Intelligent detection method for injection molding defects of bumper

Abstract

The invention provides a bumper injection molding defect intelligent detection method, and relates to the technical field of automobile part manufacturing and detection.The method comprises the steps that firstly, a multi-dimensional energy parameter set, a mold cavity form data set and a defect form data set of a bumper entity to be detected after molding in the bumper injection molding process are collected; constructing an injection molding process energy field distribution model based on the multi-dimensional energy parameter set, constructing a defect form field distribution model according to the mold cavity form data set and the defect form data set, and performing bidirectional dynamic mapping processing on the two models to generate a field domain mapping result; and finally, integrating the model and a field domain mapping result, generating a bumper injection molding defect mechanism detection report, generating an injection molding energy regulation and control scheme according to parameter characteristics of an energy field abnormal distribution area, and sending the detection report and the regulation and control scheme to an injection molding production management terminal together. Therefore, the injection molding defects of the bumper can be accurately detected in real time, the generation mechanism is analyzed, and the injection molding quality is effectively improved.

Description

Technical Field

[0001] This invention relates to the field of automotive parts manufacturing and testing technology, and more specifically, to an intelligent detection method for injection molding defects in bumpers. Background Technology

[0002] In the field of bumper injection molding, the bumper, as an important safety component of a car, directly affects the vehicle's safety performance and appearance. However, during the injection molding process of bumpers, due to the complexity of the injection molding process and the interaction of various factors, various defects are easily generated, such as surface shrinkage marks, internal porosity, and weld lines. These defects can seriously affect the performance and quality of the bumper.

[0003] Currently, the detection methods for injection molding defects in bumpers mainly rely on manual visual inspection and some simple mechanical testing methods. Manual visual inspection is not only inefficient but also easily influenced by the subjective factors of the inspectors, leading to inaccurate and incomplete results. While simple mechanical testing methods improve efficiency to some extent, they often only detect relatively obvious surface defects and have limited ability to detect internal defects, making it difficult to accurately determine the root cause of the defects.

[0004] Furthermore, most existing inspection methods are conducted after the bumper has been formed, which is a post-production inspection. This makes it impossible to detect potential problems and make adjustments in time during the injection molding process, making it difficult to fundamentally avoid defects. This results in a high defect rate during the production process, increases production costs, and reduces production efficiency. Summary of the Invention

[0005] In view of the aforementioned problems, and in conjunction with the first aspect of the present invention, embodiments of the present invention provide an intelligent detection method for injection molding defects in bumpers, the method comprising: The system collects a set of multi-dimensional energy parameters, a set of mold cavity morphology data, and a set of defect morphology data of the bumper entity to be inspected after molding during the bumper injection molding process. The set of multi-dimensional energy parameters includes thermal energy parameters, pressure energy parameters, and kinetic energy parameters at each stage of injection molding. The set of mold cavity morphology data includes real-time morphological change data of the mold cavity during injection molding. The set of defect morphology data includes morphological feature data of surface and internal defects of the bumper entity to be inspected. An energy field distribution model for the injection molding process is constructed based on the multi-dimensional energy parameter set. This model is used to present the distribution of energy in the mold cavity and inside the material at each stage of injection molding. A defect morphology field distribution model is constructed based on the mold cavity morphology data set and the defect morphology data set. The defect morphology field distribution model is used to present the spatial morphology and distribution pattern of defects in the bumper entity. The energy field distribution model of the injection molding process and the defect morphology field distribution model are subjected to bidirectional dynamic mapping to determine the correspondence between the abnormal energy field distribution area and the abnormal defect morphology field area, and to generate the field mapping result. The energy field distribution model of the injection molding process, the defect morphology field distribution model, and the field mapping results are integrated to generate a bumper injection molding defect mechanism detection report. At the same time, based on the parameter characteristics of the abnormal energy field distribution area, an injection molding energy control scheme is generated. The bumper injection molding defect mechanism detection report and the injection molding energy control scheme are sent together to the injection molding production management terminal.

[0006] In another aspect, embodiments of the present invention also provide an intelligent detection system for injection molding defects in bumpers, including a processor and a machine-readable storage medium connected to the processor. The machine-readable storage medium is used to store programs, instructions, or code, and the processor is used to execute the programs, instructions, or code in the machine-readable storage medium to implement the above-described method.

[0007] Based on the above, by collecting multi-dimensional energy parameter sets, mold cavity morphology data sets, and defect morphology data sets of the bumper entity to be inspected after molding during the bumper injection molding process, key information of the injection molding process is covered. Then, the energy field distribution model of the injection molding process constructed based on the multi-dimensional energy parameter sets can accurately present the energy distribution state of each stage of injection molding in the mold cavity and inside the material. Next, the defect morphology field distribution model constructed based on the mold cavity morphology data set and the defect morphology data set can effectively present the spatial morphology and distribution law of defects in the bumper entity. Thus, by performing bidirectional dynamic mapping processing between the energy field distribution model of the injection molding process and the defect morphology field distribution model, the correspondence between abnormal energy field distribution areas and abnormal defect morphology field areas can be determined. Furthermore, the bumper injection molding defect mechanism detection report generated by integrating the above models and field mapping results can reveal the cause of the defects. At the same time, the injection energy control scheme generated based on the parameter characteristics of the abnormal energy field distribution areas can adjust the energy parameters in the injection molding process in real time, effectively avoiding the generation of defects, improving the injection quality of the bumper, reducing the defect rate, and improving production efficiency. Attached Figure Description

[0008] Figure 1 This is a schematic diagram of the execution flow of the intelligent detection method for bumper injection molding defects provided in this embodiment of the invention.

[0009] Figure 2 This is a schematic diagram of exemplary hardware and software components of the intelligent detection system for bumper injection molding defects provided in an embodiment of the present invention. Detailed Implementation

[0010] The present invention will now be described in detail with reference to the accompanying drawings. Figure 1 This is a flowchart illustrating an intelligent detection method for injection molding defects in bumpers according to an embodiment of the present invention. The following is a detailed description of this intelligent detection method for injection molding defects in bumpers.

[0011] Step S110: Collect a set of multi-dimensional energy parameters, a set of mold cavity morphology data, and a set of defect morphology data of the bumper entity to be inspected after molding during the bumper injection molding process. The set of multi-dimensional energy parameters includes thermal energy parameters, pressure energy parameters, and kinetic energy parameters at each stage of injection molding. The set of mold cavity morphology data includes real-time morphological change data of the mold cavity during injection molding. The set of defect morphology data includes morphological feature data of surface and internal defects of the bumper entity to be inspected.

[0012] In this embodiment, the injection molding process of an automobile front bumper is used as the application scenario. In this scenario, it is first necessary to comprehensively collect various data during the injection molding process. For the collection of multi-dimensional energy parameters, this is achieved by installing corresponding sensors at different key locations on the injection molding machine. For example, a temperature sensor is installed near the barrel heating coil to collect the material melting heat energy parameters, which dynamically change with the injection sequence; thermocouples are embedded in the mold cavity wall to collect the mold cavity wall heat energy parameters; and temperature sensors are set at the inlet and outlet of the cooling water channel to obtain the material cooling heat energy parameters. Pressure energy parameters are collected by installing a pressure sensor at the nozzle to collect injection pressure energy parameters, installing a pressure transmitter in the oil circuit during the holding pressure stage to obtain holding pressure energy parameters, and setting a pressure sensor at the cavity inlet to collect cavity back pressure energy parameters. Regarding kinetic energy parameters, the screw speed data is obtained by installing an encoder at the screw drive motor to calculate the kinetic energy parameters. Displacement sensors and speed sensors are installed on the moving parts of the mold opening and closing mechanism to collect the mold opening and closing kinetic energy parameters and the ejection mechanism kinetic energy parameters. The data collected by the above sensors all include values ​​of different injection timing sequences, which together constitute a multi-dimensional energy parameter set.

[0013] The acquisition of mold cavity morphology data sets involves using high-precision laser scanning equipment to scan the mold cavity in real time during the injection molding process. This acquires data on the mold cavity wall morphology, parting surface morphology, and internal insert morphology. The scanning process follows a set injection sequence, with each scan obtaining cavity morphology data for that sequence, thus forming a mold cavity morphology data set containing morphological changes across different injection sequences.

[0014] The defect morphology data set is collected by using a 3D vision inspection system to scan the surface of the finished bumper to obtain morphological feature data of surface defects, such as the location, shape, and size of scratches, dents, and bulges. Simultaneously, industrial CT scanning technology is used to inspect the interior of the bumper, obtaining the spatial location, shape, and size data of internal defects such as bubbles, shrinkage cavities, and weld lines. These data together form the defect morphology data set. Throughout the data acquisition process, data involving potentially private or sensitive information, such as equipment operating parameters, is encrypted during transmission and storage. Sensitive identifiers are removed using data anonymization techniques, and access control is implemented to ensure data privacy and prevent leakage.

[0015] Step S120: Construct an energy field distribution model for the injection molding process based on the multi-dimensional energy parameter set. The energy field distribution model for the injection molding process is used to present the distribution of energy in the mold cavity and inside the material at each stage of injection molding.

[0016] After collecting a set of multi-dimensional energy parameters, it is necessary to construct an energy field distribution model for the injection molding process based on this set to comprehensively present the energy distribution during the injection molding process. This process requires processing thermal energy parameters, pressure energy parameters, and kinetic energy parameters separately, constructing corresponding sub-models, and then comprehensively associating and integrating them.

[0017] Step S121: Classify and extract the thermal energy parameters in the multi-dimensional energy parameter set to obtain material melting thermal energy parameters, mold cavity wall thermal energy parameters, and material cooling thermal energy parameters. Each type of thermal energy parameter includes numerical data of different injection molding sequences.

[0018] In this embodiment, thermal parameters are extracted and categorized from a multi-dimensional set of energy parameters. The material melting thermal energy parameters mainly originate from the heating of the plastic raw material by the barrel heating system. These parameters are calculated using temperature data collected from various sections of the barrel by temperature sensors, such as heating power, heating time, and the material's specific heat capacity. These melting thermal energy parameters include different time-series values ​​from the start of injection molding to the complete melting stage. The mold cavity wall thermal energy parameters are directly collected by thermocouples embedded in the cavity wall, reflecting the temperature changes of the mold wall during injection molding and including temperature values ​​at different times throughout the entire injection molding cycle. The material cooling thermal energy parameters are calculated using data such as the temperature difference between the inlet and outlet of the cooling water channels and the water flow rate. These parameters reflect the heat released by the material during the cooling stage and also include numerical data at different times during the cooling process.

[0019] Step S122: Sort the extracted thermal energy parameters according to the injection molding sequence to form a thermal energy parameter time sequence, and label the injection molding stage corresponding to each time sequence.

[0020] After extracting the material melting thermal energy parameters, mold cavity wall thermal energy parameters, and material cooling thermal energy parameters, these parameters are sorted according to the time sequence of the injection molding process. For example, the material melting thermal energy parameters are arranged sequentially from the start of injection molding, based on the values ​​collected at regular time intervals, forming a time sequence of material melting thermal energy parameters. Similarly, the mold cavity wall thermal energy parameters and material cooling thermal energy parameters are processed in a similar way, forming their own time sequences. During the sorting process, each time point is labeled with its corresponding injection molding stage, such as the material melting stage, injection stage, holding pressure stage, and cooling stage, so that the injection molding process corresponding to each parameter can be clearly identified during subsequent model construction.

[0021] Step S123: The thermal energy parameter time series is processed using the field modeling method. The three-dimensional structure of the mold cavity is used as the spatial frame. The values ​​of various thermal energy parameters in each time series are mapped to the corresponding spatial positions to generate a thermal energy field time series distribution sub-model.

[0022] Step S1231: Obtain the three-dimensional structural data of the cavity of the bumper injection mold, construct the three-dimensional structural model of the mold cavity using three-dimensional modeling software, and annotate the spatial coordinates and geometric shape of each region inside the three-dimensional structural model of the mold cavity.

[0023] In this embodiment, precise three-dimensional structural data of the bumper injection mold cavity is obtained using 3D scanning technology, including the overall dimensions of the cavity, the curvature of each surface, the wall thickness distribution, and the position of internal inserts. This data is then imported into professional 3D modeling software, such as UG or SolidWorks, to construct a 3D structural model of the mold cavity that perfectly matches the actual cavity. In this model, spatial coordinates are labeled for each internal region. For example, a three-dimensional Cartesian coordinate system is established with a vertex of the mold cavity as the origin, thereby determining the spatial coordinates of every point inside the cavity. Simultaneously, the geometric shapes of each part of the cavity, such as planar areas, curved areas, and corner areas, are labeled in detail.

