Metal processing error monitoring method and system based on situation awareness

Abstract

The invention discloses a situation awareness-based metal processing error monitoring method and system, and particularly relates to the technical field of intelligent manufacturing, and the method comprises the steps: obtaining workpiece point cloud data, executing geometric reconstruction and coordinate alignment, and forming a workpiece individual geometric expression structure which can be used for recognizing a metal processing error; the structure basis is used for subsequent difference analysis and error source judgment; performing spatial difference vector solution and difference attribute division on the workpiece individual geometric expression structure and the standard design geometric expression structure to form a difference expression structure which is consistent in structure and contains local offset information, and taking the difference expression structure as input of path adaptation; according to the method, a dynamic coupling relation between an individual geometric expression structure of a workpiece and a machining behavior variable is constructed, and a deviation situation among three elements is perceived in real time before a path is generated and in a machining process, so that early recognition and response to a machining error are realized, and the problems that the error cannot be actively recognized and the path is mismatched in a customized machining scene are solved.

Description

Technical Field

[0001] The present invention relates to the technical field of situation-aware metal processing error monitoring and management, and more specifically, to a situation-aware metal processing error monitoring method and system. Background Art

[0002] In small-batch customized metal processing scenarios, small-batch customized metal processing objects usually have individual differences in geometric dimensions, boundary morphology, and material properties;

[0003] Current CNC machining systems generally use standard CAD models and fixed process parameters to generate tool paths. Error monitoring during CNC machining primarily relies on threshold judgments for physical signals such as spindle load, vibration amplitude, and cutting force. This type of error monitoring method relies on a high degree of consistency between workpiece features and machining paths, making it suitable for high-volume, structurally consistent product manufacturing.

[0004] In customized scenarios, due to small but significant deviations between the actual state of each workpiece and the standard model, a fixed machining path may locally deviate from the actual boundary during execution. Traditional physical signals are not sensitive to such errors and cannot effectively respond, causing the system to misjudge machining behavior as "normal." Furthermore, the current system lacks a workpiece-level individual difference modeling mechanism, making it impossible to proactively identify such deviation risks during machining path generation or error monitoring.

[0005] When substantial geometric errors occur during machining, the system lacks the ability to perceive the individual state of the workpiece and fails to detect the mismatch between it and the fixed machining path. As a result, the system still maintains a "logically correct" judgment when machining errors occur, forming a systemic problem of unrecognized errors. Summary of the Invention

[0006] In order to overcome the above-mentioned defects of the prior art, an embodiment of the present invention provides a metal processing error monitoring method and system based on situational awareness, which constructs a dynamic coupling relationship between the individual geometric expression structure of the workpiece and the processing behavior variables, and perceives the deviation situation among the three elements of workpiece-path-behavior in real time before path generation and during the processing, thereby realizing early identification and response to processing errors, so as to solve the problems of inability to actively identify errors and path mismatch in customized processing scenarios proposed in the above-mentioned background technology.

[0007] To achieve the above objectives, the present invention provides the following technical solution: a metal processing error monitoring method based on situational awareness, comprising:

[0008] Acquire workpiece point cloud data and perform geometric reconstruction and coordinate alignment to form an individual workpiece geometric expression structure for identifying metal processing errors, which serves as the structural basis for difference analysis and error source determination;

[0009] The spatial difference vector is solved and the difference attributes are divided between the workpiece individual geometric expression structure and the standard design geometric expression structure to form a difference expression structure with consistent structure and local offset information as the input of path adaptation;

[0010] The standard machining path expression and the differential expression structure are mapped, and the interference detection and path correction process are performed to generate an executable path expression that meets the individual geometric state of the workpiece and the machining constraints, which serves as the input for subsequent behavior monitoring.

[0011] Structural alignment and disturbance feature extraction are performed on the set of machining behavior variables and the workpiece individual adaptation path to generate a set of behavioral response features corresponding to the path segment. The differential expression structure is then combined to determine whether the behavioral deviation is concentrated in the high-risk area, and a path segment behavioral deviation combination is output for error identification.

[0012] Error event judgment and feature vector extraction are performed on the path segment behavior offset combination, and the path generation strategy sample set is written into the strategy factor update. The updated results are used in the subsequent path constraint generation and adaptation process to complete the closed-loop evolution of the path construction strategy.

[0013] In a preferred embodiment, in metal processing error monitoring, the original spatial data of the workpiece is obtained by three-dimensional optical point cloud scanning, and density equalization, outlier elimination and boundary connectivity reconstruction are performed on the original spatial data of the workpiece to output the initial spatial expression data; the initial spatial expression data is mapped through surface continuity fitting and normal interpolation, and a surface reconstruction process is executed to output a complete individual geometric expression structure of the workpiece.

[0014] In a preferred embodiment, the coordinates of the workpiece individual geometric expression structure and the standard design geometric expression structure are aligned by a rigid registration operation, and the aligned expression structure in a unified spatial framework is output;

[0015] Determine whether there are any local difference segments in the boundary area of ​​the aligned geometric expression structure that cannot be aligned. The local difference segment is determined by whether the spatial offset between the standard design geometric expression structure and the individual workpiece geometric expression structure within the boundary range exceeds the defined local fitting tolerance. If so, execute the boundary enhancement reconstruction process; if not, retain the current geometric expression structure;

[0016] The aligned geometric expression structure is used as the input of the subsequent differential expression process to form a comparable workpiece individual geometric expression structure that is consistent with the standard design geometric expression structure space.

