In-mold step action open mold control method based on real-time feedback
By reconstructing the state of objects within the mold and adjusting the control area through real-time feedback, the problem of object mismatch in in-mold opening and demolding control was solved, ensuring the smooth release of objects from the mold.
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- SHENZHEN BSC TECHNOLOGY CO LTD
- Filing Date
- 2026-03-30
- Publication Date
- 2026-07-21
AI Technical Summary
In existing in-mold opening and demolding technologies, subsequent actions are still based on the assumption of the original object, without considering the boundary constraint relationships and force state changes caused by the previous step actions, which leads to the problem of control object mismatch.
A real-time feedback-based in-mold step-by-step action mold opening and demolding control method is adopted. By acquiring step-by-step action feedback information in real time, the state of the remaining unreleased part is reconstructed, the control area and action sequence are adjusted, and subsequent actions are ensured to match the current object state.
It achieves matching of subsequent actions with the current object state, solves the control object mismatch problem, and ensures the smooth detachment of in-model objects.
Smart Images
Figure CN122425822A_ABST
Abstract
Description
Technical Field
[0001] This invention relates to the field of in-mold demolding control technology, and in particular to an in-mold step-by-step demolding control method based on real-time feedback. Background Technology
[0002] In existing in-mold release technologies, for flexible films, gel injection-packaged parts, or in-mold objects with localized attachment characteristics, a step-by-step approach is typically used for release control. Specifically, the object to be released from the mold is first divided into initial sections, and then ejection, traction, peeling, or other release actions are applied to each section sequentially according to a preset order. In some schemes, the action parameters are adjusted based on load, displacement, or contact state information to achieve gradual release of the object from the mold.
[0003] Existing technologies typically assume that after the previous step is completed, subsequent steps will still deal with the remaining part of the original demolded object. They do not consider the changes in the boundary constraints and stress state of the remaining unreleased part caused by the previous step. This results in subsequent actions still being based on the assumption of the original object, which can easily lead to control object mismatch problems. Summary of the Invention
[0004] To overcome the above shortcomings, this invention provides an in-mold step-by-step action mold opening and demolding control method based on real-time feedback, which aims to improve the problem that subsequent actions are still based on the original object assumptions and are prone to control object mismatch.
[0005] This invention provides the following technical solution: a method for controlling in-mold step-by-step opening and demolding based on real-time feedback, comprising the following steps: S1. Divide the object to be removed from the mold into sections, establish a step-by-step mold opening and removal control model, and generate step-by-step action control instructions; S2. Control the actuator to perform the current step-by-step demolding action according to the step-by-step action control command, and obtain the corresponding real-time feedback information; S3. Based on real-time feedback information, reconstruct the state of the remaining unreleased part after the current step-by-step demolding action, determine the constraint state of the current remaining object, and determine the step-by-step action control area based on the constraint state of the current remaining object. S4. Based on real-time feedback information and the constraint status of the remaining objects, identify the release obstacle status in the current step, compare the release obstacle status in the current step with the release obstacle status in the previous step, determine the transfer result of the release obstacle status, and adjust the step-by-step action control area or action sequence according to the transfer result. S5. Monitor the status of the released area. When the reattachment of the released area is detected, control the actuator to perform a release action and correct the step-by-step actions, and determine the status of the released area. S6. Based on the current constraint state of the remaining objects, the migration result of the released obstacle state, and the state maintenance status of the released area, update the step-by-step action control instructions, and repeat steps S2 to S6 until in-mold demolding is completed.
[0006] Preferably, in S1, the partitioning of the intramural object to be detached includes: Obtain the initial contact state information and initial boundary constraint state information of the object to be detached from the model within the model; Based on the initial contact state information and the initial boundary constraint state information, identify the controlled release difference region in the object to be detached within the module; Using the controlled release difference region as the boundary, the object to be detached within the mold is divided into multiple demolding zones; Each demolding zone is established as a step-by-step action control unit, and the relationship between each step-by-step action control unit is established.
[0007] Preferably, in S1, establishing the step-by-step mold opening and demolding control model includes: Obtain the initial attachment state and initial boundary constraint state of each demolding zone; Based on the initial attachment state, initial boundary constraint state, and partition association relationship between each demolding partition, establish the partition state representation corresponding to each demolding partition. Based on the partition state representation of each demolding partition, establish the step-by-step action response relationship between each demolding partition; Based on the partition state representation of each demolding partition and the step-by-step action response relationship between each demolding partition, a step-by-step demolding control model is constructed.
[0008] Preferably, in S1, the generation of step-by-step motion control instructions includes: Read the partition state representation of each demolding partition and the step-by-step action response relationship between each demolding partition in the step-by-step demolding control model; Based on the partition status representation of each demolding partition, determine the target demolding partition corresponding to the current step action; Based on the step-by-step action response relationship between the target demolding zone and its associated demolding zones, determine the control area and action sequence of the current step-by-step action; Based on the control area and action sequence of the current step-by-step action, generate the corresponding step-by-step action control instructions.
[0009] Preferably, in S2, obtaining the corresponding real-time feedback information includes: During the execution of the current step-by-step demolding action by the actuator, the partition response information of the target demolding partition and its associated demolding partitions is collected; Based on the partition response information, determine at least one of the partition attachment change state, partition displacement change state, and partition force change state under the current step-by-step demolding action. Based on the determined status information, real-time feedback information is generated corresponding to the current step-by-step demolding action.
[0010] Preferably, in S3, the state reconstruction of the remaining unreleased portion after the current step-by-step demolding action includes: Based on the real-time feedback information corresponding to the current step-by-step demolding action, identify the release status of each demolding zone after the current step-by-step demolding action; Based on the release status of each demolding zone, determine the remaining demolding zone corresponding to the remaining unreleased portion after the current step-by-step demolding action; Based on the partition association between the remaining demolding partitions and the real-time feedback information corresponding to the current step-by-step demolding action, the boundary constraint relationship of the remaining unreleased part is redefined; Based on the redefined boundary constraints, establish the constraint state of the remaining objects.
