduction factor, and calculate the product of the first preliminary energy weight and the energy weight reduction factor to obtain the second preliminary energy weight; when the first cutting speed is greater than the preset second cutting speed, increase the first preliminary energy weight by a first set percentage to obtain the first energy weight; when the first cutting speed is not greater than the preset second cutting speed, determine the first preliminary energy weight as the first energy weight; when the first feed rate is greater than the preset second feed rate, increase the second preliminary energy weight by a second set percentage to obtain the second energy weight; when the first feed rate is not greater than the preset second feed rate, determine the second preliminary energy weight as the second energy weight.

[0011] By adopting the above technical solution, cutting speed and feed rate directly alter the cutting force distribution, affecting the vibration energy propagation path. Through threshold matching and coefficient correction, the energy weight is highly coupled with the actual tool working conditions, enhancing the physical meaning of vibration energy calculation. This multi-dimensional parameter fusion strategy retains the sensitive response of high-frequency vibration to tool micro-damage while enhancing the characterization ability of low-frequency signals for system stability. It significantly improves the accuracy of tool wear state identification under complex working conditions, providing a more reliable state perception means for intelligent machining systems.

[0012] In some embodiments of the first aspect, the calculation of the edge weights of each connecting edge directly connected to the tool includes: obtaining a first connection distance, a first connection rigidity, and a first connecting edge of a first component connected to the tool; and sequentially obtaining the connection distance and connection rigidity of all components connected to the tool, wherein the connection distance refers to the physical installation distance between the tool and each connecting component, the connection rigidity refers to the deformation resistance of the connection structure between the tool and each connecting component, and the connecting edge refers to the edge connecting the tool and each component; calculating the first connection rigidity multiplied by a preset rigidity influence coefficient to obtain a first rigidity influence value; calculating the first connection distance multiplied by a preset distance influence coefficient to obtain a first distance influence value; calculating the first rigidity influence value divided by the square of the first distance influence value to obtain a first preliminary connection weight of the first connecting edge; summing the preliminary connection weights of each connecting edge to obtain a total connection weight value; calculating the first preliminary connection weight divided by the total connection weight value to obtain a first edge weight of the first connecting edge; and sequentially calculating the edge weights of each connecting edge.

[0013] By employing the aforementioned technical solution, a side-weight calculation model based on connection rigidity and distance was constructed, revealing the propagation law of tool vibration in the mechanical system. The rigidity influence value reflects the vibration transmission efficiency, while the distance influence value reflects the energy attenuation characteristics. Through normalization, the contribution of each connecting component to tool vibration was quantified. This structural dynamics analysis method extends isolated vibration monitoring to system-level collaborative analysis, providing a theoretical basis for accurately separating tool wear from system interference.

[0014] In some embodiments of the first aspect, the calculation of the vibration influence degree and vibration flow factor of each connecting edge based on the edge weight includes: obtaining a first vibration feature vector of the tool and a second vibration feature vector of each component directly connected to the tool, the vibration feature vector including time-domain features and frequency-domain features; calculating the average of the sum of squares of each dimension value of the first vibration feature vector and the second vibration feature vector, and taking the square root of the corresponding average value to obtain the first root mean square value of the first vibration feature vector and the second root mean square value of the second vibration feature vector; calculating the product of each edge weight and the first root mean square value to obtain the tool vibration influence degree of the tool on each component, calculating the weighted sum of each tool vibration influence degree to obtain the first vibration influence degree of the tool on each component; calculating the product of each edge weight and the second root mean square value to obtain the component vibration influence degree of each connecting component on the tool, calculating the weighted sum of the component vibration influence degrees to obtain the second vibration influence degree of each connecting component on the tool; and calculating the ratio of the first vibration influence degree to the second vibration influence degree to obtain the vibration flow factor.

[0015] By employing the above technical solution, and combining edge weights and vibration eigenvectors to calculate vibration influence and flow factor, a quantitative characterization of the tool-component vibration coupling relationship is achieved. Vibration influence reflects the intensity of energy transfer, while the flow factor reveals the main propagation direction of vibration energy. When the flow factor exceeds a threshold, it indicates that the tool dominates the system vibration, resulting in a high risk of wear; conversely, it suggests component failure. This bidirectional energy flow analysis method overcomes the limitations of traditional single-point monitoring and improves the diagnostic accuracy of tool conditions under complex operating conditions.

[0016] In conjunction with some embodiments of the first aspect, in some embodiments, determining the tool wear condition based on the vibration influence degree, the vibration flow factor, and the tool vibration energy growth rate includes: if the vibration flow factor is greater than a preset first wear threshold, and the first vibration influence degree is greater than a preset second wear threshold, and the tool vibration energy growth rate is greater than a preset third wear threshold, then it is determined that the tool wear condition has reached the level requiring immediate replacement; if the vibration flow factor is not greater than the first wear threshold, and there is at least one connecting component whose vibration influence on the tool is greater than a preset fourth wear threshold, then it is determined that the tool wear condition is due to the corresponding connecting component having a vibration influence degree greater than a preset fourth wear threshold. Fault; if the vibration flow factor is greater than the first wear threshold, and the tool wear condition does not reach the level of immediate replacement, then a comprehensive risk score is calculated. The comprehensive risk score is obtained by weighted summation of the ratio of the first vibration impact to the second wear threshold and the ratio of the tool vibration energy growth rate to the third wear threshold. If the comprehensive risk score is greater than the first risk score, then the tool wear condition triggers a level one warning. If the comprehensive risk score is greater than the second risk score, then the tool wear condition triggers a level two warning. The level two warning is more severe than the level one warning.

[0017] By adopting the above technical solutions, a multi-parameter fusion-based tool wear classification and judgment mechanism was established, achieving refined identification of wear conditions. A three-level threshold system (vibration flow factor, vibration impact degree, and energy growth rate) forms a logical AND relationship, effectively filtering system interference and ensuring the reliability of emergency replacement decisions. Independent judgment of component vibration impact degree enables rapid differentiation between tool failures and system failures. The comprehensive risk scoring mechanism provides a quantitative basis for preventative maintenance, enabling the system to maximize tool life while ensuring machining quality.

[0018] In conjunction with some embodiments of the first aspect, in some embodiments, the adjustment based on the tool wear condition includes: when the tool wear condition reaches a level requiring immediate replacement, generating a first adjustment instruction, the first adjustment instruction including the type, specifications, and replacement time of the tool; when the tool wear condition indicates a fault in a connecting component, generating a second adjustment instruction, the second adjustment instruction including prioritizing the repair of the connecting component with the greatest vibration impact, and if the vibration impact of multiple components exceeds a fourth wear threshold, then arranging the repair priority in descending order of impact; when the tool wear condition triggers a first-level warning, generating a third adjustment instruction, the third adjustment instruction including fine-tuning the cutting parameters, the cutting parameters including cutting speed, feed rate, and depth of cut; when the tool wear condition triggers a second-level warning, generating a fourth adjustment instruction, the fourth adjustment instruction including reducing the cutting parameters to a set safe range and increasing the tool monitoring frequency.