[0017] In a preferred embodiment, the workpiece individual geometric expression structure and the standard design geometric expression structure in a unified coordinate system are used as input, a spatial corresponding point set difference vector solution is performed, and a spatial residual field is output;

[0018] The spatial residual field is divided into three attribute types: position offset, curvature perturbation and boundary morphological variation, and the position difference feature, curvature difference feature and boundary difference feature are output respectively. The position difference feature, curvature difference feature and boundary difference feature are collectively referred to as the difference feature set;

[0019] Determine whether there are offset distribution blocks in the position difference feature set that continuously exceed the threshold. If so, mark it as a path interference risk area for path correction.

[0020] In a preferred embodiment, the difference feature set is re-bound to the topological reference point set of the standard design geometric expression structure, a mapping relationship between the point set and the difference attribute is established, and a structurally consistent difference expression structure is output;

[0021] The differential expression structure is used as the input basis for path adaptation to constrain the path mapping range and tool position generation logic.

[0022] In a preferred embodiment, the standard machining path expression and the differential expression structure are used as inputs, a path mapping process is executed, the standard machining path is re-projected on the workpiece individual geometric expression structure expression, and an initial mapping path is output;

[0023] Perform workpiece individual geometric expression structure penetration test and boundary reachability judgment on the initial mapping path to obtain the interference area between the path and the differential expression structure;

[0024] Determine whether the interference area is caused by the combination of position difference features and curvature difference features in the differential expression structure. If so, local reconstruction is performed preferentially on the interference path segment and the path correction candidate is output; if not, it is marked as an abnormal workpiece configuration segment, triggering the path configuration exception processing logic.

[0025] In a preferred embodiment, the path correction candidate is reconstructed in a fair manner on the workpiece individual geometric expression structure, and boundary difference features in the differential expression structure are combined to perform boundary preservation constraints, and the workpiece individual adaptation path is output;

[0026] The workpiece individual adaptation path is subjected to the processing accessibility test and machine tool posture solution process to determine whether the posture is feasible and whether the trajectory is continuous. If it is feasible and continuous, the current path is retained. If it is not feasible or there is a trajectory jump, the smoothness reconstruction process is re-entered to perform iterative correction;

[0027] The path expression body that is finally confirmed to be processable is used as the structural input of the subsequent processing behavior monitoring process.

[0028] In a preferred embodiment, a set of machining behavior variables is collected during the machining process, and the set of machining behavior variables includes data such as tool load fluctuation, feed acceleration change, and contour position feedback. The set of machining behavior variables is aligned to the workpiece individual adaptation path according to the path segment structure, and a behavior response sequence corresponding to the path segment is output.

[0029] Perform disturbance pattern recognition on the behavioral response sequence, and output a behavioral response feature set by calculating the trajectory drift rate, fluctuation frequency abnormality, and local load peak change rate of the behavioral response;

[0030] Determine whether the behavioral response feature set appears concentrated in the high-risk area marked by the differential expression structure corresponding to the path segment. If so, mark it as an error trend segment; if not, determine it as a stable processing segment.

[0031] The path segments marked with error trend occurrence segments are combined with their behavioral response feature sets to form a path segment behavior offset combination, which is used in the error event identification process.

[0032] In a preferred embodiment, a behavior-structure joint deviation calculation is performed on the path segment behavior offset combination. The coupling degree between the trajectory drift rate and the position difference feature, and the correlation strength between the load peak change rate and the curvature difference feature are used to jointly determine whether an error event is constituted. If so, an error event determination label is output; if not, the error event is retained as a behavior fluctuation sample.

[0033] The path segment behavior deviation combination with the error event judgment label is used as a feature sample, and feature compression and behavior category mapping are performed to output the error type feature vector.

[0034] The error type feature vector is written into the path generation strategy sample set to participate in the update calculation of the strategy factor in the subsequent path construction;

[0035] In the new round of path construction, the path generation strategy sample set is used as one of the inputs to participate in the initial path constraint generation process in the conversion of standard path to individual path, forming a path candidate structure with dynamic adaptability.

[0036] The dynamically adaptive path candidate structure is returned to the path mapping process as the starting input of the path adaptation link, and the path adaptation mapping is executed to complete the evolution and update of the path adaptation strategy.

[0037] A metal processing error monitoring system based on situational awareness, the system includes a geometric modeling module, a difference expression module, a path adaptation module, a behavior perception module, and a strategy evolution module;

[0038] The geometric modeling module is used to obtain workpiece point cloud data and perform geometric reconstruction and coordinate alignment to form an individual geometric expression structure of the workpiece that can be used to identify metal processing errors, serving as the structural basis for subsequent difference analysis and error source determination;

[0039] The differential expression module is used to perform spatial difference vector solution and differential attribute division between the workpiece individual geometric expression structure and the standard design geometric expression structure, forming a differential expression structure with consistent structure and containing local offset information as the input of path adaptation;

[0040] The path adaptation module is used to perform mapping, interference detection and path correction processes between the standard machining path expression and the differential expression structure, and generate an executable path expression that meets the individual geometric state of the workpiece and the machining constraints as the input of the behavior monitoring process;

[0041] The behavior perception module is used to perform structural alignment and disturbance feature extraction on the set of machining behavior variables and the workpiece individual adaptation path, generate a set of behavioral response features corresponding to the path segment, and combine the differential expression structure to determine whether the behavior deviation is concentrated in the high-risk area, and output the path segment behavior deviation combination for error identification;

[0042] The strategy evolution module is used to perform error event judgment and feature vector extraction on the path segment behavior offset combination, write the path generation strategy sample set to participate in the strategy factor update, and use the update results in the subsequent path constraint generation and adaptation process to complete the evolution of the path construction strategy.

