A method for rapid optimization of linkage processes in hydraulic injection molding equipment

Abstract

This invention relates to the field of injection molding process optimization technology, and discloses a rapid optimization method for the linkage process of hydraulic injection molding equipment, comprising: S1, acquiring and extracting state characteristic parameters of the preceding process stage; S2, constructing a multi-layered inherited state; S3, determining the degree of transmission influence of each candidate parameter, screening the affected parameter subset, and determining the stage switching boundary offset trend; S4, determining the linkage constraint relationship between parameters and the collaborative adjustment relationship between parameters and the stage switching boundary; S5, performing local linkage rapid optimization based on the relationship to obtain the optimized control result; S6, acquiring the execution result and determining the mismatch type; S7, performing path correction according to the mismatch type and using it for the current or next process cycle optimization. By constructing a multi-layered inherited state and determining the affected parameter subset and the stage switching boundary offset trend, rapid local linkage optimization of subsequent process parameters and switching boundaries is achieved, improving optimization efficiency and control targeting.

Description

Technical Field

[0001] This invention relates to the field of injection molding process optimization technology, specifically a method for rapid optimization of the linkage process of hydraulic injection molding equipment. Background Technology

[0002] Hydraulic injection molding equipment refers to injection molding equipment that uses hydraulic drive to move the injection mechanism, mold closing mechanism, and related actuators to complete the processes of plastic melting injection, pressure holding molding, and mold opening and closing. It typically controls process parameters such as injection pressure, injection speed, holding pressure, holding time, and stage switching positions to meet the molding requirements of different plastic products. Rapid optimization of the linkage process in hydraulic injection molding equipment refers to the coordinated adjustment and rapid optimization of multiple process objects, such as injection, holding pressure, and stage switching boundaries, given the interrelationships and constraints among multiple process parameters during actual molding. This improves process adjustment efficiency and molding stability. Currently, when adjusting the process of hydraulic injection molding equipment, operators typically adjust parameters such as injection pressure, injection speed, holding pressure, holding time, or stage switching boundaries one by one. This is often achieved by empirically correcting individual parameters based on the current molding results and then gradually observing changes in the molding state.

[0003] However, in current technology, because the results of preceding processes continuously affect the state of subsequent processes, and there is a linkage between subsequent process parameters and between subsequent process parameters and stage switching boundaries, the item-by-item trial and adjustment method is difficult to quickly identify the truly affected parameter range and the adjustment direction of stage switching boundaries, resulting in low process optimization efficiency. Summary of the Invention

[0004] To address the shortcomings of existing technologies, this invention provides a rapid optimization method for the linkage process of hydraulic injection molding equipment, which solves the problem of low process optimization efficiency caused by the difficulty in quickly identifying the range of affected parameters and the adjustment direction of stage switching boundaries during equipment process adjustment.

[0005] To achieve the above objectives, the present invention provides the following technical solution: a method for rapid optimization of the linkage process in hydraulic injection molding equipment, comprising: S1. Obtain the operation result data of the preceding process stage and extract features to obtain the state feature parameters that characterize the output state of the preceding process. S2. Based on the state feature parameters, group the different transmission influence dimensions of the subsequent process according to the results of the preceding process, and construct a multi-layer inheritance state according to the state determination rules corresponding to the state feature parameters of each group. S3. Through the multi-layer inheritance state, determine the degree of transmission influence of each candidate parameter in the subsequent process, and determine the subset of affected parameters based on the degree of transmission influence, while determining the stage switching boundary offset trend. S4. Based on the multi-layer inheritance state and the stage switching boundary offset trend, determine the linkage constraint relationship between each parameter in the affected parameter subset and the collaborative adjustment relationship between the affected parameter subset and the stage switching boundary; S5. Based on the linkage constraint relationship and the collaborative adjustment relationship, perform local linkage rapid optimization on the affected parameter subset and stage switching boundary to obtain the optimized control result of the subsequent process; S6. Obtain the execution results of subsequent processes, and determine the mismatch type based on the deviation between the execution results and the expected results and the dominant deviation item corresponding to the deviation; S7. Based on the mismatch type, perform sub-path correction on the multi-layer inheritance state, affected parameter subset, stage switching boundary offset trend, linkage constraint relationship or collaborative adjustment relationship, and use the correction results for subsequent optimization of the current process cycle or rapid optimization of the linkage process in the next process cycle.

