An adaptive planning method and system for machining trajectory

By collecting machining status parameters in real time to identify the working condition type and dynamically correct the machining trajectory, the problem of decreased accuracy caused by tool wear, changes in workpiece stiffness, and machine tool thermal deformation during machining is solved, achieving high-precision and high-quality machining results.

CN123113386APending Publication Date: 2026-11-17JIANGSU SECURITY TECH CARRER ACADEMY
View PDF 0 Cites 0 Cited by

Patent Information

Application Number
CN202611502872.7
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2026-09-24
Publication Date
2026-11-17

AI Technical Summary

Technical Problem

In existing technologies, the machining trajectory remains unchanged during the machining process, which cannot cope with the problem of decreased machining accuracy caused by changes in working conditions such as tool wear, changes in workpiece stiffness, machine tool thermal deformation, and clamping errors.

Method used

By collecting machining status parameters in real time, such as tool wear, cutting force, cutting temperature and vibration signals, the current working condition type is identified and the deviation is calculated. The original machining trajectory is dynamically corrected, an adaptive machining trajectory is generated, and the CNC system is controlled to execute the trajectory.

Benefits of technology

It effectively compensates for machining errors caused by changes in working conditions, improves machining accuracy and surface quality, extends tool life, and reduces scrap rate.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN123113386A_ABST
    Figure CN123113386A_ABST
Patent Text Reader

Abstract

This invention discloses an adaptive machining trajectory planning method and system, solving the problems of fixed machining trajectories in existing technologies, which cannot cope with tool wear, changes in workpiece stiffness, machine tool thermal deformation, and clamping errors leading to decreased machining accuracy. The method includes: real-time acquisition of machining state parameters such as tool wear, cutting force, cutting temperature, and vibration signals; identification of working condition types based on these parameters and calculation of deviations from standard working conditions; online correction of the original machining trajectory to generate an adaptive machining trajectory; and control of the CNC system to execute the corrected trajectory. The system includes a parameter acquisition module, a working condition identification module, a trajectory correction module, and a CNC system control module. This invention can effectively compensate for machining errors caused by changes in working conditions, significantly improve machining accuracy and surface quality, extend tool life, and reduce scrap rate.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] This invention relates to the field of mechanical manufacturing, and in particular to an adaptive planning method and system for machining trajectories. Background Technology

[0002] Precision machining technology is a core pillar of modern manufacturing, widely used in aerospace, defense, medical devices, precision molds, and other fields. Machining trajectory planning, as a crucial step in CNC machining, directly determines machining accuracy, surface quality, and production efficiency. With increasingly complex and precise product structures, machining objects often include numerous complex curved surfaces, thin-walled structures, and difficult-to-machine materials, placing higher demands on the accuracy and adaptability of machining trajectories.

[0003] Adaptive machining trajectory planning is one of the core technologies for achieving high-quality machining. Specifically, machining trajectory planning relies on offline programs generated by computer-aided manufacturing software. These programs, based on a computer-aided design model of the part to be machined, and combined with selected machining strategies and cutting parameters, calculate and generate tool position data, which is then further post-processed into machining instructions recognizable by the CNC system. In actual machining, the relative motion trajectory between the tool and the workpiece determines the cutting path of the cutting edge on the workpiece surface. The accuracy of this path directly affects the geometric accuracy and surface integrity of the workpiece.

[0004] However, in existing technologies, the machining trajectory is determined before machining begins, and the entire machining process is executed according to a pre-planned fixed trajectory. This rigid trajectory mode cannot cope with the constantly changing working conditions during machining. As the cutting process continues, tool wear gradually increases, causing a deviation between the actual cutting depth and the nominal value; after material is gradually removed, the overall stiffness and local deformation characteristics of the workpiece change; during long-term operation of the machine tool, heat accumulation causes thermal deformation, resulting in coordinate accuracy drift; in addition, there are unpredictable errors in the clamping and positioning of the workpiece in the fixture. The combined effect of these factors leads to a significant deviation between the actual machining conditions and the ideal state assumed during offline programming. Since the trajectory remains unchanged during machining, these deviations cannot be compensated or corrected, the cutting conditions gradually deteriorate, ultimately resulting in decreased machining accuracy, deteriorated surface quality, shortened tool life, and in severe cases, even scrap or equipment damage. Summary of the Invention

[0005] The purpose of this invention is to provide a machining trajectory adaptive planning method and system, which can effectively solve the technical problems in the prior art where the machining trajectory remains unchanged during the machining process, and cannot cope with changes in working conditions such as tool wear, workpiece stiffness changes, machine tool thermal deformation and clamping errors, which lead to a decrease in machining accuracy.