[0043] Technical effects and advantages of the present invention:

[0044] 1. By constructing the spatial difference vector field between the workpiece's individual geometric expression structure and the standard design geometric expression structure, it is possible to perceive the errors caused by individual deviations before processing;

[0045] 2. Guide path mapping and interference correction through differential expression structure to generate a machining path that can adapt to the actual geometric state of the workpiece and avoid path expression mismatch;

[0046] 3. Synchronize the behavioral variables according to the path segment mapping structure, extract the disturbance characteristics, and enhance the ability to perceive error trends in dynamic processes;

[0047] 4. By calculating the coupling deviation between behavioral response and structural difference, we can distinguish between real errors and non-critical fluctuations, and improve the robustness of error judgment;

[0048] 5. Feedback the error event characteristics to the path construction process, update the path generation factor, and realize a dynamic adaptive path generation mechanism based on individual workpiece feedback. BRIEF DESCRIPTION OF THE DRAWINGS

[0049] Figure 1It is a flowchart of the framework of the method steps of the present invention.

[0050] Figure 2 It is a schematic diagram of the system module structure of the present invention.

[0051] Figure 3 This is a flow chart of the geometric modeling and coordinate registration of the present invention.

[0052] Figure 4 A flow chart was constructed for the differential expression structure of the present invention.

[0053] Figure 5 This is a flow chart of the path adaptation and interference correction of the present invention.

[0054] Figure 6 This is a flow chart of behavior monitoring and error trend identification of the present invention.

[0055] Figure 7 This is a flow chart of the strategy evolution of the present invention. DETAILED DESCRIPTION

[0056] The following will clearly and completely describe the technical solutions in the embodiments of the present invention in conjunction with the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative efforts are within the scope of protection of the present invention.

[0057] Refer to the instruction manual Figure 1-7 , a metal processing error monitoring method based on situational awareness according to an embodiment of the present invention:

[0058] Acquire workpiece point cloud data and perform geometric reconstruction and coordinate alignment to form an individual workpiece geometric expression structure that can be used to identify metal processing errors, serving as the structural basis for subsequent difference analysis and error source determination;

[0059] The spatial difference vector is solved and the difference attributes are divided between the workpiece individual geometric expression structure and the standard design geometric expression structure to form a difference expression structure with consistent structure and local offset information as the input of path adaptation;

[0060] The standard machining path expression and the differential expression structure are mapped, and the interference detection and path correction process is performed to generate an executable path expression that meets the individual geometric state of the workpiece and the machining constraints, which serves as the input of the subsequent behavior monitoring process;

[0061] Structural alignment and disturbance feature extraction are performed on the set of machining behavior variables and the workpiece individual adaptation path to generate a set of behavioral response features corresponding to the path segment. The differential expression structure is then combined to determine whether the behavioral deviation is concentrated in the high-risk area, and a path segment behavioral deviation combination is output for error identification.

[0062] Error event judgment and feature vector extraction are performed on the path segment behavior offset combination, and the path generation strategy sample set is written into the strategy factor update. The updated results are used in the subsequent path constraint generation and adaptation process to complete the closed-loop evolution of the path construction strategy.

[0063] It should be noted that in the formula structure involved in this solution, dimensionless terms can serve as proportionality or structural adjustment factors. When combined with quantities with units, they only play a numerical scaling role and do not introduce new physical dimensions. Therefore, they will not change or confuse the overall unit system of expression. This combination of "dimensionless terms and units" can be understood as a composite structural expression commonly used in mathematical and physical modeling, conforming to the principle of dimensional consistency and having a clear physical interpretation basis.

[0064] Secondly, in the formula structure of this scheme, if multiple variables with different physical units are involved, including but not limited to time, mass or energy variables, their joint appearance is to express the collaborative modeling relationship of multiple physical mechanisms. Each variable can be formed into a unified structure through function mapping, ratio combination or normalization adjustment. The units and meanings are clear, and the overall expression conforms to the principle of dimensional consistency and the common formula of engineering modeling.

[0065] In this solution, any design constants, weights, adjustment factors, threshold parameters, and proportional coefficients are adjustable control parameters for different application environments. Their values ​​depend on the target device configuration, data input characteristics, and performance optimization goals. During the implementation phase, they are set within a reasonable range through model verification, performance constraints, or engineering calibration. Although these parameters do not have unique preset values, they have clear adjustment logic and calculation paths, and are part of the deterministic setting process in engineering implementation. The purpose of such setting is to ensure that the solution is both universally adaptable, reproducible, and operable, without affecting its technical clarity and feasibility.

[0066] In metal processing error monitoring, the original spatial data of the workpiece is obtained through three-dimensional optical point cloud scanning, and the original spatial data of the workpiece is subjected to density balancing, outlier exclusion and boundary connectivity reconstruction to output the initial spatial expression data; the calculation logic of performing density balancing, outlier exclusion and boundary connectivity reconstruction on the original spatial data of the workpiece is: performing local spatial system division and neighborhood point number statistics on the original point cloud data in the original spatial data, and outputting a point cloud set with uniform density distribution by performing downsampling on high-density areas and interpolation processing on low-density areas; then performing k-nearest neighbor spatial consistency analysis on the point cloud set with uniform density distribution, eliminating outliers through statistical judgment of neighborhood average distance and deviation, and outputting a valid point cloud set after noise points are removed; finally, performing boundary point extraction and edge segment fitting on the valid point cloud set after outliers are excluded, generating a structurally continuous boundary point set through boundary connectivity reconstruction and interpolation, and outputting a point cloud reconstruction result with closed boundaries;

[0067] The initial spatial representation data is mapped through surface continuity fitting and normal interpolation, and the surface reconstruction process is executed to output a complete geometric representation of the workpiece. The computational logic of the surface reconstruction process is as follows: the point cloud set after boundary connectivity reconstruction is subjected to adjacency construction and curvature continuity fitting, and a continuous curvature surface is generated through surface topology reasoning and normal interpolation, which outputs a surface representation with a machinable topological structure.

