Discrete manufacturing-based order splitting method and system, and storage medium

Abstract

The invention relates to the technical field of manufacturing and order splitting, in particular to an order splitting method and system based on discrete manufacturing and a storage medium. The method comprises the following steps: obtaining task flow data of a manufacturing execution unit, collecting information of a machining part of the manufacturing unit, and executing order splitting processing of the machining part to generate order splitting data; evaluating the matching deviation condition based on the split data, detecting the dynamic fatigue accumulation condition of the cutter, and predicting the progressive imbalance trend of the function of the cutter; further measuring the cutter path deviation degree and the spindle vibration increasing condition, and calculating the part machining precision reduction degree; and finally, comprehensively analyzing the abnormal condition of the order splitting suitability, realizing the optimization management of the order splitting data, and outputting the optimized processing part order splitting data. According to the invention, the split order is optimized, so that the distribution of the manufacturing split order is more accurate.

Description

Technical Field

[0001] The present invention relates to the technical field of manufacturing order splitting, and in particular to an order splitting method, system and storage medium based on discrete manufacturing. Background Art

[0002] In the discrete manufacturing process, the task of a processing order usually needs to be split into multiple subtasks and assigned to different manufacturing units for parallel processing. This process is called order splitting. Traditional order splitting methods mostly rely on static process configuration and preset rules. They cannot dynamically respond to actual capacity matching deviations, equipment status degradation, and differentiated processing requirements of complex structural parts during the processing process, resulting in reduced production scheduling efficiency, increased fluctuations in component processing quality, and even large-scale manufacturing delays and rework problems. In the existing technology, some solutions attempt to standardize the management of manufacturing unit processing tasks through the integration of MES (Manufacturing Execution System) and ERP (Enterprise Resource Planning) systems. During long-term processing, gradual changes such as micro-crack growth, rigidity degradation, and stress fatigue on the tool surface are difficult to be perceived in a timely manner by traditional systems, resulting in processing path drift, abnormal cutting angles, and reduced processing accuracy. However, traditional discrete manufacturing order splitting has the problem of inaccurate prediction of the trend of gradual imbalance of processing tool functions, as well as inaccurate detection of the trend of gradual imbalance of processing tool functions. Summary of the Invention

[0003] Based on this, it is necessary to provide a method, system and storage medium for splitting orders based on discrete manufacturing to solve at least one of the above technical problems.

[0004] To achieve the above objectives, a method for splitting orders based on discrete manufacturing includes the following steps:

[0005] Step S1: Acquire task flow data of the manufacturing execution unit; collect manufacturing unit processing component information based on the manufacturing execution unit task flow data; perform processing component splitting processing based on the manufacturing unit processing component information to obtain processing component splitting data;

[0006] Step S2: evaluating the matching deviation of the processing component split orders based on the processing component split order data; detecting the dynamic fatigue accumulation of the processing tool based on the matching deviation of the processing component split orders; and predicting the progressive imbalance trend of the processing tool function based on the dynamic fatigue accumulation of the processing tool;

[0007] Step S3: determining the degree of tool machining trajectory deviation based on the progressive imbalance trend of the machining tool function; detecting the vibration growth of the machining spindle based on the progressive imbalance trend of the machining tool function and the degree of tool machining trajectory deviation; and calculating the degree of component machining accuracy degradation based on the vibration growth of the machining spindle;

[0008] Step S4: Analyze the abnormal adaptability of the processing order splitting based on the degree of decrease in component processing accuracy and the deviation of the processing component splitting matching; perform splitting optimization management on the processing component splitting data according to the abnormal adaptability of the processing order splitting to obtain the processing component splitting optimization data.

[0009] By acquiring and processing manufacturing execution unit task flow data, the present invention achieves standardized mapping of manufacturing tasks and process node structure analysis. This ensures that the physical matching relationship between the task flow and the manufacturing unit processing components has complete data source support, providing a logical foundation for subsequent processing component information collection and order splitting operations. By performing order splitting based on actual manufacturing unit component information, a high degree of coupling between the manufacturing entity structure characteristics and the task instruction structure is achieved. Based on the original task, an order splitting data structure consistent with the component morphology, clamping method, tool matching structure, and margin configuration is established, effectively ensuring the executability and process path consistency during the order splitting process. Matching deviation status assessment based on order splitting data enables the identification and attribution of differences between the processing path instructions and the actual processing component contour characteristics, thereby quantitatively identifying deviation phenomena such as clamping surface mismatch, interference area duplication, and tool path overlap caused by order splitting decisions. This deviation measurement is used as key basic data to reversely deduce the dynamic fatigue accumulation process of the processing tool, establishing a quantitative mapping mechanism between tool execution path load fluctuations, thermal stress disturbances, operating frequency changes, and structural micro-fission, thereby accurately identifying the tool functional state evolution process at the order splitting structure cascade feedback level. After obtaining the trend of progressive imbalance of tool function, the trend information is further applied in the trajectory control and spindle structure feedback layer. By accurately measuring the offset characteristics of the tool trajectory and structurally detecting the growth of spindle vibration, the chain path between the tool state change and the mechanical system response is opened up, and a dynamic chain of machining system response caused by tool function fluctuation is established. By quantitatively measuring the spindle vibration spectrum, displacement trend, and interference pattern, the identification and structural attribution of machining error sources are completed. Therefore, the present invention is an optimization process for the traditional discrete manufacturing order splitting, which solves the problem of inaccurate prediction of the progressive imbalance trend of machining tool function and inaccurate detection of the progressive imbalance trend of machining tool function in the traditional discrete manufacturing order splitting, and improves the accuracy of the prediction of the progressive imbalance trend of machining tool function and the accuracy of the detection of the progressive imbalance trend of machining tool function.

[0010] The present invention further provides a discrete manufacturing order splitting system for executing the discrete manufacturing order splitting method described above. The discrete manufacturing order splitting system includes:

[0011] The component splitting processing module is used to obtain the task flow data of the manufacturing execution unit; based on the task flow data of the manufacturing execution unit, it collects the manufacturing unit processing component information; based on the manufacturing unit processing component information, it performs the processing component splitting processing to obtain the processing component splitting data;

[0012] The module for assessing the progressive imbalance trend of machining tools is used to assess the matching deviation of machining parts according to the split-order data of machining parts; detect the dynamic fatigue accumulation of machining tools based on the matching deviation of machining parts; and predict the progressive imbalance trend of machining tool functions based on the dynamic fatigue accumulation of machining tools.

[0013] A module for calculating the degree of reduction in machining accuracy is used to determine the degree of deviation in the machining trajectory of the tool based on the trend of the progressive imbalance of the machining tool function; detect the growth of the vibration of the machining spindle based on the trend of the progressive imbalance of the machining tool function and the degree of deviation in the machining trajectory of the tool; and calculate the degree of reduction in the machining accuracy of the component based on the growth of the vibration of the machining spindle;

[0014] The splitting optimization management module is used to analyze the abnormal adaptability of the processing order splitting based on the degree of decline in component processing accuracy and the deviation of the processing component splitting matching; according to the abnormal adaptability of the processing order splitting, the processing component splitting data is optimized and managed to obtain the processing component splitting optimization data.

[0015] The discrete manufacturing order splitting system of the present invention can implement any discrete manufacturing order splitting method of the present invention, and is used to combine the operation and signal transmission medium between each module to complete the discrete manufacturing order splitting method. The internal modules of the system cooperate with each other, and through the collaborative analysis of the multi-dimensional data chain, the error sources in the processing part order splitting process are identified step by step and dynamically optimized, which effectively improves the adaptation accuracy between the order splitting results and the manufacturing unit capabilities.

[0016] A computer-readable storage medium stores a computer program, wherein the computer program is used to execute the discrete manufacturing order splitting method. BRIEF DESCRIPTION OF THE DRAWINGS

[0017] Figure 1 The figure is a flowchart of the steps of a split order method based on discrete manufacturing;

[0018] Figure 2 for Figure 1 Detailed implementation steps of step S3 in FIG.

[0019] Figure 3 for Figure 1 Detailed implementation steps of step S4 in FIG.

[0020] The purpose, features and advantages of the present invention will be further described with reference to the accompanying drawings and in conjunction with the embodiments. DETAILED DESCRIPTION

[0021] The following is a clear and complete description of the technical method of the present invention in conjunction with the accompanying drawings. It is obvious that the embodiments described are part of the embodiments of the present invention, but not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without making any creative efforts are within the scope of protection of the present invention.

[0022] In addition, the accompanying drawings are merely schematic illustrations of the present invention and are not necessarily drawn to scale. Identical reference numerals in the figures denote identical or similar parts, and thus repetitive descriptions thereof will be omitted. Some of the block diagrams shown in the accompanying drawings are functional entities that do not necessarily correspond to physically or logically separate entities. These functional entities may be implemented in software, in one or more hardware modules or integrated circuits, or in different network and / or processor and / or microcontroller approaches.

[0023] It should be understood that although the terms "first," "second," and the like may be used herein to describe various elements, these elements should not be limited by these terms. These terms are used solely to distinguish one element from another. For example, a first element may be referred to as a second element, and similarly, a second element may be referred to as a first element, without departing from the scope of the exemplary embodiments. The term "and / or" as used herein includes any and all combinations of one or more of the listed associated items.

