A kind of automobile mould manufacturing management system of fusion situation awareness mechanism

Abstract

The application discloses a kind of automobile mould manufacturing management systems of fusion situation awareness mechanism, specifically related to the technical field of situation awareness mechanism's automobile mould manufacturing management, including trajectory modeling module for collecting automobile mould historical manufacturing behavior data, generate normalized trajectory and execute trajectory clustering, construct reference trajectory set and behavior change rate set for subsequent comparison;Offset identification module is used to perform time normalization and derivative calculation on current manufacturing behavior data, difference comparison obtains continuous deviation section, and whether it constitutes offset state is judged based on trajectory direction;By constructing the normalized trajectory model and change rate model of manufacturing behavior variable, and based on the behavior difference identification result of current manufacturing state and historical reference trajectory, dynamic compensation control path and update plan trigger condition are generated, so as to solve the core problem that existing manufacturing system cannot identify processing condition offset within the planning time.

Description

An automotive mold manufacturing management system integrating situational awareness mechanisms Technical Field

[0001] This invention relates to the field of automotive mold manufacturing management technology based on situational awareness mechanisms, and more specifically, to an automotive mold manufacturing management system that integrates situational awareness mechanisms. Background Technology

[0002] In existing automotive mold manufacturing management, the manufacturing process is strictly embedded in a time-based preset plan. The control system usually schedules processing instructions and parameter loading logic according to the time nodes in the process plan. For example, in the mold preheating stage, the system sets a fixed time threshold, and when it is reached, it automatically switches to the finishing stage, while loading the corresponding tool compensation value, cooling parameters and thermal expansion correction model. The default time schedule of the mechanism can accurately reflect the evolution of the process state, but ignores the influence of ambient temperature, equipment load, initial material state and other factors on the process response rate during the actual manufacturing process.

[0003] When disturbances occur during the execution of the plan, such as fluctuations in ambient temperature and humidity, interruption of preheating, equipment waiting, task switching, etc., although the triggering time of the processing command is still within the original plan range, the state of the mold body may not meet the requirements of subsequent processing conditions. Under these conditions, continuing to execute the predetermined parameter configuration will lead to problems such as inaccurate thermal compensation, deviation of cooling strategy, and imbalance of structural stress, resulting in irreversible microstructure damage or form and position errors.

[0004] Because the control logic only determines whether the time node has been reached and does not have the ability to perceive the actual execution status of the plan, it cannot identify such process deviations, resulting in the manufacturing system lacking responsiveness to changes in processing conditions within the time planning mechanism;

[0005] This shows that the system makes logically correct processing decisions within the planned time, but the underlying technological conditions no longer hold, resulting in the control system's inability to identify "abnormal states in planned processing". Summary of the Invention

[0006] To overcome the aforementioned deficiencies of the prior art, embodiments of the present invention provide an automotive mold manufacturing management system that integrates a situational awareness mechanism. By constructing a normalized trajectory model and a rate of change model for manufacturing behavior variables, and based on the behavior difference identification results between the current manufacturing state and the historical baseline trajectory, the system dynamically generates compensation control paths and update plan triggering conditions, thereby solving the core problem that existing manufacturing systems cannot identify processing condition deviations within the planned time.

[0007] To achieve the above objectives, the present invention provides the following technical solution: an automotive mold manufacturing management system integrating situational awareness mechanism, including a trajectory modeling module, an offset recognition module, a path compensation module, and a plan update module;

[0008] The trajectory modeling module is used to collect historical manufacturing behavior data of automotive molds, generate normalized trajectories and perform trajectory clustering to build a set of baseline trajectories and a set of behavior change rates for subsequent comparison.

[0009] The offset recognition module is used to perform time normalization and derivative calculation on the current manufacturing behavior data, compare the differences to obtain continuous deviation segments, and determine whether an offset state is constituted based on the trajectory direction.

[0010] The path compensation module generates a compensation trajectory based on the offset state trajectory and matches it with the reference trajectory. It then performs path difference calculations to output the control parameter path, which is used to correct the current control behavior.

[0011] The plan update module is used to extract the final state quality variables and match them with the execution results of the control path to identify combinations of quality improvement intervals and update the plan trigger conditions.

[0012] In a preferred embodiment, the trajectory modeling module is used to collect and extract manufacturing behavior data from the historical manufacturing process of automotive molds, and construct a set of manufacturing behavior variables, which includes temperature trajectory, load trajectory, and cooling trajectory.

[0013] The set of manufacturing behavior variables is grouped according to the manufacturing stage, and time normalization is performed on each group of manufacturing behavior variables to generate stage normalized trajectories.

[0014] The extraction process involves slicing the manufacturing behavior data chronologically to extract the temperature change sequence, spindle load response sequence, and cooling flow record sequence for each time period. Then, for each sequence, extreme value difference, mean slope, and change amplitude calculations are performed over consecutive time periods. The calculation results are then categorized and combined according to the corresponding process segments to form a set of variables describing the temperature trajectory, load trajectory, and cooling trajectory. Finally, all variable sets are bound by stage numbers and time markers to form a set of manufacturing behavior variables.

[0015] The time normalization process maps the original trajectory to a unified time interval, performs interpolation at equally spaced nodes, and generates a normalized trajectory with consistent length and time index. The normalized trajectory is then used as the stage normalized trajectory corresponding to the current manufacturing stage.