[0068] Through the rigid registration operation, the coordinates of the workpiece individual geometric expression structure and the standard design geometric expression structure are aligned, and the aligned expression structure in the unified space framework is output;

[0069] Determine whether there are any local difference segments in the boundary area of ​​the aligned geometric expression structure that cannot be aligned. The local difference segment is determined by whether the spatial offset between the standard design geometric expression structure and the individual workpiece geometric expression structure within the boundary range exceeds the defined local fitting tolerance. If so, execute the boundary enhancement reconstruction process; if not, retain the current geometric expression structure;

[0070] The aligned geometric expression structure is used as the input of the subsequent differential expression process to form a comparable workpiece individual geometric expression structure that is consistent with the standard design geometric expression structure space.

[0071] The individual geometric expression structure of the workpiece in the unified coordinate system and the standard design geometric expression structure are used as input, and the spatial corresponding point set difference vector solution is performed to output the spatial residual field; the spatial corresponding point set difference vector solution means: in the two sets of geometric expression structures after coordinate alignment, for each point in the individual expression structure of the workpiece, by searching the standard design expression structure for a point with the smallest spatial distance and similar normal direction as the corresponding point, the position values ​​of this point and the target point on the three coordinate axes are respectively subtracted to generate a three-dimensional direction difference value group, and the three-dimensional direction difference value groups of all point pairs are combined in spatial order to form a difference vector set, which is used to express the corresponding offset of the two sets of structures in local positions;

[0072] The spatial residual field is divided into three attribute types: position offset, curvature perturbation and boundary morphological variation, and the position difference feature, curvature difference feature and boundary difference feature are output respectively. The position difference feature, curvature difference feature and boundary difference feature are collectively referred to as the difference feature set;

[0073] Determine whether there are offset distribution blocks in the position difference feature set that continuously exceed the threshold. If so, mark it as a path interference risk area for subsequent path correction;

[0074] As a further solution, the spatial difference vector is solved between the workpiece individual geometric expression structure and the standard design geometric expression structure, a local spatial offset expression is constructed and a difference feature set is output;

[0075] Formulas and models constitute the spatial difference vector field

[0076]

[0077] Point correspondence mapping function

[0078]

[0079] in:

[0080] Spatial distance term Λ d (p i ,q j ):

[0081] Λ d (p i ,q j )=||p i -q j ||2

[0082] Normal consistency term Λ n (p i ,q j ):

[0083]

[0084] Curvature difference term Λ κ (p i ,q j ):

[0085] Λ κ (p i ,q j )=|κ(p i )-κ(q j )|

[0086] Difference attribute annotation δ i :

[0087] δ i =ψ(||v i ||,κ(p i ),ρ(p i ))

[0088] Differential feature set

[0089]

[0090] where v i For point p i The spatial difference vector between the matching point and the point p i is the three-dimensional coordinate vector of the i-th workpiece point; is the point set in the geometric expression structure of the workpiece individual; the symbol | in the formula means "satisfies the condition...", which is used to introduce constraints in the set definition and can be read as "such that" or "wherein", indicating that the elements in the set are the set of all elements that meet the given conditions; is the spatial optimal point pair mapping function; q j is the jth standard model point, three-dimensional coordinate vector; is the point set in the standard design geometric expression structure; argmin represents the value of the independent variable corresponding to the minimum value of the target expression, where q is the point in the standard design geometric expression structure that minimizes the sum of the three difference terms. j The symbol Λ in the formula represents a difference calculation function, which is used to measure the degree of difference between the points in the workpiece individual geometric expression structure and the candidate points in the standard design geometric expression structure in different dimensions. The degree of difference includes spatial distance, normal direction difference, and curvature difference. For point p i ,q j The corresponding normal vector is a unit vector; κ(p i ),κ(q j ) is point p i ,q jThe local curvature value; ρ(p i ) is point p i The boundary mark value of (0 represents an internal point, 1 represents a boundary point); ψ is a multi-factor difference judgment function, and the multi-factor difference judgment function is the output category label.

[0091] Rebind the difference feature set to the topological reference point set of the standard design geometric expression structure, establish a mapping relationship between the point set and the difference attribute, and output a structurally consistent difference expression structure;

[0092] The differential expression structure is used as the input basis for path adaptation to constrain the path mapping range and tool position generation logic.

[0093] Taking the standard machining path expression and the differential expression structure as input, the path mapping process is executed to reproject the standard machining path onto the workpiece individual geometric expression structure and output the initial mapping path;

[0094] Perform workpiece individual geometric expression structure penetration test and boundary reachability judgment on the initial mapping path to obtain the interference area between the path and the differential expression structure;

[0095] Determine whether the interference region is caused by the combination of position difference features and curvature difference features in the differential expression structure. If so, perform local reconstruction on the interference path segment first and output the path correction candidate. If not, mark it as an abnormal workpiece configuration segment and trigger the path configuration exception processing logic.