[0006] To achieve the above objectives, the technical solution adopted by the present invention is: an adaptive planning method for processing trajectories, comprising the following specific steps:

[0007] S1: During the machining process, machining status parameters that characterize the current working condition are collected in real time. The machining status parameters include at least one of tool wear, cutting force, cutting temperature, and vibration signal.

[0008] S2: Based on the collected processing status parameters, identify the working condition type corresponding to the current processing status, and calculate the deviation between the working condition type and the preset standard working condition.

[0009] S3: Based on the calculated deviation, the original machining trajectory is corrected online to generate an adaptive machining trajectory that is adapted to the current working conditions;

[0010] S4: Control the CNC system to execute the adaptive machining trajectory to complete the machining of the current machining area;

[0011] S5: Repeat steps S1 to S4 until the entire part is machined.

[0012] The beneficial effects of this invention are:

[0013] This invention, by collecting machining status parameters in real time and identifying the type of working condition, can dynamically sense the influence of factors such as tool wear, workpiece stiffness changes, machine tool thermal deformation, and clamping errors during machining. Based on the deviation, it corrects the original machining trajectory online, enabling the tool to always move along a trajectory adapted to the current working condition. This effectively compensates for machining errors caused by changes in working conditions, significantly improves machining accuracy and surface quality, extends tool life, and reduces scrap rate. Attached Figure Description

[0014] Figure 1 This is a schematic diagram of the overall technical solution architecture of the adaptive planning method for processing trajectories proposed in this invention;

[0015] Figure 2 This is a schematic diagram of the core principle framework of online trajectory correction based on the deviation of processing state parameters in this invention. Detailed Implementation

[0016] The present invention will be further described below with reference to the accompanying drawings and specific embodiments. The illustrative embodiments and descriptions herein are used to explain the present invention, but are not intended to limit the present invention. Example 1

[0017] This embodiment uses precision mold cavity machining as an application scenario to explain in detail the specific implementation of the adaptive machining trajectory planning method and system proposed in this invention. In the field of precision mold manufacturing, the cavity structure is complex, the walls are thin and easily deformed, and the machining accuracy requirements are extremely high. Traditional fixed trajectory machining is difficult to guarantee product consistency and pass rate. This embodiment constructs a complete adaptive machining system architecture for this scenario and describes its workflow.

[0018] I. System Architecture Construction

[0019] In the scenario of precision mold cavity machining, the machining trajectory adaptive planning system deployed in this embodiment includes five core components: data acquisition module, working condition identification module, trajectory correction module, CNC system interface module, and control execution module. The modules establish a data interaction channel through an industrial bus to form a complete closed-loop control system.

[0020] The data acquisition module is located around the machining area and on the machine tool, specifically comprising four sub-units: a tool wear monitoring unit, a cutting force sensing unit, a temperature detection unit, and a vibration monitoring unit. The tool wear monitoring unit uses an acoustic emission sensor integrated inside the tool holder to measure wear. This sensor is positioned near the tool clip on the tool holder shank and calculates the wear on the tool's flank face by detecting the high-frequency acoustic emission signal characteristics during the cutting process. The cutting force sensing unit uses a piezoelectric force gauge embedded inside the spindle to measure three-dimensional cutting forces. This force gauge is embedded in the space between the spindle core and the bearing housing, enabling real-time acquisition of cutting force components in the X, Y, and Z axes. The temperature detection unit employs a combination of an infrared temperature sensor and a thermocouple. The infrared temperature sensor is positioned in a non-interference location above the machining area for non-contact measurement of the instantaneous temperature in the cutting zone, while the thermocouple is embedded in a designated location inside the workpiece to monitor the temperature field distribution within the workpiece. The vibration monitoring unit uses triaxial accelerometers installed on the spindle housing and workpiece fixture to collect vibration acceleration signals during the machining process. These sensors are connected to the signal conditioning circuit of the data acquisition module via shielded cables.