[0016] In a preferred embodiment, the trajectory modeling module is also used to perform clustering operations on the stage-normalized trajectories to construct a set of baseline trajectories for situational awareness.

[0017] Perform a type of derivative operation on each trajectory in the baseline trajectory set to form a set of behavior change rates;

[0018] The clustering operation unfolds each stage-normalized trajectory into a sequence of equal-length vectors in chronological order; it then performs an accumulation operation on the absolute values ​​of the differences at corresponding time nodes for each equal-length vector sequence to construct a full distance matrix between trajectories; within the full distance matrix, trajectories are classified according to the standard that the average distance does not exceed a preset merging threshold, and all trajectories that meet the preset merging threshold standard are grouped into the same cluster group; within each cluster group, the average distance value between each trajectory and the other trajectories in the group is calculated, and trajectories with average distance values ​​close to the average level of their respective groups are selected as representative trajectories; finally, all representative trajectories are combined to form a baseline trajectory set.

[0019] The aforementioned derivative operation involves subtracting the variable values ​​corresponding to two adjacent time indices in chronological order within each baseline trajectory, then dividing the difference obtained after the subtraction operation by the corresponding time step to obtain the unit change value within the current time interval. This operation is repeated segment by segment along the entire length of the trajectory to obtain the continuous rate sequence of the trajectory. The calculation process for obtaining the continuous rate sequence of the trajectory is performed on all trajectories in the baseline trajectory set, generating the corresponding rate sequence for each trajectory. The rate sequences of all trajectories are merged and arranged according to the trajectory number, and the output is a set of behavioral change rates.

[0020] In a preferred embodiment, the offset recognition module is used to collect manufacturing behavior data in the current manufacturing process of automobile molds and extract the same variables as the manufacturing behavior variable set to construct the current manufacturing behavior variable set.

[0021] Perform time normalization and second derivative operations on the current set of manufacturing behavior variables to generate a set of current behavior change rates;

[0022] The second type of derivative operation generates a unit time rate of change sequence by calculating the ratio of the difference of variable values ​​to the time step at adjacent time index positions for each normalized trajectory in the current trajectory set; all rate sequences are merged according to trajectory number to form the current rate of change set.

[0023] In a preferred embodiment, the offset recognition module is further configured to compare the current set of behavior change rates with the set of behavior change rates by difference, and output continuous deviation segments.

[0024] The system judges whether there are continuous deviation sections and whether the cooling trajectory and temperature trajectory are opposite. These are the two conditions for judgment. If both conditions are met, the deviation trajectory is output; otherwise, the current trajectory state is maintained.

[0025] The difference comparison is achieved by aligning the current set of behavior change rates with the set of behavior change rates by time index, and calculating the difference between the two at each index position; then, the difference is slid over time to generate a fixed-length time window, and the number of data points whose difference exceeds a preset deviation threshold is counted within each time window; the time period corresponding to the window in which the number of consecutive data points exceeds the preset threshold counting standard is identified as a continuous deviation segment.

[0026] In a preferred embodiment, the path compensation module performs state extraction on the current manufacturing behavior variable set based on the output offset trajectory, and constructs the offset state trajectory.

[0027] Perform a line-by-line difference calculation between the offset trajectory and the set of reference trajectories, and output candidate compensation trajectories;

[0028] The state extraction is achieved by locating the time period that matches the time index marked on the offset trajectory in the current set of manufacturing behavior variables; performing time index truncation operations on the temperature trajectory, load trajectory, and cooling trajectory in sequence to extract the variable value sequence of the corresponding interval; combining the truncated variable value sequence into trajectory segments in chronological order; and concatenating the three trajectory segments of temperature, load, and cooling under the same time index according to the variable dimension to output the offset state trajectory.

[0029] In a preferred embodiment, the path compensation module is further used to judge the candidate compensation trajectory, and to determine whether there is a trajectory in the candidate compensation trajectory whose cooling trajectory is in the same direction as the load trajectory. If there is, the trajectory whose cooling trajectory is in the same direction as the load trajectory is used as the compensation trajectory; otherwise, the trajectory whose temperature trajectory hysteresis response amplitude is less than the preset temperature trajectory hysteresis response amplitude threshold is selected from the candidate compensation trajectories as the compensation trajectory.

[0030] The difference between the compensated trajectory and the offset state trajectory is calculated, and the control parameter path driven by situational awareness is output.

[0031] In a preferred embodiment, the planning update module is used to extract mold detection data from the compensation path and construct a set of final state quality variables;

[0032] The final state quality variable set is matched with the control parameter path driven by situational awareness to generate a quality feedback correspondence.

[0033] The mold detection data extraction process involves collecting final-state temperature field distribution data, residual stress distribution data, and forming geometric deviation data along the mold surface and forming area after the compensation path execution process is completed. Spatial reconstruction, time alignment, and variable discretization mapping operations are performed on the collected data at preset spatial points and detection time points. Spatial location, time index, and measured value are combined into a set of triplets, which are then classified according to variable type and uniformly categorized into the quality dimension mapping index to form a final-state quality variable set.