[0096] As a further solution, the standard machining path expression is mapped to the workpiece individual geometric expression structure, and the spatial accessibility and structural interference of the path segment are determined based on the differential expression structure;

[0097] Path segment:

[0098]

[0099] Path segment intersection difference area:

[0100]

[0101] Interference evaluation function:

[0102]

[0103] Interference path determination:

[0104] γ k Interference segment Υ k >ξ

[0105] where γ kis the adaptive path expression of the workpiece in the kth segment, which is used to describe the actual motion path of the tool in this segment during the machining process; γ k (t) is the spatial trajectory point of the k-th processing path; Represents a three-dimensional Euclidean space, used to define the path segment γ k (t) in the time interval [0,τ k ] is in three-dimensional space; t is the expression of the workpiece individual adaptation path γ k The time parameter within the path segment is used to describe the processing time position corresponding to each trajectory point on the path; τ k is the total time range of the path segment; Ω k is the set of points in the spatial overlapping region between the pathway segment and the differentially expressed structure; p i Differentially expressed constructs The i-th structural point in is used to represent the spatial position that may interfere with the path segment; A logical quantifier indicating "existence", which is used to indicate that there is at least one point in the path segment time interval that satisfies the distance condition; t is the path segment γ k The time variable on the index path is used to index the spatial position of the path at time t; ∈ is the distance threshold for spatial proximity judgment; Υ k is the overall interference risk of the path segment; v i is the spatial difference vector of the i-th point; θ v is the position difference feature judgment threshold; κ(p i ) is the midpoint p of the hetero-expression structure i The local curvature value is used to reflect the degree of spatial curvature of the surface topography at that point in the individual geometric expression structure of the workpiece; is the path segment γ k The average curvature trend value of θ κ is the curvature difference feature determination threshold; χ(γ k ,p i ) is the directional consistency function of the path point and the structure point; where the symbol is a Boolean indicator function, the value is 1 when the condition is met, otherwise it is 0; ξ is the interference judgment threshold.

[0106] Perform fairing reconstruction on the path correction candidate on the workpiece individual geometric expression structure, combine the boundary difference features in the differential expression structure to perform boundary preservation constraints, and output the workpiece individual adaptation path;

[0107] The individual adaptive path of the workpiece is subjected to a processing accessibility test and a machine tool posture solution process to determine whether the posture is feasible and whether the trajectory is continuous. If it is feasible and continuous, the current path is retained. If it is not feasible or there is a trajectory jump, the smoothness reconstruction process is re-entered to perform iterative correction. The solution process refers to: based on the input data obtained, according to a clear mathematical relationship or physical model, through analytical deduction or numerical iteration, the geometric parameters, state variables or behavior factors are quantitatively solved to output the target results for control judgment, trajectory generation or error identification. The calculation process usually includes the establishment of mapping relationships between variables, the construction of model equations and convergence verification.

[0108] The path expression body that is finally confirmed to be processable is used as the structural input of the subsequent processing behavior monitoring process.

[0109] During the machining process, a set of machining behavior variables is collected, including data such as tool load fluctuation, feed acceleration change, and contour position feedback. The set of machining behavior variables is aligned to the workpiece individual adaptation path according to the path segment structure, and the behavior response sequence corresponding to the path segment is output.

[0110] Perform disturbance pattern recognition on the behavioral response sequence, and output the behavioral response feature set by calculating the trajectory drift rate, fluctuation frequency anomaly and local load peak change rate of the behavioral response. The calculation process includes: dividing the behavioral response sequence into time windows, extracting the time change trend of the tool trajectory position in each sequence, calculating its cumulative offset value relative to the expected trajectory as the trajectory drift rate, and then extracting the main frequency amplitude and frequency change amplitude through spectrum analysis of the spindle load or feed speed, calculating the difference from the steady-state spectrum as the fluctuation frequency anomaly, and finally extracting the local maximum load in each time window and comparing it with the mean value in the window, calculating its relative increment as the local load peak change rate. The three indicators together constitute the behavioral response feature set.

[0111] Determine whether the behavioral response feature set appears concentrated in the high-risk area marked by the differential expression structure corresponding to the path segment. If so, mark it as an error trend segment; if not, determine it as a stable processing segment.

[0112] The path segments marked with error trend occurrence segments are combined with their behavioral response feature sets to form a path segment behavior offset combination for subsequent error event identification process;

[0113] As a further solution, the processing behavior variable set is aligned according to the path segment structure to extract multi-dimensional disturbance feature quantities;

[0114] Behavioral variable sequence

[0115]

[0116] Trajectory disturbance rate Θ k :

[0117]

[0118] Frequency anomaly Φ k :

[0119]

[0120] Load response peak rate Υ k :

[0121]

[0122] Output behavior feature vector T k :

[0123] T k =[Θ k ,Φ k ,Υ k

[0124] in is the combination of machining behavior variables collected at time t corresponding to the k-th segment of the workpiece individual adaptation path. The machining behavior variable combination includes data items of multiple dimensions such as spindle load, feed acceleration, and spatial trajectory position; f t is the spindle load value at time t; a t is the feed acceleration at time t; x t ,y t ,z t is the coordinate position of the tool in three-dimensional space at time t; t is the time parameter on the path segment; τ k is the upper limit of the time interval of the k-th path; x t is the actual position of the path segment at time t; is the theoretical value of the expected reference position of the path segment at time t; is the second-order time derivative, which represents the rate of change of the trajectory at the acceleration level; dt is the differential increment of time t, which is used in integration or derivative to represent the change process at a very small time scale; ω is the frequency variable; S k (ω) is the frequency envelope spectrum of the path segment behavior; is the change of the second-order curvature of the spectrum; dω is the differential increment of the frequency variable ω, which is used to accumulate the change of the spectrum structure point by point over the entire frequency range in the integration operation, representing the frequency unit of continuous integration from low frequency to high frequency.

[0125] The path segment behavior offset combination is subjected to a behavior-structure joint deviation calculation. The coupling degree between the trajectory drift rate and the position difference feature, and the correlation strength between the load peak change rate and the curvature difference feature are used to jointly determine whether it constitutes an error event. If so, an error event determination label is output. If not, it is retained as a behavior fluctuation sample.