[0011] Preferably, in S4, identifying the release obstacle state in the current step includes: Based on the real-time feedback information corresponding to the current step-by-step demolding action, determine the release response status of each demolding zone under the control of the current step; Based on the release response status of each demolding zone, identify the demolding zones that have not formed an effective release response under the control of the current step; The demolding partition that does not generate an effective release response is identified as the region corresponding to the release obstacle state in the current step.
[0012] Preferably, in S4, adjusting the step-by-step action control area or action sequence based on the migration result includes: Based on the migration results of the released obstacle state, determine the region corresponding to the migrated released obstacle state; Determine whether the current step-by-step action control area is consistent with the area corresponding to the released obstacle state after migration; When the current step-by-step action control area is inconsistent with the area corresponding to the migrated obstacle release state, the area corresponding to the migrated obstacle release state will be determined as the new step-by-step action control area. When the area corresponding to the released obstacle state after migration is constrained by other demolding partitions, adjust the order of the step-by-step actions so that the step-by-step actions corresponding to other demolding partitions are executed before the area corresponding to the released obstacle state after migration.
[0013] Preferably, in S5, the control actuator performing the release action and correcting the step-by-step actions includes: When a reattachment is detected in a released area, the demolding partition corresponding to the released area where the reattachment occurred will be redefined as the current priority control partition. The control actuator applies a disengagement control action to the current priority control zone, causing it to re-detach from its current attachment position; Identify the subsequent actions that are related to the current priority control zone; Adjust the step-by-step actions, and continue to execute subsequent step-by-step actions after the current priority control partition is restored to the detached state.
[0014] Preferably, in S6, the update step-by-step action control instruction includes: Based on the current constraint state of the remaining objects, redetermine the control area for the next step action; Based on the migration results of the released obstacle state and the state maintenance of the released area, the action sequence of the next sub-step is re-determined; Based on the newly determined next step action control area and action sequence, generate updated step action control instructions.
[0015] The present invention has the following beneficial effects: 1. In this invention, by reconstructing the state of the remaining unreleased part based on real-time feedback information after each step of the demolding action, and redetermining the constraint state and step action control area of the current remaining object, the subsequent step actions are re-matched to the current real remaining object, thus solving the problem of control object mismatch caused by the previous action changing the state of the controlled object while the subsequent action still uses the original demolding object assumption in the prior art.
[0016] 2. In this invention, by identifying the release obstacle state under the current step, comparing the release obstacle state under the current step with that under the previous step and determining its migration result, and then adjusting the step-by-step action control area or action sequence according to the migration result, the subsequent step-by-step actions achieve targeted control of the obstructed area after migration, which solves the problem in the prior art of treating the difficult-to-remove area as a fixed position, causing subsequent control to continue to act on the failed area.
[0017] 3. In this invention, by monitoring the state of the released area and detecting that the released area has been reattached, the corresponding demolding zone is redefined as the priority control zone, the detachment control action is executed, and the relevant step actions are corrected, thereby achieving continuous maintenance of the detachment state of the released area and solving the problem of lack of timely correction control when the released area is reattached in subsequent step processes in the prior art. Attached Figure Description
[0018] Figure 1 This is a flowchart of an in-mold step-by-step action mold opening and demolding control method based on real-time feedback proposed in this invention. Detailed Implementation
[0019] The technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.
[0020] Example 1: In a first embodiment of the present invention, the present invention provides an in-mold step-by-step action mold opening and demolding control method based on real-time feedback, such as... Figure 1 As shown, it includes the following steps: S1. Divide the object to be removed from the mold into sections, establish a step-by-step mold opening and removal control model, and generate step-by-step action control instructions; Furthermore, in S1, partitioning the intramural objects to be detached includes: Obtain the initial contact state information and initial boundary constraint state information of the object to be detached from the model within the model; Based on the initial contact state information and the initial boundary constraint state information, identify the controlled release difference region in the object to be detached within the model; Using the controlled release difference region as the boundary, the object to be detached within the mold is divided into multiple demolding zones; Each demolding zone is established as a step-by-step action control unit, and the relationship between each step-by-step action control unit is established.
[0021] Furthermore, in S1, establishing the step-by-step mold release control model includes: Obtain the initial attachment state and initial boundary constraint state of each demolding zone; Based on the initial attachment state, initial boundary constraint state, and partition association relationship between each demolding partition, establish the partition state representation corresponding to each demolding partition. Based on the partition state representation of each demolding partition, establish the step-by-step action response relationship between each demolding partition; Based on the partition state representation of each demolding partition and the step-by-step action response relationship between each demolding partition, a step-by-step demolding control model is constructed.
[0022] Furthermore, in S1, generating step-by-step motion control instructions includes: Read the partition state representation of each demolding partition and the step-by-step action response relationship between each demolding partition in the step-by-step demolding control model; Based on the partition status representation of each demolding partition, determine the target demolding partition corresponding to the current step action; Based on the step-by-step action response relationship between the target demolding zone and its associated demolding zones, determine the control area and action sequence of the current step-by-step action; Based on the control area and action sequence of the current step-by-step action, generate the corresponding step-by-step action control instructions.
[0023] Specifically, S1 first divides the object to be detached from the mold into partitions, then establishes a step-by-step demolding control model based on the partitions, and generates step-by-step action control commands from this control model. The purpose of this process is not to geometrically divide the object, but to transform the object to be detached from the mold into a set of control units that can perform step-by-step control, so that subsequent step-by-step actions can be configured around the differences in attachment state, boundary constraint differences, and partition linkage relationships.