[0034] The response matching process involves performing a normalized index matching operation on each quality variable in the final state quality variable set in the time dimension to locate its corresponding normalized time period in the manufacturing process. Then, within the located normalized time period, a truncation operation is performed on each control variable in the control parameter path to extract the corresponding control parameter sequence. Subsequently, each quality variable and its corresponding control parameter sequence are combined to form candidate variable pairs. For each candidate variable pair, slope consistency judgment, fluctuation phase delay measurement, and response sensitivity differential estimation are performed to screen out a set of variable pairs that meet the requirements of consistent direction, controllable hysteresis, and non-zero response. All variable pairs in the variable pair set that meet the conditions are mapped to a one-to-one corresponding response relationship, and the overall output is a quality feedback correspondence.

[0035] In a preferred embodiment, the plan update module is further used to determine the correspondence between quality feedback and whether there is a combination of intervals for changes in cooling trajectory and load trajectory corresponding to quality improvement. If there is, the combination of intervals for changes in cooling trajectory and load trajectory corresponding to quality improvement is extracted. If there is no, the original plan triggering conditions are maintained.

[0036] The combined processing of interval combinations and stage-normalized trajectories constitutes the plan triggering conditions.

[0037] The technical effects and advantages of this invention are as follows:

[0038] 1. This solution identifies whether manufacturing behavior deviates from the baseline trajectory based on the set of behavior change rates and the comparison of differences, thus avoiding mismatch between planned time and actual state;

[0039] 2. By grouping and normalizing the trajectory during the manufacturing stage and clustering it, a set of benchmark trajectories for comparison is generated to achieve standard state representation;

[0040] 3. Perform sliding difference window calculation on the current set of change rates, and combine it with the trajectory direction relationship to determine whether an offset state is constituted, thereby improving the robustness and reliability of offset perception;

[0041] 4. In offset state, through candidate compensation trajectory filtering, output control parameter paths including temperature, load, and cooling to ensure continuous and stable manufacturing process;

[0042] 5. By combining mold inspection data with control paths, response matching is established, quality improvement interval combinations are extracted, and plan trigger conditions are updated to improve overall process management and adaptability. Attached Figure Description

[0043] Figure 1 is a schematic diagram of the system modules of the present invention;

[0044] Figure 2 is a flowchart of the manufacturing behavior variable modeling execution process in the system of the present invention;

[0045] Figure 3 is a flowchart of the offset trajectory recognition and judgment execution process of the system in this invention;

[0046] Figure 4 is a flowchart of the compensation trajectory generation and control path construction execution process of the system in this invention.

[0047] Figure 5 is a flowchart of the quality feedback identification and plan triggering adjustment execution process of the system in this invention. Detailed Implementation

[0048] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0049] Referring to Figures 1-5 in the specification, an embodiment of the present invention provides an automotive mold manufacturing management system that integrates a situational awareness mechanism, including a trajectory modeling module, an offset recognition module, a path compensation module, and a plan update module.

[0050] The trajectory modeling module is used to collect historical manufacturing behavior data of automotive molds, generate normalized trajectories and perform trajectory clustering to build a set of baseline trajectories and a set of behavior change rates for subsequent comparison.

[0051] The offset recognition module is used to perform time normalization and derivative calculation on the current manufacturing behavior data, compare the differences to obtain continuous deviation segments, and determine whether an offset state is constituted based on the trajectory direction.

[0052] The path compensation module generates a compensation trajectory based on the offset state trajectory and matches it with the reference trajectory. It then performs path difference calculations to output the control parameter path, which is used to correct the current control behavior.

[0053] The plan update module is used to extract the final state quality variables and match them with the execution results of the control path to identify combinations of quality improvement intervals and update the plan trigger conditions.

[0054] It should be noted that in the formula structure involved in this scheme, dimensionless terms can be used as proportional or structural adjustment factors. When combined with quantities with units, they only play a role in numerical scaling and do not introduce new physical dimensions. Therefore, they will not change or confuse the overall unit system. This combination of "dimensionless terms and terms with units" can be understood as a composite structural expression commonly used in mathematical physics modeling. It conforms to the principle of dimensional consistency and has a clear physical interpretation basis.

[0055] 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 form a unified structure through function mapping, ratio combination or normalization adjustment, with clear units and clear meaning. The overall expression conforms to the principle of dimensional consistency and the conventional formula of engineering modeling.

[0056] In this scheme, constants, weights, adjustment factors, threshold parameters, proportional coefficients, etc., are all adjustable control parameters for different application environments. Their values ​​depend on the target equipment configuration, data input characteristics, and performance optimization goals. During the implementation phase, they are converged within a reasonable range through model verification, performance constraints, or engineering calibration. Although these parameters do not have a unique preset value, they have clear adjustment logic and calculation paths. They belong to the deterministic setting process in engineering implementation. The purpose of this setting is to ensure that the scheme is both universally adaptable and reproducible and operable, without affecting its technical clarity and feasibility.

[0057] The trajectory modeling module is used to collect and extract manufacturing behavior data from the historical manufacturing process of automotive molds, and to construct a set of manufacturing behavior variables, which includes temperature trajectory, load trajectory, and cooling trajectory.

[0058] The set of manufacturing behavior variables is grouped according to the manufacturing stage, and time normalization is performed on each group of manufacturing behavior variables to generate stage normalized trajectories. Grouping according to the manufacturing stage means that each trajectory data in the set of manufacturing behavior variables is classified according to its specific manufacturing stage (such as roughing, finishing, heat treatment, etc.), so that each group of data corresponds to only a single stage, to ensure that the subsequent normalization and modeling steps are comparable and logically consistent within the same manufacturing stage.