[0126] The path segment behavior deviation combination with the error event judgment label is used as a feature sample, and feature compression and behavior category mapping are performed to output the error type feature vector.

[0127] The error type feature vector is written into the path generation strategy sample set to participate in the update calculation of the strategy factor in the subsequent path construction;

[0128] In the new round of path construction, the path generation strategy sample set is used as one of the inputs to participate in the initial path constraint generation process in the conversion of standard path to individual path, forming a path candidate structure with dynamic adaptability.

[0129] The dynamically adaptive path candidate structure is returned to the path mapping process as the starting input of the path adaptation phase, and the path adaptation mapping is executed to complete the closed-loop evolution and update of the path adaptation strategy.

[0130] As a further solution, the coupling energy of the path segment behavior characteristics and structural difference attributes is calculated to output the error event label; the path strategy factor is extracted from the error event and the path constructor is updated;

[0131] Coupled integral expression:

[0132]

[0133] Error path segment determination:

[0134] γ k Error segment ε k >ζ

[0135] Path strategy factor generation:

[0136]

[0137] Path generation function update:

[0138]

[0139] Update the path segment representation:

[0140]

[0141] where ε k is the error coupling energy score of the kth path; Ωk is the set of points where the k-th path overlaps in the differential expression structure; Θ k is the trajectory disturbance rate; v i is the difference vector, which represents the displacement vector between the workpiece individual point and the standard point in three-dimensional space; is the coupling function of trajectory disturbance intensity and spatial offset; k is the load response peak rate; κ(p i ) is the structural point p i The local curvature of is the path segment Y k The average curvature reference value on ; is the coupling function of load disturbance and curvature offset; dp i is the integral variable in the difference structure point set, and the integral variable represents the infinitesimal structure point; ζ is the triggering threshold of the error event; Ψ k is the strategic factor vector, used to adjust the path construction; is the behavior structure joint gradient mapping function; T k is the kth segment behavior response feature vector; δ k is the k-th segment difference feature set, which comes from the spatial offset and morphological differences between the workpiece individual and the standard design structure; is the first-order derivative of the behavior index on the k-th path in time; is the rate of change of the k-th segment difference feature in the path space; The path generation function generated for the tth iteration; is the strategy factor adjustment function, which is used to dynamically update the path generation function; is a function composition operator, indicating Synthesis with regulatory functions; Adapt path expressions for individual artifacts generated based on strategy evolution.

[0142] A metal processing error monitoring system based on situational awareness includes a geometric modeling module, a differential expression module, a path adaptation module, a behavior perception module, and a strategy evolution module:

[0143] The geometric modeling module is used to obtain workpiece point cloud data and perform geometric reconstruction and coordinate alignment to form an individual geometric expression structure of the workpiece that can be used to identify metal processing errors, serving as the structural basis for subsequent difference analysis and error source determination;

[0144] The differential expression module is used to perform spatial difference vector solution and differential attribute division between the workpiece individual geometric expression structure and the standard design geometric expression structure, forming a differential expression structure with consistent structure and containing local offset information as the input of path adaptation;

[0145] The path adaptation module is used to perform mapping, interference detection and path correction processes between the standard machining path expression and the differential expression structure, and generate an executable path expression that meets the individual geometric state of the workpiece and the machining constraints, which serves as the input of the subsequent behavior monitoring process;

[0146] The behavior perception module is used to perform structural alignment and disturbance feature extraction on the set of machining behavior variables and the workpiece individual adaptation path, generate a set of behavioral response features corresponding to the path segment, and combine the differential expression structure to determine whether the behavior deviation is concentrated in the high-risk area, and output the path segment behavior deviation combination for error identification;

[0147] The strategy evolution module is used to perform error event judgment and feature vector extraction on the path segment behavior offset combination, write the path generation strategy sample set to participate in the strategy factor update, and use the update results in the subsequent path constraint generation and adaptation process to complete the closed-loop evolution of the path construction strategy.

[0148] The following points require general explanation, including but not limited to: Traditional metal processing systems generally rely on standard design models and fixed processing paths, which are stable in high-volume processing scenarios. However, in small-batch customized processing, due to non-negligible deviations in the individual characteristics of each workpiece, such as slight differences in geometric dimensions, boundary contour deformation, or material inhomogeneity, the traditional path and the actual workpiece geometry are mismatched, resulting in errors that cannot be detected by the system in a timely manner. More seriously, the traditional error monitoring mechanism based on physical signal thresholds is not sensitive to such subtle deviations at the geometric level, resulting in error events being judged as "normal processing" even when they occur, resulting in systematic oversight.

[0149] The goal of this solution is to introduce a unified perception capability for the workpiece geometry, machining path, and machining behavior, and to build a dynamic closed loop from geometry reconstruction to path adaptation to behavior monitoring and strategy evolution, thereby identifying and responding to machining errors.

[0150] This solution includes the collaborative behavior chain construction stage:

[0151] Acquire synchronized behavioral data collected by multiple sensing terminals in the smart park, including card swipe event sequences, facial recognition results, and trajectory path sequences. Through timestamp alignment, spatial location synchronization, and identity semantic fusion, construct joint behavioral features within each time slice. Based on these features, calculate cross-modal behavioral matching, group high-matching segments into an "initial collaborative path set," and further screen "candidate trusted paths" that meet the requirements of terminal response synchronization and device type cross-talk. The intention of this stage is not to use single terminal data as the basis for behavior, but rather to judge based on the structural strength of the collaborative chain and inter-modal consistency, providing a chain-level data foundation for subsequent disturbance judgment.