[0019] By adopting the above technical solutions, a hierarchical response intelligent adjustment strategy is constructed, enabling dynamic optimization management of the tool lifecycle. Immediate replacement commands prevent equipment downtime losses, component maintenance commands eliminate systemic vibration sources, cutting parameter fine-tuning commands slow down tool wear, and safety range adjustments and high-frequency monitoring commands ensure machining safety under critical conditions. This progressive control strategy transforms passive maintenance into proactive maintenance, significantly reducing machining costs and improving the reliability and flexibility of the intelligent manufacturing system.

[0020] In a second aspect, this application provides an intelligent agent control system, which includes: one or more processors and a memory; the memory is coupled to the one or more processors, and the memory is used to store computer program code, the computer program code including computer instructions, and the one or more processors call the computer instructions to cause the intelligent agent control system to perform the method described in the first aspect and any possible implementation thereof.

[0021] Thirdly, this application provides a computer-readable storage medium including instructions that, when executed on an intelligent agent control system, cause the intelligent agent control system to perform the method described in the first aspect and any possible implementation thereof.

[0022] Fourthly, this application provides a computer program product that, when run on an intelligent agent control system, causes the intelligent agent control system to perform the method described in the first aspect and any possible implementation thereof.

[0023] One or more technical solutions provided in the embodiments of this application have at least the following technical effects or advantages:

[0024] 1. By employing a technique that divides the tool vibration signal into frequency bands and dynamically adjusts the energy weight based on tool properties and cutting parameters, the technical problem of difficulty in accurately capturing tool wear characteristics under different working conditions and the tendency to make misjudgments in existing technologies is effectively solved. This results in improving the sensitivity of early tool wear characteristic identification, providing an accurate data basis for tool wear judgment, and avoiding misjudgments based on a single parameter.

[0025] 2. By adopting a calculation model for edge weights based on connection rigidity and distance, and normalizing the weights of each connection edge, the technical problems in the prior art, such as the inability to accurately quantify the vibration transmission relationship between the tool and each connecting component, and the difficulty in separating tool wear from system interference, are effectively solved. This achieves the technical effect of revealing the propagation law of tool vibration in the mechanical system and accurately analyzing the contribution of each component to tool vibration.

[0026] 3. By adopting a tool wear classification judgment mechanism based on multi-parameter fusion and formulating graded response intelligent adjustment strategies for different wear states, the technical problems of inaccurate tool fault judgment and lack of targeted maintenance strategies in the existing technology are effectively solved. This achieves the technical effect of transforming passive maintenance into active maintenance, reducing processing costs, and improving the reliability and flexibility of intelligent manufacturing systems. Attached Figure Description

[0027] Figure 1 This is a flowchart illustrating an artificial intelligence-based adaptive tool control method in an embodiment of this application.

[0028] Figure 2 This is another flowchart illustrating the artificial intelligence-based adaptive tool control method in this application embodiment;

[0029] Figure 3 This is a schematic diagram of the physical device structure of an intelligent agent control system in the embodiments of this application. Detailed Implementation

[0030] The terminology used in the following embodiments of this application is for the purpose of describing particular embodiments only and is not intended to be limiting of this application. As used in the specification and appended claims of this application, the singular expressions “a,” “an,” “the,” “the,” “the,” and “this” are intended to include the plural expressions as well, unless the context clearly indicates otherwise. It should also be understood that the term “and / or” as used in this application refers to and includes any or all possible combinations of one or more of the listed items.

[0031] Hereinafter, the terms "first" and "second" are used for descriptive purposes only and should not be construed as implying or suggesting relative importance or implicitly indicating the number of indicated technical features. Thus, a feature defined as "first" or "second" may explicitly or implicitly include one or more of that feature, and in the description of the embodiments of this application, unless otherwise stated, "multiple" means two or more.

[0032] For ease of understanding, the method provided in this implementation is described in process below. Please refer to [link / reference]. Figure 1 This is a flowchart illustrating an artificial intelligence-based adaptive tool control method in an embodiment of this application.

[0033] S101, Calculate the growth rate of tool vibration energy;

[0034] In machining, a cutting tool is a tool used for cutting, grinding, and other operations to remove or shape material, such as a turning tool in turning and a milling cutter in milling. Vibration energy refers to the energy contained in the vibration of a cutting tool during machining due to factors such as cutting force fluctuations and mechanical structure resonance. It is closely related to physical quantities such as the amplitude, frequency, and mass of the tool vibration and is a key indicator for measuring the severity of tool vibration. The tool vibration energy growth rate refers to the proportion of increase in tool vibration energy between two adjacent time points, reflecting the trend of tool vibration energy change over time.

[0035] This step is performed when the tool is in operation and begins cutting or other machining operations, and during the continuous real-time monitoring phase of the machining process.

[0036] Specifically, the intelligent agent control system first needs to deploy a suitable sensor network to collect tool vibration data. Commonly used sensors include accelerometers, which can measure the acceleration changes of the tool in various directions, and indirectly obtain vibration velocity and displacement information through integration; and displacement sensors, which can directly measure the vibration displacement of the tool. These sensors are generally installed in the tool holder, tool post, etc., to ensure accurate acquisition of tool vibration signals.

[0037] This step is initiated when the cutting tool begins its machining task. The intelligent agent control system first controls sensors to collect tool vibration signals in real time. These signals are transmitted to the system's data acquisition module in the form of analog or digital signals. The data acquisition module performs preliminary signal processing, including noise removal and signal amplification, to improve signal quality. Next, according to a pre-set frequency band division strategy, the system uses digital signal processing technology to decompose the original vibration signal into multiple frequency bands. For each frequency band, the system employs a corresponding energy calculation method. Taking the root mean square (RMS) value calculation as an example, the signal within the frequency band is first squared, then the average value is calculated, and finally the square root is taken. The resulting RMS value is related to the energy of that frequency band. Through further conversion and calibration, an accurate energy value can be obtained. In actual machining, different machining processes (such as roughing and finishing) and tool conditions (new tools, worn tools) will cause different energy distributions in each frequency band. The system continuously records these data to provide a basis for subsequent analysis.