[0024] Step S1232: Analyze the material melting heat energy parameters for each time sequence, determine the transfer law of material melting heat energy inside the material, and allocate the material melting heat energy parameter values ​​to the corresponding spatial regions through which the material flows in the three-dimensional structural model of the mold cavity according to the material flow path.

[0025] For each time-series material molten heat energy parameter, the transfer law of molten heat energy within the material is analyzed based on the material flow simulation results of injection molding. For example, during the injection stage, the molten material enters the cavity from the gate and fills the cavity according to a certain flow path. The heat energy inside the material is gradually transferred and diffused during the flow process. Based on this flow path, the material molten heat energy parameter value for that time-series is assigned to the spatial region through which the material flows at that moment in the three-dimensional structural model of the mold cavity. For example, if in a certain time-series, the molten material has just filled 1 / 3 of the cavity, then the material molten heat energy parameter value for that time-series is assigned to the corresponding spatial region of the first 1 / 3 of the cavity.

[0026] Step S1233: Analyze the thermal energy parameters of the mold cavity wall at each time step. Based on the temperature conduction characteristics of the mold cavity wall, map the values ​​of the thermal energy parameters of the mold cavity wall to the wall space position of the three-dimensional structural model of the mold cavity, and mark the heat transfer rate of different areas of the wall.

[0027] The temperature conduction characteristics of the mold cavity wall are related to the thermal conductivity and thickness of the wall material, as well as the heat exchange with the surrounding environment. When analyzing the thermal energy parameters of the mold cavity wall at each time step, the influence of these factors on heat transfer is considered. The thermal energy parameter values ​​of the mold cavity wall at that time step are precisely mapped to the spatial positions of the wall surface in the three-dimensional structural model of the mold cavity, according to the spatial coordinates of each point on the wall. For example, if the coordinates of a point on the cavity wall are (x1, y1, z1), then the corresponding thermal energy parameter value is marked at that position in the model. Simultaneously, the heat transfer rate in different regions is calculated based on the thermal conductivity and thickness of the wall material and marked in the model; for example, the heat transfer rate is relatively fast in thinner wall regions, while it is slower in thicker wall regions.

[0028] Step S1234: Analyze the material cooling thermal energy parameters for each time sequence, and combine them with the distribution of the cooling system in the mold. Map the material cooling thermal energy parameter values ​​to the space near the cooling pipe in the three-dimensional structural model of the mold cavity to reflect the influence of the cooling process on the material thermal energy.

[0029] The distribution of the cooling system within the mold is known, including information such as the direction, diameter, and spacing of the cooling pipes. When analyzing the material cooling thermal energy parameters for each time sequence, the range of influence of the material cooling thermal energy on each region of the cavity is determined based on the distance between the cooling pipes and the cavity wall. The material cooling thermal energy parameter values ​​are mapped to the spatial region near the cooling pipes in the 3D structural model of the mold cavity. For example, the cavity region within a certain radius around the cooling pipes will be affected by the cooling thermal energy, and the material cooling thermal energy parameter values ​​in that cavity region will change accordingly. In this way, the influence of the cooling process on the distribution of material thermal energy is reflected in the model.

[0030] Step S1235: Perform spatial interpolation on the various thermal energy parameter values ​​mapped to the three-dimensional structural model of the mold cavity to make the thermal energy parameter values ​​form a continuous distribution in the spatial region and avoid field breakage caused by numerical discrepancies.

[0031] Because the data collected by sensors are discrete time-series data, directly mapping them into a 3D model results in a discrete spatial distribution of thermal parameters, leading to field discontinuities. To address this issue, spatial interpolation methods are used to process the mapped thermal parameter values. For example, for a specific spatial region within a cavity, given the thermal parameter values ​​of several discrete points within that region, interpolation algorithms such as Kriging interpolation or inverse distance weighted interpolation are used to calculate the thermal parameter values ​​of other points within that region that are not directly mapped. This ensures a continuous and smooth distribution of thermal parameters throughout the entire spatial region, guaranteeing the integrity and continuity of the thermal field.

[0032] Step S1236: Add a time series identifier to the thermal field distribution data generated for each time series to form a single time series thermal field distribution snapshot.

[0033] After completing the spatial mapping and interpolation of various thermal parameters for each time sequence, corresponding time sequence identifiers are added to the generated thermal field distribution data. These time sequence identifiers correspond to the time sequence in the injection molding process; for example, time sequence 1 corresponds to the first second after injection begins, time sequence 2 corresponds to the second second, and so on. After adding the time sequence identifiers, the thermal field distribution data for each time sequence becomes an independent snapshot of the thermal field distribution, which completely records the distribution of thermal energy within the three-dimensional space of the mold cavity at that time sequence.

[0034] Step S1237: Connect all time-series thermal field distribution snapshots in the order of injection molding to form a time-series thermal field distribution sub-model that can dynamically display changes in the thermal field.

[0035] All time-series snapshots of thermal field distribution are arranged sequentially and connected in series according to the injection molding sequence. These snapshots are then displayed dynamically using visualization software, forming a time-series thermal field distribution sub-model. This sub-model can present the dynamic changes of thermal energy within the mold cavity and material throughout the entire injection molding process, including the distribution characteristics and evolution patterns of the thermal field at different injection stages.

[0036] Step S1238: Perform spatial verification on the generated thermal energy field time-series distribution sub-model, check whether the values ​​of various thermal energy parameters match the geometric shape of the three-dimensional structure model of the mold cavity during the spatial mapping process, correct the mapping data with spatial coordinate misalignment, and make the spatial accuracy of the thermal energy field time-series distribution sub-model meet the preset standard.

[0037] When performing spatial verification on the temporal distribution sub-model of the thermal energy field, the spatial distribution of thermal parameters in the sub-model is compared with the geometry of the three-dimensional structural model of the mold cavity. For example, it checks whether the thermal parameter values ​​exceed the spatial range of the cavity, or whether there are spatial coordinate misalignments in the mapped data in areas with complex geometric features. If a problem is found, such as some thermal parameter values ​​being mapped outside the cavity or not matching the geometry of the cavity wall, the spatial coordinates of that mapped data are corrected and adjusted to the correct position. After multiple verifications and corrections, the spatial accuracy of the temporal distribution sub-model of the thermal energy field is ensured to meet the preset standards, such as the spatial coordinate error not exceeding a certain range.

[0038] Step S124: Classify and extract the pressure energy parameters in the multi-dimensional energy parameter set to obtain injection pressure energy parameters, holding pressure energy parameters, and cavity back pressure energy parameters. Each type of pressure energy parameter contains numerical data of different injection timing sequences.

[0039] Among the multi-dimensional energy parameter set, pressure energy parameters are a crucial component. By analyzing and filtering the collected raw pressure data, injection pressure energy parameters, holding pressure energy parameters, and cavity back pressure energy parameters were extracted and categorized. Injection pressure energy parameters are the pressure energy generated during the injection stage when the screw pushes the molten material into the cavity. Their values ​​vary with factors such as injection speed and material viscosity, encompassing values ​​from the start to the end of injection. Holding pressure energy parameters are the pressure energy applied during the holding stage to prevent material backflow and compensate for cavity shrinkage, including values ​​at different times within the holding stage. Cavity back pressure energy parameters are the energy parameters corresponding to the back pressure generated by the material inside the cavity against the screw, also including numerical data for different injection times.

[0040] Step S125: Sort the various pressure energy parameters in time sequence and mark the injection stage. Then, map the pressure energy parameter values ​​to the three-dimensional space frame of the mold cavity through the field modeling method to generate a pressure energy field time sequence distribution sub-model.

[0041] First, the extracted injection pressure energy parameters, holding pressure energy parameters, and cavity back pressure energy parameters are sorted according to the injection molding sequence, forming an ordered time sequence for each parameter. Then, each time point is labeled with its corresponding injection stage, such as the injection stage or the holding pressure stage. Next, using a field modeling method similar to that used in constructing the thermal field time sequence distribution sub-model, the three-dimensional structure model of the mold cavity serves as the spatial framework, mapping the values ​​of various pressure energy parameters to their corresponding spatial locations. For example, the injection pressure energy parameters are mainly related to the screw's propulsion, and their distribution within the cavity is related to the material's flow path and filling state. Based on the material's filling status within the cavity, the injection pressure energy parameter values ​​are mapped to the spatial area occupied by the material. The holding pressure energy parameters act on the entire cavity during the holding pressure stage, and their values ​​are uniformly mapped to the three-dimensional space of the cavity. The cavity back pressure energy parameters reflect the reaction force inside the cavity on the screw, and are mapped to their corresponding spatial locations within the cavity according to the pressure transmission law. Spatial interpolation is used to make the pressure energy parameter values ​​continuously distributed in space, and time sequence identifiers are added to form a snapshot of the pressure energy field distribution. Finally, all snapshots are concatenated in time sequence to generate a time sequence distribution sub-model of the pressure energy field.

[0042] Step S126: Classify and extract the kinetic energy parameters in the multi-dimensional energy parameter set to obtain material flow kinetic energy parameters, mold opening and closing kinetic energy parameters, and ejection mechanism kinetic energy parameters. Each type of kinetic energy parameter includes numerical data of different injection timing sequences.

[0043] In a multi-dimensional set of energy parameters, kinetic energy parameters are equally indispensable. By processing the collected motion-related data, kinetic energy parameters for material flow, mold opening and closing, and ejection mechanism are extracted and categorized. Material flow kinetic energy parameters are related to the flow velocity and mass of the molten material in the cavity. These parameters are obtained by calculating the flow velocity and density of the material at different time points, encompassing the values ​​at different time points during the material flow process in the cavity. Mold opening and closing kinetic energy parameters are calculated based on the motion velocity, mass, and acceleration of the mold opening and closing mechanism, including values ​​at different time points during the mold opening and closing process. Ejection mechanism kinetic energy parameters are calculated based on the motion velocity and mass of the ejection mechanism, including values ​​at different time points during the ejection process.

[0044] Step S127: Map the values ​​of various kinetic energy parameters to the corresponding spatial locations to generate a time-series distribution sub-model of the kinetic energy field.

[0045] Step S1271: Analyze the material flow kinetic energy parameters, mold opening and closing kinetic energy parameters, and ejection mechanism kinetic energy parameters extracted by category one by one, determine the transmission law of each type of kinetic energy parameter in the injection molding process, and analyze the effect of different kinetic energy parameters on material forming and cavity morphology.

[0046] This section analyzes the transmission characteristics of material flow kinetic energy parameters during the material filling process of the mold cavity. Molten material enters the cavity at a certain speed under pressure. Its kinetic energy changes during the flow process due to factors such as friction with the cavity wall and viscous dissipation within the material. Simultaneously, the material flow kinetic energy affects the filling effect; excessively fast filling speed may lead to defects such as turbulence and scorching, while excessively slow speed may result in insufficient filling. The mold opening and closing kinetic energy parameters are transmitted during mold opening and closing. Their magnitude and changes affect the smoothness of mold movement and its service life. The impact on cavity morphology is mainly reflected in the mold closing accuracy. Unstable opening and closing kinetic energy may lead to misalignment of the parting surface, affecting the cavity's sealing performance. The ejection mechanism kinetic energy parameters are transmitted during ejection. Its function is to eject the molded bumper from the cavity through the movement of the ejection mechanism. Excessive kinetic energy can cause product damage during ejection, while insufficient kinetic energy may prevent successful ejection.

[0047] Step S1272: For the material flow kinetic energy parameters, based on the flow velocity and flow path of the material inside the cavity during the injection molding process, the material flow kinetic energy parameter values ​​are allocated to the spatial area through which the material flows in the three-dimensional structural model of the mold cavity. The numerical allocation results are calibrated through material flow simulation data to ensure that the numerical allocation results are consistent with the actual material flow trajectory.

[0048] Based on the material flow velocity and path inside the cavity obtained from the material flow simulation software, the spatial region through which the material flows at each time step is determined. The kinetic energy parameters of the material flow at that time step are then allocated according to the material flow velocity distribution within that region; for example, higher kinetic energy parameter values ​​are allocated to regions with faster flow velocities, and lower values ​​are allocated to regions with slower flow velocities. The allocation results are then compared with the material flow simulation data. If a deviation is found between the allocation results and the flow trajectory in the simulation data, such as the flow position corresponding to the kinetic energy parameter values ​​in a certain region not matching the simulated trajectory, the numerical allocation for that region is adjusted until the numerical allocation results are consistent with the actual material flow trajectory.