[0059] The extraction process involves slicing the manufacturing behavior data chronologically to extract the temperature change sequence, spindle load response sequence, and cooling flow record sequence for each time period. Then, for each sequence, extreme value difference, mean slope, and change amplitude calculations are performed over consecutive time periods. The calculation results are then categorized and combined according to the corresponding process segments to form a set of variables describing the temperature trajectory, load trajectory, and cooling trajectory. Finally, all variable sets are bound by stage numbers and time markers to form a set of manufacturing behavior variables.

[0060] The time normalization process maps the original trajectory to a unified time interval, performs interpolation at equally spaced nodes, and generates a normalized trajectory with consistent length and time index. The normalized trajectory is then used as the stage normalized trajectory corresponding to the current manufacturing stage.

[0061] The trajectory modeling module is also used to perform clustering operations on stage-normalized trajectories to construct a set of baseline trajectories for situational awareness;

[0062] Perform a type of derivative operation on each trajectory in the baseline trajectory set to form a set of behavior change rates;

[0063] The clustering operation unfolds each stage-normalized trajectory into a sequence of equal-length vectors in chronological order; it then performs an accumulation operation on the absolute values ​​of the differences at corresponding time nodes for each equal-length vector sequence to construct a full distance matrix between trajectories; within the full distance matrix, trajectories are classified according to the standard that the average distance does not exceed a preset merging threshold, and all trajectories that meet the preset merging threshold standard are grouped into the same cluster group; within each cluster group, the average distance value between each trajectory and the other trajectories in the group is calculated, and trajectories with average distance values ​​close to the average level of their respective groups are selected as representative trajectories; finally, all representative trajectories are combined to form a baseline trajectory set.

[0064] The aforementioned derivative operation involves subtracting the variable values ​​corresponding to two adjacent time indices in chronological order within each baseline trajectory, then dividing the difference obtained after the subtraction operation by the corresponding time step to obtain the unit change value within the current time interval. This operation is repeated segment by segment along the entire length of the trajectory to obtain the continuous rate sequence of the trajectory. The calculation process for obtaining the continuous rate sequence of the trajectory is performed on all trajectories in the baseline trajectory set, generating the corresponding rate sequence for each trajectory. The rate sequences of all trajectories are merged and arranged according to the trajectory number, and the output is a set of behavioral change rates.

[0065] The offset recognition module is used to collect manufacturing behavior data in the current manufacturing process of automotive molds and extract the same variables as the manufacturing behavior variable set to construct the current manufacturing behavior variable set.

[0066] Perform time normalization and second derivative operations on the current set of manufacturing behavior variables to generate a set of current behavior change rates;

[0067] Define the set of current behavior change rates

[0068]

[0069] The second type of derivative operations includes:

[0070]

[0071] in The difference between two adjacent time points is divided by the time difference to obtain the unit rate of change over each time interval; Used for direction determination, extracting positive or negative features of variable trends;

[0072] in To generate behavioral variable v in trajectory number i, time interval {t} k , t k+1 The unit rate of change on}; The direction information is for the unit rate of change; i is the trajectory number, which is used to represent different trajectory instances of the same type of manufacturing behavior variable; v represents the manufacturing behavior variable, which is used to take values ​​T (temperature trajectory), L (load trajectory), and C (cooling trajectory); t k t is the normalized time scale value corresponding to index number k; t is the time scale value of the normalized trajectory; k is the discrete time index number in the normalized time series; This represents the normalized time scale value corresponding to the index k of trajectory number i under manufacturing behavior variable v in the set of manufacturing behavior variables; the symbol sign indicates that the positive and negative directions of the value are encoded.

[0073] The second type of derivative operation generates a unit time rate of change sequence by calculating the ratio of the difference of variable values ​​to the time step at adjacent time index positions for each normalized trajectory in the current trajectory set; all rate sequences are merged according to trajectory number to form the current rate of change set.

[0074] The offset recognition module is also used to compare the difference between the current set of behavior change rates and the set of behavior change rates, and output the continuous deviation segment;

[0075] The system judges whether there are continuous deviation sections and whether the cooling trajectory and temperature trajectory are opposite. These are the two conditions for judgment. If both conditions are met, the deviation trajectory is output; otherwise, the current trajectory state is maintained.

[0076] The difference comparison is achieved by aligning the current set of behavior change rates with the set of behavior change rates by time index, and calculating the difference between the two at each index position; then, the difference is slid over time to generate a fixed-length time window, and the number of data points whose difference exceeds a preset deviation threshold is counted within each time window; the time period corresponding to the window in which the number of consecutive data points exceeds the preset threshold counting standard is identified as a continuous deviation segment.

[0077] The path compensation module extracts the state of the current manufacturing behavior variable set when the offset trajectory is output, and constructs the offset state trajectory.

[0078] Perform line-by-line difference calculations between the offset trajectory and the reference trajectory set, and output candidate compensation trajectories; define the candidate compensation trajectories. in and This indicates that the difference between the offset trajectory and the set of reference trajectories is calculated line by line.