[0038] After dividing the tool vibration signal into frequency bands and calculating the energy of each band, the intelligent agent control system compares the frequency range of each band with the set frequency band threshold. For frequency bands with an upper frequency limit greater than the set threshold, the system retrieves the corresponding first energy weight from a pre-established weight database or calculates it using a specific algorithm. The weight database stores weight configuration schemes for different types of tools, machining materials, and process combinations, allowing the system to quickly match appropriate weights based on the parameters of the current machining task. If no matching scheme is found, the system uses a machine learning-based algorithm, combined with historical machining data and current vibration signal characteristics, to dynamically calculate the first energy weight. For frequency bands with an upper frequency limit not greater than the set threshold, the system determines the second energy weight in a similar manner. Furthermore, the system dynamically adjusts the weights based on real-time feedback during machining, such as tool wear and cutting force changes. For example, when increased tool wear is detected, the first energy weight of the high-frequency band may be appropriately increased to more sensitively capture changes in tool vibration and promptly detect potential faults. The set frequency band threshold is a frequency limit value determined comprehensively based on factors such as tool material characteristics, machining process requirements, and machine tool performance. It is used to distinguish the importance of different frequency bands in the assessment of tool vibration status. The first energy weight is a weighting coefficient assigned to the energy of a frequency band when the frequency of the tool vibration signal is greater than the set frequency band threshold, used to highlight the importance of high-frequency energy in the calculation of the total energy of tool vibration. The second energy weight is a weighting coefficient assigned to the energy of low-frequency bands when the frequency is not greater than the set frequency band threshold.

[0039] After determining the energy weights for each frequency band, the intelligent agent control system traverses all frequency bands. For frequency bands with energy greater than a set threshold, its first energy weight is multiplied by the corresponding frequency band energy to obtain a first product. For frequency bands with energy less than or equal to the set threshold, a second product is calculated. The system sums all the first products to obtain a sum of the first products, and sums all the second products to obtain a sum of the second products. Finally, these two sums are added together to obtain the total vibration energy of the tool. To improve computational efficiency, the system employs parallel computing technology to process the product calculations of multiple frequency bands simultaneously. The first product refers to the value obtained by multiplying the first energy weight by the corresponding frequency band energy, used to quantify the contribution of high-frequency energy to the total vibration energy of the tool. The second product is the result of multiplying the second energy weight by the corresponding frequency band energy, reflecting the contribution of low-frequency energy. The total vibration energy of the tool is a quantitative indicator reflecting the overall intensity of tool vibration after comprehensively considering the energy and weights of all frequency bands.

[0040] After calculating the total energy of tool vibration, short-term fluctuations in the total vibration energy need to be eliminated to obtain stable energy characteristics. Specifically, the intelligent agent control system collects the total energy data of tool vibration in real time according to a preset sampling frequency and stores the data in a buffer. When the number of data in the buffer reaches the number of sampling points corresponding to the preset time window, the system calculates the arithmetic mean of these data to obtain the average total vibration energy within that time window. To achieve real-time monitoring, the system adopts a sliding window mechanism, that is, as new data is continuously collected, the window slides forward, removing the earliest sampling point data, adding the latest data, and recalculating the mean. The system also dynamically adjusts the size of the preset time window according to changes in the machining process. In the early stage of machining, a smaller time window may be set to quickly capture the initial state of tool vibration; in the stable stage of machining, the time window will be appropriately increased to obtain a more stable average energy value. In addition, the system filters the calculated average total vibration energy, such as using moving average filtering and median filtering, to further remove noise interference and make the mean data more accurately reflect the actual vibration energy state of the tool. The average total vibration energy is calculated by averaging the total vibration energy of all sampling points within the time window. This value can smooth out short-term fluctuations and more stably reflect the average state of tool vibration energy.

[0041] After calculating the average total vibration energy for each time window, the system monitors the changing trend of tool vibration energy in real time to promptly detect abnormal tool wear or changes in machining conditions. Specifically, after obtaining the average total vibration energy of two adjacent time windows, the intelligent agent control system calculates the tool vibration energy growth rate according to the formula: Tool vibration energy growth rate = (Average total vibration energy of the later time window - Average total vibration energy of the previous time window) ÷ Average total vibration energy of the previous time window × 100%. The system compares the calculated growth rate with a pre-set threshold. If the growth rate exceeds the threshold, it indicates that the tool vibration energy is increasing too rapidly, which may indicate problems such as accelerated tool wear, unreasonable cutting parameters, or changes in workpiece material hardness. The system will immediately issue an early warning signal and take corresponding measures according to the preset emergency plan, such as adjusting cutting parameters and prompting operators to check the tool.

[0042] In some embodiments, the first energy weight and the second energy weight can also be determined by: firstly acquiring the attribute data of the currently used tool, the first cutting speed, and the first feed rate. The attribute data of the currently used tool represents the inherent physical characteristic parameters of the tool participating in the machining task, and is an important basis for measuring tool performance and applicability. Material hardness is used to represent the tool material's ability to resist local deformation, especially plastic deformation, indentation, or scratching. Wear resistance is a quantitative indicator of the tool's ability to resist wear during cutting, generally obtained through experimental testing or technical parameters provided by the tool manufacturer; the higher the value, the more wear-resistant the tool. The first cutting speed refers to the linear velocity of a selected point on the tool's cutting edge relative to the main motion of the workpiece, and the first feed rate refers to the displacement of the tool relative to the workpiece in the feed motion direction. This is done before the tool is installed and ready to begin the machining task, and when changing tools or adjusting cutting parameters during machining. The intelligent agent control system first acquires the attribute data of the currently used tool by interacting with the tool management system or tool identification device. This data may be stored in the tool's electronic tag, which the system retrieves and parses using a reading device; alternatively, it may be pre-entered into the tool database of the machine tool control system, where the system retrieves the corresponding data based on the tool number or installation location. For the first cutting speed and first feed rate, the system obtains them from the machine tool's CNC system parameter setting interface, which are preset by the operator according to the machining process requirements.

[0043] After acquiring the tool material hardness data, the intelligent agent control system compares the acquired tool material hardness value with each material hardness threshold interval in the preset tool attribute threshold table. Starting from the first interval in the threshold table, the system determines whether the material hardness falls within that interval. If it does, it acquires the first preliminary energy weight corresponding to that interval; otherwise, it continues comparing with the next interval until a matching interval is found. If the material hardness exceeds the range of any interval in the threshold table, the system will respond according to preset processing rules. The preset tool attribute threshold table is a rule table pre-stored in the intelligent agent control system, used to associate tool attribute parameters with energy weights. It contains multiple attribute threshold intervals and their corresponding weight values.

[0044] After determining the first preliminary energy weight, the energy weight calculation based on tool attributes needs to be further refined. The intelligent agent control system matches the acquired tool wear resistance value with the wear resistance threshold range in the preset tool attribute threshold table. The matching process is similar to material hardness matching, finding the corresponding range by comparing one by one, thereby obtaining the energy weight reduction coefficient corresponding to that range. After obtaining the reduction coefficient, the system multiplies the first preliminary energy weight by the reduction coefficient to calculate the second preliminary energy weight. As the tool's usage and wear condition change, the system can dynamically update the wear resistance value based on real-time monitored tool wear data and recalculate the second preliminary energy weight, making the weight value more closely reflect the actual state of the tool.