[0006] Preferably, feature extraction of the running result data in S1 specifically includes: Fluctuation amplitude analysis and fluctuation frequency analysis were performed on the pressure data, velocity data and displacement data of the preceding process stage to obtain the corresponding fluctuation characteristic parameters; The rate of change and direction of change of the stage duration data of the preceding process stage and the change data before and after the switch are analyzed to obtain the corresponding trend characteristic parameters. Based on the fluctuation characteristic parameters and the trend characteristic parameters, state characteristic parameters are extracted to characterize the output state of the preceding process.

[0007] Preferably, in step S2, the grouping of different transmission influence dimensions of the subsequent process based on the results of the preceding process specifically includes: Based on the influence of each state characteristic parameter on the stability of subsequent processes, the corresponding state characteristic parameters are divided into stability parameter groups. Based on the influence of each state characteristic parameter on the switching conditions of subsequent process stages, the corresponding state characteristic parameters are divided into boundary offset parameter groups. Based on the influence of each state characteristic parameter on the probability of subsequent process anomalies, the corresponding state characteristic parameters are divided into risk transmission parameter groups.

[0008] Preferably, the construction of a multi-level inheritance state in S2 specifically includes: Each state characteristic parameter in the stability parameter group is compared with the corresponding stability judgment threshold, and the output stability of the preceding process stage is determined based on the comparison result to generate a stable output state. Each state feature parameter in the boundary offset parameter group is compared with the corresponding boundary offset judgment threshold, and the boundary offset direction and degree at the end of the preceding process stage are determined based on the comparison result to generate the stage boundary offset state. The state characteristic parameters in the risk transmission parameter group are weighted according to the preset parameter weights to obtain the abnormal risk evaluation value. The abnormal risk evaluation value is then compared with the corresponding risk judgment threshold, and an abnormal risk transmission state is generated based on the comparison result.

[0009] Preferably, determining the subset of affected parameters in subsequent processes in step S3 specifically includes: Based on the stable output state, stage boundary offset state, and abnormal risk transmission state, calculate the degree of transmission influence of each candidate parameter in the subsequent process. The degree of transmission influence corresponding to each candidate parameter is compared with a preset influence threshold, and subsequent process parameters with a degree of transmission influence greater than the preset influence threshold are selected to form a set of affected parameters; Based on the degree of influence correlation between each subsequent process parameter in the affected parameter set, subsequent process parameters with an influence correlation greater than a preset correlation threshold are selected as a subset of affected parameters.

[0010] Preferably, determining the stage switching boundary offset trend of subsequent processes in S3 specifically includes: Obtain the target switching conditions corresponding to the subsequent process stage, and extract the boundary state parameters related to the target switching conditions at the end of the preceding process stage. The boundary state parameters include the end pressure deviation, the end displacement deviation, and the end velocity deviation. The boundary state parameters are compared with the target switching conditions to obtain the current boundary state deviation value; Based on the magnitude and direction of the current boundary state deviation, a boundary offset evaluation value is calculated according to a preset evaluation function. The boundary offset evaluation value is then compared with a preset boundary trend determination threshold range. Based on the comparison result, the offset trend of the subsequent process stage switching boundary—whether it moves forward, backward, or remains unchanged—is determined.