[0079]

[0080] Where g s Candidate compensation trajectories for trajectory number s; E s E represents the difference matching error between the candidate compensation trajectory with trajectory number s and the offset state trajectory; s represents the number of one of the candidate compensation trajectories; j represents the trajectory number traversed through all candidate compensation trajectories; j The difference matching error between the candidate compensation trajectory with trajectory number j and the offset state trajectory; ∈ represents the error nearest neighbor threshold, which is used to control the range of candidate trajectories; symbol For minimization operator; s (T) (t k ) represents the normalized time scale value of the temperature trajectory in the offset state trajectory at index number k; s (L) (t k ) represents the normalized time scale value of the load trajectory at index number k in the offset state trajectory; s (C) (t k The value is the normalized time scale value of the cooling trajectory at index number k in the offset state trajectory. The normalized time scale value of the temperature trajectory in the baseline trajectory set at index number k; The normalized time scale value of the load trajectory in the baseline trajectory set at index number k; The normalized time scale value of the cooling trajectory in the baseline trajectory set at index number k;

[0081] The state extraction is achieved by locating the time period that matches the time index marked on the offset trajectory in the current set of manufacturing behavior variables; performing time index truncation operations on the temperature trajectory, load trajectory, and cooling trajectory in sequence to extract the variable value sequence of the corresponding interval; combining the truncated variable value sequence into trajectory segments in chronological order; and concatenating the three trajectory segments of temperature, load, and cooling under the same time index according to the variable dimension to output the offset state trajectory.

[0082] The path compensation module is also used to judge the candidate compensation trajectory. It determines whether there is a trajectory in the candidate compensation trajectory whose cooling trajectory is in the same direction as the load trajectory. If there is, the trajectory whose cooling trajectory is in the same direction as the load trajectory is used as the compensation trajectory. Otherwise, the trajectory whose temperature trajectory hysteresis response amplitude is less than the preset temperature trajectory hysteresis response amplitude threshold is selected from the candidate compensation trajectories as the compensation trajectory.

[0083] The difference between the compensation trajectory and the offset state trajectory is calculated, and the control parameter path driven by situational awareness is output.

[0084] Define the control parameter path driven by situational awareness Where c (v) (t k )-s (v) (t k This indicates that the difference between the compensated trajectory and the offset state trajectory will be calculated.

[0085]

[0086] P (v) (t k )=(c (v) (t k )-s (v) (t k ))·θ (v) (t k )

[0087]

[0088] Where P (v) (t k P represents the normalized time scale value of the manufacturing behavior variable v at index k; (T) (t k ) represents the normalized time scale value of the temperature trajectory at index number k; P (L) (t k ) represents the normalized time scale value of the load trajectory at index number k; P (C) (t k ) represents the normalized time scale value of the cooling trajectory at index number k; c(v) (t k ) represents the normalized time scale value of the manufacturing behavior variable v in the candidate compensation trajectory at index k; s (v) (t k ) represents the normalized time scale value corresponding to index number k in the offset state trajectory; θ (v) (t k ) is the moderating factor, which represents the degree of response of the manufacturing behavior variable v to the intensity of control; θ (C) (t k ) represents the normalized adjustment factor of the temperature change rate on the cooling change rate, which indicates the dependence weight of cooling behavior on temperature behavior; d (T) (t k ) represents the normalized time scale value of the temperature trajectory at index number k; d (C) (t k ) represents the normalized time scale value of the cooling trajectory at index number k.

[0089] The plan update module is used to extract mold inspection data from the compensation path and construct a set of final-state quality variables;

[0090] The final state quality variable set is matched with the control parameter path driven by situational awareness to generate a quality feedback correspondence.

[0091] The mold detection data extraction process involves collecting final-state temperature field distribution data, residual stress distribution data, and forming geometric deviation data along the mold surface and forming area after the compensation path execution process is completed. Spatial reconstruction, time alignment, and variable discretization mapping operations are performed on the collected data at preset spatial points and detection time points. Spatial location, time index, and measured value are combined into a set of triplets, which are then classified according to variable type and uniformly categorized into the quality dimension mapping index to form a final-state quality variable set.

[0092] The response matching process involves performing a normalized index matching operation on each quality variable in the final state quality variable set in the time dimension to locate its corresponding normalized time period in the manufacturing process. Then, within the located normalized time period, a truncation operation is performed on each control variable in the control parameter path to extract the corresponding control parameter sequence. Subsequently, each quality variable and its corresponding control parameter sequence are combined to form candidate variable pairs. For each candidate variable pair, slope consistency judgment, fluctuation phase delay measurement, and response sensitivity differential estimation are performed to screen out a set of variable pairs that meet the requirements of consistent direction, controllable hysteresis, and non-zero response. All variable pairs in the variable pair set that meet the conditions are mapped to a one-to-one corresponding response relationship, and the overall output is a quality feedback correspondence.

[0093] The plan update module is also used to determine the correspondence between quality feedback and the interval combination of cooling trajectory and load trajectory changes corresponding to quality improvement. If it exists, the interval combination of cooling trajectory and load trajectory changes corresponding to quality improvement is extracted. If it does not exist, the original plan triggering conditions are maintained. The interval combination of cooling trajectory and load trajectory changes corresponding to quality improvement refers to aligning the change interval of the quality feedback variable with the change interval of the cooling trajectory and load trajectory in time. Then, for each overlapping time period after time alignment, it is determined whether the value of the quality feedback variable is within the system's preset improvement magnitude threshold range, and whether the changes of the cooling trajectory and load trajectory meet the conditions of consistent direction or linkage. If all three conditions are met simultaneously in the same time period, then the time period is considered to constitute a quality improvement interval combination.