[0045] After determining the initial energy weight, the final determination of the initial energy weight is made in conjunction with the cutting speed condition. The intelligent agent control system compares the acquired first cutting speed with the preset second cutting speed. If the first cutting speed is greater than the preset second cutting speed, the system adjusts the initial energy weight upwards by a first set percentage, that is, multiplying the initial energy weight by (1 + the first set percentage) to obtain the first energy weight; if the first cutting speed is not greater than the preset second cutting speed, the initial energy weight is directly used as the first energy weight. In actual machining, the preset second cutting speed and the first set percentage can be flexibly adjusted according to different tool types, machining materials, and process requirements. The system also records and analyzes changes in the first energy weight. For example, during machining, when a change in cutting speed leads to an adjustment of the first energy weight, the system analyzes the impact of this adjustment on tool vibration and machining quality, providing a basis for subsequent optimization of cutting parameters and weight calculation rules. The first set percentage is a pre-set value used to adjust the initial energy weight when the first cutting speed is greater than the preset second cutting speed.

[0046] After determining the second preliminary energy weight, the final determination of the second energy weight is made in conjunction with the feed rate condition. The intelligent agent control system compares the acquired first feed rate with the preset second feed rate. If the first feed rate is greater than the preset second feed rate, the system adjusts the second preliminary energy weight upwards by a second set percentage, that is, multiplying the second preliminary energy weight by (1 + the second set percentage) to obtain the second energy weight; if the first feed rate is not greater than the preset second feed rate, the second preliminary energy weight is directly used as the second energy weight. Similar to the influence of cutting speed, the preset second feed rate and the second set percentage can also be adjusted according to the actual machining situation. Among them, the second set percentage is preset and is the proportional value used to adjust the second preliminary energy weight when the first feed rate is greater than the preset second feed rate.

[0047] S102. Calculate the edge weights of each connection edge directly connected to the tool;

[0048] In machining equipment, the connecting edges directly connected to the cutting tool represent the physical connection between the cutting tool and directly connected components (such as tool holders, tool shanks, spindles, etc.). This connection can be a rigid connection (such as through threaded fastening) or an elastic connection (such as using elastic bushings). The connection distance refers to the physical installation distance between the cutting tool and each connecting component. This distance affects the efficiency of vibration transmission between components; the greater the distance, the greater the potential attenuation during vibration transmission. Connection rigidity refers to the resistance to deformation of the connection structure between the cutting tool and each connecting component. The stronger the connection rigidity, the less likely the connection structure is to deform under external forces or vibrations, which also affects the vibration transmission characteristics. Edge weight is a quantitative indicator used to measure the degree of influence of the connecting edge on the vibration characteristics of the cutting tool during vibration transmission; a larger value indicates a more significant influence of the connecting edge on the tool vibration. For example, in milling equipment, the milling cutter is mounted on the spindle via a heat-shrink tool holder, forming connecting edges between the milling cutter and the tool holder, and between the tool holder and the spindle. If the connection distance between the milling cutter and the tool holder is 50mm, the connection rigidity is relatively strong. If the connection distance between the spindle and the tool holder is 100mm, the connection rigidity is relatively weak. Through subsequent calculations, the edge weight of the connection between the milling cutter and the tool holder may be 0.6, and the edge weight of the connection between the spindle and the tool holder may be 0.4.

[0049] This step is performed after the tool is installed on the machining equipment and debugged, and periodically during machining (e.g., after each batch of parts is machined) or when the system detects abnormal tool vibration. Specifically, the intelligent agent control system first obtains information about each component connected to the tool through the equipment's 3D modeling information, assembly drawings, or built-in sensor network, thereby determining all connection edges directly connected to the tool. For each connection edge, the system obtains its corresponding first connection distance and first connection rigidity data. The connection distance can be directly obtained from the equipment's geometric parameters, while the acquisition of connection rigidity data is more complex. It may involve obtaining the theoretical rigidity value by consulting the component's technical manual, or by applying a small excitation to the connecting component and measuring its deformation while the equipment is running under no-load, and then using mechanical formulas to deduce the actual rigidity value. After acquiring the data, the system performs a series of calculations: First, the first connection rigidity is multiplied by a preset rigidity influence coefficient to obtain the first rigidity influence value. This rigidity influence coefficient is preset based on a large amount of experimental data and theoretical analysis, and is used to amplify or reduce the influence of connection rigidity on edge weights. Next, the first connection distance is calculated and multiplied by a preset distance influence coefficient to obtain the first distance influence value. The distance influence coefficient is also calibrated and set to reflect the weight of the connection distance on vibration transmission. Then, the first rigidity influence value is calculated and divided by the square of the first distance influence value to obtain the first preliminary connection weight of the first connection edge. This calculation method is based on vibration transmission theory, considering that connection rigidity is positively correlated with vibration transmission capacity, while the square of the connection distance is related to the degree of vibration attenuation. Subsequently, the system adds the preliminary connection weights of each connection edge to obtain the total connection weight value. Finally, the first preliminary connection weight is calculated and divided by the total connection weight value to obtain the first edge weight of the first connection edge, and the edge weights of each connection edge are calculated sequentially according to the same calculation process. During the calculation process, the system records calculation data and intermediate results in real time. If an anomaly is detected in the calculation result of a certain connection edge (such as a negative edge weight or one that is far outside the reasonable range), the system will automatically backtrack and check the data acquisition and calculation process to investigate whether there are any data errors or improper parameter settings. At the same time, the system will also dynamically adjust the rigidity influence coefficient and distance influence coefficient according to changes in the equipment's operating status and processing technology to ensure the accuracy of edge weight calculation.

[0050] S103. Calculate the vibration influence degree and vibration flow direction factor of each connected edge based on the edge weight;

[0051] Among them, vibration influence degree is a relative index used to measure the degree of influence of each connecting edge on the vibration state of the tool. By processing the edge weights, the importance of each connecting edge in the vibration transmission process of the entire tool-component system can be clarified. Vibration flow direction factor is a parameter reflecting the direction and intensity of vibration propagation between the components at both ends of the connecting edge. Its positive and negative values ​​indicate the direction of vibration propagation (positive for propagation from the tool to other components, negative for propagation from other components to the tool), and the magnitude of the value reflects the propagation intensity. Vibration feature vector is a set of data used to describe the vibration characteristics of the tool or component, including time-domain features (such as the change curve of vibration displacement over time, the peak value of vibration velocity, etc.) and frequency-domain features (such as the frequency component distribution of the vibration signal, the energy proportion at each frequency, etc.). For example, in a certain lathe machining, the vibration influence degree of a certain connecting edge is 0.3, which means that the influence of this connecting edge on the tool vibration state accounts for 30% of the influence of all connecting edges; if the vibration flow direction factor of this connecting edge is -0.2, it means that the vibration propagates from the component connected to this connecting edge to the tool, and the propagation intensity is relatively weak.