[0021] The working condition identification module comprises three parts: a signal preprocessing unit, a feature extraction unit, and a working condition classification unit, all integrated into an industrial control computer. The signal preprocessing unit is equipped with a multi-channel synchronous sampling card, capable of synchronously acquiring and filtering the raw signal transmitted from the data acquisition module at a sampling rate of no less than 10 kHz, effectively removing power frequency interference and random noise. The feature extraction unit employs a combination of time-domain statistical features and frequency-domain spectral features to extract feature parameters such as root mean square value, peak factor, and frequency band energy proportion from the preprocessed signal, forming a feature vector representing the current processing state. The working condition classification unit classifies and identifies the feature vector based on a pre-trained working condition identification model. This model, built using a support vector machine algorithm, can distinguish five basic working condition types: normal wear, severe wear, workpiece stiffness reduction, machine tool thermal deformation, and clamping error. It outputs the current working condition type label and the deviation value from the standard working condition.

[0022] The trajectory correction module also runs within the industrial control computer and includes a trajectory interpolation unit and a compensation calculation unit. The trajectory interpolation unit reads the original machining trajectory data file, stored on the system hard drive, which contains the tool position sequence generated by the computer-aided manufacturing software and the corresponding tool posture information. The compensation calculation unit calculates the spatial compensation vector to be applied to each tool position based on the deviation value output by the working condition identification module and a pre-established error compensation mapping relationship. This compensation vector includes a component along the tool feed direction, a normal component perpendicular to the machined surface, and a temperature compensation component for compensating for thermal deformation. The trajectory correction module smooths and corrects the original trajectory using a cubic spline interpolation algorithm, generating adaptive machining trajectory data adapted to the current working condition. This data is stored in a standard CNC machining code format and transmitted to the CNC system via a network.

[0023] The CNC system interface module is implemented using a dedicated machining communication gateway. This gateway is positioned between the industrial control computer and the CNC system, responsible for protocol conversion and data forwarding. One end of the gateway connects to the industrial control computer via an Ethernet interface, receiving adaptive machining trajectory data generated by the trajectory correction module; the other end connects to the CNC system via a dedicated data interface, converting the adaptive machining trajectory data into an instruction format recognizable by the CNC system. During data transmission, the CNC system interface module is also responsible for real-time monitoring of the CNC system's status information, including the currently executing program line number, spindle speed, feed rate, and other parameters, and feeding this information back to the host computer system for decision-making reference.

[0024] The control execution module is integrated within the CNC system and includes a trajectory analysis unit and a servo drive control unit. The trajectory analysis unit is responsible for parsing the adaptive machining trajectory data transmitted from the CNC system interface module and converting it into task instructions executable by the internal interpolator. The servo drive control unit controls the servo motor to drive the tool along the adaptive machining trajectory according to the parsed instructions, thus achieving the actual cutting of the workpiece. Throughout the machining process, the control execution module updates the tool position in real time with millisecond-level control cycles, ensuring that the deviation between the actual cutting path and the adaptive machining trajectory is controlled within acceptable limits.

[0025] II. Workflow Deduction

[0026] The adaptive machining trajectory planning system in this embodiment executes its workflow according to the following time sequence logic to achieve real-time adaptive adjustment of the machining trajectory.

[0027] During the machining startup phase, the operator first inputs the original machining trajectory file path and relevant process parameters of the part to be machined through the host computer software. After the system loads the file, it completes the initial configuration. At this time, the data acquisition module begins to work, and each sensor powers on and performs self-tests to confirm that the signal transmission channel is normal. The working condition recognition module loads the pre-trained working condition recognition model and error compensation mapping table to complete the algorithm initialization. The CNC system interface module establishes a communication connection with the CNC system to confirm that the data transmission channel is available. After everything is ready, the system enters the formal machining cycle phase.

[0028] At the start of each machining cycle, the data acquisition module executes step S1, acquiring machining status parameters characterizing the current working condition in real time. Specifically, the tool wear monitoring unit continuously acquires the raw signal output from the acoustic emission sensor and transmits it to the signal preprocessing unit for amplification and filtering; the cutting force sensing unit acquires the triaxial force signal output from the piezoelectric force gauge in real time, amplifies it through a charge amplifier, and then transmits it to the data acquisition module; the temperature detection unit simultaneously acquires temperature data from the infrared temperature sensor and thermocouple to obtain the instantaneous temperature of the cutting zone and the internal temperature distribution of the workpiece; the vibration monitoring unit acquires the vibration signal from the triaxial accelerometer to obtain the dynamic response characteristics during the machining process. The data acquired by these sensors is stored in a structured format in a data buffer, awaiting subsequent processing.