[0049] Step S1273: Based on the mold opening and closing kinetic energy parameters, and combined with the motion trajectory of the mold opening and closing mechanism and its positional correspondence with the cavity, the mold opening and closing kinetic energy parameter values ​​are mapped to the spatial region near the mold opening and closing components in the three-dimensional structural model of the mold cavity. The mapping results are verified by mold motion simulation data to reflect the influence of mold opening and closing motion on the energy distribution around the cavity.

[0050] The motion trajectory of the mold opening and closing mechanism is known. Based on its trajectory and its positional correspondence with the cavity, the influence range of the mold opening and closing motion on the surrounding area of ​​the cavity is determined. The kinetic energy parameters of the mold opening and closing mechanism are mapped to the spatial region near the mold opening and closing components in the 3D structural model of the mold cavity, such as the cavity region near the parting surface. Then, the mold opening and closing process is simulated using mold motion simulation software to obtain data such as the motion velocity and acceleration of the mold opening and closing components, thereby verifying the reasonableness of the mapping results of the kinetic energy parameters. For example, if the simulation shows that the motion velocity of the mold opening and closing components is large at a certain time sequence, the corresponding kinetic energy parameter value of the mapped region should also be large. If there is an inconsistency, it is corrected to reflect the influence of the mold opening and closing motion on the energy distribution around the cavity.

[0051] Step S1274: For the kinetic energy parameters of the ejection mechanism, based on the installation position of the ejection mechanism in the mold and the cavity area corresponding to the ejection action, the kinetic energy parameter values ​​of the ejection mechanism are mapped to the spatial area in contact with the ejection mechanism in the three-dimensional structural model of the mold cavity. The mapping results are calibrated by the ejection action simulation data to reflect the effect of the kinetic energy generated by the ejection action on the local area.

[0052] The ejector mechanism's installation position in the mold is fixed, and its ejection action directly acts on a specific area of ​​the product within the cavity. Based on the ejector mechanism's installation position and the cavity area corresponding to the ejection action, the kinetic energy parameters of the ejector mechanism are mapped to the spatial area in the mold cavity's three-dimensional structural model that contacts the ejector mechanism—that is, the area near the ejected product location. The ejection process is simulated using ejection simulation software to obtain data such as the ejector mechanism's speed and ejection force. The mapping results of the kinetic energy parameters are then calibrated based on this data. For example, if the simulation shows that the ejector mechanism's ejection speed is relatively fast at a certain time, the corresponding kinetic energy parameters in the mapped area should be adjusted accordingly to accurately reflect the effect of the kinetic energy generated by the ejection action on that local area.

[0053] Step S1275: Perform spatial interpolation on the material flow kinetic energy parameters, mold opening and closing kinetic energy parameters, and ejection mechanism kinetic energy parameters mapped to the three-dimensional structural model of the mold cavity, so that the various kinetic energy parameters form a continuous distribution in the corresponding spatial region, avoiding the discontinuous distribution of kinetic energy field caused by numerical discreteness.

[0054] Similar to the processing of thermal and pressure energy parameters, spatial interpolation is performed on the numerical values ​​of various kinetic energy parameters mapped to the three-dimensional structural model of the mold cavity. Using appropriate interpolation algorithms, such as linear interpolation or spline interpolation, discrete kinetic energy parameter points are interpolated to obtain the kinetic energy parameter values ​​at adjacent points, thus creating a continuous and smooth distribution of various kinetic energy parameter values ​​within the corresponding spatial region. This avoids breaks or discontinuities in the kinetic energy field distribution caused by numerical discreteness, ensuring the integrity and accuracy of the kinetic energy field.

[0055] Step S1276: Add a time sequence identifier to the processed kinetic field distribution data for each time sequence to form a snapshot of the kinetic field distribution for a single time sequence. The snapshot of the kinetic field distribution includes the distribution state of various kinetic energy parameters in the cavity space under that time sequence.

[0056] After completing the spatial mapping and interpolation of various kinetic energy parameters for each time series, a time series identifier is added to the processed kinetic energy field distribution data. This time series identifier corresponds to the injection molding time series, clearly defining the injection molding time point corresponding to the kinetic energy field distribution data. After adding the time series identifier, the kinetic energy field distribution data for each time series becomes a snapshot of the kinetic energy field distribution for a single time series. This snapshot of the kinetic energy field distribution completely records the distribution state of the material flow kinetic energy parameters, mold opening and closing kinetic energy parameters, and ejection mechanism kinetic energy parameters in the three-dimensional space of the mold cavity under that time series.

[0057] Step S1277: Connect all the time-series kinetic energy field distribution snapshots in the order of injection molding to form a kinetic energy field time-series distribution sub-model that can dynamically display the changes in kinetic energy field at different injection stages.

[0058] All snapshots of the kinetic energy field distribution with time-series identifiers are arranged sequentially and connected in series according to the injection molding sequence. Using visualization technology, these snapshots are presented dynamically, forming a kinetic energy field time-series distribution sub-model. This sub-model dynamically displays the distribution and changes of kinetic energy in the mold cavity and surrounding area at different stages from the start to the end of injection molding, helping to intuitively understand the evolution of the kinetic energy field.

[0059] Step S1278: Perform spatial verification on the generated kinetic energy field time-series distribution sub-model, check whether the spatial mapping position of various kinetic energy parameter values ​​matches the geometric structure of the mold cavity three-dimensional structure model and the corresponding mechanism motion range, correct the mapping data of spatial coordinate deviation, so that the kinetic energy field time-series distribution sub-model can accurately reflect the distribution law of kinetic energy in the cavity and surrounding area.

[0060] Spatial verification is performed on the temporal distribution sub-model of the kinetic energy field. The spatial distribution of kinetic energy parameters in the sub-model is compared with the geometry of the 3D structure model of the mold cavity and the motion range of the corresponding mechanism. For example, it is checked whether the mapping position of the material flow kinetic energy parameters is within the cavity area through which the material may actually flow, whether the mapping position of the mold opening and closing kinetic energy parameters is within the motion range of the mold opening and closing mechanism, and whether the mapping position of the ejection mechanism kinetic energy parameters matches the installation position and operating area of ​​the ejection mechanism. If spatial coordinate deviations are found in the mapping data, such as kinetic energy parameter values ​​being mapped to positions outside the motion range of the mechanism or mismatched with the cavity geometry, the spatial coordinates of the above data are corrected to ensure that the temporal distribution sub-model of the kinetic energy field can accurately reflect the distribution law of kinetic energy in the cavity and surrounding areas.

[0061] Step S128: Determine the energy field weight allocation rules. Based on the degree of influence of various types of energy on material molding at different injection stages, assign dynamic weights to the thermal energy field time-series distribution sub-model, the pressure energy field time-series distribution sub-model, and the kinetic energy field time-series distribution sub-model.

[0062] The influence of thermal energy, pressure energy, and kinetic energy on material molding varies at different injection stages. In the injection stage, pressure energy and kinetic energy play a major role in material filling, and their weights are relatively high. In the holding pressure stage, pressure energy is more crucial for maintaining cavity pressure, thus the pressure energy field has a higher weight. In the cooling stage, the thermal energy field has a significant impact on material solidification and molding quality, and its weight increases accordingly. Based on the influence of various energy types at these different stages, an energy field weight allocation rule is established. For example, in the injection stage, the weight of the pressure energy field is set to 0.4, the weight of the kinetic energy field to 0.3, and the weight of the thermal energy field to 0.3; in the holding pressure stage, the weight of the pressure energy field increases to 0.5, the weight of the thermal energy field to 0.3, and the weight of the kinetic energy field to 0.2; in the cooling stage, the weight of the thermal energy field increases to 0.5, the weight of the pressure energy field to 0.3, and the weight of the kinetic energy field to 0.2. These weight values ​​are dynamically adjusted according to the injection stage, forming a dynamic weight allocation rule.

[0063] Step S129: The thermal energy field time-series distribution sub-model, the pressure energy field time-series distribution sub-model, and the kinetic energy field time-series distribution sub-model are integrated and correlated according to dynamic weights to generate an injection molding process energy field distribution model that includes temporal and spatial dimensions. The integrated correlation refers to associating the spatial distribution data of the thermal energy field time-series distribution sub-model, the pressure energy field time-series distribution sub-model, and the kinetic energy field time-series distribution sub-model under the same time sequence to form a multi-physics coupling model. The relative importance of each energy field in the multi-physics coupling model is adjusted by dynamic weights, rather than directly superimposing the values. The injection molding process energy field distribution model is used to present the comprehensive distribution state of each energy in the mold cavity and inside the material under different time sequences.

[0064] In the comprehensive integration process, the data from the thermal energy field time-series distribution sub-model, the pressure energy field time-series distribution sub-model, and the kinetic energy field time-series distribution sub-model at the same time sequence are first correlated. For each time sequence point, the energy distribution data of the three sub-models at the corresponding positions in the three-dimensional space of the mold cavity are extracted. Then, according to the dynamic weight allocation rules, the energy distribution data of each sub-model are assigned corresponding weights. For example, at a certain time sequence in the injection stage, the pressure energy field has a weight of 0.4, the kinetic energy field has a weight of 0.3, and the thermal energy field has a weight of 0.3. The spatial distribution data of the three sub-models at this time sequence are multiplied by their corresponding weights and then correlated and combined to form the multiphysics coupling model data at that time sequence. In this way, instead of directly superimposing the values ​​of each energy field, their relative importance in the comprehensive model is adjusted through weights. By concatenating all the multiphysics coupling model data at all time sequences, a final energy field distribution model of the injection molding process containing both temporal and spatial dimensions is generated. This energy field distribution model for the injection molding process can present the comprehensive distribution of thermal energy, pressure energy, and kinetic energy within the mold cavity and material under different time sequences, reflecting the interaction and influence between various energy fields.

[0065] Step S130: Construct a defect morphology field distribution model based on the mold cavity morphology data set and the defect morphology data set. The defect morphology field distribution model is used to present the spatial morphology and distribution pattern of defects in the bumper entity.

[0066] Step S131: Classify and extract the data in the mold cavity morphology data set to obtain cavity wall morphology data, cavity parting surface morphology data, and cavity internal insert morphology data. Each type of morphology data includes morphology change data for different injection molding sequences.

[0067] The mold cavity morphology data set contains rich information on cavity morphology. By classifying and processing the collected raw morphology data, cavity wall morphology data, cavity parting surface morphology data, and cavity insert morphology data are extracted. Cavity wall morphology data reflects the geometric changes of the cavity inner wall, such as the presence of wear or deformation, and includes scan data under different injection molding sequences. Cavity parting surface morphology data focuses on the flatness and fit of the parting surface, and also includes morphological change data under different injection molding sequences. Cavity insert morphology data involves the position, orientation, and other morphological information of the inserts, and also includes change data under different injection molding sequences.

[0068] Step S132: Sort the extracted cavity morphology data of each type according to the injection molding sequence to form a cavity morphology time sequence, label the injection molding stage corresponding to each time sequence, and record the deviation data between the cavity morphology of each time sequence and the standard cavity morphology.

[0069] The extracted cavity wall morphology data, cavity parting surface morphology data, and cavity insert morphology data are sorted according to the injection molding sequence, forming an ordered cavity morphology sequence for each type of morphology data. Then, each sequence point is labeled with the corresponding injection stage, such as mold closing stage, injection stage, and mold opening stage. Simultaneously, the cavity morphology data for each sequence is compared with preset standard cavity morphology data, and the deviations are calculated and recorded, such as positional deviations and shape deviations. These deviations reflect the difference between the actual cavity morphology and the ideal morphology.

[0070] Step S133: The three-dimensional morphological modeling method is used to process the cavity morphology time sequence. Based on the standard cavity three-dimensional structure, the cavity morphology deviation data of each time sequence are mapped to the corresponding spatial position to generate the cavity morphology deviation time sequence sub-model.