[0024] To achieve this, please refer to Figures 1 to 3 , a method for splitting orders based on discrete manufacturing, comprising the following steps:

[0025] Step S1: Acquire task flow data of the manufacturing execution unit; collect manufacturing unit processing component information based on the manufacturing execution unit task flow data; perform processing component splitting processing based on the manufacturing unit processing component information to obtain processing component splitting data;

[0026] In an embodiment of the present invention, in a discrete manufacturing system environment, the task flow data within the current cycle manufacturing execution unit is obtained through the task scheduling control module integrated into the workshop MES (manufacturing execution system). The task flow data includes: task number, processing unit identification, task start and end time, specified process path, material tracking code, and corresponding component production process sequence information. This data is uploaded to the database of the central control server in real time by the on-site industrial network acquisition equipment, and is connected to the edge computing terminal through the PLC to ensure that the task flow data is stored in the order of timestamps. Based on the obtained manufacturing execution unit task flow data, the process flow interface is further called to obtain the identification of the components currently actually involved in the processing of each manufacturing unit. The component information is extracted by reading the clamping station QR code recognition module, including the component number, material code, process segment to which it belongs, process parameters (such as cutting depth, feed speed, etc.), processing tool number, and initial dimensional tolerance. The collected manufacturing unit processing component information is associated with the task flow data and analyzed to form a component and process segment correspondence table. Subsequently, the processing component splitting operation is performed based on the manufacturing unit processing component information. The order splitting process combines a static feature matching algorithm with a process-level time interval segmentation method to break down component tasks according to different process sections, different equipment capabilities, and processing rhythms. For example, if a part needs to undergo three processes: turning, milling, and heat treatment, its order splitting process will split the tasks based on the equipment distribution and process processing time intervals corresponding to the three sections, and form multiple independent subtasks. The result after processing is the processing component order splitting data, including the subtask number, split node information, processing section identification, required tool number, component base shape description, and corresponding equipment identification. This step outputs the processing component order splitting data.

[0027] Step S2: evaluating the matching deviation of the processing component split orders based on the processing component split order data; detecting the dynamic fatigue accumulation of the processing tool based on the matching deviation of the processing component split orders; and predicting the progressive imbalance trend of the processing tool function based on the dynamic fatigue accumulation of the processing tool;

[0028] In an embodiment of the present invention, the processing component splitting data generated in step S1 is used to evaluate the matching deviation between the current splitting task and the preset process splitting solution during the processing execution process according to the actual scheduling timeline set in the task scheduling system. The matching deviation evaluation takes the processing displacement trajectory of the tool corresponding to the task, the time difference between the start and completion of the task, and the difference between the actual feed rate and the calibrated rate as input to construct a matching deviation vector. Among them, the displacement trajectory is obtained by reading the real-time tool position data in the CNC system, the time difference is obtained by collecting and comparing the start and end timestamps of the processing task, and the feed rate is obtained by the feedback data from the output end of the CNC machine tool. After the matching deviation vector is generated, the vector is further combined with the historical task standard trajectory library for difference analysis to obtain the change in the abnormal tool load coefficient caused by each subtask during the execution process. The dynamic fatigue accumulation of the processing tool is obtained by difference calculation. The fatigue value is statistically calculated based on the frequency of the actual tool load exceeding the set threshold during the processing process, and is quantified in the form of a "processing fatigue factor" with a unit of N times / minute. Based on the fatigue accumulation value and the machining path complexity factor (determined by the tool path point density and direction change frequency), the tool's functional progressive imbalance trend is deduced. This deduction process uses the load evolution data of the same tool model under the same process path from the tool life database. The load growth rate of the current tool under this machining path is integrated linearly, the tool functional degradation factor is calculated, and the current tool's imbalance level (classified into five levels, L1 to L5) is output. Step S2 outputs the machining tool's functional progressive imbalance trend level information and the corresponding imbalance factor.

[0029] Step S3: determining the degree of tool machining trajectory deviation based on the progressive imbalance trend of the machining tool function; detecting the vibration growth of the machining spindle based on the progressive imbalance trend of the machining tool function and the degree of tool machining trajectory deviation; and calculating the degree of component machining accuracy degradation based on the vibration growth of the machining spindle;

[0030] In this embodiment of the present invention, based on the progressive imbalance trend of the machining tool function derived in step S2, real-time path trajectory data from the tool machining process is retrieved and vector difference analysis is performed against the standard CAD machining trajectory. The specific method is as follows: the coordinates (X, Y, Z) of each tool position recorded by the CNC control system during the machining task are read and compared point by point with the original tool position CAD path to calculate the trajectory offset vector. The trajectory offset values are then gridded and averaged by block according to the machining area to determine the overall trajectory offset in μm. Simultaneously, the vibration signal of the tool during the spindle machining process is collected using a triaxial accelerometer mounted on the spindle end. The acquisition frequency is set to 10 kHz, and the vibration data is processed using a fast Fourier transform (FFT) to extract the peak value of the main frequency and the number of corresponding subharmonics. The vibration intensity is normalized by the sum of the main frequency amplitude and the average harmonic amplitude to form the spindle vibration growth factor. Next, the degree of decline in component machining accuracy is assessed using the trajectory offset and the spindle vibration growth factor. The machining accuracy degradation assessment uses a dimensional deviation deduction mechanism. This involves measuring key dimensions of completed parts (using an online laser measurement system), calculating the difference between the measured and standard dimensions, analyzing the deviation distribution trends, and outputting the degree of machining accuracy degradation for the current task phase, expressed in μm and its standard deviation. The outputs of this step include the machining trajectory offset (μm), the spindle vibration growth factor, and the component machining accuracy degradation and deviation distribution.

[0031] Step S4: Analyze the abnormal adaptability of the processing order splitting based on the degree of decrease in component processing accuracy and the deviation of the processing component splitting matching; perform splitting optimization management on the processing component splitting data according to the abnormal adaptability of the processing order splitting to obtain the processing component splitting optimization data.

[0032] In an embodiment of the present invention, based on the component processing accuracy reduction value obtained in step S3 and the split order matching deviation vector in step S2, the overall split order adaptability anomaly of the processing order is evaluated. The adaptability anomaly analysis relies on the task-process path mapping model to establish a matching library between the task ID and the geometric shape, dimensional requirements and material processing behavior it should complete. The processing accuracy reduction data and the split order deviation data are then cross-analyzed. The specific processing flow includes: comparing the corresponding dimensional deviation value in each subtask with the upper limit of the allowable deviation. If it exceeds the upper limit, it is marked as a first-level adaptation anomaly; if it does not exceed the limit but the cumulative error of the split order deviation vector is greater than 20% of the equipment setting error margin, it is marked as a second-level adaptation anomaly; if there is an increasing trend of trajectory deviation or abnormal spindle vibration frequency in three consecutive subtasks, it is marked as a third-level adaptation anomaly. According to the above-mentioned abnormality level identification results, the original split order data is optimized and managed based on the processing path redistribution method. The operation method is: merge or split the tasks that were originally split according to the process sequence again according to the process load balancing priority; call the idle equipment load information through the task rearrangement module, rematch the processing unit resources, and adjust the original task allocation path; at the same time, regenerate the processing path code, and output the new splitting task data set as the processing part splitting optimization data, including the optimized task ID, processing segment path reconstruction data, matching error control parameters and task execution sequence correction table, providing the input basis for the subsequent production scheduling system loading and execution.

[0033] Preferably, step S1 includes the following steps:

[0034] Step S11: Acquire manufacturing execution unit task flow data;

[0035] In an embodiment of the present invention, in a discrete manufacturing scenario, a manufacturing execution unit generally includes hardware systems such as a CNC lathe, a vertical machining center, and a flexible manufacturing unit. To obtain the task flow data of the manufacturing execution unit, it is necessary to access the task issuing module in the manufacturing execution control platform (MES) and extract the task flow log file therefrom. The task flow data is composed of fields such as the manufacturing work order number, the operation step number, the corresponding manufacturing unit identification, the planned processing time, the processing sequence identification code, the material binding code, and the process path number. In specific operations, by deploying an industrial gateway based on an industrial Ethernet communication protocol (such as Modbus TCP / IP) or an OPC UA protocol, the task data interface of the MES system is connected to the manufacturing unit control system, and the real-time data in the task flow instruction cache area is continuously read with a sampling period of 5 seconds. All read data is transferred through a local cache server, and data is sorted using a double index rule of timestamp and task number. Cleaning operations are performed on task flow data, including removing records with missing fields and correcting inconsistent field formats (for example, converting time fields to the ISO 8601 standard format). The data is then written to the database using PostgreSQL storage to build a standardized manufacturing execution unit task flow dataset for subsequent analysis and processing.

[0036] Step S12: collecting manufacturing unit processing component information based on the manufacturing execution unit task flow data;

[0037] In an embodiment of the present invention, based on the task flow data obtained in step S11, it is necessary to parse the bound processing unit codes and operation step numbers one by one, and establish a one-to-one mapping relationship between the task flow instructions and the actual processing parts. By calling the embedded process data interface of the manufacturing unit, the part information associated with each processing task is obtained, including part number, blank type, original external dimensions, pre-processing shape, process requirements, clamping method and expected surface roughness. During the data acquisition process, an industrial PLC (such as Siemens S7-1500 series) is used to read the sensor and control register values connected to the manufacturing equipment control logic, and extract the clamp state changes, spindle working status and tool switching events. During the manufacturing execution process, the visual recognition module (deployment of industrial CCD camera + OpenCV image analysis framework) is used to identify the parts to be processed on the clamping device, and the consistency is checked with the material code in the task flow through the QR code identification (DPM code). After obtaining preliminary information, the processing component information is structured and parsed according to the component design drawings (standard STP file format) and the manufacturing process parameter file (G code or tool path definition file), and a processing component information record with fields such as component number, geometric parameters, material type, rough processing-finishing process sequence, etc. is generated to support subsequent internal structure analysis operations.

[0038] Step S13: dividing the internal structure of the processing component into the processing component information of the manufacturing unit to obtain the internal structure division data of the processing component;

[0039] In an embodiment of the present invention, the goal of dividing the internal structure of a processing component is to identify each geometric feature area in the component and to perform module division according to the processing sequence, tool contact path, cutting depth and process section. This step calls the three-dimensional modeling data (input in STP format) and uses a CAD / CAE integrated platform (such as PTC Creo Parametric or SiemensNX) to load the component model. In the modeling platform, the built-in feature recognition module (Feature Recognition) is used to automatically analyze structural elements such as holes, grooves, steps, surfaces, threads, and ribs. By setting the geometric recognition threshold (such as a small deep hole is defined as a circular hole with a diameter less than 10mm and a depth greater than 5mm) and the material removal strategy (based on the number of cutting layers in the Z-axis direction and the coincidence of the tool profile), each component is divided into several structural processing units. Further combined with the tool path segment of the G code in the process file, the processing path is mapped to each structural element, and the processing time, required tool number and clamping method corresponding to each structural unit are calculated. Use the rule engine to divide these structural units into different split sections, such as "rough machining of outer contour in area A", "fine hole machining in area B", "thread tapping in area C", etc., and use the fields such as each structural unit number, position index, structural type, processing technology number, processing tool category, processing time, etc. to form a structural division data table.