[0024] In practice, the initial contact state information and initial boundary constraint state information of the object to be detached within the mold are first obtained. The initial contact state information reflects the contact distribution between the object and the in-mold attachment surface, while the initial boundary constraint state information reflects the constrained distribution of the object at edges, support parts, and adjacent regions. The object to be detached within the mold is discretized into multiple adjacent local elements. Contact feature values and constraint feature values are extracted for each local element, and a local state vector is constructed. ; in, Indicates the first The local state vector of each local unit. Indicates the first Initial contact characteristic values of each local unit, Indicates the first The initial boundary constraint eigenvalues of each local element. For any two adjacent local elements... and Calculate the state difference: ; in, Representing adjacent local units and The difference in state between them. When When the difference is less than a preset threshold, the two local units are grouped into the same controlled release difference region; when... When the difference is not less than a preset threshold, the two local units are assigned to different controlled release difference regions. Thus, multiple demolding partitions are formed with the controlled release difference regions as boundaries, and each demolding partition is established as a step-by-step action control unit.
[0025] After partitioning is established, the relationships between each step-by-step action control unit are further established. These relationships reflect whether a step-by-step action received by one control unit will cause changes in the contact, constraint, or release states of another control unit. Subsequently, the initial attachment state and initial boundary constraint state of each demolding partition are acquired, and combined with the partition relationships, partition state representations and step-by-step action response relationships are established for each demolding partition. The partition state representation expresses the current attachment and constraint states of the corresponding demolding partition, while the step-by-step action response relationship expresses the state transmission relationship of a demolding partition to associated demolding partitions after a controlled action. Based on these two types of information, a step-by-step demolding control model is constructed, ensuring that the control model simultaneously includes the partition's own state and the action linkage relationships between partitions.
[0026] When generating step-by-step motion control commands, the partition status representation and step-by-step motion response relationship of each demolding partition are first read, and then the target demolding partition corresponding to the current step-by-step motion is determined. After the target demolding partition is determined, the control area of the current step-by-step motion is determined according to the step-by-step motion response relationship between the target demolding partition and its associated demolding partitions. The control area can be represented as: ; in, Indicates the target demolding partition. The current step-by-step action control area is formed around the center. Indicates the current target demolding partition. Indicates any demolding partition. Indicates the target demolding partition Demolding partition Does a step-by-step action-response relationship exist? If a response relationship exists, then... ,otherwise After determining the control area, the action sequence is determined based on the step-by-step action response relationship. When the release of the target demolding partition depends on the preceding state change of the associated demolding partition, the actions corresponding to the associated demolding partition are arranged first, followed by the actions corresponding to the target demolding partition. Finally, step-by-step action control commands are generated based on the control area and action sequence, enabling the actuator to perform step-by-step demolding actions according to the current control unit configuration. In this way, S1 completes the entire process from area division and control model establishment to control command generation.
[0027] S2. Control the actuator to perform the current step-by-step demolding action according to the step-by-step action control command, and obtain the corresponding real-time feedback information; Furthermore, in S2, obtaining the corresponding real-time feedback information includes: During the execution of the current step-by-step demolding action by the actuator, the partition response information of the target demolding partition and its associated demolding partitions is collected; Based on the partition response information, determine at least one of the partition attachment change state, partition displacement change state, and partition force change state under the current step-by-step demolding action. Based on the determined status information, real-time feedback information is generated corresponding to the current step-by-step demolding action.
[0028] Specifically, S2 is the process of executing the current step-by-step demolding action and simultaneously generating real-time feedback information after the step-by-step action control command has been generated. This process does not simply record the execution result, but continuously collects the response status related to the current control area during the action implementation, and converts these response statuses into feedback information that can be directly invoked for subsequent state reconstruction and control correction. In actual operation, the actuator performs the current step-by-step demolding action on the target demolding partition according to the current step-by-step action control command; when there is a step-by-step action response relationship between the target demolding partition and the associated demolding partition, the response changes of the associated demolding partition under the action of the current action are also collected simultaneously.
[0029] The collection of partition response information revolves around the target demolding partition and its associated demolding partitions. This partition response information includes at least response quantities that reflect interface changes, local deformation changes, and local force changes after the current action. For ease of subsequent processing, any demolding partition... The partition response information under the current step-by-step action is denoted as: ; in, Indicates demolding partition Partition response information, Indicates demolding partition The amount of attachment response, Indicates demolding partition The displacement response, Indicates demolding partition The force response is measured. The attachment response is used to characterize the change in contact between the partition and the in-mold attachment surface before and after the current action. The displacement response is used to characterize the local displacement change of the partition under the action. The force response is used to characterize the local mechanical change of the partition under the action. After the actuator starts moving, the above partition response information is continuously collected from the target demolding partition and the associated demolding partition to form a response sequence synchronized with the current step-by-step demolding action.
[0030] After collecting the partition response information, at least one of the following is determined based on the partition response information: the partition attachment change state, the partition displacement change state, and the partition stress change state under the current step-by-step demolding action. Specifically, any demolding partition can be... The change in attachment before and after the current action is defined as: ; in, Indicates demolding partition The amount of attachment change. This indicates the demolding zone before the current step-by-step demolding action is executed. The amount of attachment response, This indicates the demolding partition after the current step-by-step demolding action is executed. The attachment response. The displacement change and force change are defined as follows: ; ; in, Indicates demolding partition The amount of displacement change. Indicates the demolding partition before the action is executed. The displacement response, Indicates the demolding partition after the action is executed. The displacement response; Indicates demolding partition The change in force, Indicates the demolding partition before the action is executed. The force response quantity, Indicates the demolding partition after the action is executed. The stress response quantities. Based on the adhesion change, displacement change, and stress change quantities, the adhesion change state, displacement change state, and stress change state of the corresponding demolding zone under the current action can be determined respectively.