[0152] This solution includes the disturbance response behavior construction phase:

[0153] For the initially trusted path, the system injects interference requests into the corresponding terminal. Interference methods include not only trajectory perturbations (such as path deviations), but also identity confidence perturbations (simulating misidentification) and trigger timing perturbations (controlling response time differences). By recording the terminal's behavioral feedback trajectory under interference, the response time, behavioral deviation trajectory, and state adjustment frequency of each path during the recovery process are extracted to construct a "disturbance resilience indicator vector group." The key innovation here is that an attacker can synchronously construct a consistent path but cannot truly reproduce the path's resilience under interference conditions. By constructing a perturbation-recovery mapping relationship, it is possible to infer which path behaviors are merely superficially coordinated and lack true behavioral inertia, thereby identifying "elastic response anomaly chains."

[0154] This solution includes the reverse behavior generation path derivation stage:

[0155] For identified abnormal chain paths, the system reversely extracts the behavior generation process and constructs a "time-state trajectory diagram" to analyze whether each node conforms to the natural generation logic. This judgment is made by calculating the transition logic factors between behavior nodes. For example, it determines whether the response time difference between adjacent behaviors falls within the device response window, whether the identities are continuous, and whether the behavior triggers are reasonable. This part uses "generated path inversion" to distinguish between truly natural user behavior chains and false paths spliced ​​by attackers, strengthening the inherent verification of the authenticity of the collaborative path expression.

[0156] This solution includes behavioral trust decay and revalidation phases:

[0157] If a behavior path is identified as a "pseudo-consistency chain," the system doesn't immediately reject it. Instead, it applies a "trust weakening process" to it. Specifically, the path behavior is mapped to the actual task execution record. If no behavior chain supporting its triggering can be found in the non-pseudo-consistency chain, the trust of the pseudo-consistency path is weakened. The system then reinjects low-intensity perturbations and collects its recovery trajectory to further establish a perturbation recovery strength score. This demonstrates that the solution does not rely on a rigid "one-step fatality" strategy, but rather incorporates multiple rounds of dynamic feedback verification, improving the false positive recovery rate and system resilience.

[0158] This solution includes the trust path reconstruction and pseudo-chain marking stages:

[0159] The system performs multiple recovery constraint judgments on the recovery scoring results, including a joint judgment of indicators such as the fluctuation amplitude of the disturbance recovery time, the trend of the behavior deviation residual change and the number of path interruptions; when the path maintains a stable recovery mode within all constraints, it is judged as a "trusted recoverable path" and re-included in the trusted path map; otherwise, a "pseudo-consistency chain risk label" is output for subsequent monitoring systems to alarm or intervene; this stage constitutes the closing mechanism of the system security judgment, ensuring that only paths that have truly experienced the test of interference and restored stability are confirmed as trustworthy, thereby realizing the paradigm shift from "surface collaborative judgment" to "process trust verification".

[0160] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of the present invention should be included in the scope of protection of the present invention.

Claims

1. A metal processing error monitoring method based on situational awareness, characterized in that: include: Acquire workpiece point cloud data and perform geometric reconstruction and coordinate alignment to form an individual workpiece geometric expression structure for identifying metal processing errors, which serves as the structural basis for difference analysis and error source determination; The spatial difference vector is solved and the difference attributes are divided between the workpiece individual geometric expression structure and the standard design geometric expression structure to form a difference expression structure with consistent structure and local offset information as the input of path adaptation; The standard machining path expression and the differential expression structure are mapped, and the interference detection and path correction process are performed to generate an executable path expression that meets the individual geometric state of the workpiece and the machining constraints, which serves as the input for subsequent behavior monitoring. Structural alignment and disturbance feature extraction are performed on the set of machining behavior variables and the workpiece individual adaptation path to generate a set of behavioral response features corresponding to the path segment. The differential expression structure is then combined to determine whether the behavioral deviation is concentrated in the high-risk area, and a path segment behavioral deviation combination is output for error identification. Error event judgment and feature vector extraction are performed on the path segment behavior offset combination, and the path generation strategy sample set is written into the strategy factor update. The updated results are used in the subsequent path constraint generation and adaptation process to complete the closed-loop evolution of the path construction strategy.

2. The metal processing error monitoring method based on situational awareness according to claim 1, characterized in that: In metal processing error monitoring, the original spatial data of the workpiece is obtained through three-dimensional optical point cloud scanning, and density equalization, outlier elimination and boundary connectivity reconstruction are performed on the original spatial data of the workpiece to output the initial spatial expression data; the initial spatial expression data is mapped through surface continuity fitting and normal interpolation, and the surface reconstruction process is executed to output the complete individual geometric expression structure of the workpiece.

3. The metal processing error monitoring method based on situational awareness according to claim 2, characterized in that: Through the rigid registration operation, the coordinates of the workpiece individual geometric expression structure and the standard design geometric expression structure are aligned, and the aligned expression structure in the unified space framework is output; Determine whether there are any local difference segments in the boundary area of ​​the aligned geometric expression structure that cannot be aligned. The local difference segment is determined by whether the spatial offset between the standard design geometric expression structure and the individual workpiece geometric expression structure within the boundary range exceeds the defined local fitting tolerance. If so, execute the boundary enhancement reconstruction process; if not, retain the current geometric expression structure; The aligned geometric expression structure is used as the input of the subsequent differential expression process to form an individual geometric expression structure of the workpiece that is consistent with the standard design geometric expression structure space.