[0040] Step S14: performing splitting of the processing components based on the internal structure division data of the processing components to obtain splitting data of the processing components.

[0041] In this embodiment of the present invention, based on the structural partitioning data generated in step S13, a splitting rule engine is used to perform processing task decomposition. The splitting rule engine consists of a pre-set logic control module, and its classification criteria include: 1) whether the structure can be processed continuously with the same tool; 2) whether fixture replacement or repositioning is required; 3) whether there are thermal deformation-sensitive areas spanning multiple process stages; and 4) whether the spindle load corresponding to each process stage exceeds the rated threshold of the manufacturing unit. The splitting process is as follows: a structural dependency graph is established for all structural units based on the processing sequence and geometric position. Structural units with a process logic sequence relationship are connected with directed edges. Next, a graph partitioning operation is performed on the structural dependency graph based on tool type, number of clamping times, and processing heat distribution. A depth-first traversal is used during the graph partitioning process, marking the points on each path where clamping or tool switching is required as splitting breakpoints. After all breakpoints are marked, the structural units are grouped, with each group forming a processing subtask. Each subtask defines fields such as the set of processing structural units, estimated processing time, tool number, processing sequence number, and corresponding processing unit number. All subtask data is aggregated to form part-split data, exported in CSV format, and imported into the manufacturing planning system as the basis for issuing subtasks. This data will guide subsequent steps in detecting and analyzing split-split matching deviations, tool fatigue accumulation, and spindle vibration trends.

[0042] Preferably, step S13 includes the following steps:

[0043] Step S131: extracting material parameters of the processing component according to the processing component information of the manufacturing unit;

[0044] In an embodiment of the present invention, during the manufacturing execution process, the manufacturing unit processing component information is composed of manufacturing task flow data and processing component list data, wherein the component information field includes the component number, the work order number, the structure code, the material number, etc. During the implementation process, the original data on which the processing component material parameters are extracted is the material number in the component information field. The extraction operation is based on the manufacturing resource basic database for field mapping. The database presets a one-to-one correspondence table between all material codes and their physical property parameters (yield strength, elastic modulus, thermal conductivity, thermal expansion coefficient, etc.). During the implementation operation, the SQL statement is called to match and search the material number in each component information in the database, and the returned fields include yield strength (unit MPa), Brinell hardness HB, work hardening index n, etc. After the query is completed, the ETL tool is used to perform field-level data fusion with the original component information through the component number, and a component material parameter table with clear fields is output. Each row of data corresponds to a manufacturing component and contains its complete material physical property parameters.

[0045] Step S132: evaluating the processing difficulty coefficient of the material of the processing component when the yield strength of the material parameter of the processing component exceeds 355 MPa;

[0046] In an embodiment of the present invention, the evaluation process is based on the national metal material cutting processing standards, and uses the empirical data table in the "hardness-cutting resistance-processing difficulty" sequence as a reference to set the processing difficulty coefficient for different yield strength intervals. The processing difficulty coefficient is defined as a dimensionless coefficient that reflects the comprehensive reflection of the processing force, cutting heat and tool wear rate, and the value range is usually 0.7 to 1.2. In the evaluation operation, based on the range of the yield strength, the built-in static parameter mapping table is called to return the corresponding processing difficulty coefficient. For example, when the yield strength is between 355 and 420 MPa, the corresponding processing difficulty coefficient is set to 0.927; if the yield strength exceeds 550 MPa, the corresponding coefficient is set to 1.035. The processing result forms a three-field data table of "processing part number-yield strength-processing difficulty coefficient".

[0047] Step S133: collecting the three-dimensional structure of the manufacturing unit processing component based on the manufacturing unit processing component information;

[0048] In an embodiment of the present invention, the acquisition of three-dimensional structural information of the processed parts relies on the combined analysis of CAD files and actual manufacturing path files. During the implementation process, the corresponding 3D modeling file is located by parsing the structural coding field in the part information, and the file format is limited to the STEP format (.stp) of the ISO 10303 standard. In the processing scheduling system, these files are imported in batches using CreoParametric software, and a boundary representation (BoundaryRepresentation, abbreviated as B-Rep) model is generated for each part. In the B-Rep model, three-dimensional coordinates, topological relationships, and geometric body types (such as holes, bosses, grooves, surfaces, etc.) are extracted to generate a face-edge-vertex topological index structure. At the same time, the NC program file (G code) associated with the part is read through the tool path analysis module to identify the tool contact area and processing contour line, and to determine the processing direction and accessible area. In addition, an industrial visual recognition module is integrated, a CCD camera device is deployed at the key workstations of the production line, and a data channel is established with the visual server. By fitting the physical part's projection profile with the B-Rep model, the structural consistency between the CAD model and the physical part is assessed, and data offsets are corrected within a ±0.2mm acquisition error range. The acquired 3D structure is output in a structure format, including fields such as geometry type, bounding box parameters, Z-projection area, machining surface identifier, and material removal volume ratio.

[0049] Step S134: identifying spatial connectivity of the processing components according to the three-dimensional structure of the processing components of the manufacturing unit;

[0050] In an embodiment of the present invention, the spatial connectivity of the processing component is identified based on the B-Rep topological structure data generated in step S133. This operation is based on the boundary volume topology map, constructing a spatial graph structure, performing triangular mesh generation on all surfaces, using the Delaunay triangulation algorithm to divide each surface region, and constructing a matrix of normal vector angles between adjacent faces. Based on the angle threshold between triangulated meshes, it is determined whether the surface has a continuous processing path. An angle less than 5° is considered a continuous region. On this basis, a depth-first traversal algorithm is used to search for connected subgraphs in the topological map, grouping all subsets of structures that can be continuously processed into a spatially connected unit. Each connected region is defined as a "structure processing domain" and is numbered (e.g., C1, C2, etc.). Through connectivity analysis, it is possible to determine whether the structure has "deep cavities," "blind holes," "interlocking structures," or "hollow areas," and to mark areas that require special processing paths (such as undercut processing, staged clamping, etc.). The output spatial connectivity data includes fields such as structure number, connected domain number, processing continuity status, and whether multi-axis linkage is required, providing a basis for subsequent path evaluation.

[0051] Step S135: Evaluate component processing path requirement data based on the three-dimensional structure of the processing component of the manufacturing unit and the spatial connectivity of the processing component;

[0052] In an embodiment of the present invention, a processing path requirement assessment model is established for each component. The processing path requirement assessment is based on the general rules of the NC program, and a path complexity scoring system is constructed. The scoring criteria are composed of the following dimensions: 1) the number of connected domains; 2) the number of times the tool needs to be switched; 3) the angle deviation of each connected area; 4) the number of deep hole / groove structures. According to the required processing direction of each processing surface in the structure, the tool incident angle range is calculated, and it is determined whether it can be completed by the current manufacturing unit (if five-axis linkage or special tools are required, the path complexity score is increased). At the same time, the shortest length of the tool path, the number of required cutting times and the expected processing heat load are calculated for all structural domains to form a path requirement quantification matrix to output the "path requirement score" (value range 0 to 1) of each component, which is used to indicate the complexity of the processing path; at the same time, the processing sequence number and recommended incident direction of each processing structural domain are output.

[0053] Step S136: Based on the component processing path requirement data and the component material processing difficulty coefficient exceeding 0.927, the internal structure of the processing component is divided to obtain the internal structure division data of the processing component.

[0054] In this embodiment of the present invention, component records with a processing difficulty coefficient greater than 0.927 obtained in step S132 are merged with the path requirement score output in step S135. The processing logic is that when the processing difficulty coefficient of a component is higher than 0.927 and its path requirement score is higher than 0.65 (indicating a complex structure and high processing energy consumption), the structure partitioning module is entered. The structure partitioning module calls the B-Rep structure, structural processing domain partitioning, and path requirement matrix generated in the previous steps. The partitioning operation is performed according to the following two rules: 1) Structural domains that must be replaced, clamped, or repositioned in the processing sequence are individually divided into structural units; 2) Structural surfaces that require switching tool types or directions are aggregated and divided into structural units. The partitioning method uses graph theory partitioning, marking "processing breakpoints" in the spatial topology diagram and using the breakpoints as the basis for partitioning, forming multiple processing sub-unit structural domains. The parameters of each structural unit include its connected domain, tool incidence direction, estimated processing time, required cutting force, and clamping method, forming a complete processing structure partitioning table, which is output in CSV format and serves as the key input for the next step of order splitting.

[0055] Preferably, the evaluation of the matching deviation of the processing parts in step S2 includes:

[0056] Count the mismatch of component processing machinery capabilities based on the data of component splitting orders;

[0057] In an embodiment of the present invention, the processing parameters of each processing task and the equipment capability parameters of the docking manufacturing unit are extracted through the processing component splitting data set obtained from the manufacturing execution system MES. The processing parameters involved include the spindle speed, feed speed, tool type, clamping form and cooling method required for the process; the equipment capability parameters include the equipment spindle speed limit range, maximum cutting torque, tool position number of the tool magazine, automatic tool change speed, worktable travel range and load bearing capacity, etc. A linear comparison algorithm based on field precision matching is used to compare the process parameter requirements and equipment capability limits one by one in the database to form a mechanical capability matching relationship matrix. The capability overflow value is recorded in each comparison item. When the overflow value is positive, it means that the equipment capability is insufficient. By counting the ratio of the number of items with non-zero overflow values to the total number of comparison items in the matrix, the mechanical capability mismatch rate data is obtained for subsequent step calls.

[0058] Detect insufficient adaptation of processing equipment functions based on component processing machinery capability mismatch;

[0059] In an embodiment of the present invention, the mechanical capability mismatch rate data generated in the previous step is used as an input parameter, and further combined with the functional configuration description table recorded in the equipment function library (including the number of supported tools, positioning methods, fixture types and switching methods, spindle cooling functions, laser tool setting or visual recognition assisted positioning and other functional modules) for cross comparison. Functional set analysis is performed using conditional difference set cross logic to extract configuration items that are not included or mutually exclusive between the current task requirements and equipment functions, and mark them as missing functional adaptation. The weight of each missing function is calculated according to the process criticality level set by the manufacturing system, and summarized into an overall processing equipment functional adaptation deficiency score value. This value is used to reflect the degree to which the current equipment function meets the requirements of the order splitting task.