[0071] Using a standard cavity 3D structural model as a benchmark, the cavity wall morphology deviation data, cavity parting surface morphology deviation data, and cavity internal insert morphology deviation data for each time series are mapped to their corresponding spatial locations. For example, if the actual shape of a point on the cavity wall deviates from the standard shape, this deviation value is marked at the spatial location of that point in the standard cavity 3D structural model. For shape deviations, the normal or tangential deviations at each point on the surface are calculated by comparing the actual scanned surface with the standard surface and mapped to their corresponding spatial locations. In this way, the cavity morphology deviation data for each type of cavity is visualized in 3D space. Then, a time series identifier is added to the deviation data for each time series, forming a snapshot of the cavity morphology deviation distribution. Finally, all snapshots are concatenated in time series to generate a cavity morphology deviation time series sub-model. This sub-model can dynamically display the spatial distribution and changes of cavity morphology deviations at different time series.

[0072] Step S134: Classify and extract the data in the defect morphology data set to obtain surface defect morphology data and internal defect morphology data. The surface defect morphology data includes the surface location, shape, and size data of the defect, and the internal defect morphology data includes the internal spatial location, shape, and size data of the defect.

[0073] The defect morphology dataset contains defect information from both the surface and interior of the bumper. By classifying the data in this dataset, surface defect morphology data and internal defect morphology data are extracted. Surface defect morphology data mainly includes information on scratches, dents, bulges, color differences, and other defects on the bumper surface. For each surface defect, its specific location coordinates on the bumper surface, its geometric shape description (e.g., circular, striped, irregular), and dimensional data (e.g., length, width, depth) are extracted. Internal defect morphology data involves defects inside the bumper such as bubbles, shrinkage cavities, weld lines, and inclusions. Similarly, the internal spatial location coordinates, shape, and dimensional data of each internal defect are extracted.

[0074] Step S135: Perform spatial positioning processing on the surface defect morphology data. Using the three-dimensional model of the formed bumper as a spatial frame, map the position, shape, and size data of each surface defect to the corresponding surface spatial coordinates to generate a surface defect morphology sub-model.

[0075] Step S1351: Obtain the 3D scanning data of the formed bumper entity to be inspected, construct a 3D model of the bumper entity using 3D reconstruction software, and annotate the spatial coordinates of each area on the surface of the 3D model of the bumper entity.

[0076] A high-precision 3D scanning device is used to perform a comprehensive scan of the car's front bumper to obtain 3D point cloud data of its surface. This point cloud data is then imported into 3D reconstruction software, such as Geomagic DesignX, for point cloud processing, surface fitting, and model construction, ultimately resulting in a 3D model that perfectly matches the actual bumper. Within the constructed 3D model of the bumper, a 3D coordinate system is established with a specific feature point as the origin, and the spatial coordinates of different areas on the model's surface are precisely labeled, such as the coordinate ranges of the front, side, and top surfaces of the bumper.

[0077] Step S1352: Analyze the surface defect morphology data in the defect morphology data set one by one, and extract the approximate location description of each surface defect on the surface of the bumper.

[0078] Surface defect morphology data in a defect morphology dataset typically includes textual descriptions or image annotations of the defect locations. The morphology data for each surface defect is analyzed individually. For example, for a description of a "dent defect located in the middle of the left side of the bumper," its approximate location is extracted as "middle of the left side." Through this method, the approximate distribution area of ​​each surface defect on the bumper's surface is determined.

[0079] Step S1353: Based on the approximate location description of the surface defect, determine the corresponding surface area range in the 3D model of the bumper entity, and use this surface area range as the preliminary spatial positioning area of ​​the surface defect.

[0080] Based on the approximate location description of the surface defect obtained from the analysis, a search and matching process is performed in the 3D model of the bumper to determine the corresponding surface area range. For example, for the approximate location description of "left side center," the middle part of the left side of the bumper is found in the 3D model, and a rectangular or polygonal area encompassing this region is defined as the initial spatial location area for the surface defect. The size of the initial location area can be set according to the maximum possible size of the defect to ensure that the defect can be contained within this area.

[0081] Step S1354: Use a high-resolution image acquisition device to acquire a surface image of the preliminary spatial positioning area, and use image analysis technology to extract the pixel coordinates and pixel size data of surface defects in the image.

[0082] A high-resolution industrial camera is used to photograph the initial spatially located area, acquiring a clear surface image of the region. The image resolution should be high enough to ensure accurate identification and extraction of detailed information about surface defects. Then, image analysis techniques are used to process the acquired surface image, such as grayscale conversion, edge detection, and thresholding, to separate the surface defects from the background. The pixel coordinates of the surface defects in the image are determined using the image coordinate system, such as the pixel coordinates of the defect's boundary points. Simultaneously, based on the image scaling, the pixel dimensions of the defects in the image are converted into actual physical size data, such as the number of pixels corresponding to the length and width of the defects in the image, and then multiplied by the pixel equivalent to obtain the actual size.

[0083] Step S1355: Establish the mapping relationship between the surface coordinates of the three-dimensional model of the bumper and the pixel coordinates of the high-resolution image. Based on this mapping relationship, convert the pixel coordinates of the surface defects into the spatial coordinates of the three-dimensional model surface, and at the same time convert the pixel size data into the actual physical size data.

[0084] Camera calibration technology is used to establish a mapping relationship between the surface coordinates of the 3D model of the bumper and the pixel coordinates of the high-resolution image. During camera calibration, the camera's intrinsic and extrinsic parameters, as well as the transformation matrix between the image coordinate system and the world coordinate system (i.e., the 3D model coordinate system), are acquired. Using this transformation matrix, the pixel coordinates of surface defects in the image are converted into spatial coordinates of the surface of the 3D model of the bumper. For example, if the pixel coordinates of a boundary point of a defect are (u, v), its spatial coordinates in the 3D model are calculated as (x, y, z) through the mapping relationship. Simultaneously, based on the pixel equivalent of the image, the previously obtained defect pixel size data is converted into actual physical size data, such as converting pixel length to millimeters.

[0085] Step S1356: Based on the converted spatial coordinates and actual physical size data, construct a three-dimensional morphological model of the defect at the corresponding surface position of the bumper solid three-dimensional model.

[0086] Based on the converted spatial coordinates of the surface defects and their actual physical dimensions, a 3D morphological model of the defects is constructed at the corresponding surface locations of the bumper's solid 3D model. For regularly shaped defects, such as circular dents, a cylinder or hemisphere can be constructed to simulate the defect shape based on the center coordinates, diameter, and depth data. For irregularly shaped defects, a 3D model consistent with the actual defect shape is constructed using methods such as surface fitting, based on the spatial coordinates of the defect boundary points. This 3D morphological model accurately reflects the spatial shape of the defects on the bumper surface.

[0087] Step S1357: Label the attributes of each constructed three-dimensional morphological model of surface defects, including defect type, size parameters, and morphological features.

[0088] After constructing a 3D morphological model of each surface defect, its attributes are labeled. The labels include defect type, such as scratches, dents, bulges, etc.; dimensional parameters, such as specific values ​​for length, width, depth, area, etc.; and morphological features, such as descriptive information like whether the defect edges are smooth and whether there are branches. These attribute labels help to classify and analyze defects.

[0089] Step S1358: Integrate the three-dimensional morphological models of all surface defects into the three-dimensional model of the bumper entity according to their spatial coordinates to form a surface defect morphological sub-model. The surface defect morphological sub-model can clearly present the spatial distribution and morphological characteristics of all surface defects on the surface of the bumper entity.

[0090] All the constructed 3D morphological models of surface defects are integrated into the 3D model of the bumper entity according to their respective spatial coordinates. During the integration process, it is ensured that each defect model is accurately superimposed on the corresponding surface position of the bumper entity model. After integration, a surface defect morphological sub-model is formed. This sub-model is based on the 3D model of the bumper entity, superimposing the 3D morphology of all surface defects, and can present the spatial distribution of all surface defects on the surface of the bumper entity and their respective morphological characteristics.

[0091] Step S136: Spatial positioning processing is performed on the internal defect morphology data. The internal defect imaging data obtained by non-destructive testing technology is used to map the spatial location, shape, and size data of the internal defects to the internal spatial coordinates of the bumper solid three-dimensional model to generate an internal defect morphology sub-model.

[0092] Spatial localization of internal defect morphology data is primarily based on imaging data obtained from industrial CT scans. Industrial CT scans can penetrate the bumper body to obtain tomographic images of internal defects. These tomographic images are analyzed and processed to extract the spatial location information of each internal defect. By transforming the coordinate system of the CT images to the coordinate system of the bumper's 3D model, the spatial location of the internal defects is mapped to the internal spatial coordinates of the bumper's 3D model. Simultaneously, based on the morphology of the defects in the tomographic images, the shape and size data of the internal defects are determined, and a 3D morphological model of the internal defects is constructed at the corresponding location in the 3D model. All 3D morphological models of internal defects are integrated into the bumper's 3D model to generate an internal defect morphological sub-model. This sub-model can display the spatial distribution and morphological characteristics of the internal defects within the bumper body.

[0093] Step S137: Analyze the spatial relationship between the cavity morphology deviation time sequence sub-model, the surface defect morphology sub-model, and the internal defect morphology sub-model, and determine the corresponding positional relationship between the cavity morphology deviation region and the defect morphology region.

[0094] The temporal sub-model of cavity morphological deviation is spatially compared and analyzed with the surface defect morphological sub-model and the internal defect morphological sub-model. First, in the same three-dimensional coordinate system, the position of the cavity morphological deviation region (i.e., the region where the cavity shape deviates from the standard shape) is compared with the surface defect region and the internal defect region. For example, the location of a surface defect on the bumper surface is checked to determine which area in the mold cavity it corresponds to, and whether there is a morphological deviation in that cavity area. Similarly, for internal defects, their corresponding areas in the cavity are located, and the cavity morphological deviation in that area is analyzed. Through the above comparison, the corresponding positional relationship between the cavity morphological deviation region and the defect morphological region is determined. For example, a concave deviation region on a cavity wall corresponds to a bulge defect region on the bumper surface, or a dimensional deviation in a certain area of ​​the cavity corresponds to an internal bubble defect region, etc.

[0095] Step S138: Integrate the cavity morphology deviation time sequence sub-model, surface defect morphology sub-model, and internal defect morphology sub-model according to spatial correlation, and label the cavity morphology deviation data and time sequence information corresponding to each defect area to generate a defect morphology field distribution model. The defect morphology field distribution model is used to present the spatial distribution of defects in the bumper entity and its correlation with the cavity morphology deviation.

[0096] Based on the established spatial relationships, the cavity morphology deviation time-series sub-model, surface defect morphology sub-model, and internal defect morphology sub-model are integrated. In the integrated model, each defect region is associated with a corresponding cavity morphology deviation region, and the cavity morphology deviation data (such as deviation type, deviation value, etc.) and the timing information of the deviation are labeled for that defect region. For example, for a surface depression defect region, the model labels the corresponding cavity wall protrusion deviation data and the injection timing of the deviation. Through the above integration and labeling, a defect morphology field distribution model is generated. This defect morphology field distribution model can not only present the spatial distribution of defects in the bumper entity, but also show the correlation between defects and mold cavity morphology deviations, helping to analyze the causes of defect generation.

[0097] Step S140: Perform bidirectional dynamic mapping processing on the energy field distribution model of the injection molding process and the defect morphology field distribution model to determine the correspondence between the abnormal energy field distribution area and the abnormal defect morphology field area, and generate the field mapping result.

[0098] Step S141: Extract energy field distribution data for each time series from the energy field distribution model of the injection molding process, identify regions in the energy field distribution data that exceed the preset normal energy range, mark them as abnormal energy field distribution regions, and record the energy parameter values, spatial coordinates, and time series information of the abnormal energy field distribution regions.

[0099] The energy field distribution model for the injection molding process includes comprehensive distribution data of thermal energy, pressure energy, and kinetic energy at different time sequences. First, energy field distribution data for each time sequence is extracted from this model. Then, based on a preset normal energy range—obtained through statistical analysis of a large amount of normal injection molding production data, encompassing the normal value ranges of various energy parameters in different injection stages and cavity regions—the extracted energy field distribution data is compared with the normal energy range to identify areas exceeding the normal range; these areas are designated as abnormal energy field distribution regions. For each abnormal energy field distribution region, the specific values ​​of its energy parameters (such as abnormal thermal energy, abnormal pressure energy, and abnormal kinetic energy), their coordinate positions in the three-dimensional space of the mold cavity, and the timing information of the occurrence and duration of the anomaly are recorded.

[0100] Step S142: Extract defect morphology distribution data from the defect morphology field distribution model, identify areas with defects in the defect morphology distribution data, mark them as abnormal areas of the defect morphology field, and record the defect type, morphological parameters, spatial coordinates, and temporal correlation information of the abnormal areas of the defect morphology field.