[0031] To unify the state changes of multiple partitions into real-time feedback information that can be invoked by the current action, the state changes of the target demolding partition and its associated demolding partitions are further combined. The real-time feedback information corresponding to the current step-by-step demolding action can be denoted as: ; in, This indicates the real-time feedback information corresponding to the current step-by-step demolding action. This indicates the control area for the current step-by-step demolding action. Indicates demolding partition The set of state changes, and ;in, Reflecting demolding zones The changes in attachment, displacement, and force under the current action. This real-time feedback information is not a single sensor value, but a collection of state changes covering multiple demolding zones within the current control area. Subsequent steps can directly utilize this information when reconstructing the state of the remaining unreleased portions. The set of state changes of each demolding partition is used to identify which partitions have been released, which partitions remain attached, which partitions have obvious linkage responses, and which partitions have not formed an effective release response under the current action.
[0032] In application, once the current step-by-step demolding action is executed, feedback acquisition is not completed all at once after the action ends, but is carried out continuously as the action is implemented. When the adhesion change of the target demolding zone increases significantly and the associated demolding zone shows a synchronous change in displacement or force, the real-time feedback information indicates that the current action has already responded to the associated area. When the force change of the target demolding zone continues to change while the adhesion and displacement changes are not significant, the real-time feedback information indicates that the zone has not yet formed an effective release response under the current action. In this way, S2 not only completes the execution of the current step-by-step demolding action by the actuator, but also simultaneously forms the real-time feedback basis required for subsequent control, enabling the entire step-by-step demolding process to continue into the state reconstruction, obstacle identification, and step-by-step action correction stages.
[0033] S3. Based on real-time feedback information, reconstruct the state of the remaining unreleased part after the current step-by-step demolding action, determine the constraint state of the current remaining object, and determine the step-by-step action control area based on the constraint state of the current remaining object. Furthermore, in S3, the state reconstruction of the remaining unreleased portion after the current step-by-step demolding action includes: Based on the real-time feedback information corresponding to the current step-by-step demolding action, identify the release status of each demolding zone after the current step-by-step demolding action; Based on the release status of each demolding zone, determine the remaining demolding zone corresponding to the remaining unreleased portion after the current step-by-step demolding action; Based on the partition association between the remaining demolding partitions and the real-time feedback information corresponding to the current step-by-step demolding action, the boundary constraint relationship of the remaining unreleased part is redefined; Based on the redefined boundary constraints, establish the constraint state of the remaining objects.
[0034] Specifically, S3 is the process of re-identifying and remodeling the remaining unreleased portion after the current step-by-step demolding action is completed. The object processed in this step is no longer the entire object to be detached within the initial mold, but rather the portion that remains detached after the current action. Because the previous step-by-step action has changed the local attachment state, local displacement state, and local stress state, the linkage relationships between the constrained boundaries and partitions corresponding to the remaining portion have also changed. Therefore, it is necessary to redetermine the current controlled object based on the current real-time feedback information, rather than continuing to use the initial object state.
[0035] During actual operation, the system first reads the real-time feedback information corresponding to the current step-by-step demolding action and identifies the release status of each demolding section after the current action. The release status is used to distinguish whether the corresponding demolding section has been released, partially released, or is still in a non-released state after the current action. To facilitate unified judgment, any demolding section can be... Construct a release state discriminant: ; in, Indicates demolding partition Release state discrimination quantity, Indicates demolding partition The amount of attachment change before and after the current step-by-step demolding action. Indicates demolding partition The amount of displacement change before and after the current step-by-step demolding action. Indicates demolding partition The change in force before and after the current step-by-step demolding action. , , These represent the discrimination coefficients for the attachment change, displacement change, and force change terms, respectively. Increased attachment and displacement changes indicate that the corresponding zone is closer to release; persistent force changes coupled with insufficient attachment and displacement changes suggest that the corresponding zone may still remain in a restricted attachment state. Based on the relationship between the release state discrimination coefficients and preset discrimination conditions, the release state of each demolding zone is identified.
[0036] After obtaining the release status of each demolding partition, the remaining unreleased partitions corresponding to the remaining unreleased portion after the current step-by-step demolding action are further determined. Released partitions are no longer included in the current round of remaining objects; unreleased partitions and partially released partitions constitute the current remaining unreleased portion. In this way, the current controlled object is transformed from the initial overall object into the current remaining object composed of the remaining demolding partitions.
[0037] After the remaining demolding zones are determined, the boundary constraints of the remaining unreleased portions are redefined based on the zone relationships between them and the current real-time feedback information. These boundary constraints reflect whether there are still force transfers, contact connections, or displacement restrictions between the remaining demolding zones. This can be represented as a constraint matrix: ; in, This represents the boundary constraint matrix of the remaining objects. Indicates the remaining demolding partitions With remaining demolding zones The constraint relationship is marked after the current action; when real-time feedback information indicates demolding partitioning. Changes in state will still cause demolding partitioning. When the force, displacement, or adhesion response changes, take Otherwise take In this way, constraints that have been released in the previous action are removed, while the constraints that still exist are retained, thus completing the reconstruction of the remaining unreleased boundary constraints.
[0038] After the boundary constraints are redefined, the constraint state of the remaining objects is further determined. The constraint state of the remaining objects characterizes the constraint transmission relationship between the main restricted areas of the remaining unreleased portion and the remaining demolding zones. Based on this, the control system determines which remaining demolding zones the main restricted positions of the current round are located in, and identifies the corresponding areas of these zones as the control areas for the next round of step-by-step actions. Thus, S3 completes the transformation from "real-time feedback after the current action" to "the constraint state of the current remaining objects" and "the control area for the next round," enabling subsequent step-by-step actions to control the currently existing unreleased objects.