[0052] This step is performed after the edge weight calculation is completed, and is continuously calculated and updated in real time throughout the machining process. Specifically, the intelligent agent control system first collects vibration data in real time through sensors (such as accelerometers and displacement sensors) installed on the tool and various connecting components. Signal processing techniques (such as Fast Fourier Transform and Wavelet Transform) are used to convert the collected time-domain vibration signals into vibration feature vectors containing both time-domain and frequency-domain features. For the first vibration feature vector of the tool and the second vibration feature vectors of each component, the system calculates the average of the sum of squares of each dimension value and takes the square root of the corresponding average value to obtain the first root mean square value of the first vibration feature vector and the second root mean square value of the second vibration feature vector. The root mean square value comprehensively reflects the intensity of the vibration signal. Next, the system calculates the vibration influence degree: it calculates the product of each edge weight and the first root mean square value to obtain the tool vibration influence degree of the tool on each component, which represents the degree to which the tool vibration affects each component through the connecting edges; then, it calculates the weighted sum of the tool vibration influence degrees to obtain the first vibration influence degree of the tool on each component, which is used to measure the comprehensive impact of the overall tool vibration on other components. Similarly, by calculating the product of each side weight and the second root mean square value, the vibration influence degree of each connecting component on the tool is obtained. Then, the weighted sum of the vibration influence degrees of each component is calculated to obtain the second vibration influence degree of each connecting component on the tool, which reflects the comprehensive influence of the connecting component vibration on the tool. Finally, the system calculates the ratio of the first vibration influence degree to the second vibration influence degree to obtain the vibration flow direction factor. The sign and magnitude of this factor are used to determine the propagation direction and intensity of vibration between the tool and the components.

[0053] S104. Determine the tool wear condition based on the vibration influence, the vibration flow factor, and the tool vibration energy growth rate.

[0054] This step is performed after calculating the vibration impact degree, vibration flow factor, and tool vibration energy growth rate, and is continuously judged in real time during the machining process.

[0055] Specifically, the intelligent control system first compares the calculated vibration flow factor, first vibration influence degree, and tool vibration energy growth rate with preset first wear threshold, second wear threshold, and third wear threshold, respectively. If the vibration flow factor is greater than the preset first wear threshold, the first vibration influence degree is greater than the preset second wear threshold, and the tool vibration energy growth rate is greater than the preset third wear threshold, this indicates that the tool vibration energy is growing rapidly and the vibration transmission between the tool and the component is abnormal. At this time, the system determines that the tool wear has reached a level requiring immediate replacement and issues an emergency alarm to the operator. Simultaneously, it automatically suspends the processing task to prevent serious degradation of processing quality or equipment damage due to excessive tool wear. If the vibration flow factor is not greater than the first wear threshold, but at least one connecting component has a component vibration influence degree on the tool greater than the preset fourth wear threshold, this indicates that although the vibration transmission direction is not obviously abnormal, individual connecting components have an excessive impact on tool vibration. The system determines that the tool wear is due to a fault in the corresponding connecting component, such as loose connection or component damage. At this time, the system will issue a fault prompt, suggesting that the operator check the relevant connecting components. If the vibration flow factor is greater than the first wear threshold, but the tool wear is not severe enough to warrant immediate replacement, the system calculates a comprehensive risk score. This score is obtained by weighting and summing the ratio of the first vibration impact to the second wear threshold and the ratio of the tool vibration energy growth rate to the third wear threshold using certain weighting coefficients. The system compares this comprehensive risk score with the first and second risk scores. If the comprehensive risk score is greater than the first risk score, a level one warning is triggered, indicating an increased risk of tool wear. Operators can adjust machining parameters appropriately, such as reducing cutting speed or feed rate, to mitigate tool wear. If the comprehensive risk score is greater than the second risk score, a level two warning is triggered. Since the second risk score is greater than the first, the level two warning is more severe, and the system strongly recommends that operators replace the tool as soon as possible to avoid machining accidents caused by tool wear. Throughout the judgment process, the system continuously learns and optimizes the judgment rules and threshold settings. By analyzing a large amount of historical machining data and tool wear cases, the system adjusts the threshold values ​​and weighting coefficients to improve the accuracy and reliability of tool wear judgment. The first, second, third, and fourth wear thresholds are pre-set limits for judging tool wear based on factors such as tool type, machining process, and equipment performance. The comprehensive risk score is a value calculated by weighting and summing the ratio of the first vibration impact to the second wear threshold and the ratio of the tool vibration energy growth rate to the third wear threshold, and is used to quantify the comprehensive risk level of tool wear. The first-level and second-level warnings are alarm levels of different severity set according to the comprehensive risk score, used to prompt operators to take appropriate measures.For example, in a certain processing scenario, the first wear threshold is preset to be 0.5, the second wear threshold is 0.4, and the third wear threshold is 20%. If the vibration flow factor is 0.6, which is greater than the first wear threshold, the first vibration influence is 0.5, which is greater than the second wear threshold, and the tool vibration energy growth rate is 30%, which is greater than the third wear threshold, then it can be determined that the tool wear condition has reached the point where it needs to be replaced immediately.

[0056] S105. Adjust according to the wear condition of the tool.

[0057] The tool wear status refers to the assessment of the current tool condition based on a comprehensive analysis of parameters such as the tool vibration energy growth rate, the vibration influence of the connecting edge, and the vibration flow direction factor. This assessment covers different states, including normal tool condition, parameter adjustment required, component failure, and immediate replacement. Adjustment refers to the intelligent agent control system taking corresponding measures based on the tool wear status to restore or optimize the machining process, ensuring machining quality and efficiency. For example, if the tool wear is determined to be slight, requiring parameter adjustment, the system may reduce tool wear by decreasing the cutting speed. The specific adjustment process will be detailed in steps S201-S204 and will not be repeated here.

[0058] In this embodiment, the intelligent control system accurately calculates the tool vibration energy growth rate, the edge weight of the connecting edge, the vibration influence degree, and the vibration flow direction factor, and scientifically judges the tool wear based on these parameters, and then makes targeted adjustments. Therefore, it can grasp the working status of the tool in real time and accurately, effectively solving the problems of difficulty in timely detection of tool wear, inability to accurately judge the cause of wear, and untimely adjustment of processing parameters leading to decreased processing quality and excessive tool wear in traditional processing. This results in improved processing accuracy and product quality, extended tool life, increased processing efficiency, reduced production costs, and intelligentization of the machining process.

[0059] Based on the above, the following is a more detailed description of the process provided in this implementation. Please refer to [link / reference]. Figure 2 This is another flowchart illustrating the artificial intelligence-based adaptive tool control method in this application embodiment.

[0060] S201. When the tool wear reaches a level that requires immediate replacement, a first adjustment command is generated. The first adjustment command includes the type, specifications, and replacement time of the tool.

[0061] The phrase "tool wear reaches the level requiring immediate replacement" indicates that, based on a comprehensive analysis of parameters such as tool vibration energy growth rate, connection edge vibration impact, and vibration flow factor, the tool is severely worn, and continued use would seriously affect machining quality and equipment safety, necessitating immediate replacement. The first adjustment instruction is a set of instructions generated by the intelligent agent control system to guide the tool replacement operation in response to this situation. The "tool type" refers to the material and application classification of the new tool, such as carbide tools or high-speed steel tools; the "specifications" indicate the tool's detailed dimensional parameters, such as diameter, length, and number of teeth; and the "replacement time" refers to the specific moment when tool replacement is recommended, generally immediate execution or execution immediately after completing the current urgent machining task.