[0060] Assess the extent of excess coverage of processing equipment based on insufficient adaptation of processing equipment functions and mismatch of component processing machinery capabilities;

[0061] In an embodiment of the present invention, the mechanical capability mismatch rate obtained in step S21 and the functional adaptation deficiency score obtained in step S22 are used as dual evaluation inputs to enter the coverage range evaluation module. The evaluation module uses the equipment workspace model in three-dimensional space as a reference standard, and performs coverage space analysis in combination with the component geometric dimensions, outermost contour boundaries, and processing angle requirements in the splitting task. In the specific operation, a multi-boundary bounding box (Bounding Box) modeling method is adopted to analyze whether the workpiece boundary required by the task exceeds the effective range of the equipment worktable travel and the tool rotation path. If there is an overflow area, the proportion of the overflow area volume to the total processing volume of the task is defined as the coverage range overlimit degree index, and the overlimit score value is calculated in combination with the above-mentioned scoring weights.

[0062] Identify the collapse of the device's structural alignment capability based on the insufficient adaptation of the processing device's functions and the degree of excess coverage of the processing device;

[0063] In an embodiment of the present invention, based on the functional adaptation deficiency score and the coverage excess score, a joint analysis is conducted on whether the processing equipment can stably complete the alignment operation in actual operation. During the analysis, a three-dimensional registration coordinate system of the clamping system and the component positioning point is established, and the fixture installation structure is docked through the K coordinate set, and three-dimensional fitting is performed with the positioning hole structure required in the processing task. If there is a fitting error that exceeds the maximum error tolerance threshold of the equipment alignment system (such as ±0.2mm), it is marked as a structural alignment collapse point. At the same time, the risk points of deformation offset in the actual positioning process of the fixture are evaluated, and the finite element static analysis method is used to simulate the displacement of the fixture structure nodes in the clamping state. The displacement exceeds the predetermined critical value and is recorded as an alignment capability collapse index. This indicator constructs an alignment capability collapse status score based on the number of collapse points and the proportion of the total contact points of the coverage task.

[0064] Measure the abnormal clamping force of the device according to the disintegration of the device structure's alignment ability;

[0065] In an embodiment of the present invention, the alignment ability collapse status score value is combined with the real-time clamping data of the equipment clamping system to enter the clamping force analysis process. In this process, the real-time clamping data is derived from the feedback value of the clamping force sensor array (such as piezoelectric or strain gauge) collected by the equipment control system. The clamping pressure of each clamping claw node is sampled in time series and the load change curve is drawn. At the same time, the offset between the initial clamping force value and the actual force peak value is extracted to form a clamping abnormality strength index. If the force offset at the alignment collapse position exceeds 5%, it is recorded as a structural stress concentration point. The ratio of the number of statistical concentration points to the force abnormality area is the clamping force abnormality degree score value.

[0066] Calculate the device clamping posture drift parameters based on the abnormal clamping force and the collapse of the device structure alignment ability;

[0067] In an embodiment of the present invention, the clamping force abnormality score and the alignment ability collapse score are input together into the posture offset analysis module. This module constructs a multi-rigid body dynamics simulation system based on the geometric parameters of the fixture (such as the number of jaws, opening angle, jaw arm length, joint stiffness, etc.), applies the corresponding clamping load and torque in an offline state, and simulates the change in deflection angle during the clamping process. The Euler angle difference method is used to measure the rotational offset angle of the clamping fixture between the initial set posture and the clamping completion state to obtain the clamping posture drift parameter, and the posture offset is expressed in angular units (°). This parameter serves as an important basis for evaluating the stability of the instrument clamping.

[0068] The matching deviation of the processed parts is evaluated based on the instrument clamping posture drift parameters and the abnormal force of the instrument clamping.

[0069] In an embodiment of the present invention, the posture drift parameters and the clamping force abnormality score are comprehensively introduced into the deviation assessment module. The module sets multiple combined threshold intervals to determine the deviation level of the processed parts caused by unstable instrument posture or uneven clamping during the splitting and matching process. A weighted fusion function is used to perform numerical normalization and weighted summation on the posture drift angle and the clamping abnormality strength, and the processing part splitting and matching deviation score is output. The higher the score, the worse the physical adaptability between the instrument and the splitting task, and the larger the deviation, the more the score is output for use in subsequent process adjustments and equipment scheduling strategy optimization of the processing flow.

[0070] Preferably, the detection of dynamic fatigue accumulation of the machining tool in step S2 includes:

[0071] Detect the excessive cutting force of machining tools based on the deviation of the matching of machining parts;

[0072] In an embodiment of the present invention, based on the processing component splitting matching deviation score value obtained in the previous step, the score value is matched with the rated threshold value of the tool cutting force. The tool cutting force is collected in real time by the spindle power sensor and the cutting force three-component sensor (Fx, Fy, Fz), with a sampling frequency of 1000Hz, and the force signal is subjected to noise reduction processing using an average filtering algorithm with a sliding time window of 0.5 seconds. The system reads the three-dimensional force vector recorded by the sensor in each cutting task, and combines it with the maximum allowable cutting force data provided by the tool manufacturer to determine whether there is an event in which the cutting force instantaneously exceeds the rated upper limit through the vector modulus calculation method. The number of times the cutting force exceeds the limit for each processing tool in a splitting task is counted, and the ratio of this number to the total time length of the processing section is used to form a cutting force exceedance rate, which is used as input data for subsequent judgment of the growth of tool mechanical impact.

[0073] Detect the growth degree of mechanical impact of tool according to the excessive cutting force of tool in processing equipment;

[0074] In an embodiment of the present invention, the cutting force excess rate data generated in the previous step is imported into the impact detection module, and the impact identification standard within the cutting cycle is established in the module. The acceleration mutation point per unit time is detected by the first-order derivative calculation method of the cutting force three-component curve, and the position and intensity of the instantaneous impact are determined. The derivative value at the moment of each force mutation is subtracted from the average value of the derivative of the stable cutting section of the previous cycle to obtain the impact amplitude; at the same time, the frequency of occurrence of impact events per unit time is counted to form an impact frequency parameter. The impact growth degree index is calculated by multiplying the impact amplitude and the impact frequency. This index is used to reflect the frequency and intensity of non-steady-state loading to which the tool is subjected during the actual machining process, and is used for load step analysis in subsequent steps.

[0075] Determine the step load growth condition of the machining tool according to the growth degree of the tool mechanical impact and the excessive cutting force of the machining tool;

[0076] In an embodiment of the present invention, a load superposition curve is constructed based on the cutting force excess rate of step S21 and the impact growth degree index of step S22. The load superposition curve is drawn with the total cutting load (in Newtons) of each processing time period as the vertical axis and the processing time as the horizontal axis. The piecewise linear regression analysis method is applied to determine the sudden increase breakpoints in the curve, and the load slope change multiple is recorded. If the load slope increases by more than twice before and after a certain time period, the segment is determined to be a load step segment. The ratio of the length and number of all step segments to the total processing time is weighted to form a load step score value, which is used as a measurement basis for the evolution of the nonlinear load of the processing tool and is provided to the subsequent stress assessment module for use.

[0077] Calculate the stress growth degree of the machining tool based on the step-type load growth condition of the machining tool;

[0078] In an embodiment of the present invention, the obtained load step score is used in combination with the tool structure data (including tool body length, spindle interface diameter, material elastic modulus, and effective force area of the tool tip) to establish a static force model of the tool. Static stress simulation calculation is performed using the finite element uniaxial compression-shear coupling method, and the actual force data under different load step conditions is applied to simulate the internal stress distribution of the tool under the combined force of spindle rotation and cutting feed. The maximum equivalent stress point stress value is selected from the simulation output results, and the stress growth rate is calculated by ratio with the yield limit of the tool material. The stress growth rate represents the degree of nonlinear growth of structural stress caused by load fluctuations during the processing process, which directly affects the subsequent development trend of microcracks.

[0079] Monitor the growth trend of micro cracks in machining tools based on the stress growth degree and step load growth condition of machining tools;

[0080] In an embodiment of the present invention, the stress growth rate calculated in loading step S24 and the load step score value in step S23 are used to detect the development path of the tool microcracks. The detection tool is a high-frequency vibration detection sensor installed between the tool handle and the spindle. The signal acquisition frequency is 10kHz. Fourier transform analysis is performed in combination with the vibration spectrum of the tool at the moment of cutting in and out in the processing section. If the stability decreases, the spectrum line width increases, and the harmonic peak is enhanced in the high-frequency band, it can be determined that there is crack propagation behavior on the tool surface. By performing time fitting on the crack propagation velocity, a microcrack growth trend curve is obtained, and the crack length growth rate is output as a trend parameter to provide a basic judgment basis for the subsequent roughness evolution analysis.

[0081] Identify the growth trend of machining tool surface roughness based on the growth trend of machining tool microcracks;

[0082] In an embodiment of the present invention, after the microcrack growth trend parameters are determined, the surface roughness data collected by the surface profiler of the corresponding processing surface is read, and the time correlation coefficient between the roughness parameters of the tool processing surface and the crack growth rate is calculated using surface profile statistical standards (such as Ra, Rz, Rp, etc.). If the correlation coefficient is higher than 0.8, it is determined that the microcrack propagation behavior significantly affects the stability of the tool cutting trajectory, thereby causing a trend of surface roughness degradation. Using the Ra mean growth rate as the roughness growth trend indicator, a sequence comparison is performed between processing batches to construct a tool life degradation trend sequence. This trend data serves as the basis for estimating micro-deformation, and further extends the judgment of the change path of tool structure stability.