[0039] S4. Based on real-time feedback information and the constraint status of the remaining objects, identify the release obstacle status in the current step, compare the release obstacle status in the current step with the release obstacle status in the previous step, determine the transfer result of the release obstacle status, and adjust the step-by-step action control area or action sequence according to the transfer result. Furthermore, in S4, identifying the release barrier state in the current step includes: Based on the real-time feedback information corresponding to the current step-by-step demolding action, determine the release response status of each demolding zone under the control of the current step; Based on the release response status of each demolding zone, identify the demolding zones that have not formed an effective release response under the control of the current step; The demolding partition that does not generate an effective release response is identified as the region corresponding to the release obstacle state in the current step.
[0040] Furthermore, in S4, adjusting the step-by-step action control area or action sequence based on the migration results includes: Based on the migration results of the released obstacle state, determine the region corresponding to the migrated released obstacle state; Determine whether the current step-by-step action control area is consistent with the area corresponding to the released obstacle state after migration; When the current step-by-step action control area is inconsistent with the area corresponding to the migrated obstacle release state, the area corresponding to the migrated obstacle release state will be determined as the new step-by-step action control area. When the area corresponding to the released obstacle state after migration is constrained by other demolding partitions, adjust the order of the step-by-step actions so that the step-by-step actions corresponding to other demolding partitions are executed before the area corresponding to the released obstacle state after migration.
[0041] Specifically, S4 is the process of identifying the release obstacle state under the current step after the current step-by-step demolding action is completed, based on real-time feedback information and the constraint state of the remaining object, and adjusting the control area and action sequence of subsequent steps accordingly. The object processed in this step is not a generally difficult-to-demold area, but rather a demolding zone that has not formed an effective release response under the control of the current step. This avoids continuing to use control areas that have already failed and shifts subsequent actions to the current actual obstructed location.
[0042] During actual operation, the system first reads the real-time feedback information corresponding to the current step-by-step demolding action, obtaining the attachment change state, displacement change state, and force change state of each demolding section in the current step. For any demolding section... Record the release response determination value under its current step as .in, Indicates demolding partition An effective release response has been achieved. Indicates demolding partition No valid release response was generated. The release response is determined using a rule-based method: when the demolding partition... Adhesion variation The attachment release criteria are met, or the displacement change is... The release displacement judgment condition is met and the change in force is When it indicates that the restricted state of the partition has been lifted, take In other cases, take .in, Indicates the demolding partition before and after the current step. The amount of attachment change. Indicates the demolding partitions before and after the current step. The amount of displacement change. Indicates the demolding partitions before and after the current step. The change in force. From all satisfying The demolding partitions constitute the set of regions corresponding to the release obstacle state in the current step: ; in, Indicates the current step The set of regions corresponding to the release barrier state.
[0043] After identifying the release obstacle state in the current step, it is compared with the release obstacle state in the previous step to determine the transition result of the release obstacle state. The set of regions corresponding to the release obstacle state in the previous step is denoted as... The newly added obstacle area in the current step is defined as follows: ; in, This represents the set of regions corresponding to the newly added barrier release states in the current step compared to the previous step; the symbol " " represents the set difference operation. The barrier region retained in the current step is defined as: ; in, This represents the set of regions corresponding to the release obstacle states that coexist in the current step and the previous step; the symbol " " represents the intersection operation of sets. When When it is an empty set, it indicates that the state of releasing the obstacle has not migrated to a new region; when If the value is not empty, it indicates that the barrier release state has been migrated to the new demodulation partition, and the migration result is... The corresponding demolding zones are provided.
[0044] After the migration results are determined, further adjust the step-by-step action control area or action sequence. Record the current step-by-step action control area as... .like and This indicates that the current step-by-step action control area does not cover the area corresponding to the released obstacle state after migration. In this case, the control area for the next step-by-step action will be updated as follows: ; in, The symbol "" indicates the step-by-step action control area for the next step. "" indicates a set union operation. In this way, subsequent actions directly target the currently blocked area.
[0045] If the current constraint state of the remaining objects indicates that the region corresponding to the released obstacle state after migration is still constrained by other demolding partitions, then the action order needs to be adjusted. The set of demolding partitions that constrain the obstacle region after migration is denoted as: ; in, This represents the set of demodulation partitions that impose constraints on the region corresponding to the released obstacle state after migration. This represents the demodulation partition in the current remaining object constraint matrix. Demolding partition Constraint relationship markers, This indicates the existence of constraints. When adjusting the order of actions, the set is executed first. The step-by-step actions corresponding to each demolding section are then executed in a set. The step-by-step actions corresponding to the area where the obstacle release state is located after migration are performed. Through the above process, S4 completes the identification of the obstacle release state, the determination of obstacle migration, and the correction of the control area and action sequence.
[0046] S5. Monitor the status of the released area. When the reattachment of the released area is detected, control the actuator to perform a release action and correct the step-by-step actions, and determine the status of the released area. Furthermore, in S5, controlling the actuator to perform release actions and modify step-by-step actions includes: When a reattachment is detected in a released area, the demolding partition corresponding to the released area where the reattachment occurred will be redefined as the current priority control partition. The control actuator applies a disengagement control action to the current priority control zone, causing it to re-detach from its current attachment position; Identify the subsequent actions that are related to the current priority control zone; Adjust the step-by-step actions, and continue to execute subsequent step-by-step actions after the current priority control partition is restored to the detached state.
[0047] Specifically, S5 is used to continuously monitor the released area and, upon detecting reattachment of the released area, immediately switch the control object, execute a disengagement action, and adjust subsequent step actions. The object processed in this step is not the currently released target area, but rather the demolding section that has already been released in the previous step or several steps. Because subsequent step actions can cause localized springback, boundary entanglement, or force redistribution, the released area may still reattach to the in-mold surface; therefore, the released area needs to be included in continuous monitoring.