[0094] The interval combination and the stage normalized trajectory are jointly processed to form the plan triggering condition, which replaces the original plan triggering condition.

[0095] Define plan trigger conditions

[0096]

[0097] Where [t] a ,t b [R] represents a continuous time interval within the target evaluation phase; ij (t) represents the dynamic value of the feedback response intensity on the normalized trajectory at time scale t; S (C) (t) represents the rate of change of the cooling trajectory on the normalized trajectory's time scale value t; D (L) (t) represents the direction of change of the load trajectory on the time scale value t of the normalized trajectory; G z ′(t) is the perturbation derivative of the stage normalized trajectory, which represents the dynamic degree of perturbation in the global manufacturing stage; η,ζ are system thresholds, which are used to define the control response and perturbation intensity; the symbol ∧ means that "and" or "and" must be satisfied at the same time for the overall logical judgment to be valid.

[0098] It should be noted that the traditional automotive mold manufacturing process is based on static time-planning control logic, which automatically triggers parameter loading and process switching according to a preset time schedule. This mechanism assumes that the manufacturing process is stable, but ignores the actual process response lag caused by fluctuations in ambient temperature, changes in equipment load, and differences in material condition.

[0099] When the plan is executed on time, the mold may not have been preheated, cooled sufficiently, or reached load balance, leading to a decrease in subsequent finishing accuracy or abnormal thermal expansion behavior of the mold, resulting in structural damage or geometric deviations. This type of "planned process mismatch" is a significant hidden danger that is difficult to identify in time-oriented manufacturing systems.

[0100] Therefore, a manufacturing management solution integrating situational awareness mechanism needs to be constructed so that the system can perceive the deviation of manufacturing status in real time before executing the plan, dynamically correct the control path according to the deviation, and finally feed back the correction results to optimize future plans.

[0101] This solution includes a trajectory modeling phase:

[0102] By collecting historical manufacturing process data, three types of manufacturing behavior variables—temperature, load, and cooling—are extracted, categorized by manufacturing stage, and normalized over time to form stage-normalized trajectories. Subsequently, cluster analysis is performed on the stage-normalized trajectories to construct a set of baseline trajectories that reflect the evolution characteristics of standard operating conditions. Then, derivative operations are performed on the trajectories within the set of baseline trajectories to extract the dynamic behavior of each variable in the trajectory over time, forming a set of behavior change rates.

[0103] This stage is used to build a foundation of multi-dimensional manufacturing behavior trajectory data that is structurally consistent, scale-uniform, and highly comparable, to support subsequent offset identification and status judgment.

[0104] This solution includes an offset identification stage:

[0105] By collecting current manufacturing process data and extracting three types of variables corresponding to the historical variable set, a current manufacturing behavior variable set is generated. The current manufacturing behavior variable set is first normalized by time and the derivative is calculated to generate a current behavior change rate set. Then, the difference is compared with the behavior change rate set according to the time index, and the continuous over-threshold time window of the rate difference is extracted to form a continuous deviation segment. If there is a continuous deviation segment in the current behavior trajectory, and the cooling trajectory is opposite to the temperature trajectory, it is judged as an off-trajectory state.

[0106] This stage is used to sense whether the evolution of manufacturing behavior deviates from the standard evolution path, trigger the identification of the deviation state, and provide a basis for judgment for the compensation logic;

[0107] This solution includes a path compensation phase:

[0108] Trajectory data within the offset time period is extracted from the current manufacturing behavior variables to form the offset state trajectory; the offset state trajectory and the reference trajectory set are subtracted to generate a candidate compensation trajectory set; if there is a trajectory in the candidate set that is consistent with the cooling and load directions, it is selected first; otherwise, the trajectory with a hysteresis response amplitude less than the threshold is selected as the compensation trajectory; finally, the compensation trajectory and the offset trajectory are subtracted point by point to construct the control parameter path driven by situational awareness.

[0109] This stage generates a multivariable control path based on the trajectory offset state, forming a dynamic control response to ensure real-time adjustment of the manufacturing process.

[0110] This plan includes a planning update phase:

[0111] After the control path is executed, mold detection data is collected, spatial variable distribution is reconstructed, and a final state quality variable set is formed. The final state quality variable set is matched with the control parameter path execution process to generate a quality feedback correspondence. If there is a time period combination where the change of control variables is consistent with the improvement direction of quality variables, and the improvement magnitude reaches the threshold, it is extracted as a quality improvement interval combination. The quality improvement interval combination is analyzed together with the normalized trajectory to form a new plan trigger condition, replacing the original static time node.

[0112] This stage utilizes control execution feedback to form a reverse perception mechanism, enabling proactive adjustments of the manufacturing status to the planning logic, thereby addressing the "in-plan mismatch" problem at its root.

[0113] The above description is merely 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 within the protection scope of the present invention.