[0083] Determine the degree of micro deformation accumulation of machining tools based on the growth trend of machining tool surface roughness and the growth trend of machining tool micro cracks;

[0084] In an embodiment of the present invention, a tool micro-deformation accumulation model is constructed by combining the roughness growth trend parameter of step S26 and the crack propagation rate of step S25. With the tool end as the coordinate reference and compared with the initial geometric dimensions, a laser displacement sensor is used to perform a multi-point scan of the tool end point after processing is completed to detect the offset between it and the theoretical geometric contour. The maximum displacement value in the multi-point scan is recorded as the micro-deformation index value, and a three-dimensional surface fitting relationship is established with the crack growth rate and the roughness growth rate. The deformation path corresponding to the maximum offset coordinate point in the fitting surface is extracted and quantified as a tool micro-deformation accumulation degree index. This index reflects the microscopic deformation stability trend of the tool after long-term stress.

[0085] The dynamic fatigue accumulation of the machining tool is detected based on the cumulative degree of micro-deformation of the machining tool and the growth trend of the surface roughness of the machining tool.

[0086] In an embodiment of the present invention, a micro-deformation accumulation index and a surface roughness growth trend parameter are used as input to construct an integral function of fatigue loss of a machining tool. This function forms a cumulative fatigue loss value by superimposing the area integral of the micro-deformation on the time axis with the integral of the slope of the roughness curve. Fatigue tolerance thresholds are set for different types of tools (such as carbide end mills, indexable inserts, drill bits, etc.), and the current tool fatigue state is determined to be low, medium, or high based on the relative relationship between the cumulative value and the threshold. The fatigue state data is output to the tool management system for tool replacement scheduling or maintenance reminders, completing the full process detection closed loop of the dynamic fatigue accumulation of the machining tool.

[0087] Preferably, the prediction of the progressive imbalance trend of the machining tool function in step S2 includes:

[0088] Detect the degree of hardness reduction of the surface layer of the machining tool material based on the dynamic fatigue accumulation of the machining tool;

[0089] In one embodiment of the present invention, after identifying the dynamic fatigue accumulation of a machining tool, the accumulated fatigue data is mapped to a machining history database using the tool number as an index. Key parameters corresponding to the tool number, such as the machining batch, material properties, cutting path, cutting time, and spindle speed, are extracted. This data is then input into a machining tool surface condition detection device, which consists of a laser rebound wave velocity measurement module, a microtexture comparison and analysis module, and a thermal decay response scanning module. The laser rebound wave velocity measurement module emits a short pulse laser beam at a frequency of 10 MHz, measuring changes in the material's surface hardness based on the rebound time difference. The microtexture comparison and analysis module compares and analyzes the tool surface's original manufacturing state with its current state. If the per-unit area roughness peak-to-valley deviation increases by more than 0.5 μm, it is considered a decrease in surface material hardness. The thermal decay response scanning module applies a standard thermal load and monitors the local thermal diffusion rate on the tool surface. If the diffusion coefficient decreases by more than 8%, it is also flagged as indirect evidence of hardness degradation. The three test data are normalized and scored to form a determination of the degree of tool surface hardness degradation, which is output as a percentage to downstream processes.

[0090] Detect the rigidity attenuation of the machining tool when the surface hardness of the machining tool material decreases by more than 10%;

[0091] In an embodiment of the present invention, when the hardness drop of the surface layer of the tool material exceeds a threshold of 10%, the system calls the historical rigidity measurement record associated with the tool number and starts the machining tool rigidity detection system for real-time detection. The system consists of a laser interferometer rigidity measurement device, a stress-rebound response comparison module, and a precision load response stand. When the tool is loaded with three fixed loads of 10N, 20N, and 50N, the laser interferometer measures the unit deformation of the top of the tool and outputs a rigidity curve; the stress-rebound response module detects the residual stress ratio within the unit loading-unloading cycle. If the residual deformation accounts for more than 3% of the total deformation, it is marked as rigidity attenuation; combined with the historical curve, the percentage change between the current rigidity and the factory reference rigidity is calculated. If the attenuation exceeds 10%, the cutting force fluctuation trend analysis step is triggered.

[0092] Analyze the growth trend of cutting force fluctuation of machining tools when the rigidity attenuation of machining tools drops by more than 10% and the degree of decrease in the surface hardness of machining tool materials;

[0093] In an embodiment of the present invention, the percentage decrease data of the current tool rigidity and material hardness are obtained from the first two steps, and the cutting force fluctuation growth trend analysis is performed in combination with the force signal curve collected in real time by the cutting force sensor in the current processing task. The measuring device used is a three-dimensional torque sensor device, which is arranged between the spindle tool holder and the machine tool worktable to collect transverse, longitudinal and normal three-dimensional cutting forces. The frequency distribution and peak number of the current cutting force fluctuation curve are compared with the historical reference curve. If the number of peaks in the low frequency band (0-10Hz) increases by more than 15%, and the mean fluctuation amplitude in the high frequency band (100-500Hz) increases by more than 8%, it is determined that the cutting force fluctuation trend is increasing. The relevant data is normalized after Fourier transformation, and the percentage value of the fluctuation growth amplitude is output as a prediction indicator.

[0094] Predict the intermittent chipping of machining tools based on the fluctuation growth trend of cutting force of machining tools;

[0095] In an embodiment of the present invention, when it is detected that the amplitude of the cutting force fluctuation growth trend exceeds the 12% threshold, the system accesses the chipping particle image monitoring channel in the processing task. The particle collection system is used to separate the particles through the coolant in the cutting area, and then a scanning electron microscope imaging device (magnification 20,000 times) is used to collect the microscopic morphology of the chipping particles, and the laser particle size distribution analyzer is used to extract parameters such as particle size, edge sharpness, and crack initiation angle. If the number of chipping particles surges per unit time and the average particle size exceeds 3 times the grain size of the processed material, it is judged to be an intermittent chipping state; in addition, a statistical analysis is performed on the end face scanning profiles of multiple batches of processed products. If the fluctuation frequency of the roughness Ra value is more than 5 times / minute, it can also prove the existence of intermittent chipping. The system combines and encodes the above two types of data and outputs a chipping probability judgment value.

[0096] Detect the torque coupling force distortion of machining tools based on the intermittent chipping of machining tools;

[0097] In an embodiment of the present invention, after the intermittent chipping state is confirmed, the torque data of the tool spindle drive unit bound to the processing task is automatically called. The spindle rotation torque is collected in real time by a high-precision dynamic torque sensor (sampling frequency 2000Hz), and the original torque curve is extracted within a 5-second time window before and after the tool chipping. The torque change rate is analyzed by first-order difference processing, and coupled with the processing path direction for analysis. If the angle between the main component of the torque and the feed direction is between 90°±15°, and abnormal fluctuations of more than 20% still occur, it is judged to be a torque coupling force distortion phenomenon. The distortion frequency and distortion amplitude are output, and the distortion density index is calculated as a reference for subsequent overload judgment.

[0098] Determine the degree of tool body overload based on the torque coupling force distortion of the tool and the intermittent chipping of the tool;

[0099] In an embodiment of the present invention, a tool body load distribution surface diagram is constructed by combining the distortion density index output from the fifth step with the intermittent chipping probability from the fourth step. The distribution of the machining axial force and radial force at different points on the tool surface is used to calculate the maximum deviation of the resultant force offset distance from the standard center axis. If the offset distance exceeds 20% of the tool edge radius, the presence of overload is confirmed. The overload frequency is further calculated based on the ratio of the overload occurrence period to the total machining time in each machining cycle, and the overload frequency is combined with the offset amplitude to output the overload degree value to distinguish between mild, moderate or severe overload categories.

[0100] The progressive imbalance trend of machining tool function is predicted based on the torque coupling force distortion of machining tool and the degree of tool body overload.

[0101] In an embodiment of the present invention, based on the classification of the degree of off-load obtained in the sixth step and the value of the distortion density in the fifth step, the distortion frequency is greater than 3 times per minute and the degree of off-load is moderate or above as the starting threshold for functional imbalance prediction in the prediction logic. The cumulative load imbalance index (calculated by the ratio of the cumulative off-load force to the total number of processing times) is used as the leading variable. Combined with the current abnormal quality records of processed products (such as dimensional deviation exceeding the upper tolerance limit or increased surface defect density), the functional progressive imbalance degree output is set to a risk level score of 0-100 to generate a processing tool functional progressive imbalance trend assessment result, and is marked as high, medium, and low risk levels for controlling the system to schedule tool replacement time and optimize the order splitting strategy.

[0102] Preferably, step S3 includes:

[0103] Step S31: determining the degree of tool machining trajectory deviation according to the progressive imbalance trend of the machining tool function;

[0104] In an embodiment of the present invention, the output risk level score of the progressive imbalance trend of the machining tool function is used as an input parameter, and the actual machining trajectory of the corresponding tool in the specific machining task is retrieved from the machining task scheduling system. The machining trajectory is recorded by a high-precision trajectory recording module in the embedded five-axis linkage machining center. The module collects the coordinate data of the machining path points in real time through photoelectric position sensors and encoders, collecting data every 1ms, covering the entire machining process. The actual trajectory data is aligned with the theoretical trajectory set in the machining task setting and the difference analysis is performed. For each machining section, the offset value in the X, Y, and Z directions is calculated based on the tool end trajectory coordinates. The Euclidean distance formula is used to obtain the machining offset at each moment. Subsequently, when the functional imbalance risk level exceeds the threshold of 70, a local weighted analysis is performed on the offset value to calculate the maximum trajectory offset degree during the machining process. The offset value data is output with millimeter-level accuracy and presented in the form of a thermal distribution map on the machining console, indicating the offset area of each segment of the trajectory.

[0105] Step S32: calculating the machining tool cutting angle difference data according to the tool machining trajectory deviation degree and the progressive imbalance trend of the machining tool function;

[0106] In an embodiment of the present invention, the trajectory offset value data and the functional imbalance trend risk level generated in step S31 are used to start the angle difference calculation module. The module uses the initial angle of the tool entering the workpiece as the reference angle, and combines the trajectory offset direction and the angle variation to perform spatial posture inversion analysis. The specific method is: in a three-dimensional coordinate system, a normal vector is constructed with the main direction of the tool cutting edge, and then the angle between the two vectors is calculated with the direction vector formed by the offset trajectory in a unit time period. The angle is the entry angle difference. In this process, a posture capture unit installed on the spindle head is used. The unit uses a gyro inertial navigation unit and a three-axis acceleration sensor to output posture change data in real time. At the same time, the functional imbalance risk level parameter is called for risk weighting and the angle change range is corrected. If the continuous angle difference growth rate is detected to exceed 5°, the segment is recorded as an abnormal entry area. The angle difference data of each processing segment is identified and stored with the processing task number. The output format includes a timestamp, a position index and an angle difference value.