[0048] During actual operation, first process each released demodulated partition... Continuously acquire information on changes in its attachment state, displacement state, and stress state, and construct state retention criteria. .in, Indicates demolding partition Remain in a detached state. Indicates demolding partition A re-attachment has occurred. The determination rule uses a state combination method: when the demolding partition... When the attachment state becomes continuous again, the displacement state returns to the vicinity of the attachment position, or the force state reverts to a restricted state, take... In other cases, take By satisfying all The demolding zones constitute a set of reattachment areas: ; in, Indicates the current step The set of released demodulated partitions that have been reattached and reattached.
[0049] when If this occurs, it indicates that a released area has been reattached in the current step. In this case, the subsequent steps will not be executed in the original preset order, but the collection will be... The corresponding demodulation partitions are redefined as the current priority control partition set. ,Right now ; in, This represents the set of demolding partitions that are prioritized for processing in the current step. The control system then processes the set... Each demolding zone is subjected to a release control action to re-detach it from its current attachment position. This release control action refers to the control action that causes the reattached area to re-establish a release displacement, release the current attachment contact, and weaken the locally restricted force. Its goal is to restore the area that has re-attached to its released state.
[0050] While executing the de-control action, it's also necessary to identify which subsequent steps might again affect the currently prioritized control partition. To do this, partition associations are read to determine their relationship with the set. The set of subsequent steps that are related: ; in, This represents the set of demodulated partitions corresponding to subsequent step actions associated with the current priority control partition. Indicates demolding partition Will the step-by-step actions affect the demolding partition? Association relationship markers, This indicates the existence of an influence relationship. (Regarding sets) The corresponding subsequent steps are adjusted, including delaying execution, pausing execution, or executing after the current priority control partition is restored to its detached state, so as to avoid the area being pulled open and then re-attached.
[0051] After completing the disengagement action, then reassemble. Each demolding zone in the middle is used for recovery determination. Let the recovery mark be... ,in, Indicates demolding partition It has been removed from its current attachment position. This indicates that the problem has not yet been resolved. (When the set...) All demolding zones in the middle meet the requirements When the current priority control zone has been restored to the uncontrolled state, the adjusted subsequent step-by-step actions are executed. Through the above process, S5 achieves continuous monitoring of the status of the released area and completes the switching of priority control zones, execution of uncontrolled actions, and correction of subsequent step-by-step actions when reattachment occurs.
[0052] S6. Based on the current constraint state of the remaining objects, the migration result of the released obstacle state, and the state maintenance status of the released area, update the step-by-step action control instructions, and repeat steps S2 to S6 until in-mold demolding is completed.
[0053] Furthermore, in S6, updating the step-by-step motion control instructions includes: Based on the current constraint state of the remaining objects, redetermine the control area for the next step action; Based on the migration results of the released obstacle state and the state maintenance of the released area, the action sequence of the next sub-step is re-determined; Based on the newly determined next step action control area and action sequence, generate updated step action control instructions.
[0054] Specifically, S6 is used to update the control instructions for the next step action after the current step-by-step demolding action is completed, based on the constraint state of the remaining object, the transition results of the released obstacle state, and the state maintenance status of the released area, and drives the control process to enter the next round of execution. This step is not a simple repetition of the previous round of action generation process, but rather, based on the fact that the current round of actions has changed the object boundary constraints, the distribution of released obstacles, and the stable state of the released area, it redetermines the control area and action sequence of the next round of step actions, so that subsequent control always corresponds to the current real state.
[0055] During runtime, the constraint state of the remaining objects is first read. This constraint state reflects the boundary propagation relationships, local confinement relationships, and partition linkage relationships that still exist within the remaining unreleased portions. Based on the constraint state of the remaining objects, the control region for the next step action is redefined. This next step action control region can be denoted as a set. ,in, Indicate steps The step-by-step action control area. If a remaining demolding partition belongs to the main restricted area in the current constraint state, or if there is still a clear constraint transit relationship between it and other remaining demolding partitions, then the demolding partition is included in the set. Therefore, the control area for the next sub-step action is no longer fixed and continues to use the control area of the previous step, but is reselected based on the current constraint state of the remaining objects.
[0056] After redefining the control area for each step of the action, the sequence of actions for the next step is determined based on the transition results of the obstacle release state and the state maintenance of the released areas. This sequence depends not only on the transitioned obstacle release area but also on whether the released areas can maintain stable detachment in subsequent actions. Therefore, the set of areas corresponding to the transitioned obstacle release state is denoted as... The set of released regions that have been reattached or are still in an unstable state in the current step is denoted as... Then construct the priority set for the next step: ; in, Indicate steps The set of demodulation partitions that are executed first in the middle. Indicate steps The set of new obstacle regions formed after the release and migration of obstacles. Indicate steps The set of released regions that remain unstable in the middle state, symbolized by " The colon ("") represents the union operation on sets. The demolding partition in the process is given priority to enter the next step action sequence, which is used to ensure that the obstructed area after migration and the released but unstable area are processed first in the next round of control.
[0057] After determining the priority set for execution, the order of actions needs to be adjusted based on the constraint states of the remaining objects. This will involve adjusting the set... The set of demolding partitions that have constraints in the demolding partition is denoted as: ; in, Indicates the steps The set of constraint demodulation partitions that needs to be processed first. This represents the demodulation partition in the current remaining object constraint matrix. Demolding partition Constraint relationship markers, Indicates demolding partition Demolding partition The constraints still apply. When the order of actions is determined, the set should be arranged first. The steps corresponding to the demolding section are then arranged and assembled. The steps corresponding to the demolding section are arranged in the final assembly. The remaining demolding partitions correspond to the step-by-step actions. Thus, the reordering of actions is not a simple change of sequence, but a new execution sequence formed based on constraints, obstacle migration results, and state preservation.