[0062] This step is triggered when the intelligent agent control system determines, based on step S104, that the tool wear has reached a level requiring immediate replacement. It is applicable to various machining and manufacturing scenarios. Specifically, after the intelligent agent control system completes its assessment of the tool wear and confirms the need for immediate tool replacement, it first filters suitable tool types and specifications from the system's built-in tool database based on the requirements of the current machining task (such as machining materials, machining processes, and machining accuracy requirements). The database stores a large amount of information on tools of different brands, materials, and specifications, as well as the applicable machining scenarios and parameter ranges for each tool. The system uses a preset matching algorithm to comprehensively consider factors such as tool cutting performance, durability, and cost to recommend the most suitable tool. After determining the tool type and specifications, the system determines a reasonable replacement time based on the progress of the current machining task. If the current machining task cannot be interrupted, the system will issue a warning to the operator a certain time in advance (e.g., 5 minutes) before the task is about to end, reminding them to prepare for tool replacement. If the current task can be stopped immediately, the system will directly send a stop command to the machine tool control system and generate the first adjustment command, which will be displayed to the operator through a human-machine interface (such as the operation panel display, mobile application, etc.). The command content includes detailed tool replacement guidance information, such as tool installation steps and precautions. In addition, the system will automatically record the tool replacement history, including replacement time, tool status before and after replacement, etc., for subsequent data analysis and tool management optimization.

[0063] S202. When the tool wear condition indicates a fault in the connecting component, a second adjustment instruction is generated. The second adjustment instruction includes prioritizing the connecting component with the greatest vibration impact. If the vibration impact of multiple components exceeds the fourth wear threshold, the maintenance priority is arranged in descending order of impact.

[0064] The second adjustment instruction is generated by the intelligent agent control system in response to this situation to guide the repair of connecting components. The component vibration impact degree refers to the degree of influence of the connecting component on the tool vibration state; a higher value indicates a more significant impact. The fourth wear threshold is a pre-set limit value used to determine whether a connecting component is faulty. When the component vibration impact degree exceeds this threshold, the component is considered to be potentially faulty. For example, during lathe machining, the system detects abnormal tool vibration. Analysis reveals that the vibration impact degree of the spindle-tool holder connecting component reaches 0.6, exceeding the preset fourth wear threshold of 0.5. The system then determines that the tool wear indicates a fault in the connecting component and generates the second adjustment instruction, prioritizing the repair of this spindle-tool holder connecting component. If the vibration impact degree of the tool holder-tool shank connecting component also exceeds the threshold and is 0.7, then the components are prioritized for repair according to their impact degree in descending order.

[0065] This step is initiated when the intelligent control system determines that the tool wear is caused by a fault in the connecting component. Specifically, once the system determines that the tool wear is caused by a fault in the connecting component, it first sorts the vibration impact of all components directly connected to the tool. The system retrieves the vibration impact data of each component from the stored calculation results, compares it with a fourth wear threshold, and filters out components whose impact exceeds the threshold. For the filtered components, they are arranged in descending order of vibration impact to determine the maintenance priority. Then, the system generates a second adjustment instruction. In addition to specifying the connecting component to be prioritized for maintenance, the instruction also provides detailed maintenance guidance information, such as disassembly steps for the component, common troubleshooting methods, and recommended maintenance tools. This information is displayed to maintenance personnel through the system's human-machine interface and is also sent to the relevant maintenance management system to facilitate maintenance personnel in scheduling maintenance plans.

[0066] S203. When tool wear triggers a first-level warning, a third adjustment command is generated. The third adjustment command includes fine-tuning the cutting parameters, which include cutting speed, feed rate, and depth of cut.

[0067] The first-level warning triggered by tool wear indicates that, based on analysis of parameters such as the tool vibration energy growth rate, the vibration influence of the connecting edge, and the vibration flow factor, the risk of tool wear has increased but has not yet reached a severe level, requiring certain measures to mitigate tool wear. The third adjustment command is generated by the intelligent agent control system in response to the first-level warning to adjust cutting parameters. Fine-tuning cutting parameters refers to making small adjustments to parameters such as cutting speed, feed rate, and depth of cut to optimize the cutting process and reduce tool wear rate. Cutting speed refers to the linear velocity of a selected point on the tool's cutting edge relative to the main motion of the workpiece; feed rate refers to the displacement of the tool relative to the workpiece in the feed direction; depth of cut refers to the depth to which the tool penetrates the workpiece.

[0068] This step is executed when the intelligent control system detects tool wear and triggers a level one warning. Specifically, after the system detects tool wear and triggers a level one warning, it first determines a fine-tuning scheme for the cutting parameters based on preset adjustment strategies and an experience database. The system analyzes factors such as the current machining process, tool type, and workpiece material, and combines this with the parameter adjustment effects under similar conditions in historical data to calculate appropriate adjustment values ​​for cutting speed, feed rate, and depth of cut. Generally, to mitigate tool wear, the cutting speed and feed rate are appropriately reduced, while the depth of cut adjustment depends on the specific machining conditions. After determining the adjustment scheme, the system generates a third adjustment command and sends it to the machine tool's CNC system. Upon receiving the command, the CNC system automatically adjusts the machine tool's operating status according to the new parameter values. During the adjustment process, the system monitors the tool's vibration state and machining quality in real time, collecting tool vibration data and machining surface quality data (such as surface roughness and dimensional accuracy) through sensors. If the adjusted parameters still fail to effectively control tool wear or cause a decrease in machining quality, the system will automatically re-evaluate and adjust the parameters until the tool wear is improved and the machining quality meets the requirements.

[0069] S204. When tool wear triggers a level 2 warning, a fourth adjustment command is generated. The fourth adjustment command includes reducing the cutting parameters to the set safe range and increasing the tool monitoring frequency.

[0070] Among them, triggering a level two warning due to tool wear indicates a high risk of tool wear and a more serious situation, requiring stricter measures to prevent excessive tool wear or damage. The fourth adjustment command is a command generated by the intelligent agent control system in response to the level two warning, used for emergency adjustments to the machining process. Reducing cutting parameters to a set safe range means lowering parameters such as cutting speed, feed rate, and depth of cut to a pre-set range that ensures safe tool operation and maintains machining progress as much as possible; increasing the tool monitoring frequency means shortening the time interval between tool status data acquisition and analysis to more promptly grasp changes in tool wear.