Claims

1. An automotive mold manufacturing management system integrating situational awareness mechanisms, comprising a trajectory modeling module, an offset recognition module, a path compensation module, and a plan update module, characterized in that: The trajectory modeling module is used to collect historical manufacturing behavior data of automotive molds, generate normalized trajectories and perform trajectory clustering to build a set of baseline trajectories and a set of behavior change rates for subsequent comparison. The offset identification module performs time normalization and derivative calculation on the current manufacturing behavior data, compares the differences to obtain continuous deviation segments, and determines whether an offset state is constituted based on the trajectory direction; the path compensation module generates a compensation trajectory based on the offset state trajectory and matches the baseline trajectory, performs path difference calculation to output the control parameter path, and corrects the current control behavior; the plan update module extracts the final state quality variables and performs response matching with the control path execution results, identifies the combination of quality improvement intervals, and updates the plan trigger conditions. By collecting historical manufacturing process data, three types of manufacturing behavior variables—temperature, load, and cooling—are extracted, categorized by manufacturing stage, and normalized over time to form stage-normalized trajectories. Subsequently, cluster analysis is performed on the stage-normalized trajectories to construct a set of baseline trajectories that reflect the evolution characteristics of standard operating conditions. Then, derivative operations are performed on the trajectories within the set of baseline trajectories to extract the dynamic behavior of each variable in the trajectory over time, forming a set of behavior change rates. By collecting current manufacturing process data and extracting three types of variables corresponding to the historical variable set, a current manufacturing behavior variable set is generated; The current set of manufacturing behavior variables is first normalized by time and the derivative is calculated to generate the current behavior change rate set. Then, the difference is compared with the behavior change rate set according to the time index. The continuous over-threshold time window of the rate difference is extracted to form a continuous deviation segment. If there is a continuous deviation segment in the current behavior trajectory and the cooling trajectory is opposite to the temperature trajectory, it is judged as an off-trajectory state. Trajectory data within the offset time period is extracted from the current manufacturing behavior variables to form the offset state trajectory; the offset state trajectory and the baseline trajectory set are used to perform a difference operation to generate a candidate compensation trajectory set. If a trajectory exists in the candidate set that aligns with the direction of cooling and load, it should be selected first. Otherwise, the trajectory with a lag response amplitude less than the threshold is selected as the compensation trajectory; finally, the compensation trajectory and the offset trajectory are subjected to point-by-point difference calculation to construct the control parameter path driven by situational awareness; after the control path is executed, mold detection data is collected to reconstruct the spatial variable distribution and form the final state quality variable set; the final state quality variable set is matched with the control parameter path execution process to generate a quality feedback correspondence; if there is a time period combination where the change of control variables is consistent with the improvement direction of quality variables, and the improvement magnitude reaches the threshold, it is extracted as a quality improvement interval combination; the quality improvement interval combination is analyzed together with the normalized trajectory to form a new plan trigger condition to replace the original static time node.

2. The automotive mold manufacturing management system integrating situational awareness mechanism according to claim 1, characterized in that: The trajectory modeling module is used to collect and extract manufacturing behavior data from the historical manufacturing process of automotive molds, and to construct a set of manufacturing behavior variables, which includes temperature trajectory, load trajectory, and cooling trajectory. The manufacturing behavior variable set is grouped according to manufacturing stages, and time normalization is performed on each group of manufacturing behavior variables to generate stage-normalized trajectories. The extraction process involves slicing the manufacturing behavior data in chronological order to extract the temperature value change sequence, spindle load response sequence, and cooling flow record sequence within each time period. Then, extreme value difference calculation, mean slope calculation, and change amplitude calculation are performed on each sequence within a continuous time period. The calculation results are then categorized and combined according to the corresponding process segment to form a variable set for describing the temperature trajectory, load trajectory, and cooling trajectory. All variable sets are bound by stage number and time marker to form the manufacturing behavior variable set. The time normalization process maps the original trajectory to a unified time interval and performs interpolation at equally spaced nodes to generate a normalized trajectory with consistent length and unified time index. The normalized trajectory is used as the stage-normalized trajectory corresponding to the current manufacturing stage.

3. The automotive mold manufacturing management system integrating situational awareness mechanism according to claim 2, characterized in that: The trajectory modeling module is also used to perform clustering operations on stage-normalized trajectories to construct a set of baseline trajectories for situational awareness; perform a type of derivative operation on each trajectory in the set of baseline trajectories to form a set of behavior change rates; the clustering operation expands each stage-normalized trajectory into a sequence of equal-length vectors in chronological order; performs an accumulation operation on the absolute value of the difference at the corresponding time node on each sequence of equal-length vectors to construct a full distance matrix between trajectories; and classifies trajectories in the full distance matrix according to the standard that the average distance does not exceed a preset merging threshold, and groups all trajectories that meet the preset merging threshold into the same cluster group; Within each cluster, the average distance between each trajectory and the other trajectories in the group is calculated, and the trajectory whose average distance is close to the average level of its group is selected as the representative trajectory. All representative trajectories are combined to form a baseline trajectory set. The first type of derivative operation is performed by subtracting the variable values ​​corresponding to two adjacent time indices in chronological order in each baseline trajectory, and then dividing the difference obtained after the subtraction operation by the corresponding time step to obtain the unit change value within the current time interval. This operation is repeated segment by segment along the entire length of the trajectory to obtain the continuous rate sequence of the trajectory. The calculation process for obtaining the continuous rate sequence of the trajectory is performed on all trajectories in the baseline trajectory set to generate the corresponding rate sequence for each trajectory. The rate sequences of all trajectories are merged and arranged according to the trajectory number, and the output is a set of behavioral change rates.