[0058] After determining the control area and action sequence, updated step-by-step action control instructions are generated. These updated instructions include at least the next step-by-step action control area information and the next step-by-step action sequence information, serving as the basis for the next round of actuator action invocation. The actuator executes the next step of the demolding action based on these updated instructions, subsequently re-entering the process of real-time feedback acquisition, state reconstruction, obstacle release identification, reattachment processing, and control instruction update. Thus, S6 forms a complete closed-loop update mechanism: the previous round of actions outputs real-time feedback and state results; the current round re-determines the control area and action sequence based on these state results, generates new step-by-step action control instructions, and enters the next round of execution, until all objects to be demolded within the mold have completed demolding.
[0059] Example 2: In the in-mold step-by-step demolding process of gel injection packaged parts, after the previous step is completed, the boundary constraints and stress state of the remaining unreleased portion have changed. However, subsequent control still uses the original preset action sequence and control area, resulting in a mismatch between subsequent actions and the current actual remaining object. To solve the above problem, this invention provides an in-mold step-by-step action demolding control method based on real-time feedback, the structure of which is as follows: Figure 1 As shown. The specific implementation process of this method is as follows: The gel injection package to be detached from the mold is divided into multiple demolding zones, and a corresponding step-by-step demolding control model is established. This control model records the initial attachment state, initial boundary constraint state, and the inter-zone relationships between each demolding zone, and generates initial step-by-step action control commands based on this. The initial step-by-step action control commands include the target demolding zone for the current step, the control area, and the action sequence.
[0060] Initially, the packaged component is divided into an edge zone, a central transition zone, and a central zone. Based on the initial attachment distribution and boundary constraint distribution, the control system first selects the edge zone as the target demolding zone corresponding to the first step of the action, and controls the actuator to perform the current step-by-step demolding action on that edge zone. During the action, the partition response information of the target demolding zone and its associated demolding zones is simultaneously collected, forming real-time feedback information corresponding to the current step-by-step demolding action. This real-time feedback information reflects at least the attachment change state, displacement change state, and force change state under the current action.
[0061] After the first step is completed, the control system does not directly continue to the next partition according to the initial preset sequence. Instead, it first identifies the release status of each demolding partition based on real-time feedback information. If the attachment state of a demolding partition has been clearly released and the displacement state has reached the condition for separation after the current action, it is determined that the demolding partition has been released. If a demolding partition has only undergone local changes but has not yet reached the separation condition, it is determined to be partially released. If a demolding partition remains continuously attached and is in a restricted stress state after the current action, it is determined to be not released. Based on the release status of each demolding partition, the remaining demolding partitions that still constitute the remaining unreleased part after the current action are selected.
[0062] After identifying the remaining demolding zones, the control system further determines the boundary constraints of the remaining unreleased portions based on the partition relationships between the remaining demolding zones and the current real-time feedback information. For example, after the first step action is completed, the central transition zone, which was originally connected to the edge zone, may have its boundary connections partially released because the edge zone has detached, while the central zone and the central transition zone may become the new primary constraint relationship. At this point, the control system re-establishes the constraint state of the current remaining object and treats the re-identified remaining unreleased portions as new control objects, rather than considering them as simple remaining parts of the original whole object.
[0063] The control system redetermines the step-by-step action control area based on the constraint state of the remaining object. If the reconstruction results indicate that the originally preset next control area no longer corresponds to the main restricted area in the current remaining object, the original preset control area is abandoned, and the demolding partition with the most concentrated constraint relationships and the greatest influence on the subsequent demolding process is selected as the new step-by-step action control area. In this way, the control object for subsequent actions switches from the "original preset area" to the "main restricted area in the current actual remaining object".
[0064] After the first step action is completed, although the next step in the original preset action sequence should still act on another adjacent edge partition, real-time feedback information indicates that this area is no longer the main unreleased area, and a new main constraint relationship has been formed between the middle transition partition and the center partition. Therefore, the control system redefines the middle transition partition and the linkage area between it and the center partition as the new step action control area, and adjusts the subsequent action sequence so that the actuator prioritizes the second step demolding action to be performed on this new control area.
[0065] During the second step of the action, new real-time feedback information is acquired again, and the processes of release status recognition, remaining object reconstruction, constraint status update, and control area redeter determination are repeated. As each step of the action is completed, the control system continuously updates the step-by-step action control instructions based on the actual remaining object status after the current action, and controls the actuator to implement the next step of demolding action, until all demolding zones are completely detached.
[0066] In this embodiment, by reconstructing the state of the remaining unreleased part after each step action, and redetermining the constraint state and step action control area of the current remaining object based on the reconstruction result, the subsequent step actions no longer use the failed original preset action sequence and original control area, but continuously match the current real remaining object, thereby completing the in-mold step opening and demolding control.
[0067] Finally, it should be noted that the above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art can still modify the technical solutions described in the foregoing embodiments or make equivalent substitutions for some of the technical features. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the protection scope of the present invention.
Claims
1. A method for controlling in-mold step-by-step opening and demolding actions based on real-time feedback, characterized in that, Includes the following steps: S1. Divide the object to be removed from the mold into sections, establish a step-by-step mold opening and removal control model, and generate step-by-step action control instructions; S2. Control the actuator to perform the current step-by-step demolding action according to the step-by-step action control command, and obtain the corresponding real-time feedback information; S3. Based on real-time feedback information, reconstruct the state of the remaining unreleased part after the current step-by-step demolding action, determine the constraint state of the current remaining object, and determine the step-by-step action control area based on the constraint state of the current remaining object. S4. Based on real-time feedback information and the constraint status of the remaining objects, identify the release obstacle status in the current step, compare the release obstacle status in the current step with the release obstacle status in the previous step, determine the transfer result of the release obstacle status, and adjust the step-by-step action control area or action sequence according to the transfer result. S5. Monitor the status of the released area. When the reattachment of the released area is detected, control the actuator to perform a release action and correct the step-by-step actions, and determine the status of the released area. S6. Based on the current constraint state of the remaining objects, the migration result of the released obstacle state, and the state maintenance status of the released area, update the step-by-step action control instructions, and repeat steps S2 to S6 until in-mold demolding is completed.