[0011] Preferably, the step of determining the linkage constraint relationship, the collaborative adjustment relationship, and performing local linkage rapid optimization specifically includes: Based on the influence correlation between each subsequent process parameter in the affected parameter subset and the offset trend of the stage switching boundary, the target adjustment amount, adjustment order and adjustment range of each subsequent process parameter are calculated to form the linkage constraint relationship between each parameter in the affected parameter subset. Based on the offset trend of the stage switching boundary, the target adjustment amount of the subsequent process parameters corresponding to the stage switching boundary is associated and matched with the boundary correction amount of the stage switching boundary to form a collaborative adjustment relationship between the affected parameter subset and the stage switching boundary. According to the linkage constraint relationship and the collaborative adjustment relationship, the subsequent process parameters and the stage switching boundary within the affected parameter subset are synchronously adjusted, and local linkage rapid optimization is performed within the parameter range corresponding to the affected parameter subset and the boundary range corresponding to the stage switching boundary to obtain the optimized control result of the subsequent process.

[0012] Preferably, in step S6, determining the mismatch type based on the deviation between the execution result and the expected result specifically includes: The actual execution trajectory, actual stage switching result, and actual parameter response result of the subsequent process are compared with the corresponding expected execution trajectory, expected stage switching result, and expected parameter response result to obtain the trajectory deviation value, boundary deviation value, and parameter deviation value. The trajectory deviation value, boundary deviation value, and parameter deviation value are compared with the corresponding mismatch judgment thresholds, and the deviation item that exceeds the corresponding mismatch judgment threshold and has the largest deviation amplitude is determined as the dominant deviation item. When the dominant deviation term corresponds to the state determination result of the multi-level inheritance state, it is determined to be an inheritance state mismatch; When the dominant deviation term corresponds to the correction result of the stage switching boundary, it is determined to be a stage switching boundary mismatch; When the dominant deviation term corresponds to the screening result of the affected parameter subset, it is determined to be a parameter subset mismatch.

[0013] Preferably, the path correction in step S7 based on the mismatch type specifically includes: When the mismatch type is inherited state mismatch, the state evaluation results corresponding to each state feature parameter are recalculated, and the multi-layer inherited state is regenerated based on the recalculated state evaluation results. When the mismatch type is stage switching boundary mismatch, the current boundary state deviation value is recalculated, and the stage switching boundary offset trend is redetermined based on the recalculated current boundary state deviation value. When the mismatch type is parameter subset mismatch, the degree of transmission influence and the degree of influence correlation of each candidate parameter in the subsequent process are re-determined, and the affected parameter subset is re-selected based on the re-determined degree of transmission influence and degree of influence correlation.

[0014] Preferably, in step S7, the correction result is used for subsequent optimization of the current process cycle or for rapid optimization of the linked process in the next process cycle, specifically including: The corrected multi-level inheritance state, affected parameter subset, stage switching boundary offset trend, linkage constraint relationship and collaborative adjustment relationship are stored to form a template for subsequent optimization calls; When there is an unexecuted subsequent process stage in the current process cycle, the subsequent optimization call template is invoked to perform rapid optimization of the unexecuted subsequent process stage in the current process cycle. At the start of the next process cycle, the subsequent optimization call template is invoked as the initial optimization basis for constructing multi-layer inheritance states, determining the affected parameter subset, determining the stage switching boundary offset trend, and performing local linkage rapid optimization in the next process cycle.

[0015] This invention provides a method for rapid optimization of the linkage process in hydraulic injection molding equipment. It has the following beneficial effects: 1. This invention constructs a multi-layered inheritance state and determines the affected parameter subset and stage switching boundary offset trend accordingly, thereby achieving rapid local linkage optimization of subsequent process parameters and switching boundaries, avoiding repeated trial and error adjustments of all parameters, and improving process optimization efficiency and control targeting.

[0016] 2. This invention groups and judges the results of preceding processes according to different influence dimensions such as stability, boundary offset, and risk transmission, so that the transmission influence of preceding processes on subsequent processes can be characterized in a hierarchical manner, which is beneficial to improving the accuracy of subsequent parameter screening and boundary identification.