4. The automotive mold manufacturing management system integrating situational awareness mechanism according to claim 3, characterized in that: The offset recognition module is used to collect manufacturing behavior data in the current manufacturing process of automotive molds and extract variables that are the same as those in the manufacturing behavior variable set to construct the current manufacturing behavior variable set; time normalization and second derivative operation are performed on the current manufacturing behavior variable set to generate the current behavior change rate set; the second derivative operation calculates the ratio of the difference of variable values ​​to the time step size at adjacent time index positions for each normalized trajectory in the current trajectory set to generate a unit time change rate sequence; all rate sequences are merged according to trajectory number to form the current change rate set.

5. The automotive mold manufacturing management system integrating situational awareness mechanism according to claim 4, characterized in that: The offset recognition module is also used to compare the difference between the current set of behavior change rates and the set of behavior change rates, and output the continuous deviation segment; The system judges whether there are continuous deviation segments and whether the cooling trajectory and temperature trajectory are opposite in direction. These are the two conditions for judgment. If both conditions are met, the offset trajectory is output; otherwise, the current trajectory state is maintained. The difference comparison is performed by aligning the current behavior change rate set and the behavior change rate set by time index and calculating the difference between the two at each index position. Then, the difference is slid over time to generate a fixed-length time window. Within each time window, the number of data points whose difference exceeds a preset deviation threshold is counted. The time period corresponding to the window where the number of consecutive data points exceeds the preset threshold counting standard is identified as a continuous deviation segment.

6. The automotive mold manufacturing management system integrating situational awareness mechanism according to claim 5, characterized in that: The path compensation module performs state extraction on the current manufacturing behavior variable set when the offset trajectory is output, and constructs the offset state trajectory; it performs line-by-line difference calculation between the offset state trajectory and the reference trajectory set, and outputs candidate compensation trajectories; the state extraction is performed by locating the time period in the current manufacturing behavior variable set that is consistent with the time index marked by the offset trajectory; the time index truncation operation is performed on the temperature trajectory, load trajectory and cooling trajectory in sequence to extract the variable value sequence of the corresponding interval; the truncated variable value sequence is combined into trajectory segments according to the time order; the three trajectory segments of temperature, load and cooling are spliced ​​together according to the variable dimension under the same time index, and the output is the offset state trajectory.

7. The automotive mold manufacturing management system integrating situational awareness mechanism according to claim 6, characterized in that: The path compensation module is also used to judge the candidate compensation trajectory, and to determine whether there is a trajectory in the candidate compensation trajectory whose cooling trajectory and load trajectory are in the same direction. If so, the trajectory whose cooling trajectory and load trajectory are in the same direction is selected as the compensation trajectory. Otherwise, the trajectory whose temperature trajectory hysteresis response amplitude is less than the preset temperature trajectory hysteresis response amplitude threshold is selected from the candidate compensation trajectories as the compensation trajectory. The difference between the compensation trajectory and the offset state trajectory is calculated, and the control parameter path driven by situational awareness is output.

8. The automotive mold manufacturing management system integrating situational awareness mechanism according to claim 7, characterized in that: The planning update module is used to extract mold detection data from the compensation path and construct a set of final state quality variables; the set of final state quality variables is matched with the control parameter path driven by situational awareness to generate a corresponding quality feedback relationship. The mold detection data extraction involves collecting final-state temperature field distribution data, residual stress distribution data, and forming geometric deviation data along the mold surface and forming area after the compensation path execution process. Spatial reconstruction, time alignment, and variable discretization mapping operations are performed on the collected data at preset spatial points and detection time points. The spatial location, time index, and measured value form a triplet set, which is then categorized according to variable type and uniformly assigned to the quality dimension mapping index, integrating to form a final-state quality variable set. The response matching involves performing a normalized index matching operation on the time dimension for each quality variable in the final-state quality variable set to locate its corresponding normalized time period in the manufacturing process. Then, within the located normalized time period, a truncation operation is performed on each control variable in the control parameter path to extract the corresponding control parameter sequence. Subsequently, each quality variable and its corresponding control parameter sequence are combined to form candidate variable pairs. For each candidate variable pair, variable slope consistency judgment, fluctuation phase delay measurement, and response sensitivity differential estimation are performed to screen out a set of variable pairs that meet the requirements of consistent direction, controllable hysteresis, and non-zero response. Map all variable pairs that meet the conditions in the variable pair set to a one-to-one correspondence of response relationships, and the overall output is the quality feedback correspondence.

9. The automotive mold manufacturing management system integrating situational awareness mechanism according to claim 7, characterized in that: The plan update module is also used to determine the correspondence between quality feedback and whether there is a combination of intervals for changes in cooling trajectory and load trajectory corresponding to quality improvement. If there is, the combination of intervals for changes in cooling trajectory and load trajectory corresponding to quality improvement is extracted. If there is no such combination, the original plan triggering conditions are maintained. The interval combination and the stage normalized trajectory are jointly processed to form the plan triggering conditions.

Images