[0101] The defect morphology field distribution model contains defect distribution information for both the surface and interior of the bumper entity. Defect morphology distribution data is extracted from this model, including the spatial location, shape, and size of all surface and internal defects. Based on this data, areas with defects are identified and marked as defect morphology field anomalous areas. For each defect morphology field anomalous area, its defect type (e.g., surface scratches, internal bubbles), morphological parameters (e.g., defect length, width, volume), spatial coordinates in the 3D model of the bumper entity, and the temporal sequence information of the cavity morphology deviation corresponding to the defect obtained from the defect morphology field distribution model are recorded. This indicates which temporal cavity morphology deviation might have caused the defect, thus forming temporal correlation information.

[0102] Step S143: Establish a spatial coordinate comparison mechanism between the abnormal energy field distribution area and the abnormal defect morphology field area, match the spatial coordinates of the abnormal energy field distribution area with the spatial coordinates of the abnormal defect morphology field area one by one, and calculate the spatial overlap between the two.

[0103] Because there is a correspondence between the mold cavity and the formed bumper entity—that is, the shape of the cavity determines the shape of the bumper—there is a spatial correspondence between the abnormal energy field distribution area (located in the cavity space) and the abnormal defect morphology field area (located in the bumper entity space). A spatial coordinate comparison mechanism is established to transform the spatial coordinates of the abnormal energy field distribution area to the coordinate system of the bumper entity's 3D model, or to transform the spatial coordinates of the abnormal defect morphology field area to the coordinate system of the mold cavity, ensuring both are in the same coordinate system. Then, the spatial coordinates of the abnormal energy field distribution area and the abnormal defect morphology field area are matched one by one, and their spatial overlap is calculated. The spatial overlap can be calculated by the ratio of the intersection volume to the union volume of the two areas; the larger the ratio, the higher the degree of spatial overlap between the two areas.

[0104] Step S144: Perform temporal correlation analysis on the energy field anomaly distribution area and the defect morphology field anomaly area where the spatial overlap reaches a preset threshold, compare the temporal information of the energy field anomaly distribution area with the temporal correlation information of the defect morphology field anomaly area, and determine the synchronicity of the two in the time dimension.

[0105] A spatial overlap threshold, such as 0.7, is preset. When the spatial overlap between the abnormal energy field distribution area and the abnormal defect morphology field area reaches or exceeds this threshold, the two areas are considered to be potentially related. Temporal correlation analysis is performed on these areas, comparing the abnormal temporal information (start time, end time, duration, etc.) of the abnormal energy field distribution area with the temporal correlation information (time of corresponding cavity morphology deviation) of the abnormal defect morphology field area. The analysis examines whether there is temporal synchronicity between the two, for example, whether the timing of the energy field anomaly coincides with the timing of the cavity morphology deviation corresponding to the defect, or whether the duration of the energy field anomaly matches the time process of defect formation. Through this comparison, the temporal correlation between the energy field anomaly and the defect morphology anomaly is determined.

[0106] Step S145: Analyze the parameter correlation between pairs of abnormal regions that meet the synchronization requirements, including the correlation between the abnormal values ​​of thermal energy parameters in the abnormal energy field distribution area and the changes in defect size parameters in the abnormal defect morphology field area, and the correlation between the abnormal values ​​of pressure energy parameters in the abnormal energy field distribution area and the changes in defect shape parameters in the abnormal defect morphology field area.

[0107] Step S1451: For a set of abnormal energy field distribution areas and abnormal defect morphology field areas that meet the synchronization requirements, extract the abnormal thermal energy parameter value sequence of the abnormal energy field distribution area. The abnormal thermal energy parameter value sequence includes the abnormal thermal energy parameter values ​​of each time sequence within the abnormal time period.

[0108] For a pair of anomalous regions that meet the synchronization requirements, namely, an anomalous energy field distribution region and its corresponding defect morphology field anomalous region, the abnormal values ​​of thermal parameters within the anomalous energy field distribution region during the anomalous time period are first extracted from the energy field distribution model of the injection molding process. The anomalous time period refers to the temporal range from the start to the end of the energy field anomalous event. Within this time period, each time sequence has a corresponding abnormal value of thermal parameters, and these values ​​are arranged chronologically to form a sequence of abnormal thermal parameter values.

[0109] Step S1452: Simultaneously extract the defect size parameter sequence of the defect morphology field anomaly region, the defect size parameter sequence containing the defect size parameters of each time series corresponding to the thermal energy parameter anomaly period.

[0110] From the defect morphology field distribution model, the defect size parameters of the abnormal region of the defect morphology field within the time range corresponding to the abnormal period of thermal energy parameters are extracted. Since the formation of defects is a dynamic process related to the injection molding process, the size of the defects may change at different time sequences. Therefore, the defect size parameters (such as length, width, depth, etc.) of each time sequence are extracted to form a defect size parameter sequence.

[0111] Step S1453: Align the abnormal thermal energy parameter value sequence with the defect size parameter sequence according to the same time sequence to form parameter association data pairs. Each parameter association data pair contains the abnormal thermal energy parameter value and defect size parameter in the same time sequence.

[0112] The sequences of abnormal thermal parameters and defect size parameters are aligned one-to-one according to time sequence. For example, the abnormal thermal parameter values ​​in time sequence 1 correspond to the defect size parameters in time sequence 1, the abnormal thermal parameter values ​​in time sequence 2 correspond to the defect size parameters in time sequence 2, and so on. In this way, each time sequence forms a parameter association data pair, which contains the abnormal thermal parameter values ​​and defect size parameters in the same time sequence.

[0113] Step S1454: Observe the accompanying changes in defect size parameters when abnormal values ​​of thermal energy parameters change in the parameter correlation data pair, and record the direction and amount of change of defect size parameters each time abnormal values ​​of thermal energy parameters change.

[0114] The aligned parameter correlation data pairs are analyzed one by one to observe how the defect size parameter changes when the abnormal value of the thermal energy parameter changes. For example, when the abnormal value of the thermal energy parameter increases, does the defect size parameter increase or decrease, and by what magnitude? When the abnormal value of the thermal energy parameter decreases, how does the defect size parameter change? Record the direction (increase or decrease) and amount (change value relative to the previous time series) of the defect size parameter change caused by each abnormal value of the thermal energy parameter.

[0115] Step S1455: Extract the abnormal numerical sequence of pressure energy parameters in the abnormal energy field distribution area, and at the same time extract the defect shape parameter sequence in the abnormal defect morphology field area.

[0116] Similarly, from the energy field distribution model of the injection molding process, the abnormal value sequence of pressure energy parameters in the abnormal distribution area of ​​the energy field during the abnormal period is extracted. This abnormal value sequence of pressure energy parameters includes the abnormal values ​​of pressure energy parameters for each time sequence during the abnormal period. From the defect morphology field distribution model, the defect shape parameter sequence of the abnormal region of the defect morphology field within the corresponding time sequence is extracted. The defect shape parameters may include parameters describing the shape characteristics of the defect, such as curvature, angle, and irregularity.

[0117] Step S1456: Align the abnormal pressure energy parameter value sequence with the defect shape parameter sequence according to time sequence to form another set of parameter correlation data pairs, and record the correlation between the abnormal pressure energy parameter value changes and the defect shape parameter changes.

[0118] The abnormal pressure energy parameter sequence and the defect shape parameter sequence are aligned according to the same time sequence to form new parameter association data pairs. Each data pair contains the abnormal pressure energy parameter value and the defect shape parameter in the same time sequence. Then, the accompanying changes in the defect shape parameter when the abnormal pressure energy parameter value changes are observed. The direction and amount of change of the defect shape parameter are recorded for each change in the abnormal pressure energy parameter value, such as how the curvature and angle of the defect change when the abnormal pressure energy value increases.

[0119] Step S1457: Summarize the correlation between the recorded thermal energy parameters and defect size parameters, and the correlation between the pressure energy parameters and defect shape parameters, and summarize the correlation rules of parameter changes.

[0120] The correlation between recorded abnormal values ​​of thermal energy parameters and changes in defect size parameters, as well as the correlation between abnormal values ​​of pressure energy parameters and changes in defect shape parameters, were analyzed and summarized. For example, it was found that when the abnormal value of thermal energy parameters increases by a certain amount, the defect size parameter increases accordingly; or when the abnormal value of pressure energy parameters changes within a certain range, the defect shape parameter exhibits a specific change pattern. Through the above summarization, the correlation rules between the parameter changes of the two were concluded.

[0121] Step S1458: Combine the summarized correlation patterns with the spatial location information and temporal information of the corresponding abnormal region pairs to form a complete parameter correlation description of the corresponding abnormal region pairs.

[0122] The summarized correlation patterns of parameter changes are combined with the spatial location information (spatial coordinates and relative positions of the abnormal energy field region and the abnormal defect morphology field region) and temporal information (time of occurrence and duration of the anomaly) of the abnormal region to form a complete description of the parameter correlation relationship of the abnormal region. For example, in a specific spatial region (coordinate range) of the mold cavity, at a certain time stage of injection molding, when the thermal energy parameter abnormally increases by a certain value range, it will cause the defect size parameter at the corresponding position of the bumper to increase according to a certain pattern. At the same time, abnormal changes in the pressure energy parameter will cause the defect shape parameter to change according to another pattern, and so on.

[0123] Step S146: Construct a bidirectional mapping rule base and formulate bidirectional mapping rules for energy field anomaly characteristics and defect morphology anomaly characteristics based on parameter correlation.

[0124] Based on the parameter correlations of multiple pairs of abnormal regions obtained from the analysis, a bidirectional mapping rule base is constructed. These rules include mapping rules from abnormal energy field characteristics to abnormal defect morphology characteristics, and mapping rules from abnormal defect morphology characteristics to abnormal energy field characteristics. For example, a rule might stipulate that when a certain area of ​​the mold cavity exhibits abnormally high thermal parameters during the injection stage, and this persists for a certain period, it will lead to a bulge defect on the corresponding surface area of ​​the bumper, with the size of the bulge being proportional to the magnitude of the elevated thermal parameters. Conversely, when a bulge defect of a specific size and shape appears on the bumper surface, it can be inferred that the corresponding area of ​​the mold cavity exhibits abnormally high thermal parameters during the injection stage. These rules are then organized and stored in the bidirectional mapping rule base.

[0125] Step S147: Based on the bidirectional mapping rule base, perform bidirectional verification on all energy field anomaly distribution areas and defect morphology field anomaly areas to confirm the correspondence of anomaly areas that conform to the mapping rules.

[0126] Using rules from the bidirectional mapping rule base, all identified energy field anomaly distribution regions and defect morphology field anomaly regions are bidirectionally verified. For each energy field anomaly distribution region, the possible defect morphology anomaly characteristics are predicted according to the rules, and then compared with the actual defect morphology field anomaly region to see if there are any defect regions that match the predicted characteristics. For each defect morphology field anomaly region, the possible energy field anomaly characteristics are inferred according to the rules, and then compared with the actual energy field anomaly distribution region. Through this bidirectional verification, anomaly region correspondences that conform to the mapping rules are filtered out, and anomaly region pairs that, although having high spatial overlap and temporal synchronization, do not conform to the parameter association rules are excluded.

[0127] Step S148: Integrate the confirmed abnormal region correspondence, spatial overlap data, temporal synchronization data, parameter correlation data, and bidirectional mapping rule application to generate a field mapping result. The field mapping result can clearly define the correspondence between how energy field anomalies lead to defect morphological anomalies.

[0128] The system integrates information such as the correspondence between the energy field anomaly distribution region and the defect morphology field anomaly region confirmed through bidirectional verification, spatial overlap data (overlap values ​​for each corresponding region pair), temporal synchronization data (description of the degree of synchronization between the two in time), parameter correlation data (description of the correlation between thermal energy and defect size, pressure energy and defect shape, etc.), and the application of bidirectional mapping rules (which rules were applied to which region pairs for verification). This information is organized into a structured form, containing multiple entries, each corresponding to a set of detailed mapping information for anomaly region pairs, thus generating a field mapping result. This field mapping result clarifies the correspondence between the energy field anomaly distribution region and the defect morphology field anomaly region, as well as the process and rules by which energy field anomalies lead to defect morphology anomalies through parameter changes.