[0071] This step is initiated immediately when the intelligent control system detects tool wear and triggers a level-two warning. Specifically, once the system determines that tool wear has triggered a level-two warning, it quickly retrieves the appropriate cutting parameter safety range from a pre-set safety parameter library. The safety parameter library stores lower safety limits for cutting parameters for different tool types, workpiece materials, and machining process combinations. These values ​​are derived from extensive experiments and practical production experience. The system compares the current cutting speed, feed rate, and depth of cut with the safety range, calculates the parameter values ​​that need to be reduced, generates a fourth adjustment command, and sends it to the machine tool's CNC system. Upon receiving the command, the CNC system immediately adjusts the machine tool's operating status according to the new parameter values. Simultaneously, the system issues an alarm to the operator, indicating severe tool wear and that emergency parameter adjustment measures have been taken. While reducing cutting parameters, the system adjusts the settings of the data acquisition module, increasing the tool monitoring frequency. The system controls sensors installed on the tool and machine tool to collect data such as tool vibration signals, temperature signals, and cutting force signals at a higher frequency, and accelerates data transmission and processing speed. The collected data will be analyzed in real time. Once the tool condition is found to be deteriorating further or an abnormal situation occurs, the system will immediately take more stringent measures, such as suspending the machining task.

[0072] In this embodiment, the intelligent agent control system accurately calculates the tool vibration energy growth rate, determines the frequency band energy weight by comprehensively considering tool attributes and cutting parameters, scientifically calculates the edge weight, vibration influence degree, and vibration flow direction factor of the connecting edge, and establishes a multi-dimensional tool wear judgment system based on these parameters. Therefore, it can capture subtle changes in the tool's operating state in real time and accurately, effectively solving the problems of lagging tool wear monitoring and high tool wear costs in traditional machining, thereby improving the level of intelligence in the machining process.

[0073] The intelligent agent control system in the embodiments of this invention is described below from the perspective of hardware processing. Please refer to [link / reference needed]. Figure 3 This is a schematic diagram of the physical device structure of an intelligent agent control system in an embodiment of this application.

[0074] It should be noted that, Figure 3 The structure of the intelligent agent control system shown is merely an example and should not impose any limitations on the functionality and scope of use of the embodiments of the present invention.

[0075] like Figure 3 As shown, the intelligent agent control system includes a central processing unit (CPU) 301, which can perform various appropriate actions and processes based on programs stored in read-only memory (ROM) 302 or programs loaded from storage section 308 into random access memory (RAM) 303, such as executing the methods described in the above embodiments. The RAM 303 also stores various programs and data required for system operation. The CPU 301, ROM 302, and RAM 303 are interconnected via a bus 304. An input / output (I / O) interface 305 is also connected to the bus 304.

[0076] The following components are connected to I / O interface 305: input section 306 including audio input devices, push-button switches, etc.; output section 307 including a liquid crystal display (LCD) and audio output devices, indicator lights, etc.; storage section 308 including a hard disk, etc.; and communication section 309 including a network interface card such as a LAN (Local Area Network) card, modem, etc. Communication section 309 performs communication processing via a network such as the Internet. Drive 310 is also connected to I / O interface 305 as needed. Removable media 311, such as a disk, optical disk, magneto-optical disk, semiconductor memory, etc., are installed on drive 310 as needed so that computer programs read from them can be installed into storage section 308 as needed.

[0077] In particular, according to embodiments of the present invention, the processes described above with reference to the flowcharts can be implemented as computer software programs. For example, embodiments of the present invention include a computer program product comprising a computer program carried on a computer-readable medium, the computer program containing computer programs for performing the methods shown in the flowcharts. In such embodiments, the computer program can be downloaded and installed from a network via communication section 309, and / or installed from removable medium 311. When the computer program is executed by central processing unit (CPU) 301, it performs the various functions defined in the present invention.

[0078] It should be noted that specific examples of computer-readable storage media may include, but are not limited to: electrical connections having one or more wires, portable computer disks, hard disks, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM), flash memory, optical fiber, portable compact disc read-only memory (CD-ROM), optical storage devices, magnetic storage devices, or any suitable combination thereof. In this invention, a computer-readable storage medium can be any tangible medium containing or storing a program that can be used by or in conjunction with an instruction execution system, apparatus, or device.

[0079] The flowcharts and block diagrams in the accompanying drawings illustrate the architecture, functionality, and operation of possible implementations of systems, methods, and computer program products according to various embodiments of the present invention. Each block in a flowchart or block diagram may represent a module, program segment, or portion of code, which contains one or more executable instructions for implementing a specified logical function. It should also be noted that in some alternative implementations, the functions indicated in the blocks may occur in a different order than those shown in the drawings.

[0080] Specifically, the intelligent agent control system of this embodiment includes a processor and a memory. The memory stores a computer program. When the computer program is executed by the processor, it implements the artificial intelligence-based adaptive tool control method provided in the above embodiment.

[0081] In another aspect, the present invention also provides a computer-readable storage medium, which may be included in the intelligent agent control system described in the above embodiments; or it may exist independently and not incorporated into the intelligent agent control system. The storage medium carries one or more computer programs that, when executed by a processor of the intelligent agent control system, cause the intelligent agent control system to implement the artificial intelligence-based adaptive tool control method provided in the above embodiments.

[0082] The above-described embodiments are only used to illustrate the technical solutions of this application, and are not intended to limit it. Although this application has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features. Such modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the scope of the technical solutions of the embodiments of this application.

[0083] As used in the above embodiments, depending on the context, the term "when..." can be interpreted as meaning "if...", "after...", "in response to determining...", or "in response to detecting...". Similarly, depending on the context, the phrase "when determining..." or "if (the stated condition or event) is interpreted as meaning "if determining...", "in response to determining...", "when (the stated condition or event) is detected", or "in response to detecting (the stated condition or event)".

[0084] Those skilled in the art will understand that all or part of the processes in the methods of the above embodiments can be implemented by a computer program instructing related hardware. This program can be stored in a computer-readable storage medium, and when executed, it can include the processes described in the above method embodiments. The aforementioned storage medium includes various media capable of storing program code, such as ROM or random access memory (RAM), magnetic disks, or optical disks.

Claims

1. An artificial intelligence-based adaptive tool control method, applied to an intelligent agent control system, characterized in that, The method includes: Calculate the growth rate of tool vibration energy; Calculate the edge weights of each connection edge directly connected to the tool; The vibration influence degree and vibration flow direction factor of each connected edge are calculated by combining the edge weights. The tool wear condition is determined based on the vibration influence degree, the vibration flow direction factor, and the tool vibration energy growth rate. Adjustments are made based on the wear condition of the cutting tool.