[0017] 3. This invention improves the stability and adaptability of the optimization process under continuous operation by performing path correction on the multi-layer inheritance state, affected parameter subset and stage switching boundary offset trend according to the mismatch type, and applying the correction results to the current process cycle or the next process cycle. Attached Figure Description

[0018] Figure 1 This is a flowchart of a rapid optimization method for the linkage process of a hydraulic injection molding equipment according to the present invention. Detailed Implementation

[0019] The technical solution of the present invention will now be clearly and completely described 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] Please see the appendix Figure 1 This invention provides a method for rapid optimization of the linkage process in hydraulic injection molding equipment, comprising: S1. Obtain the operation result data of the preceding process stage and extract features to obtain the state feature parameters that characterize the output state of the preceding process. Furthermore, feature extraction is performed on the results data in S1, specifically including: Fluctuation amplitude analysis and fluctuation frequency analysis were performed on the pressure data, velocity data and displacement data of the preceding process stage to obtain the corresponding fluctuation characteristic parameters; The rate of change and direction of change of the stage duration data of the preceding process stage and the change data before and after the switch are analyzed to obtain the corresponding trend characteristic parameters. Based on the fluctuation characteristic parameters and trend characteristic parameters, state characteristic parameters are extracted to characterize the output state of the preceding process.

[0021] Specifically, the preceding process stage can be the plasticizing stage, the pre-injection stage, or the pre-pressure holding stage before the subsequent process. By collecting pressure data, speed data, displacement data, stage duration data, and change data before and after switching within the preceding process stage, raw data that characterizes the operation of the preceding process can be obtained, providing a basis for the subsequent extraction of state characteristic parameters. After acquiring the operational results data, fluctuation amplitude analysis and fluctuation frequency analysis are performed on the pressure data, velocity data, and displacement data of the preceding process stage to obtain the corresponding fluctuation characteristic parameters. In specific implementation, the above data can be segmented according to the preset sampling time window, and the fluctuation amplitude and fluctuation frequency within each sampling time window can be extracted to obtain the pressure fluctuation characteristic parameters, velocity fluctuation characteristic parameters, and displacement fluctuation characteristic parameters, so as to characterize the dynamic fluctuation of the preceding process stage during operation. After obtaining the fluctuation characteristic parameters, further analysis of the change rate and change direction of the stage duration data of the preceding process stage and the change data before and after the switch is performed to obtain the corresponding trend characteristic parameters. In specific implementation, the actual duration of the preceding process stage can be compared with the preset standard duration to determine the change of duration. At the same time, the data change trend within the preset time range before and after the switch is analyzed to form the stage duration trend characteristic parameters and the trend characteristic parameters before and after the switch, so as to characterize the overall evolution trend of the preceding process stage and the change state at the end of the stage. After extracting the fluctuation and trend characteristic parameters, the fluctuation and trend characteristic parameters are classified, filtered, and standardized to form a set of state characteristic parameters that characterize the output state of the preceding process. The original operating result data of the preceding process stage is transformed into a representation quantity that can be directly called in the subsequent multi-level inherited state construction, thereby providing a data foundation for the subsequent determination of the affected parameter subset, the identification of the stage switching boundary offset trend, and the rapid optimization of the linkage process.

[0022] S2. Based on state feature parameters, group the different transmission influence dimensions of subsequent processes according to the results of previous processes, and construct multi-layer inheritance states according to the state determination rules corresponding to the state feature parameters of each group. Furthermore, in step S2, the effects of preceding process results on subsequent processes are grouped according to different dimensions of transmission, specifically including: Based on the influence of each state characteristic parameter on the stability of subsequent processes, the corresponding state characteristic parameters are divided into stability parameter groups. Based on the influence of each state characteristic parameter on the switching conditions of subsequent process stages, the corresponding state characteristic parameters are divided into boundary offset parameter groups. Based on the influence of each state characteristic parameter on the probability of subsequent process anomalies, the corresponding state characteristic parameters are divided into risk transmission parameter groups.

[0023] Specifically, after obtaining the state feature parameters that characterize the output state of the preceding process, the state feature parameters are grouped according to the different dimensions of the transmission influence of the preceding process results on the subsequent process. Since different state feature parameters have different effects on the subsequent process, grouping them first can make the subsequent state determination more targeted and provide a classification basis for the construction of multi-level inherited states. Based on the influence of each state characteristic parameter on the stability of subsequent processes, the corresponding state characteristic parameters are divided into stability parameter groups. State characteristic parameters that mainly reflect the output fluctuation and operational stability of the preceding process stages can be included in the stability parameter group for subsequent judgment on whether the output of the preceding process is in a stable state.