[0129] Step S149: The method further includes a step of verifying and optimizing the field mapping results, including: Step S1491: Obtain field mapping result data for similar historical bumper injection molding scenarios. The field mapping result data for similar historical bumper injection molding scenarios includes abnormal energy field distribution areas, abnormal defect morphology field areas, correspondence between abnormal areas, and parameter correlation rules in historical cases.

[0130] We collected field mapping results data generated during the injection molding production process of bumpers of the same model or similar structure as the current car front bumper. The above historical data came from previous production batches and included information such as abnormal energy field distribution areas, corresponding abnormal defect morphology areas, the correspondence between abnormal areas and parameters under different production conditions.

[0131] Step S1492: Cross-compare the currently generated field mapping result with the field mapping result data of the same type of bumper injection molding scenario in the past, and analyze the consistency between the currently generated field mapping result and the field mapping result data of the same type of bumper injection molding scenario in terms of the correspondence of abnormal areas and the correlation of parameters, and mark the parts with differences.

[0132] The current field mapping results, including the correspondence between abnormal regions and the correlation patterns of parameters, are compared item by item with the field mapping results of similar historical scenarios. For example, the correspondence between a certain group of abnormal energy field regions and defect morphology regions is compared to see if similar situations exist in historical data, and whether the corresponding parameter correlation patterns are consistent. Any discrepancies, such as differences between the current parameter correlation patterns and those in historical data, or inconsistencies in the correspondence between abnormal regions, are marked.

[0133] Step S1493: For the differences in the markings, combine the actual parameter data of the current injection molding process, mold status data and defect morphology data to analyze the cause of the differences and determine whether the differences are caused by scene differences or mapping rule deviations.

[0134] For the discrepancies in the markings, a thorough analysis of their causes is needed. Collect actual parameter data for the current injection molding process, such as the injection molding machine's settings and material performance parameters; mold status data, such as the number of times the mold has been used, maintenance records, and current wear status; and more detailed defect morphology data. Combining this data, determine whether the discrepancies are due to objective differences between the current and historical scenarios (such as different material batches, different mold wear levels, or changes in injection molding machine performance), or due to deviations or imperfections in the bidirectional mapping rules used when constructing the current field mapping results.

[0135] Step S1494: If the difference is caused by the deviation of the mapping rule, re-examine the corresponding rule in the bidirectional mapping rule base, and modify the rule content in combination with the actual relationship of the current scenario so that the modified rule can accurately reflect the corresponding logic of energy field anomaly and defect morphology anomaly.

[0136] If the analysis determines that the discrepancy is caused by a deviation in the mapping rules—that is, some rules in the currently used bidirectional mapping rule base are not accurately applicable to the actual situation in the current scenario—then these corresponding rules should be re-examined. Combined with the actual correlation between energy field anomalies and defect morphology anomalies in the current scenario (obtained through in-depth analysis of the current data), the rules should be revised. For example, the description of the relationship between parameter variation magnitude and defect characteristic variation in the rules may be adjusted, or new rules may be added to adapt to the specific circumstances of the current scenario, ensuring that the revised rules accurately reflect the correspondence between energy field anomalies and defect morphology anomalies.

[0137] Step S1495: Based on the revised bidirectional mapping rule, re-verify the correlation between the abnormal energy field distribution area and the abnormal defect morphology field area, and adjust the correspondence of the abnormal areas so that the correspondence of the abnormal areas meets the requirements of the revised rule.

[0138] Using the revised bidirectional mapping rule base, the correlation verification between the current energy field anomaly distribution region and the defect morphology field anomaly region is re-performed. According to the new rules, the previous anomaly region correspondences are checked and adjusted; correspondences that do not conform to the new rules are corrected, and correspondences that do conform to the new rules are supplemented to ensure that the final anomaly region correspondences meet the requirements of the revised rules.

[0139] Step S1496: Calculate the matching accuracy of the correspondence between abnormal regions in the optimized field mapping result. The matching accuracy is determined by the ratio of the number of abnormal region pairs that conform to the mapping rules to the total number of abnormal region pairs.

[0140] After adjusting the field mapping results, calculate the matching accuracy of the correspondence between abnormal regions. Statistically calculate the number of abnormal region pairs conforming to the bidirectional mapping rules in the optimized field mapping results, as well as the total number of abnormal region pairs. The matching accuracy equals the number of abnormal region pairs conforming to the mapping rules divided by the total number of abnormal region pairs.

[0141] Step S1497: If the matching accuracy does not meet the preset requirements, repeat the cross-matching, difference analysis, rule correction and association verification steps until the matching accuracy meets the preset requirements.

[0142] A preset matching accuracy threshold, such as 0.9, is established. If the calculated matching accuracy does not reach this threshold, the process returns to step S1492 to repeat the steps of historical data cross-comparison, difference analysis, mapping rule correction, and association verification. Through multiple iterative optimizations, the mapping rules and the correspondence between abnormal regions are continuously adjusted until the matching accuracy reaches the preset requirement.

[0143] Step S1498: Record key data throughout the verification and optimization process, integrate the key data with the optimized field mapping results to form the final field mapping results, and ensure that the accuracy and reliability of the final field mapping results meet the preset standards. The key data includes historical comparison data, difference analysis results, rule correction content, and verification results.

[0144] During the verification and optimization process, detailed records of key data are kept, such as specific differences from historical comparisons, conclusions of difference analysis, modifications to the mapping rules (rules before and after modification), and results of each association verification (number of region pairs conforming to the rules, total number of region pairs, matching accuracy, etc.). These key data are then integrated with the optimized field mapping results to form the final field mapping result. This final field mapping result not only includes core information such as the correspondence between abnormal regions but also includes a detailed record of the verification and optimization process, ensuring its accuracy and reliability meet preset standards.

[0145] Step S150: Integrate the energy field distribution model of the injection molding process, the defect morphology field distribution model, and the field mapping results to generate a bumper injection molding defect mechanism detection report. At the same time, based on the parameter characteristics of the abnormal energy field distribution area, generate an injection molding energy control scheme. Send the bumper injection molding defect mechanism detection report and the injection molding energy control scheme to the injection molding production management terminal.

[0146] This paper integrates the energy field distribution model, defect morphology field distribution model, and field mapping results of the injection molding process. The energy field distribution model provides the dynamic distribution of the energy field, the defect morphology field distribution model shows the spatial distribution and morphological characteristics of defects, and the field mapping results reveal the correspondence between energy field anomalies and defect morphology anomalies. By integrating the information from these three sources, the mechanism of defect generation is analyzed, such as which abnormal energy field distributions lead to which defects through which parameter relationships. Based on the above mechanism analysis, a bumper injection molding defect mechanism detection report is generated, which includes a detailed analysis of the defect type, distribution, and causes (energy field anomalies and cavity morphology deviations, etc.).

[0147] Step S151: Based on the parameter characteristics of the abnormal energy field distribution area, generate an injection molding energy control scheme, including: Step S1511: Extract parameter features of the abnormal energy field distribution area from the field mapping result, extract information related to the abnormal energy type of each abnormal energy field distribution area, determine the abnormal energy type of each abnormal energy field distribution area, and simultaneously extract information related to the abnormal parameter value range, abnormal duration time range, and corresponding defect type of each abnormal energy field distribution area.

[0148] From the field mapping results, parameter features are extracted for each energy field anomaly distribution region. First, the type of anomalous energy is determined, i.e., whether the energy anomaly in this region is a thermal energy anomaly, a pressure energy anomaly, or a kinetic energy anomaly. Then, the numerical range of the anomalous energy parameters is extracted, i.e., the minimum and maximum values ​​of the anomalous values; the temporal range of the anomaly's duration, i.e., the start and end times of the anomaly; and the type of defect corresponding to the energy field anomaly distribution region, such as whether the energy anomaly causes surface scratches, internal bubbles, or other types of defects. The above information together constitutes the parameter features of the energy field anomaly distribution region.

[0149] Step S1512: Obtain the standard energy parameter range for the injection molding of the corresponding model of bumper. The standard energy parameter range includes the normal value range and time sequence variation law of various energy parameters in different injection molding stages.

[0150] By consulting injection molding production process databases or technical documents, the standard energy parameter range for injection molding of the corresponding model of the currently produced automotive front bumper can be obtained. This standard energy parameter range is based on extensive experimental and production experience. For different injection stages (such as injection, holding pressure, and cooling), the normal value ranges of thermal energy parameters, pressure energy parameters, and kinetic energy parameters are specified, as well as the normal variation patterns of these parameters under different time sequences. For example, during the injection stage, the pressure energy should gradually increase to a peak and then remain stable, while during the cooling stage, the thermal energy parameter should gradually decrease.

[0151] Step S1513: Compare the range of abnormal parameter values ​​for each energy field anomaly distribution region with the range of standard energy parameters for the corresponding stage, calculate the deviation of the abnormal parameter values ​​from the standard energy parameter range, and analyze the timing nodes where the deviation occurs.

[0152] The abnormal parameter values ​​for each energy field anomaly distribution area are compared with the standard energy parameter range for that area during the injection molding stage. For example, if an abnormal energy field distribution area is in the holding pressure stage, its abnormal pressure energy value range is P1-P2, while the standard pressure energy range for the holding pressure stage is the lower limit of standard P to the upper limit of standard P. The deviation of the abnormal parameter values ​​from the standard range is calculated, such as the difference between the average value of the abnormal values ​​and the center value of the standard range, or the maximum extent by which the abnormal values ​​exceed the standard range. Simultaneously, the specific time points at which the deviation occurs are analyzed, i.e., at which time point the energy parameters begin to deviate from the standard range, at which time point the deviation reaches its maximum, and at which time point the deviation ends.

[0153] Step S1514: Based on the type and magnitude of the abnormal energy, and in conjunction with the principle of injection molding energy control, determine the control direction for the abnormal distribution area of ​​the energy field.

[0154] Based on the type of abnormal energy (thermal, pressure, or kinetic energy) and the calculated deviation range, and in accordance with the basic principles of injection molding energy control, the control direction is determined. For example, if the thermal energy parameter is abnormally high and the deviation range is large, the control direction is to reduce the thermal energy input in that area; if the pressure energy parameter is abnormally low, the control direction is to increase the corresponding pressure; if the kinetic energy parameter fluctuates abnormally, the control direction is to stabilize the change in the kinetic energy parameter. Determining the control direction requires considering the impact of energy parameters on material molding to ensure that the control measures can effectively improve defects.

[0155] Step S1515: Consult the injection molding energy control case library to find historical control cases that are the same as or similar to the current abnormal energy type, deviation range, and corresponding defect type. Extract the control methods, adjustment range of control parameters, control implementation sequence, and control effect data from the historical control cases.

[0156] The injection molding energy control case library stores past cases of handling various energy anomalies. By consulting this library, based on the type of abnormal energy, the magnitude of the deviation, and the corresponding defect type in the current energy field anomaly distribution area, similar or identical historical control cases can be found. From these cases, control methods are extracted, such as adjusting heating temperature, changing injection pressure, and optimizing screw speed; the adjustment range of control parameters, such as how many degrees the temperature is reduced or how many Pascals the pressure is increased; the timing of the control implementation, i.e., at which injection stage the adjustment is made; and the post-control effect data, such as whether defects are reduced and to what extent.

[0157] Step S1516: Based on the control data of historical control cases and combined with the actual operating status of the current injection molding equipment, determine the specific control measures for each abnormal energy field distribution area.

[0158] While referencing historical control methods and parameter adjustment ranges, adjustments must be made based on the actual operating status of the current injection molding equipment. The actual operating status of the current injection molding equipment includes its model, performance parameters, current workload, and wear condition. For example, historical cases may have used increased injection pressure to address certain types of pressure energy anomalies, but the maximum injection pressure of the current injection molding machine may be limited. Therefore, it is necessary to adjust the pressure increase range according to the actual equipment conditions, or select other feasible control methods, such as adjusting the holding time. Through the above methods, specific and feasible control methods can be determined for each area of ​​abnormal energy field distribution.

[0159] Step S1517: Determine the implementation sequence of each control measure so that the control operation is implemented in the corresponding injection molding stage.

[0160] Based on the duration of the energy field anomaly and the corresponding injection molding stage, the timing of each control measure should be determined. For example, if the energy anomaly occurs in the middle of the injection stage, the control measures should be implemented in the middle or slightly earlier of the injection stage to ensure timely intervention when the anomaly occurs. The timing of the control measures should correspond to the stages of the injection molding process to ensure that the control effect is accurately applied to the anomaly area.

[0161] Step S1518: Develop monitoring indicators for the control effect. For each control measure, set corresponding energy parameter monitoring indicators and defect improvement monitoring indicators.