4. The metal processing error monitoring method based on situational awareness according to claim 3 is characterized in that: The workpiece individual geometric expression structure and the standard design geometric expression structure in the unified coordinate system are used as input, the difference vector of the spatial corresponding point set is solved, and the spatial residual field is output; The spatial residual field is divided into three attribute types: position offset, curvature perturbation and boundary morphological variation, and the position difference feature, curvature difference feature and boundary difference feature are output respectively. The position difference feature, curvature difference feature and boundary difference feature are collectively referred to as the difference feature set; Determine whether there are offset distribution blocks in the position difference feature set that continuously exceed the threshold. If so, mark it as a path interference risk area for path correction.

5. The method for monitoring metal processing errors based on situational awareness according to claim 4, characterized in that: Rebind the difference feature set to the topological reference point set of the standard design geometric expression structure, establish a mapping relationship between the point set and the difference attribute, and output a structurally consistent difference expression structure; The differential expression structure is used as the input basis for path adaptation to constrain the path mapping range and tool position generation logic.

6. The method for monitoring metal processing errors based on situational awareness according to claim 5, characterized in that: Taking the standard machining path expression and the differential expression structure as input, the path mapping process is executed to reproject the standard machining path onto the workpiece individual geometric expression structure and output the initial mapping path; Perform workpiece individual geometric expression structure penetration test and boundary reachability judgment on the initial mapping path to obtain the interference area between the path and the differential expression structure; Determine whether the interference area is caused by the combination of position difference features and curvature difference features in the differential expression structure. If so, local reconstruction is performed preferentially on the interference path segment and the path correction candidate is output; if not, it is marked as an abnormal workpiece configuration segment, triggering the path configuration exception processing logic.

7. The method for monitoring metal processing errors based on situational awareness according to claim 6, characterized in that: Perform fairing reconstruction on the path correction candidate on the workpiece individual geometric expression structure, combine the boundary difference features in the differential expression structure to perform boundary preservation constraints, and output the workpiece individual adaptation path; The workpiece individual adaptation path is subjected to the processing accessibility test and machine tool posture solution process to determine whether the posture is feasible and whether the trajectory is continuous. If it is feasible and continuous, the current path is retained. If it is not feasible or there is a trajectory jump, the smoothness reconstruction process is re-entered to perform iterative correction; The path expression body that is finally confirmed to be processable is used as the structural input of the subsequent processing behavior monitoring process.

8. The method for monitoring metal processing errors based on situational awareness according to claim 7, characterized in that: During the machining process, a set of machining behavior variables is collected, including data such as tool load fluctuation, feed acceleration change, and contour position feedback. The set of machining behavior variables is aligned to the workpiece individual adaptation path according to the path segment structure, and the behavior response sequence corresponding to the path segment is output. Perform disturbance pattern recognition on the behavioral response sequence, and output a behavioral response feature set by calculating the trajectory drift rate, fluctuation frequency abnormality, and local load peak change rate of the behavioral response; Determine whether the behavioral response feature set appears concentrated in the high-risk area marked by the differential expression structure corresponding to the path segment. If so, mark it as an error trend segment; if not, determine it as a stable processing segment. The path segments marked with error trend occurrence segments are combined with their behavioral response feature sets to form a path segment behavior offset combination, which is used in the error event identification process.

9. The metal processing error monitoring method based on situational awareness according to claim 8, characterized in that: The path segment behavior offset combination is subjected to a behavior-structure joint deviation calculation. The coupling degree between the trajectory drift rate and the position difference feature, and the correlation strength between the load peak change rate and the curvature difference feature are used to jointly determine whether it constitutes an error event. If so, an error event determination label is output. If not, it is retained as a behavior fluctuation sample. The path segment behavior deviation combination with the error event judgment label is used as a feature sample, and feature compression and behavior category mapping are performed to output the error type feature vector. The error type feature vector is written into the path generation strategy sample set to participate in the update calculation of the strategy factor in the subsequent path construction; In the new round of path construction, the path generation strategy sample set is used as one of the inputs to participate in the initial path constraint generation process in the conversion of standard path to individual path, forming a path candidate structure with dynamic adaptability. The dynamically adaptive path candidate structure is returned to the path mapping process as the starting input of the path adaptation link, and the path adaptation mapping is executed to complete the evolution and update of the path adaptation strategy.

10. A metal processing error monitoring system based on situational awareness, comprising the metal processing error monitoring method based on situational awareness according to claim 9, wherein the system comprises a geometric modeling module, a differential expression module, a path adaptation module, a behavior perception module, and a strategy evolution module, and is characterized in that: The geometric modeling module is used to obtain workpiece point cloud data and perform geometric reconstruction and coordinate alignment to form an individual workpiece geometric expression structure for identifying metal processing errors, which serves as the structural basis for subsequent difference analysis and error source determination; The differential expression module is used to perform spatial difference vector solution and differential attribute division between the workpiece individual geometric expression structure and the standard design geometric expression structure, forming a differential expression structure with consistent structure and containing local offset information as the input of path adaptation; The path adaptation module is used to perform mapping, interference detection and path correction processes between the standard machining path expression and the differential expression structure, and generate an execution path expression that meets the individual geometric state of the workpiece and the machining constraints as the input of the behavior monitoring process; The behavior perception module is used to perform structural alignment and disturbance feature extraction on the set of machining behavior variables and the workpiece individual adaptation path, generate a set of behavioral response features corresponding to the path segment, and combine the differential expression structure to determine whether the behavior deviation is concentrated in the high-risk area, and output the path segment behavior deviation combination for error identification; The strategy evolution module is used to perform error event judgment and feature vector extraction on the path segment behavior offset combination, write the path generation strategy sample set to participate in the strategy factor update, and use the update results in the subsequent path constraint generation and adaptation process to complete the evolution of the path construction strategy.

Images