[0107] Step S33: estimating the unstable contact of the machining tool according to the machining tool cutting angle difference data;

[0108] In one embodiment of the present invention, based on the tool engagement angle difference data acquired in the previous step, this difference data is input as the primary input parameter into the cutting path data sequence recorded by the tool attitude sensor. A frame-by-frame comparative analysis of the incident contact points between the tool and the workpiece surface is performed. By setting a three-dimensional reference frame within a multi-axis machining coordinate system, the ideal engagement angle and the current actual engagement angle are triangulated to obtain the difference between the tangential and normal forces. Based on this, a three-dimensional vector angle error analysis method is introduced to integrate the perturbation of the contact surface. If the tangential force fluctuation caused by the engagement angle difference exceeds a critical offset of 2.8 N / mm, contact instability is determined in that section. A laser displacement sensor and a high-frequency strain gauge are used to collect the contact point drift frequency per unit time between the tool and the workpiece as a verification method. Contact point stability is then reconfirmed. All periods with a contact point perturbation frequency exceeding 5 times / s are marked as contact unstable regions, and a data sequence of contact instability is output as input for subsequent steps.

[0109] Step S34: detecting the degree of friction growth of the machining tool based on the unstable contact condition of the machining tool;

[0110] In an embodiment of the present invention, the contact instability data is used as the basic information source, and the infrared thermal imaging monitoring module deployed in the processing area monitors the heat energy release trend of the tool cutting end and the workpiece surface in real time. According to the rate of change of the thermal radiation intensity, combined with the number of contacts per unit time measured in the previous sequence, the friction heat growth coefficient per unit time is calculated. The degree of friction growth is expressed in the form of the friction power growth rate. The cutting speed, the normal force change value and the heat flux per unit area parameter are introduced into the formula. The friction power curve is inverted by the area difference of the infrared spectrum grayscale curve. Through this method, the degree of friction growth is quantified as percentage change data. Parallel comparison and verification are carried out on multiple groups of tool samples. When the friction growth rate of a certain section exceeds 15% and the frequency of contact point disturbances shows an increasing trend, the friction growth degree of this time period is marked as a high-risk area, which is used as input data for the assessment of the aggravation of tool wear.

[0111] Step S35: detecting the vibration growth of the machining spindle based on the degree of friction growth of the machining tool and the unstable contact of the machining tool;

[0112] In an embodiment of the present invention, the friction growth degree data and contact instability state index data obtained in step S34 are received and input into the machining spindle vibration monitoring module. The three-axis acceleration sensor embedded in the module is arranged on the spindle bearing seat housing and the machining platform bracket to collect the vibration acceleration data during spindle machining, and the spindle vibration power spectrum density is calculated using the time-frequency domain analysis method in combination with the tool overload degree and the friction force change trend. In particular, for the acceleration response in the low-frequency (30-60Hz) and medium-frequency (60-200Hz) frequency bands, the vibration peak characteristics that are highly coupled with the abnormal contact behavior of the tool are extracted. If periodic vibration peak growth occurs and its vibration power amplitude exceeds 1.5 times the baseline value, it is recorded as the occurrence of spindle vibration growth, and the root mean square acceleration value (mm / s 2 ) and vibration peak frequency (Hz) are output as data and sent to the next step of machining accuracy evaluation calculation process.

[0113] Step S36: Calculate the degree of decrease in component machining accuracy based on the vibration growth of the machining spindle and the difference in the cutting angle of the machining tool.

[0114] In this embodiment of the present invention, the component machining accuracy change calculation process begins based on the spindle vibration growth index output in step S33 and the cut-in angle difference data for each machining segment in step S32. This process relies on an offline inspection platform to perform three-dimensional contour scanning and accuracy measurement on the finished component. The equipment used is a combined system of a blue-light 3D scanner and a precision contact cylindrical probe to acquire point cloud data of the machined surface contour. The contact probe is used to accurately measure key geometric features such as aperture, end face perpendicularity, and axis concentricity. Standard deviation analysis is performed on the measurement errors of each key dimensional point, and the deviation ratio relative to the process benchmark is calculated. The spindle vibration growth index and cut-in angle difference for each segment are correlated with the dimensional error of the corresponding machining segment. Using a deviation attribution mapping matrix, the proportional coefficients of the impact of vibration and angle changes on accuracy are deduced. The accuracy reduction per unit vibration increase or angle difference is calculated, and the overall component machining accuracy reduction data is summarized. The average accuracy error growth rate is output in microns, and a machining quality heat map is generated based on the specific machining trajectory segment.

[0115] It is particularly important that step S33 includes the following steps:

[0116] Step S331: predicting the aggravation of machining tool wear based on the degree of machining tool friction growth and the unstable contact of the machining tool;

[0117] In an embodiment of the present invention, the edge profile of the cutting edge line of the machining tool is digitally reconstructed by establishing a differential analysis chain of the wear profile image at multiple time points, taking the friction growth rate and contact instability data as input conditions. A three-dimensional profile scanning device is used to collect the edge morphology of the initial and target time points, and the morphology matching method is used to calculate the wear width of a single tool edge line at 0.1mm intervals. The wear process is fitted using a time series linear interpolation algorithm, and the degree of superlinear deviation of the actual wear rate is analyzed under the premise that the friction growth rate is a constant. When the wear growth rate exceeds the reference rate (for example, 0.02mm / min) by more than 30%, the tool state is defined as a wear aggravation state. The wear aggravation trend map and the corresponding timestamp are output as the input basis for the clearance growth trend detection.

[0118] Step S332: detecting the growth trend of the machining tool clearance according to the aggravation of the machining tool wear;

[0119] In an embodiment of the present invention, based on the wear profile image and its aggravation trend data, the real-time position offset data of the tool in the working state is obtained through the micro-gap monitoring device at the connection between the tool handle and the spindle (using laser interferometry technology). 0.01mm is set as the micro-gap reference critical value, the tool handle swing amplitude under different load conditions is compared, and the spindle rotation angle synchronization error comparison data is introduced to analyze the degree of deviation of the tool rotation center. The circular runout value of the tool off the center is sampled with a resolution of 0.001mm, and the gap growth trend under the time series is calculated. If the gap growth amplitude exceeds 0.05mm in 30 minutes of continuous sampling and there is an overlapping interval with the wear aggravation time node, the gap growth trend data sequence is output as the basis for subsequent cutting vibration analysis.

[0120] Step S333: estimating the degree of increase in tool cutting vibration based on the growth trend of tool clearance and tool friction;

[0121] In an embodiment of the present invention, the tool clearance growth trend and the friction growth degree are used as coupling inputs, and the machine tool's built-in vibration accelerometer is used to collect three-axis vibration acceleration data in real time at the connection position between the spindle and the tool holder. The vibration spectrum is deconstructed by Fourier transform, and the main frequency change data in the range of 10Hz to 5000Hz is extracted. The clearance change is used as the input parameter to modulate the frequency drift, and the friction growth degree is introduced as an amplitude adjustment factor to generate a theoretical vibration growth curve, which is then fitted with the actual collected data. If the fitting deviation is less than 3%, the vibration growth trend is confirmed to be credible, and a tool cutting vibration growth degree data table is formed, which includes three core parameters: the main vibration frequency, the peak acceleration, and the rate of change.

[0122] Step S334: identifying the tool cutting vibration transmission condition based on the tool cutting vibration growth degree;

[0123] In an embodiment of the present invention, a vibration propagation path mapping relationship is constructed based on the vibration growth data and the spindle structural parameters. Multiple groups of MEMS acceleration sensors are arranged in the middle of the spindle, the bearing interface and the workbench structural parts to synchronously collect the vibration responses of different structural levels during the processing. The phase difference and amplitude attenuation of the vibration signals at different monitoring points are calculated by the waveform phase delay analysis method, and then the main direction of the vibration conduction path is identified. If the vibration signal intensity of the upper section of the spindle is highly matched with the waveform frequency of the cutting point, and the phase difference is less than 0.02 seconds, it is confirmed that the vibration has been effectively transmitted from the tool to the spindle body. The output includes vibration conduction identification data of the conduction path, phase difference, and energy attenuation ratio.

[0124] Step S335: Detecting the vibration growth of the machining spindle according to the tool cutting vibration transmission condition and the tool cutting vibration growth degree.

[0125] In an embodiment of the present invention, based on the existing vibration conduction data and the degree of vibration growth, the total vibration energy at the initial stage of the process and the current stage is compared by the incremental analysis method of the total amount of spindle vibration (the energy index is calculated by integrating the square of the acceleration). Combined with the overlap between the natural frequency of the spindle structure and the cutting frequency, it is evaluated whether the spindle has entered the critical resonance range. When the spindle vibration amplitude is higher than the reference vibration value by more than 20% for 10 consecutive minutes, and the frequency deviation rate reaches more than 1.5%, it is confirmed that the spindle vibration state has entered the growth range and a spindle vibration growth report is output. The content covers the average vibration amplitude of the spindle, the frequency change trend and the duration of the vibration, which is used to drive the downstream precision attenuation analysis.

[0126] It is particularly important that step S34 includes the following steps:

[0127] Step S341: Calculating the change of the axial force component according to the cutting angle difference data of the machining tool;

[0128] In an embodiment of the present invention, based on the cutting tool angle difference data measured in the previous step, the static decomposition method is used to calculate the change of the axial force component in the three-axis coordinate system. The real-time offset of the cutting incident angle is recorded synchronously by the laser displacement interferometer and the tool posture recording device, and the cutting angle error Δθ is calculated by the angle between the tool centerline and the normal line of the workpiece surface. Based on this error angle, the cutting resultant force Fcut is decomposed into two components: the normal force Fmethod and the axial force Faxis using the force vector decomposition formula, where Faxis = Fcut*sin(Δθ). The load acquisition module in the CNC machining system reads the real-time power output of the spindle and obtains the cutting resultant force through numerical integration. Based on this value, the axial force component is calculated in real time. The axial force change value is archived in time series and a change curve is drawn. At the same time, 1N is set as the unit change threshold. If the axial force change range exceeds 3N within 3 consecutive minutes, it is recorded as the axial force component fluctuation area and the axial force change data is output to judge the stability of the structural force during the machining process.