2. The in-mold step-by-step action mold opening and demolding control method based on real-time feedback according to claim 1, characterized in that, In S1, the partitioning of the intramural object to be detached includes: Obtain the initial contact state information and initial boundary constraint state information of the object to be detached from the model within the model; Based on the initial contact state information and the initial boundary constraint state information, identify the controlled release difference region in the object to be detached within the module; Using the controlled release difference region as the boundary, the object to be detached within the mold is divided into multiple demolding zones; Each demolding zone is established as a step-by-step action control unit, and the relationship between each step-by-step action control unit is established.
3. The in-mold step-by-step action mold opening and demolding control method based on real-time feedback according to claim 1, characterized in that, In S1, establishing the step-by-step mold opening and demolding control model includes: Obtain the initial attachment state and initial boundary constraint state of each demolding zone; Based on the initial attachment state, initial boundary constraint state, and partition association relationship between each demolding partition, establish the partition state representation corresponding to each demolding partition. Based on the partition state representation of each demolding partition, establish the step-by-step action response relationship between each demolding partition; Based on the partition state representation of each demolding partition and the step-by-step action response relationship between each demolding partition, a step-by-step demolding control model is constructed.
4. The in-mold step-by-step action mold opening and demolding control method based on real-time feedback according to claim 1, characterized in that, In S1, the generation of step-by-step action control instructions includes: Read the partition state representation of each demolding partition and the step-by-step action response relationship between each demolding partition in the step-by-step demolding control model; Based on the partition status representation of each demolding partition, determine the target demolding partition corresponding to the current step action; Based on the step-by-step action response relationship between the target demolding zone and its associated demolding zones, determine the control area and action sequence of the current step-by-step action; Based on the control area and action sequence of the current step-by-step action, generate the corresponding step-by-step action control instructions.
5. The in-mold step-by-step action mold opening and demolding control method based on real-time feedback according to claim 1, characterized in that, In S2, obtaining the corresponding real-time feedback information includes: During the execution of the current step-by-step demolding action by the actuator, the partition response information of the target demolding partition and its associated demolding partitions is collected; Based on the partition response information, determine at least one of the partition attachment change state, partition displacement change state, and partition force change state under the current step-by-step demolding action. Based on the determined status information, real-time feedback information is generated corresponding to the current step-by-step demolding action.
6. The in-mold step-by-step action mold opening and demolding control method based on real-time feedback according to claim 1, characterized in that, In S3, the state reconstruction of the remaining unreleased portion after the current step-by-step demolding action includes: Based on the real-time feedback information corresponding to the current step-by-step demolding action, identify the release status of each demolding zone after the current step-by-step demolding action; Based on the release status of each demolding zone, determine the remaining demolding zone corresponding to the remaining unreleased portion after the current step-by-step demolding action; Based on the partition association between the remaining demolding partitions and the real-time feedback information corresponding to the current step-by-step demolding action, the boundary constraint relationship of the remaining unreleased part is redefined; Based on the redefined boundary constraints, establish the constraint state of the remaining objects.
7. The in-mold step-by-step action mold opening and demolding control method based on real-time feedback according to claim 1, characterized in that, In S4, identifying the release obstacle state in the current step includes: Based on the real-time feedback information corresponding to the current step-by-step demolding action, determine the release response status of each demolding zone under the control of the current step; Based on the release response status of each demolding zone, identify the demolding zones that have not formed an effective release response under the control of the current step; The demolding partition that does not generate an effective release response is identified as the region corresponding to the release obstacle state in the current step.
8. The in-mold step-by-step action mold opening and demolding control method based on real-time feedback according to claim 1, characterized in that, In S4, adjusting the step-by-step action control area or action sequence based on the migration result includes: Based on the migration results of the released obstacle state, determine the region corresponding to the migrated released obstacle state; Determine whether the current step-by-step action control area is consistent with the area corresponding to the released obstacle state after migration; When the current step-by-step action control area is inconsistent with the area corresponding to the migrated obstacle release state, the area corresponding to the migrated obstacle release state will be determined as the new step-by-step action control area. When the area corresponding to the released obstacle state after migration is constrained by other demolding partitions, adjust the order of the step actions so that the step actions corresponding to other demolding partitions are executed before the area corresponding to the released obstacle state after migration.
9. The in-mold step-by-step action mold opening and demolding control method based on real-time feedback according to claim 1, characterized in that, In S5, the control actuator performs the release action and corrects the step-by-step actions, including: When a re-attachment is detected in a released area, the demolding partition corresponding to the released area where the re-attachment occurred will be redefined as the current priority control partition. The control actuator applies a disengagement control action to the current priority control zone, causing it to re-detach from its current attachment position; Identify the subsequent actions that are related to the current priority control zone; Adjust the step-by-step actions and continue executing subsequent step-by-step actions after the current priority control partition is restored to the detached state.
10. The in-mold step-by-step action mold opening and demolding control method based on real-time feedback according to claim 1, characterized in that, In S6, the update step-by-step action control instructions include: Based on the current constraint state of the remaining objects, redetermine the control area for the next step action; Based on the migration results of the released obstacle state and the state maintenance of the released area, the action sequence of the next sub-step is re-determined; Based on the newly determined next step action control area and action sequence, generate updated step action control instructions.