[0029] After data acquisition, the working condition identification module executes step S2, identifying the working condition type corresponding to the current machining state based on the acquired machining state parameters and calculating the deviation. The signal preprocessing unit first performs digital filtering on the raw signal from the data acquisition module to remove high-frequency noise and power frequency interference, obtaining a clean signal waveform. Subsequently, the feature extraction unit performs time-domain and frequency-domain analysis on the preprocessed signal, extracting statistical features such as root mean square value, peak value, skewness, and kurtosis of each channel signal, as well as spectral features such as dominant frequency and frequency band energy distribution, forming a complete feature vector. This feature vector is input into the support vector machine model of the working condition classification unit, and the model outputs the working condition type label of the current machining state. The system queries a preset error compensation mapping table according to the working condition type to obtain the deviation of the current working condition relative to the standard working condition. This deviation includes multiple components such as tool radial compensation, axial compensation, temperature compensation, and clamping offset.

[0030] After acquiring the working condition type and deviation information, the trajectory correction module executes step S3, which corrects the original machining trajectory online based on the calculated deviation, generating an adaptive machining trajectory that fits the current working condition. The compensation calculation unit first reads the actual position coordinates of the tool in the machining space to determine the range of tool positions requiring correction. Subsequently, the system calculates the spatial compensation vector to be applied to each relevant tool position based on the deviation value. This compensation vector calculation comprehensively considers factors such as cutting depth deviation caused by tool wear, elastic tool relief caused by changes in workpiece stiffness, coordinate drift caused by machine tool thermal deformation, and positional deviation caused by clamping errors. The trajectory interpolation unit uses a cubic spline curve fitting method to smooth the compensated tool positions, ensuring the continuity and smoothness of the tool movement trajectory and avoiding impacts and vibrations. The corrected adaptive machining trajectory data is transmitted to the CNC system in real time through the CNC system interface module.

[0031] After receiving the adaptive machining trajectory data generated in step S3, the control execution module executes step S4, controlling the CNC system to execute the adaptive machining trajectory and complete the machining of the current machining area. The trajectory parsing unit parses the adaptive machining trajectory data into specific interpolation instructions, including position instructions, speed instructions, and acceleration instructions for each axis. The servo drive control unit controls the servo motors of the X-axis, Y-axis, Z-axis, and rotary axes according to these instructions, driving the tool to move along the adaptive machining trajectory to achieve cutting of the workpiece. During the cutting process, the spindle speed and feed rate are adaptively adjusted according to the real-time acquired cutting force signal. When the cutting force exceeds the set threshold, the system automatically reduces the feed rate to protect the tool; when the cutting force is below the lower limit, the feed rate is appropriately increased to improve machining efficiency.

[0032] After completing the machining of the current machining area, the system determines whether the entire part is finished. If not, it returns to step S1 to begin a new cycle of data acquisition, identification, correction, and execution. The system continuously executes the above process according to a preset sampling period, typically triggering a complete adaptive adjustment cycle at each tool position transition to ensure the machining trajectory always matches the current working conditions. When the system detects that all machining areas have been cut, it automatically exits the machining cycle, outputs machining completion status information, and generates a machining process data report for subsequent analysis.

[0033] Through the above workflow, the machining trajectory adaptive planning system of this embodiment can perceive the tool wear state, cutting force changes, temperature field distribution and vibration characteristics in real time during the machining of precision mold cavities, identify the current working condition type and calculate the deviation from the standard working condition, and dynamically correct the original machining trajectory according to the deviation, so that the tool always moves along the path adapted to the current working condition, effectively compensates for the machining error caused by the change of working condition, and significantly improves the machining accuracy and surface quality. Example 2

[0034] This embodiment takes the processing of aero-engine blades as an application scenario. Based on the first embodiment, it further illustrates the specific implementation of the present invention in the precision machining of complex curved surfaces, and focuses on the working condition identification method of multi-source parameter fusion and the trajectory correction strategy under variable stiffness conditions.

[0035] In the field of aero-engine blade manufacturing, blades are typical complex curved thin-walled structures, mostly made of high-temperature alloys or titanium alloys, making them difficult to machine and requiring high precision. During the cutting process, the blade's local stiffness continuously changes due to material removal, and the effects of tool wear and machine tool thermal deformation are particularly significant. Traditional machining methods struggle to guarantee the high precision requirements of the blade profile. This embodiment optimizes the system architecture and workflow specifically for this particular scenario.