[0129] Step S342: detecting the instability of the component processing force based on the change of the axial force component;

[0130] In an embodiment of the present invention, the axial force change data obtained in step S341 is used as input, combined with the parameters of the force conduction structure of the processing fixture, and a calculation and judgment is performed through the mechanical stability index. The workpiece is installed on a high-precision force-sensitive platform with a six-point distribution. The platform has a piezoelectric ceramic piece embedded in it, which can sense slight changes in external forces in real time. The axial force change value recorded per second during the processing is analyzed by ratio with the fixture stiffness K fixture, and the force fluctuation degree under unit stiffness is defined as P fluctuation = ΔF axis / K fixture. If the fluctuation value continuously exceeds the threshold value of 0.35 within a one-minute window, it is judged to be a force unstable area. The displacement response of the fixture surface during the unstable period is synchronously collected, and the spatial force field is mapped through a distributed strain fiber Bragg grating sensing network to output a three-dimensional force disturbance distribution map and the corresponding force fluctuation data.

[0131] Step S343: determining the component processing subtle displacement disturbance based on the component processing force instability;

[0132] In an embodiment of the present invention, in the area where unstable force is detected in step S342, a laser micro-interference displacement measurement system is arranged on the surface of the workpiece, and a laser triangulation ranging module is used to perform non-contact real-time tracking of the tiny displacement of the workpiece processing surface. The sampling accuracy is set to 0.01μm, and the maximum displacement amplitude and its change frequency in each contact cycle are recorded. The system synchronously transmits the displacement data obtained from the multi-point laser measurement device to the central processing unit, and establishes a processing displacement disturbance time history curve through the time domain reconstruction method. The dynamic frequency response analysis method is introduced to calculate the interference frequency distribution density and extract the main disturbance frequency band. The disturbance data is compared with the force fluctuation data to confirm whether there is a frequency resonance phenomenon in the same time window, so as to verify whether the disturbance is caused by unstable force. The displacement disturbance amplitude curve and disturbance frequency band statistical data are output to provide quantitative indicators for subsequent precision degradation calculations.

[0133] Step S344: detecting the spindle center position drift according to the vibration growth of the machining spindle;

[0134] In an embodiment of the present invention, the spindle vibration growth obtained in the previous step is used as a trigger condition to deploy a spindle center position tracking system, which is composed of a high-precision eddy current sensor array configured at the end of the spindle, which is three-dimensionally distributed in 360-degree directions around the spindle and arranged at a pitch of 0.1 mm, with a total of 32 sensor points. The system collects the dynamic concentricity changes and radial eccentricity of the spindle during operation in real time. The theoretical trajectory of the spindle center is set as the reference benchmark, and the actual trajectory of the spindle center point is analyzed by eddy current signals to construct a real-time spindle center drift trajectory diagram. The coordinate center of gravity migration algorithm is applied to cluster statistics of the center point trajectory offset values within the continuous time window. If the total amplitude of the spindle center displacement exceeds 1.8 times the initial static eccentricity within a 5-minute processing cycle, and the trajectory change shows a linear offset trend, it is judged that there is a center position drift condition in this cycle. Output the spindle drift trajectory data and the center position change curve.

[0135] Step S345: Calculate the degree of component machining accuracy degradation based on the spindle center position drift and component machining fine displacement disturbance.

[0136] In an embodiment of the present invention, the fine displacement disturbance amplitude curve in step S343 and the spindle center position drift trajectory data in step S344 are combined to construct an error superposition analysis chain to coordinately map the local micro-displacement disturbance area of the workpiece with the spindle center offset vector field, and calculate the superposition error area through the spatial coincidence analysis method. The superposition value of the displacement amplitude and the offset trajectory in the error area is regarded as the local machining accuracy loss factor. The surface area integration method is used to integrate the error point density in the entire machining area, and the degree of machining accuracy reduction is expressed as the unit area machining accuracy reduction coefficient D precision, which is defined as the average geometric offset per unit area of the machining surface. 0.015mm is set as the minimum tolerance line. If the D precision exceeds the critical value, a machining accuracy reduction alarm signal is output to form a machining accuracy reduction report, which includes a three-dimensional error distribution diagram, an average error curve and a reduction degree assessment result, which serves as the basis for the splitting process adjustment or process compensation input.

[0137] Preferably, step S4 includes the following steps:

[0138] Step S41: evaluating the failure trend of the component manufacturing process based on the degree of component processing accuracy degradation and the matching deviation of the processed component split orders;

[0139] In this embodiment of the present invention, the component processing precision degradation index output from step S34 is used, and the split-order matching deviation data recorded in the discrete manufacturing task system is introduced. The split-order matching deviation data is output by the assembly precision comparison module, which quantifies the deviation of the assembly interface of the processed components in different time periods under the same process path. The grating alignment platform and digital micro-displacement probe are used to obtain the component assembly surface parallelism, hole pitch fit error, and surface contact characteristic deviation values. Parts processed from the same batch are matched and compared one by one to obtain a standard deviation index. The component processing precision degradation and the split-order matching deviation are synchronized and superimposed in a time window to create a corresponding matrix between each processing batch and each component matching interface. This matrix is then analyzed using the manufacturing node trend analysis module to extract sections along the same process path that experience systematic offsets and repeatability defects, and the frequency and severity of occurrence are recorded. This analysis generates a trend curve based on the time axis, outputting a component manufacturing process failure trend index, annotating the failure probability and potential impact range of each section by section number, and storing it in the failure trend database as a basis for subsequent process decisions.

[0140] Step S42: detecting component manufacturing order splitting process defect data based on component manufacturing process failure trends;

[0141] In this embodiment of the present invention, the component manufacturing process failure trend index obtained in step S41 is used as an input parameter to activate the splitting process flow comparison and analysis unit. Based on the process layout file within the splitting task, this unit compares the standard process path with the task execution path in the failure trend concentrated area, section by section. For process sections where the failure trend index exceeds 0.6, the unit accesses the operation logs and processing status data stored in the Manufacturing Execution System (MES) to compare the actual splitting execution path, identifying splitting-level process defects such as task scheduling deviation, fixture failure, and tooling selection discrepancy. Defect data is three-dimensionally labeled using timestamp, workstation number, and process task number. The influencing parameters corresponding to each defect type are extracted, including the percentage reduction in process repeatability, interface alignment failure frequency, and clamping slip ratio. The time periods where the splitting execution path deviates from the standard path are combined with the time-series overlap rate of machining accuracy defects to form a splitting process defect dataset. This dataset serves as the basis for subsequent abnormal adaptability analysis. All defect data is stored in CSV format, with fields including defect type, occurrence workstation, corresponding task ID, impact indicator value, and corresponding failure trend index value.

[0142] Step S43: analyzing abnormalities in the adaptability of the processing order splitting based on the defect data of the component manufacturing splitting process and the failure trend of the component manufacturing process;

[0143] In an embodiment of the present invention, step S43 takes the process defect data set generated in step S42 as input, and refers to the fault trend index output by step S41, and uses the order splitting adaptability analysis engine to perform deviation mapping on the execution status of the current order splitting path and the standard process path. The adaptability analysis engine constructs an order splitting logic graph, with processing nodes as graph nodes and process execution dependencies as graph edges, and evaluates the degree of impact on the critical path in the order splitting logic graph in combination with the node failure records in the defect data set. The engine scores the effectiveness of the execution of each order splitting path, and the scoring is based on indicators such as the proportion of defective nodes, the impact ratio of key processes, and the number of path interruptions. If there are three consecutive process segments in the order splitting path that simultaneously meet the fault trend index higher than 0.7 and there are more than two order splitting process defect records, they are marked as adaptability abnormal paths. The adaptability abnormal path is mapped to the order dimension, and the analysis results are output as an order ID and adaptability score list, while also indicating the abnormal process segment ID and abnormal type label. This analysis process generates a structured exception list for all abnormal paths, which is used for subsequent data-driven order splitting optimization management operations. The abnormal path information is synchronously transmitted to the process scheduling center and task adjustment module.

[0144] Step S44: performing splitting optimization management on the processing component splitting data according to the abnormality of the processing order splitting adaptability to obtain the processing component splitting optimization data.

[0145] In an embodiment of the present invention, a split-order task optimization management controller uses a generated list of exceptions to its adaptability as input to reconfigure the scheduling path. This controller, relying on a manufacturing resource status monitoring platform, analyzes existing machine tool resources, processing load, tool life status, and the queue sequence of pending tasks. The system determines the optimal split-order adjustment strategy based on task importance and the priority of the impact on path length. There are two methods for split-order optimization: one is to migrate task nodes with excessively high failure trends in the abnormal path to low-risk sections or adjacent production lines; the other is to introduce alternative fixtures and different cutting parameter solutions through alternative processing strategies to avoid repeated triggering of overlapping risk nodes. After optimization, the system regenerates the split-order execution path, rebinding each path segment to the processing machine number, execution time period, and corresponding tool number, and updating the process flow file. The optimization results form a split-order optimization dataset, which includes a comparison of the paths before and after optimization, beat adjustment data, and station resource reallocation records. This data is output to the split-order control center system and simultaneously updates the split-order path map in the MES system. The system will also cross-validate the optimization data with historical processing exception records to ensure that the optimization plan does not introduce new risk paths. The output order splitting optimization data is recorded in a JSON structure and synchronized to each subsystem to complete the order splitting execution preparation.