[0162] To evaluate the effectiveness of control measures, corresponding monitoring indicators need to be established. For each control measure, energy parameter monitoring indicators should be set, such as the range within which thermal energy parameters should stabilize after control, and the fluctuation range of pressure energy parameters should be controlled within a certain range. Simultaneously, defect improvement monitoring indicators should be set, such as the percentage by which the number of corresponding defect types should be reduced, and the extent to which defect size should be reduced. These monitoring indicators should be specific and quantifiable to facilitate effectiveness evaluation after control measures are implemented.

[0163] Step S1519: Organize and sort the control direction, specific control methods, implementation sequence, and control effect monitoring indicators corresponding to each abnormal energy field distribution area according to the injection molding stage to form a complete injection molding energy control scheme. The injection molding energy control scheme is used to guide the precise control of energy field anomalies during the production process to reduce the generation of defects.

[0164] The control direction, specific control methods (such as adjusting the heating coil temperature to a certain range, increasing the injection pressure by a certain amount), implementation sequence (such as starting adjustment at time X of the injection stage), and monitoring indicators for the control effect (such as controlling the fluctuation range of thermal parameters within ±ΔT, reducing the number of surface bulge defects by Y%) for each abnormal energy field distribution area are organized and sorted according to the order of the injection molding stages. For example, control measures belonging to the injection stage are grouped together, and control measures belonging to the holding pressure stage are grouped together. Through the above organization and sorting, a clear and complete injection molding energy control scheme is formed. This injection molding energy control scheme can guide production personnel to perform precise control operations for energy field anomalies occurring in different injection molding stages during actual production, thereby effectively reducing the generation of defects.

[0165] Finally, the generated bumper injection molding defect mechanism detection report and injection energy control plan are sent to the injection molding production management terminal via network transmission or data interface. Production managers can view the report and plan through this terminal to understand the causes of defects and adjust and optimize the injection molding production process according to the control plan.

[0166] Figure 2 The diagram illustrates exemplary hardware and software components of an intelligent bumper injection molding defect detection system 100, which can implement the ideas of this application, according to some embodiments of this application. For example, a processor 120 can be used in the intelligent bumper injection molding defect detection system 100 and to perform the functions described in this application.

[0167] For example, the intelligent detection system 100 for bumper injection molding defects may include a network port 110 connected to a network, one or more processors 120 for executing program instructions, a communication bus 130, and various forms of storage media 140, such as a disk, ROM, or RAM, or any combination thereof. Exemplarily, the intelligent detection system 100 for bumper injection molding defects may also include program instructions stored in ROM, RAM, or other types of non-transitory storage media, or any combination thereof. The methods of this application can be implemented according to the aforementioned program instructions. The intelligent detection system 100 for bumper injection molding defects also includes an I / O interface 150 between the computer and other input / output devices.

[0168] Furthermore, this embodiment of the invention also provides a readable storage medium, wherein computer-executable instructions are preset in the readable storage medium, and when the processor executes the computer-executable instructions, the above-mentioned intelligent detection method for bumper injection molding defects is implemented.

[0169] It should be noted that, in order to simplify the description of the present invention and thus help to understand one or more embodiments of the invention, multiple features may sometimes be grouped into one embodiment, drawing or description thereof in the foregoing description of the embodiments of the present invention.

Claims

1. A bumper injection molding defect intelligent detection method, characterized in that, The method comprises: Collecting a multi-dimensional energy parameter set, a mold cavity shape data set, and a defect shape data set of a to-be-detected bumper entity after molding during a bumper injection molding process, the multi-dimensional energy parameter set comprising thermal energy parameters, pressure energy parameters, and kinetic energy parameters at each stage of injection molding, the mold cavity shape data set comprising real-time shape change data of the mold cavity during injection molding, and the defect shape data set comprising shape feature data of surface and internal defects of the to-be-detected bumper entity; Constructing an injection molding process energy field distribution model based on the multi-dimensional energy parameter set, the injection molding process energy field distribution model being used to present the distribution state of energy at each stage of injection molding in the mold cavity and the material interior; Constructing a defect shape field distribution model according to the mold cavity shape data set and the defect shape data set, the defect shape field distribution model being used to present the spatial shape and distribution law of defects in the bumper entity; Performing bidirectional dynamic mapping processing on the injection molding process energy field distribution model and the defect shape field distribution model to determine the corresponding relationship between the energy field abnormal distribution area and the defect shape field abnormal area, and generating a field domain mapping result; Integrating the injection molding process energy field distribution model, the defect shape field distribution model, and the field domain mapping result to generate a bumper injection molding defect mechanism detection report, and generating an injection molding energy regulation scheme according to the parameter characteristics of the energy field abnormal distribution area, and sending the bumper injection molding defect mechanism detection report and the injection molding energy regulation scheme to an injection molding production management terminal.

2. The bumper injection molding defect intelligent detection method of claim 1, wherein, The method comprises: Classifying and extracting thermal energy parameters in the multi-dimensional energy parameter set to obtain material melting thermal energy parameters, mold cavity wall thermal energy parameters, and material cooling thermal energy parameters, each type of thermal energy parameter comprising numerical data at different injection molding time sequences; Sorting each type of thermal energy parameter according to the injection molding time sequence to form a thermal energy parameter time sequence, and labeling the injection molding stage corresponding to each time sequence; Processing the thermal energy parameter time sequence by using a field domain modeling method, taking a three-dimensional structure of the mold cavity as a spatial framework, mapping each type of thermal energy parameter value at each time sequence to a corresponding spatial position to generate a thermal energy field time sequence distribution sub-model; Classifying and extracting pressure energy parameters in the multi-dimensional energy parameter set to obtain injection pressure energy parameters, holding pressure energy parameters, and cavity back pressure energy parameters, each type of pressure energy parameter comprising numerical data at different injection molding time sequences; Time-sequentially sorting each type of pressure energy parameter and labeling the injection molding stage, and then mapping the pressure energy parameter values to the three-dimensional spatial framework of the mold cavity by using the field domain modeling method to generate a pressure energy field time sequence distribution sub-model; Classifying and extracting kinetic energy parameters in the multi-dimensional energy parameter set to obtain material flow kinetic energy parameters, mold opening and closing kinetic energy parameters, and ejection mechanism kinetic energy parameters, each type of kinetic energy parameter comprising numerical data at different injection molding time sequences; Mapping each type of kinetic energy parameter value to a corresponding spatial position to generate a kinetic energy field time sequence distribution sub-model; Determine the energy field weight distribution rule, according to the influence degree of different injection stages on the material forming, assign dynamic weights to the thermal energy field time sequence distribution sub-model, the pressure energy field time sequence distribution sub-model and the kinetic energy field time sequence distribution sub-model; Integrating the thermal energy field time sequence distribution sub-model, the pressure energy field time sequence distribution sub-model and the kinetic energy field time sequence distribution sub-model according to the dynamic weights to generate an injection process energy field distribution model containing time sequence dimension and space dimension, wherein the integrating refers to correlating the space distribution data of the thermal energy field time sequence distribution sub-model, the pressure energy field time sequence distribution sub-model and the kinetic energy field time sequence distribution sub-model at the same time sequence to form a multi-physical field coupling model, and adjusting the relative importance of each energy field in the multi-physical field coupling model through dynamic weights without directly superimposing numerical values, and the injection process energy field distribution model is used to present the comprehensive distribution state of each energy in the mold cavity and the material interior at different time sequences.

3. The bumper injection molding defect intelligent detection method of claim 2, wherein, The field modeling method is used to process the thermal energy parameter time sequence, take the three-dimensional structure of the mold cavity as a spatial framework, map each type of thermal energy parameter value of each time sequence to the corresponding spatial position to generate a thermal energy field time sequence distribution sub-model, including: Obtain the three-dimensional structure data of the bumper injection mold cavity, construct a mold cavity three-dimensional structure model through a three-dimensional modeling software, and label the spatial coordinates and geometric shapes of each region inside the mold cavity three-dimensional structure model; Analyze the material melting thermal energy parameters of each time sequence, determine the transfer rule of the material melting thermal energy inside the material, and distribute the material melting thermal energy parameter values to the spatial regions corresponding to the material flow in the mold cavity three-dimensional structure model according to the material flow path; Analyze the mold cavity wall thermal energy parameters of each time sequence, map the mold cavity wall thermal energy parameter values to the wall spatial positions of the mold cavity three-dimensional structure model according to the temperature conduction characteristics of the mold cavity wall, and label the thermal energy transfer rates of different regions of the wall; Analyze the material cooling thermal energy parameters of each time sequence, combine the distribution positions of the cooling system in the mold, map the material cooling thermal energy parameter values to the spatial regions close to the cooling pipeline in the mold cavity three-dimensional structure model to reflect the influence of the cooling process on the material thermal energy; Perform spatial interpolation processing on each type of thermal energy parameter value mapped to the mold cavity three-dimensional structure model to form a continuous distribution of the thermal energy parameter values in the spatial region, avoiding field rupture caused by discrete numerical values; Add time sequence identifiers to the thermal energy field distribution data generated for each time sequence to form a thermal energy field distribution snapshot of a single time sequence; Serially connect the thermal energy field distribution snapshots of all time sequences in the order of injection time sequence to form a thermal energy field time sequence distribution sub-model capable of dynamically displaying the changes of the thermal energy field. Perform spatial verification on the generated thermal energy field time sequence distribution sub-model to check whether each type of thermal energy parameter value is matched with the geometric shape of the mold cavity three-dimensional structure model in the spatial mapping process, correct the mapping data with spatial coordinate misplacement, and make the spatial accuracy of the thermal energy field time sequence distribution sub-model meet the preset standard.

4. The bumper injection molding defect intelligent detection method of claim 1, wherein, The defect morphology field distribution model is constructed according to the mold cavity morphology data set and the defect morphology data set, and comprises: The data in the mold cavity morphology data set is classified and extracted to obtain cavity wall surface morphology data, cavity parting surface morphology data and cavity internal insert morphology data, and each type of morphology data contains morphology change data at different injection times; The extracted cavity morphology data of each type is sorted according to the injection time sequence to form a cavity morphology time sequence, and the injection stage corresponding to each time sequence is labeled, and the deviation data of each time sequence cavity morphology and the standard cavity morphology are recorded; The three-dimensional morphology modeling method is used to process the cavity morphology time sequence, and the standard cavity three-dimensional structure is taken as the reference, the cavity morphology deviation data of each type at each time sequence is mapped to the corresponding spatial position, and a cavity morphology deviation time sequence sub-model is generated; The data in the defect morphology data set is classified and extracted to obtain surface defect morphology data and internal defect morphology data, the surface defect morphology data contains surface position, shape and size data of the defect, and the internal defect morphology data contains internal space position, shape and size data of the defect; The surface defect morphology data is processed by spatial positioning, and the three-dimensional model of the formed bumper entity is taken as the spatial framework, the position, shape and size data of each surface defect are mapped to the corresponding surface spatial coordinates to generate a surface defect morphology sub-model; The internal defect morphology data is processed by spatial positioning, and the internal defect imaging data obtained by the non-destructive testing technology is used to map the spatial position, shape and size data of the internal defect to the internal spatial coordinates of the bumper entity three-dimensional model to generate an internal defect morphology sub-model; The spatial correlation between the cavity morphology deviation time sequence sub-model and the surface defect morphology sub-model and the internal defect morphology sub-model is analyzed to determine the corresponding position relationship between the cavity morphology deviation region and the defect morphology region; The cavity morphology deviation time sequence sub-model, the surface defect morphology sub-model and the internal defect morphology sub-model are integrated according to the spatial correlation, and the cavity morphology deviation data and the time sequence information corresponding to each defect region are labeled to generate a defect morphology field distribution model, which is used to present the spatial distribution of the defect in the bumper entity and the correlation state with the cavity morphology deviation.