[0024] Based on the influence of each state characteristic parameter on the switching conditions of subsequent process stages, the corresponding state characteristic parameters are divided into boundary offset parameter groups. State characteristic parameters that mainly reflect the end-of-stage changes of the preceding process stage and the changing trends before and after switching can be included in the boundary offset parameter group for subsequent judgment on whether there is an offset trend in the switching boundary of the subsequent process stage. Meanwhile, based on the influence of each state characteristic parameter on the probability of subsequent process anomalies, the corresponding state characteristic parameters are divided into risk transmission parameter groups. In specific implementation, state characteristic parameters that mainly reflect abnormal fluctuations in the preceding process stage, the accumulation of end deviations, and the trend of adverse state continuation can be included in the risk transmission parameter group for subsequent judgment on whether the anomaly in the preceding process will be transmitted to the subsequent process.

[0025] Furthermore, S2 constructs a multi-level inheritance state, specifically including: Each state characteristic parameter in the stability parameter group is compared with the corresponding stability judgment threshold, and the output stability of the preceding process stage is determined based on the comparison results to generate a stable output state. Each state feature parameter in the boundary offset parameter group is compared with the corresponding boundary offset judgment threshold, and the boundary offset direction and degree at the end of the preceding process stage are determined based on the comparison results to generate the stage boundary offset state. The abnormal risk assessment value is obtained by weighting each state characteristic parameter in the risk transmission parameter group according to the preset parameter weights. The abnormal risk assessment value is then compared with the corresponding risk judgment threshold, and the abnormal risk transmission status is generated based on the comparison result.

[0026] Specifically, after grouping the state feature parameters according to different dimensions of transmission influence, the state determination process is further performed on different parameter groups to form corresponding multi-level inheritance states. The multi-level inheritance states here are not a simple summary of the results of the previous process stages, but rather the results of the previous processes are transformed into state results that can characterize the impact of subsequent process stability, the impact of stage switching boundaries, and the impact of abnormal transmission risks. This provides a direct basis for determining the subset of affected parameters and identifying the trend of stage switching boundary offset. During the formation of a stable state, a corresponding stability judgment threshold range can be pre-set for each state characteristic parameter in the stability parameter group, and each state characteristic parameter can be compared with the corresponding stability judgment threshold. In specific implementation, the pressure fluctuation characteristic parameter, velocity fluctuation characteristic parameter, displacement fluctuation characteristic parameter, and stage duration offset characteristic parameter can be compared with the corresponding allowable fluctuation range to determine whether the preceding process stage is in a stable state during the output process. When all state characteristic parameters are within the corresponding stability judgment threshold range, the output of the preceding process stage can be considered stable, forming a stable output state. When at least some state characteristic parameters exceed the corresponding stability judgment threshold range, it indicates that the preceding process stage has an unstable output situation. The characteristic parameters that originally reflected the fluctuation situation are further transformed into a stable state result, so that subsequent steps can directly use the state result to determine whether the preceding process has an impact on the stable execution of the subsequent process. During the formation of the stage boundary offset state, the boundary change characteristics at the end of the preceding process stage can be used for judgment processing. That is, each state characteristic parameter in the boundary offset parameter group is compared with the corresponding boundary offset judgment threshold, and the offset of the end of the preceding process stage relative to the preset switching condition is determined by combining the deviation direction of each state characteristic parameter. For example, the end pressure deviation, end speed deviation, end displacement deviation and the change trend before and after the switch can be combined to determine whether the end condition of the preceding process stage tends to be earlier, later or basically consistent with the target switching condition of the subsequent process. At the same time, the degree of boundary offset can be determined by the magnitude of the deviation exceeding the threshold. In this way, the change characteristics at the end of the preceding process stage can be transformed into the stage boundary offset state, providing a state basis for the identification of the boundary offset trend and boundary correction of the subsequent stage switching. During the formation of an abnormal risk transmission state, the state characteristic parameters in the risk transmission parameter group can be comprehensively evaluated to determine the degree of adverse impact that may be transmitted from the preceding process stage to the subsequent process. In specific implementation, the state characteristic parameters in each risk transmission parameter group can be weighted according to preset parameter weights to obtain the abnormal risk evaluation value. The calculation method can be expressed as follows: ; in, Indicates the abnormal risk assessment value. This indicates the number of state characteristic parameters in the risk propagation parameter group. Indicates the first Preset parameter weights corresponding to each state feature parameter Indicates the first The values ​​of state characteristic parameters in each risk transmission parameter group. This can correspond to data characterizing the tendency of abnormal transmission in the preceding process stages, such as abnormal pressure abrupt changes, abnormal velocity attenuation, displacement end deviation, and abnormal changes before and after switching. The influence of each state characteristic parameter on the probability of subsequent process anomalies can be preset. After the anomaly risk assessment value is calculated, it is compared with the corresponding risk judgment threshold. When the anomaly risk assessment value is greater than the risk judgment threshold, it can be determined that there is a high anomaly risk transmission in the preceding process stage, forming an anomaly risk transmission state. When the anomaly risk assessment value does not exceed the risk judgment threshold, it can be determined that the risk of the preceding process stage transmitting anomalies to the subsequent process is low.