[0036] In terms of system architecture, this embodiment adds an ultrasonic thickness measurement unit and a strain measurement unit to the data acquisition module to monitor the thickness changes and local deformation of the blade during processing. The ultrasonic thickness measurement unit uses a high-frequency ultrasonic probe placed on the non-processed area on the back of the blade to measure the blade wall thickness by emitting ultrasonic waves, thereby acquiring the remaining wall thickness information after material removal in real time. The strain measurement unit uses foil strain gauges attached to designated locations on the blade to measure the strain distribution during processing, which is used to calculate the current local stiffness matrix of the blade. In addition, the working condition identification module of this embodiment uses a deep learning network instead of the support vector machine model in Embodiment 1. It uses a convolutional neural network to perform joint feature extraction and working condition classification on multi-channel sensor signals, which can more accurately identify complex working condition combinations, such as the combined working condition of tool wear and workpiece stiffness reduction.

[0037] Regarding the trajectory correction strategy, this embodiment introduces a variable stiffness compensation algorithm. Based on the real-time wall thickness data measured by the ultrasonic thickness measuring unit and the strain field information obtained by the strain measurement unit, the local stiffness of the blade at different machining positions is dynamically calculated, thereby predicting the elastic deflection amount at that position. The trajectory correction module performs reverse compensation on the machining trajectory based on the predicted deflection amount, ensuring that the actual cutting position falls precisely on the target design surface. Simultaneously, the temperature compensation strategy in this embodiment is more refined. It not only monitors the instantaneous temperature of the machining area but also establishes a three-dimensional temperature field model of the workpiece, calculating the thermal deformation compensation amount based on the temperature field distribution to achieve more accurate thermal error compensation.

[0038] During the manufacturing process, the system follows the basic workflow described in Example 1, but adds real-time wall thickness updates and stiffness matrix recalculation steps in each loop. When the system detects a significant change in blade wall thickness, it automatically updates the geometric parameters in the finite element analysis model, recalculates the local stiffness matrix, and adjusts the trajectory compensation strategy accordingly. In this way, this embodiment can effectively address common variable stiffness and severe thermal deformation problems in aero-engine blade manufacturing, ensuring that the blade profile machining accuracy meets design requirements. Example 3

[0039] This embodiment uses the milling of large structural components as an application scenario, and focuses on illustrating the system configuration and implementation method of the present invention under heavy-duty cutting and large-size workpiece machining conditions.

[0040] In the machining of large structural components such as aircraft fuselage beams and ship propeller impellers, the large workpiece size, high cutting forces, and long machining cycles make workpiece stiffness variations and machine tool thermal deformation issues particularly prominent. Furthermore, the machining of these products is typically performed on large CNC gantry milling machines, placing higher demands on system integration and real-time performance. This embodiment adapts the system architecture to this scenario.

[0041] In terms of system hardware configuration, the data acquisition module in this embodiment adopts a distributed sensor network architecture, with multiple data acquisition terminals distributed across different locations on the machine tool and connected to the central processing unit via industrial Ethernet. Cutting force measurement utilizes a wireless force gauge mounted on the tool holder, avoiding the inconvenience of extensive wiring; temperature monitoring employs an array-type infrared thermal imager to achieve comprehensive coverage of the temperature field in the machining area; vibration monitoring uses ICP-type accelerometers mounted on the crossbeam and column to monitor the dynamic response of the machine tool structure. Regarding data processing, this embodiment employs a multi-core processor parallel computing architecture, with tasks such as signal preprocessing, feature extraction, working condition identification, and trajectory correction distributed across different processor cores to ensure real-time performance is met even under complex computational loads.

[0042] In terms of working condition identification, the working condition classification unit in this embodiment adopts a feature selection method based on recursive feature elimination to filter out the feature subset most sensitive to the processing state from numerous sensor signal features, thereby improving the accuracy of working condition identification. Simultaneously, the system establishes a working condition evolution prediction model, predicting the trend of working condition changes based on historical processing data sequences, and adjusting the processing trajectory in advance to achieve feedforward control rather than simple feedback control.

[0043] Regarding trajectory correction, this embodiment employs a holistic optimization method for spatial curves rather than a point-by-point compensation method. After acquiring the current working condition deviation, the trajectory correction module replans the entire trajectory to be processed, seeking the globally optimal trajectory shape while satisfying constraints, ensuring both processing accuracy and efficiency. The corrected trajectory is transmitted to the CNC system in real time via a high-speed data channel. The CNC system uses a look-ahead interpolation algorithm to pre-read multiple segments of trajectory data, achieving smooth speed transitions and precise position control.