[0146] The present invention further provides a discrete manufacturing order splitting system for executing the discrete manufacturing order splitting method described above. The discrete manufacturing order splitting system includes:

[0147] The component splitting processing module is used to obtain the task flow data of the manufacturing execution unit; based on the task flow data of the manufacturing execution unit, it collects the manufacturing unit processing component information; based on the manufacturing unit processing component information, it performs the processing component splitting processing to obtain the processing component splitting data;

[0148] The module for assessing the progressive imbalance trend of machining tools is used to assess the matching deviation of machining parts according to the split-order data of machining parts; detect the dynamic fatigue accumulation of machining tools based on the matching deviation of machining parts; and predict the progressive imbalance trend of machining tool functions based on the dynamic fatigue accumulation of machining tools.

[0149] A module for calculating the degree of reduction in machining accuracy is used to determine the degree of deviation in the machining trajectory of the tool based on the trend of the progressive imbalance of the machining tool function; detect the growth of the vibration of the machining spindle based on the trend of the progressive imbalance of the machining tool function and the degree of deviation in the machining trajectory of the tool; and calculate the degree of reduction in the machining accuracy of the component based on the growth of the vibration of the machining spindle;

[0150] The splitting optimization management module is used to analyze the abnormal adaptability of the processing order splitting based on the degree of decline in component processing accuracy and the deviation of the processing component splitting matching; according to the abnormal adaptability of the processing order splitting, the processing component splitting data is optimized and managed to obtain the processing component splitting optimization data.

[0151] A computer-readable storage medium stores a computer program, wherein the computer program is used to execute the discrete manufacturing order splitting method.

[0152] The foregoing description is intended only to provide specific embodiments of the present invention, which will enable those skilled in the art to understand and implement the present invention. Various modifications to these embodiments will be readily apparent to those skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of the present invention. Therefore, the present invention is not intended to be limited to the embodiments shown herein, but is to be construed in the widest possible manner consistent with the principles and novel features disclosed herein.

Claims

1. A method for splitting orders based on discrete manufacturing, characterized in that: The following steps are involved: Step S1: Acquire task flow data of the manufacturing execution unit; Collect manufacturing unit processing component information based on manufacturing execution unit task flow data; Perform splitting of processing parts based on the processing parts information of the manufacturing unit to obtain splitting data of processing parts; Step S2: evaluating the matching deviation of the processing component split orders based on the processing component split order data; detecting the dynamic fatigue accumulation of the processing tool based on the matching deviation of the processing component split orders; and predicting the progressive imbalance trend of the processing tool function based on the dynamic fatigue accumulation of the processing tool; Step S3: determining the degree of tool machining trajectory deviation based on the progressive imbalance trend of the machining tool function; detecting the vibration growth of the machining spindle based on the progressive imbalance trend of the machining tool function and the degree of tool machining trajectory deviation; and calculating the degree of component machining accuracy degradation based on the vibration growth of the machining spindle; Step S4: Analyze the abnormal adaptability of the processing order splitting based on the degree of decrease in component processing accuracy and the deviation of the processing component splitting matching; perform splitting optimization management on the processing component splitting data according to the abnormal adaptability of the processing order splitting to obtain the processing component splitting optimization data.

2. The order splitting method based on discrete manufacturing according to claim 1 is characterized in that: Step S1 includes the following steps: Step S11: Acquire manufacturing execution unit task flow data; Step S12: collecting manufacturing unit processing component information based on the manufacturing execution unit task flow data; Step S13: dividing the internal structure of the processing component into the processing component information of the manufacturing unit to obtain the internal structure division data of the processing component; Step S14: performing splitting of the processing components based on the internal structure division data of the processing components to obtain splitting data of the processing components.

3. The order splitting method based on discrete manufacturing according to claim 2, characterized in that: Step S13 includes the following steps: Step S131: extracting material parameters of the processing component according to the processing component information of the manufacturing unit; Step S132: evaluating the processing difficulty coefficient of the material of the processing component when the yield strength of the material parameter of the processing component exceeds 355 MPa; Step S133: collecting the three-dimensional structure of the manufacturing unit processing component based on the manufacturing unit processing component information; Step S134: identifying spatial connectivity of the processing components according to the three-dimensional structure of the processing components of the manufacturing unit; Step S135: Evaluate component processing path requirement data based on the three-dimensional structure of the processing component of the manufacturing unit and the spatial connectivity of the processing component; Step S136: Based on the component processing path requirement data and the component material processing difficulty coefficient exceeding 0.927, the internal structure of the processing component is divided to obtain the internal structure division data of the processing component.

4. The order splitting method based on discrete manufacturing according to claim 1, characterized in that: The evaluation of the matching deviation of the processing parts in step S2 includes: Count the mismatch of component processing machinery capabilities based on the data of component splitting orders; Detect insufficient adaptation of processing equipment functions based on component processing machinery capability mismatch; Assess the extent of excess coverage of processing equipment based on insufficient adaptation of processing equipment functions and mismatch of component processing machinery capabilities; Identify the collapse of the device's structural alignment capability based on the insufficient adaptation of the processing device's functions and the degree of excess coverage of the processing device; Measure the abnormal clamping force of the device according to the disintegration of the device structure's alignment ability; Calculate the device clamping posture drift parameters based on the abnormal clamping force and the collapse of the device structure alignment ability; The matching deviation of the processed parts is evaluated based on the instrument clamping posture drift parameters and the abnormal force of the instrument clamping.

5. The order splitting method based on discrete manufacturing according to claim 1, characterized in that: The dynamic fatigue accumulation detection of the machining tool in step S2 includes: Detect the excessive cutting force of machining tools based on the deviation of the matching of machining parts; Detect the growth degree of mechanical impact of tool according to the excessive cutting force of tool in processing equipment; Determine the step load growth condition of the machining tool according to the growth degree of the tool mechanical impact and the excessive cutting force of the machining tool; Calculate the stress growth degree of the machining tool based on the step-type load growth condition of the machining tool; Monitor the growth trend of micro cracks in machining tools based on the stress growth degree and step load growth condition of machining tools; Identify the growth trend of machining tool surface roughness based on the growth trend of machining tool microcracks; Determine the degree of micro deformation accumulation of machining tools based on the growth trend of machining tool surface roughness and the growth trend of machining tool micro cracks; The dynamic fatigue accumulation of the machining tool is detected based on the cumulative degree of micro-deformation of the machining tool and the growth trend of the surface roughness of the machining tool.

6. The order splitting method based on discrete manufacturing according to claim 1, characterized in that: The prediction of the progressive imbalance trend of the machining tool function in step S2 includes: Detect the degree of hardness reduction of the surface layer of the machining tool material based on the dynamic fatigue accumulation of the machining tool; Detect the rigidity attenuation of the machining tool when the surface hardness of the machining tool material decreases by more than 10%; Analyze the growth trend of cutting force fluctuation of machining tools when the rigidity attenuation of machining tools drops by more than 10% and the degree of decrease in the surface hardness of machining tool materials; Predict the intermittent chipping of machining tools based on the fluctuation growth trend of cutting force of machining tools; Detect the torque coupling force distortion of machining tools based on the intermittent chipping of machining tools; Determine the degree of tool body overload based on the torque coupling force distortion of the tool and the intermittent chipping of the tool; The progressive imbalance trend of machining tool function is predicted based on the torque coupling force distortion of machining tool and the degree of tool body overload.

7. The order splitting method based on discrete manufacturing according to claim 1, characterized in that: Step S3 includes: Step S31: determining the degree of tool machining trajectory deviation according to the progressive imbalance trend of the machining tool function; Step S32: calculating the machining tool cutting angle difference data according to the tool machining trajectory deviation degree and the progressive imbalance trend of the machining tool function; Step S33: estimating the unstable contact of the machining tool according to the machining tool cutting angle difference data; Step S34: detecting the degree of friction growth of the machining tool based on the unstable contact condition of the machining tool; Step S35: detecting the vibration growth of the machining spindle based on the degree of friction growth of the machining tool and the unstable contact of the machining tool; Step S36: Calculate the degree of decrease in component machining accuracy based on the vibration growth of the machining spindle and the difference in the cutting angle of the machining tool.

8. The order splitting method based on discrete manufacturing according to claim 1, characterized in that: Step S4 includes the following steps: Step S41: evaluating the failure trend of the component manufacturing process based on the degree of component processing accuracy degradation and the matching deviation of the processed component split orders; Step S42: detecting component manufacturing order splitting process defect data based on component manufacturing process failure trends; Step S43: analyzing abnormalities in the adaptability of the processing order splitting based on the defect data of the component manufacturing splitting process and the failure trend of the component manufacturing process; Step S44: performing splitting optimization management on the processing component splitting data according to the abnormality of the processing order splitting adaptability to obtain the processing component splitting optimization data.

9. A discrete manufacturing order splitting system, characterized in that: For executing the discrete manufacturing order splitting method according to claim 1, the discrete manufacturing order splitting system comprises: The component splitting processing module is used to obtain the task flow data of the manufacturing execution unit; based on the task flow data of the manufacturing execution unit, it collects the manufacturing unit processing component information; based on the manufacturing unit processing component information, it performs the processing component splitting processing to obtain the processing component splitting data; The module for assessing the progressive imbalance trend of machining tools is used to assess the matching deviation of machining parts according to the split-order data of machining parts; detect the dynamic fatigue accumulation of machining tools based on the matching deviation of machining parts; and predict the progressive imbalance trend of machining tool functions based on the dynamic fatigue accumulation of machining tools. A module for calculating the degree of reduction in machining accuracy is used to determine the degree of deviation in the machining trajectory of the tool based on the trend of the progressive imbalance of the machining tool function; detect the growth of the vibration of the machining spindle based on the trend of the progressive imbalance of the machining tool function and the degree of deviation in the machining trajectory of the tool; and calculate the degree of reduction in the machining accuracy of the component based on the growth of the vibration of the machining spindle; The splitting optimization management module is used to analyze the abnormal adaptability of the processing order splitting based on the degree of decline in component processing accuracy and the deviation of the processing component splitting matching; according to the abnormal adaptability of the processing order splitting, the processing component splitting data is optimized and managed to obtain the processing component splitting optimization data.

10. A computer-readable storage medium storing a computer program, characterized in that: When the computer program is executed, the discrete manufacturing order splitting method according to any one of claims 1 to 8 is implemented.

Images