5. The bumper injection molding defect intelligent detection method of claim 4, wherein, The surface defect morphology data is processed by spatial positioning, and the three-dimensional model of the formed bumper entity is taken as the spatial framework, the position, shape and size data of each surface defect are mapped to the corresponding surface spatial coordinates to generate a surface defect morphology sub-model, and comprises: Three-dimensional scanning data of the formed bumper entity to be detected is obtained, a bumper entity three-dimensional model is constructed by three-dimensional reconstruction software, and the spatial coordinates of each region on the surface of the bumper entity three-dimensional model are labeled; The surface defect morphology data in the defect morphology data set is analyzed one by one, and for each surface defect, its approximate position description on the surface of the bumper entity is extracted; According to the approximate position description of the surface defect, the corresponding surface region range in the bumper entity three-dimensional model is determined, and the surface region range is taken as the preliminary spatial positioning region of the surface defect; The surface image of the preliminary spatial positioning area is acquired by using a high-resolution image acquisition device, and pixel coordinates and pixel size data of the surface defects in the image are extracted by using image analysis technology; A mapping relationship between the surface coordinates of the bumper entity three-dimensional model and the pixel coordinates of the high-resolution image is established, and the pixel coordinates of the surface defects are converted into the spatial coordinates of the surface of the three-dimensional model according to the mapping relationship, and the pixel size data is converted into the actual physical size data; According to the converted spatial coordinates and the actual physical size data, a three-dimensional morphology model of the defect is constructed at the corresponding surface position of the bumper entity three-dimensional model; Each three-dimensional morphology model of the surface defect is labeled with attributes, including defect type, size parameter, and morphology feature; All three-dimensional morphology models of the surface defects are integrated into the bumper entity three-dimensional model according to their spatial coordinates to form a surface defect morphology sub-model, which clearly presents the spatial distribution and morphology features of all surface defects on the surface of the bumper entity.

6. The bumper injection molding defect intelligent detection method of claim 1, wherein, The energy field distribution model of the injection molding process and the defect morphology field distribution model are bidirectionally dynamically mapped to determine the correspondence between the energy field abnormal distribution area and the defect morphology field abnormal area, and a field mapping result is generated, including: Energy field distribution data of each time sequence is extracted from the injection molding process energy field distribution model, and the area outside the preset normal energy range in the energy field distribution data is identified and marked as an energy field abnormal distribution area, while the energy parameter value, spatial coordinates, and time sequence information of the energy field abnormal distribution area are recorded; Defect morphology distribution data is extracted from the defect morphology field distribution model, and the area with defects in the defect morphology distribution data is identified and marked as a defect morphology field abnormal area, while the defect type, morphology parameter, spatial coordinates, and time sequence correlation information of the defect morphology field abnormal area are recorded; A spatial coordinate comparison mechanism of the energy field abnormal distribution area and the defect morphology field abnormal area is established, and the spatial coordinates of the energy field abnormal distribution area and the spatial coordinates of the defect morphology field abnormal area are matched one by one, and the spatial overlap degree is calculated; The energy field abnormal distribution area and the defect morphology field abnormal area with a spatial overlap degree reaching a preset threshold are analyzed in time sequence, and the time sequence information of the energy field abnormal distribution area and the time sequence correlation information of the defect morphology field abnormal area are compared to determine the synchronization of the two in the time dimension; The parameter correlation between the abnormal area pairs meeting the requirements of synchronization is analyzed, including the change correlation between the thermal energy parameter abnormal value of the energy field abnormal distribution area and the defect size parameter of the defect morphology field abnormal area, and the change correlation between the pressure energy parameter abnormal value of the energy field abnormal distribution area and the defect shape parameter of the defect morphology field abnormal area; A bidirectional mapping rule library is constructed, and bidirectional mapping rules of energy field abnormal features and defect morphology abnormal features are formulated based on the parameter correlation; According to the bidirectional mapping rule library, all energy field abnormal distribution areas and defect morphology field abnormal areas are bidirectionally verified to confirm the corresponding relationship of the abnormal areas meeting the mapping rules. The confirmed abnormal area corresponding relationship, spatial overlap data, time sequence synchronization data, parameter correlation data and bidirectional mapping rule application are integrated to generate a field mapping result, which can clearly determine the corresponding relationship between the energy field anomaly and the defect morphology anomaly.

7. The bumper injection molding defect intelligent detection method of claim 6, wherein, The parameter correlation between the abnormal areas that meet the synchronization requirements is analyzed, including the correlation between the thermal energy parameter abnormal value of the energy field abnormal distribution area and the defect size parameter of the defect morphology field abnormal area, the correlation between the pressure energy parameter abnormal value of the energy field abnormal distribution area and the defect shape parameter of the defect morphology field abnormal area, including: For a certain group of energy field abnormal distribution areas and defect morphology field abnormal areas that meet the synchronization requirements, the thermal energy parameter abnormal value sequence of the energy field abnormal distribution area is extracted, and the thermal energy parameter abnormal value sequence includes the thermal energy parameter abnormal value of each time sequence in the abnormal period; At the same time, the defect size parameter sequence of the defect morphology field abnormal area is extracted, and the defect size parameter sequence includes the defect size parameter of each time sequence corresponding to the thermal energy parameter abnormal period; The thermal energy parameter abnormal value sequence and the defect size parameter sequence are aligned according to the same time sequence to form a parameter correlation data pair, and each parameter correlation data pair includes the thermal energy parameter abnormal value and the defect size parameter of the same time sequence; In the parameter correlation data pair, the accompanying change of the defect size parameter when the thermal energy parameter abnormal value changes is observed, and the change direction and amount of the defect size parameter when the thermal energy parameter abnormal value changes are recorded; The pressure energy parameter abnormal value sequence of the energy field abnormal distribution area is extracted, and the defect shape parameter sequence of the defect morphology field abnormal area is extracted at the same time; The pressure energy parameter abnormal value sequence and the defect shape parameter sequence are aligned according to the time sequence to form another group of parameter correlation data pairs, and the correlation between the pressure energy parameter abnormal value change and the defect shape parameter change is recorded; The correlation between the recorded thermal energy parameter and the defect size parameter, and the correlation between the pressure energy parameter and the defect shape parameter are summarized to summarize the correlation rule of parameter change; The summarized correlation rule is combined with the spatial position information and time sequence information of the corresponding abnormal area pair to form a complete parameter correlation description of the corresponding abnormal area pair.

8. The bumper injection molding defect intelligent detection method of claim 1, wherein, The parameter characteristics of the energy field abnormal distribution area are extracted from the field mapping result, the abnormal energy type related information of each energy field abnormal distribution area is extracted, the abnormal energy type of each energy field abnormal distribution area is determined, the abnormal parameter value range, the abnormal duration time sequence range and the corresponding defect type related information of each energy field abnormal distribution area are extracted, and the abnormal parameter value range, the abnormal duration time sequence range and the corresponding defect type of each energy field abnormal distribution area are determined. The standard energy parameter range of the corresponding type of bumper injection is obtained, and the standard energy parameter range includes the normal value interval and time sequence change rule of various energy parameters in different injection stages. ​ The abnormal parameter value range of each energy field abnormal distribution area is compared with the standard energy parameter range of the corresponding stage, the deviation amplitude of the abnormal parameter value and the standard energy parameter range is calculated, and the time sequence node of the deviation is analyzed; According to the abnormal energy type and the deviation amplitude, the control direction for the energy field abnormal distribution area is determined combined with the injection energy control principle; The injection energy control case library is consulted to find historical control cases similar to the current abnormal energy type, deviation amplitude, and corresponding defect type, and the control means, control parameter adjustment amplitude, control implementation time sequence, and control effect data in the historical control cases are extracted; According to the control data of the historical control cases, the specific control means for each energy field abnormal distribution area is determined combined with the actual running state of the current injection equipment; The implementation time sequence of each control means is determined to make the control operation implemented in the corresponding injection stage; The control effect monitoring index is formulated, and for each control means, the corresponding energy parameter monitoring index and defect improvement monitoring index are set; The control direction, specific control means, implementation time sequence, and control effect monitoring index corresponding to each energy field abnormal distribution area are sorted according to the injection stage to form a complete injection energy control scheme, which is used to guide the precise control of the energy field abnormality in the production process to reduce defects.

9. The bumper injection molding defect intelligent detection method of claim 2, wherein, The mapping of various kinetic energy parameter values to corresponding spatial positions to generate a kinetic energy field time sequence distribution sub-model includes: The material flow kinetic energy parameters, mold opening and closing kinetic energy parameters, and ejection mechanism kinetic energy parameters classified and extracted are analyzed one by one to determine the transmission law of various kinetic energy parameters in the injection molding process, and the action mode of different kinetic energy parameters on material forming and cavity shape is analyzed; For material flow kinetic energy parameters, according to the flow speed and flow path of the material in the cavity during the injection molding process, the material flow kinetic energy parameter values are distributed to the spatial regions through which the material flows in the mold cavity three-dimensional structure model, and the value distribution result is calibrated through material flow simulation data to make the value distribution result consistent with the actual material flow trajectory; For mold opening and closing kinetic energy parameters, combined with the motion trajectory of the mold opening and closing mechanism and the position correspondence with the cavity, the mold opening and closing kinetic energy parameter values are mapped to the spatial regions near the mold opening and closing components in the mold cavity three-dimensional structure model, and the mapping result is verified through mold motion simulation data to reflect the influence of mold opening and closing motion on the energy distribution around the cavity; For ejection mechanism kinetic energy parameters, according to the installation position of the ejection mechanism in the mold and the cavity area corresponding to the ejection action, the ejection mechanism kinetic energy parameter values are mapped to the spatial regions in contact with the ejection mechanism in the mold cavity three-dimensional structure model, and the mapping result is calibrated through ejection action simulation data to reflect the effect of kinetic energy generated by the ejection action on the local area. The kinetic energy parameter values of the material flow, the mold opening and closing, and the ejection mechanism mapped to the three-dimensional structure model of the mold cavity are subjected to spatial interpolation processing, so that the kinetic energy parameter values form a continuous distribution in the corresponding spatial region, avoiding discontinuous kinetic energy field distribution caused by discrete values; Add a time sequence identifier to the kinetic energy field distribution data processed at each time sequence to form a kinetic energy field distribution snapshot for a single time sequence, which includes the distribution of various kinetic energy parameters in the cavity space at that time sequence; Serially connect the kinetic energy field distribution snapshots of all time sequences in the order of injection time sequence to form a kinetic energy field time sequence distribution sub-model that can dynamically display the changes in the kinetic energy field at different injection stages; Perform spatial verification on the generated kinetic energy field time sequence distribution sub-model to check whether the spatial mapping positions of various kinetic energy parameter values match the geometric structure of the mold cavity three-dimensional structure model and the corresponding mechanism movement range, and correct the mapping data with spatial coordinate deviations, so that the kinetic energy field time sequence distribution sub-model can accurately reflect the distribution of kinetic energy in the cavity and the surrounding area.

10. The bumper injection molding defect intelligent detection method of claim 6, wherein, After the step of applying the confirmed abnormal area corresponding relationship, spatial overlap data, time sequence synchronicity data, parameter correlation data, and bidirectional mapping rules to generate the field mapping result, the method further includes a step of verifying and optimizing the field mapping result, including: Obtain field mapping result data under historical similar bumper injection scenarios, which includes energy field abnormal distribution areas, defect morphology field abnormal areas, abnormal area corresponding relationships, and parameter correlation rules in historical cases; Cross-compare the currently generated field mapping result with the field mapping result data under historical similar bumper injection scenarios to analyze the consistency of the currently generated field mapping result with the field mapping result data under historical similar bumper injection scenarios in terms of abnormal area corresponding relationship and parameter correlation rule, and mark the parts that differ; For the marked difference parts, analyze the causes of the difference in combination with the actual parameter data, mold state data, and defect morphology data of the current injection process, and determine whether the difference is caused by scenario difference or mapping rule deviation; If the difference is caused by mapping rule deviation, re-examine the corresponding rules in the bidirectional mapping rule library and modify the rule content in combination with the actual correlation of the current scenario, so that the modified rule can accurately reflect the corresponding logic of energy field abnormality and defect morphology abnormality; According to the modified bidirectional mapping rule, re-associate and verify the energy field abnormal distribution area and the defect morphology field abnormal area, and adjust the abnormal area corresponding relationship so that it meets the requirements of the modified rule; Calculate the matching accuracy of the abnormal area corresponding relationship in the optimized field mapping result, which is determined by the ratio of the number of abnormal area pairs that meet the mapping rule to the total number of abnormal area pairs; If the matching accuracy does not meet the preset requirements, repeat the cross-comparison, difference analysis, rule modification, and association verification steps until the matching accuracy meets the preset requirements; Key data in the whole verification optimization process is recorded, and the key data is integrated with the optimized field mapping result to form a final field mapping result, so that the accuracy and reliability of the final field mapping result meet preset standards, and the key data includes historical comparison data, difference analysis results, rule correction content and verification results.

Images