2. The method according to claim 1, characterized in that, The calculation of the tool vibration energy growth rate includes: The vibration signal of the cutting tool is acquired, and the vibration signal is divided into multiple frequency bands according to a preset frequency band division strategy, and the energy of each frequency band is calculated. When the frequency band of the tool vibration signal is greater than the set frequency band threshold, the first energy weight of the corresponding frequency band is determined; when the frequency band of the tool vibration signal is not greater than the set frequency band threshold, the second energy weight of the corresponding frequency band is determined. Calculate the first product of the first energy weight and the energy of the corresponding frequency band, calculate the second product of the second energy weight and the energy of the corresponding frequency band, and calculate the total vibration energy of the tool, wherein the total vibration energy is the sum of the sum of each first product and the sum of each second product; Calculate the average total vibration energy of multiple sampling points within a preset time window. The average total vibration energy is the arithmetic mean of the total vibration energy of each sampling point within the time window. The preset time window contains at least two consecutive sampling points. Calculate the tool vibration energy growth rate for adjacent time windows, whereby the tool vibration energy growth rate is the ratio of the difference between the average total vibration energy of adjacent time windows to the average total vibration energy of the previous time window.

3. The method according to claim 2, characterized in that, The step of determining a first energy weight for the corresponding frequency band when the frequency band of the tool vibration signal is greater than a set frequency band threshold, and determining a second energy weight for the corresponding frequency band when the frequency band of the tool vibration signal is not greater than the set frequency band threshold, specifically includes: Acquire the attribute data of the currently used tool, the first cutting speed, and the first feed rate. The attribute data includes the material hardness and wear resistance values ​​of the tool. Based on the material hardness, the corresponding material hardness threshold range is matched in the preset tool attribute threshold table to determine the corresponding first preliminary energy weight; Based on the wear resistance value, the corresponding wear resistance threshold range is matched in the preset tool attribute threshold table to determine the corresponding energy weight reduction coefficient, and the product of the first preliminary energy weight and the energy weight reduction coefficient is calculated to obtain the second preliminary energy weight. When the first cutting speed is greater than the preset second cutting speed, the first preliminary energy weight is increased by a first set percentage to obtain the first energy weight; When the first cutting speed is not greater than the preset second cutting speed, the first preliminary energy weight is determined as the first energy weight; When the first feed amount is greater than the preset second feed amount, the second preliminary energy weight is increased by a second set percentage to obtain the second energy weight; When the first feed amount is not greater than the preset second feed amount, the second preliminary energy weight is determined as the second energy weight.

4. The method according to claim 1, characterized in that, The calculation of the edge weights of each connection edge directly connected to the tool includes: The first connection distance, first connection rigidity, and first connection edge of the first component connected to the tool are obtained, and the connection distance and connection rigidity of all components connected to the tool are obtained in sequence. The connection distance refers to the physical installation distance between the tool and each connecting component, the connection rigidity refers to the deformation resistance of the connection structure between the tool and each connecting component, and the connection edge refers to the edge connecting the tool and each component. The first rigidity influence value is obtained by multiplying the first connection rigidity by a preset rigidity influence coefficient; The first distance influence value is obtained by multiplying the first connection distance by a preset distance influence coefficient. Calculate the first rigidity influence value and divide it by the square of the first distance influence value to obtain the first preliminary connection weight of the first connecting edge; The initial connection weights of each of the connected edges are added together to obtain the total connection weight value; The first preliminary connection weight is calculated by dividing the total connection weight value to obtain the first edge weight of the first connection edge, and the edge weight of each connection edge is calculated in sequence.

5. The method according to claim 1, characterized in that, The calculation of the vibration influence degree and vibration flow direction factor of each connected edge in combination with the edge weight includes: The first vibration feature vector of the cutting tool and the second vibration feature vectors of each component directly connected to the cutting tool are obtained. The vibration feature vectors include time-domain features and frequency-domain features. The average of the sum of squares of the values ​​of each dimension of the first vibration feature vector and the second vibration feature vector is calculated, and the square root of the corresponding average value is taken to obtain the first root mean square value of the first vibration feature vector and the second root mean square value of the second vibration feature vector. Calculate the product of each edge weight and the first root mean square value to obtain the tool vibration influence degree of the tool on each component; calculate the weighted sum of each tool vibration influence degree to obtain the first vibration influence degree of the tool on each component. Calculate the product of each edge weight and the second root mean square value to obtain the component vibration influence degree of each connecting component on the tool. Calculate the weighted sum of the vibration influence degrees of each component to obtain the second vibration influence degree of each connecting component on the tool. The vibration flow direction factor is obtained by calculating the ratio of the first vibration influence degree to the second vibration influence degree.

6. The method according to claim 1, characterized in that, The method of determining tool wear based on the vibration influence degree, the vibration flow direction factor, and the tool vibration energy growth rate includes: If the vibration flow direction factor is greater than a preset first wear threshold, the first vibration influence is greater than a preset second wear threshold, and the tool vibration energy growth rate is greater than a preset third wear threshold, then it is determined that the tool wear condition has reached the point where it needs to be replaced immediately. If the vibration flow direction factor is not greater than the first wear threshold, and at least one connecting component has a greater influence on the vibration of the tool than the preset fourth wear threshold, then the tool wear condition is determined to be that the corresponding connecting component is faulty. If the vibration flow factor is greater than the first wear threshold and the tool wear condition does not reach the level of immediate replacement, then a comprehensive risk score is calculated. The comprehensive risk score is obtained by weighted summation of the ratio of the first vibration impact to the second wear threshold and the ratio of the tool vibration energy growth rate to the third wear threshold. If the comprehensive risk score is greater than the first risk score, the tool wear condition is determined to trigger a level one warning. If the comprehensive risk score is greater than the second risk score, the tool wear condition is determined to trigger a level two warning. The second risk score is greater than the first risk score, and the severity of the level two warning is higher than that of the level one warning.

7. The method according to claim 1, characterized in that, The adjustment based on the tool wear condition includes: When the wear of the cutting tool reaches a point where it needs to be replaced immediately, a first adjustment command is generated. The first adjustment command includes the type, specifications, and replacement time of the cutting tool. When the wear condition of the tool indicates that the connecting component is faulty, a second adjustment instruction is generated. The second adjustment instruction includes prioritizing the repair of the connecting component with the greatest vibration impact. If the vibration impact of multiple components exceeds the fourth wear threshold, the repair priority is arranged in descending order of impact. When the tool wear condition triggers a first-level warning, a third adjustment command is generated. The third adjustment command includes fine-tuning the cutting parameters, which include the cutting speed, feed rate, and depth of cut. When the tool wear condition triggers a secondary warning, a fourth adjustment command is generated. The fourth adjustment command includes reducing the cutting parameters to a set safe range and increasing the tool monitoring frequency.

8. An intelligent agent control system, characterized in that, The intelligent agent control system includes: one or more processors and a memory; the memory is coupled to the one or more processors, the memory is used to store computer program code, the computer program code including computer instructions, and the one or more processors call the computer instructions to cause the intelligent agent control system to perform the method as described in any one of claims 1-7.

9. A computer-readable storage medium comprising instructions, characterized in that, When the instructions are executed on the intelligent agent control system, the intelligent agent control system performs the method as described in any one of claims 1-7.

10. A computer program product, characterized in that, When the computer program product is run on the intelligent agent control system, the intelligent agent control system performs the method as described in any one of claims 1-7.

Images