[0044] Through the above technical solution, this embodiment can meet the special requirements of heavy-duty precision machining of large structural components, continuously compensate for errors caused by changes in workpiece stiffness and machine tool thermal deformation during long-cycle machining, and ensure the overall machining accuracy of large and complex workpieces.

[0045] In summary, the adaptive machining trajectory planning method and system provided by this invention, by real-time acquisition of machining state parameters such as tool wear, cutting force, cutting temperature, and vibration signals, identifies the current working condition type and calculates the deviation from the standard working condition. Based on the deviation, the original machining trajectory is dynamically corrected to generate an adaptive machining trajectory adapted to the current working condition, and the CNC system is controlled to execute this adaptive trajectory to complete the part machining. The entire system forms a complete closed loop of perception-identification-decision-execution, effectively coping with factors such as tool wear, workpiece stiffness changes, machine tool thermal deformation, and clamping errors during machining, significantly improving machining accuracy and surface quality, while extending tool life and reducing scrap rate.

[0046] The technical solutions of the present invention are not limited to the specific embodiments described above. Any technical modifications made in accordance with the technical solutions of the present invention fall within the protection scope of the present invention.

Claims

1. A method for adaptive planning of machining trajectories, characterized in that, Includes the following steps: S1: During the machining process, machining status parameters that characterize the current working condition are collected in real time. The machining status parameters include at least one of tool wear, cutting force, cutting temperature, and vibration signal. S2: Based on the collected processing status parameters, identify the working condition type corresponding to the current processing status, and calculate the deviation between the working condition type and the preset standard working condition. S3: Based on the calculated deviation, the original machining trajectory is corrected online to generate an adaptive machining trajectory that matches the current working conditions; S4: Control the CNC system to execute the adaptive machining trajectory to complete the machining of the current machining area; S5: Repeat steps S1 to S4 until the entire part is machined.

2. The adaptive planning method for machining trajectory according to claim 1, characterized in that: In step S1, the tool wear is collected using an acoustic emission sensor installed inside the tool holder, the cutting force is collected using a piezoelectric force gauge installed inside the spindle, the cutting temperature is collected using an infrared temperature sensor and a thermocouple, and the vibration signal is collected using a triaxial acceleration sensor installed on the spindle housing and the workpiece fixture.

3. The adaptive planning method for machining trajectory according to claim 1, characterized in that: In step S2, the working condition type is identified using a working condition identification model based on support vector machine or deep learning network. This model can distinguish between normal wear working condition, severe wear working condition, workpiece stiffness reduction working condition, machine tool thermal deformation working condition, and clamping error working condition.

4. The adaptive planning method for machining trajectory according to claim 1, characterized in that: In step S2, the calculated deviation includes the tool radial compensation, axial compensation, temperature compensation, and clamping offset.

5. The adaptive planning method for machining trajectory according to claim 1, characterized in that: In step S3, the spatial compensation vector to be applied to each tool position is calculated based on the deviation, and the original trajectory is smoothed and corrected using a cubic spline interpolation algorithm.

6. A machining trajectory adaptive planning method according to any one of claims 1 to 5, characterized in that: Step S2 further includes performing digital filtering on the acquired raw signal to extract time-domain statistical features and frequency-domain spectral features, forming a feature vector.

7. A machining trajectory adaptive planning system for implementing the method of any one of claims 1 to 5, characterized in that, include: A data acquisition module is used to collect processing status parameters in real time; A working condition identification module is used to identify the working condition type corresponding to the current processing state and calculate the deviation. A trajectory correction module is used to correct the original machining trajectory online based on the deviation, and generate an adaptive machining trajectory; A CNC system interface module is used to transmit adaptive machining trajectory data to the CNC system; A control execution module is used to control the CNC system to execute adaptive machining trajectories.

8. The adaptive planning system for machining trajectory according to claim 7, characterized in that: The data acquisition module includes a tool wear monitoring unit, a cutting force sensing unit, a temperature detection unit, and a vibration monitoring unit, with each unit located around the machining area and on the machine tool.

9. The adaptive planning system for machining trajectory according to claim 7, characterized in that: The operating condition identification module includes a signal preprocessing unit, a feature extraction unit, and an operating condition classification unit, each of which runs in an industrial control computer.

10. The adaptive planning system for machining trajectory according to claim 7, characterized in that: The trajectory correction module includes a trajectory interpolation unit and a compensation calculation unit, and uses a cubic spline interpolation algorithm to generate adaptive processing trajectory data.