[0027] S3. By using multi-level inheritance states, determine the degree of transmission influence of each candidate parameter in the subsequent process, and determine the subset of affected parameters based on the degree of transmission influence, while also determining the trend of stage switching boundary offset. Furthermore, S3 identifies a subset of affected parameters for subsequent processes, specifically including: Based on the stable output state, the stage boundary offset state, and the abnormal risk transmission state, the degree of transmission influence of each candidate parameter in the subsequent process is calculated. The degree of transmission influence of each candidate parameter is compared with the preset influence threshold, and subsequent process parameters with a degree of transmission influence greater than the preset influence threshold are selected to form a set of affected parameters. Based on the degree of influence correlation between each subsequent process parameter in the affected parameter set, subsequent process parameters with an influence correlation greater than a preset correlation threshold are selected as a subset of affected parameters.

[0028] Specifically, after completing the construction of the multi-layer inheritance state, the influence analysis of each candidate parameter in the subsequent process is further carried out using the stable output state, stage boundary offset state, and abnormal risk transmission state to identify the actual transmission range of the results of the preceding process to the subsequent process. The candidate parameters here can be injection pressure, injection speed, holding pressure, holding time, switching position, or other process parameters related to the molding process in the subsequent process. Since different candidate parameters are affected by the preceding process stages to different degrees, the transmission influence of each candidate parameter is first quantified and calculated, and then the parameters that really need to participate in the subsequent linkage optimization are selected. This helps to avoid making overall adjustments to all subsequent process parameters, thereby improving the targeting of subsequent optimization. In calculating the degree of transmission impact, the stable output state, the stage boundary offset state, and the abnormal risk transmission state can be converted into corresponding state evaluation quantities. Then, by combining the sensitivity of each candidate parameter to the responses of the three types of states, the degree of transmission impact corresponding to each candidate parameter can be calculated. Specifically, the degree of transmission impact can be calculated using the following formula: ; in, Indicates the first The degree of propagation of each candidate parameter for subsequent processes. This represents the state evaluation quantity corresponding to the stable output state. This represents the state evaluation quantity corresponding to the stage boundary offset state. This represents the state evaluation quantity corresponding to the abnormal risk transmission state. , , They represent the first The weights of each candidate parameter on the stable output state, the stage boundary offset state, and the abnormal risk propagation state. It can be obtained by comparing the stable state parameter set with the stability determination threshold. It can be obtained by comparing the boundary offset parameter set with the boundary offset determination threshold. It can be determined by the aforementioned abnormal risk assessment calculation results. In this way, the influence of the preceding process stage on different candidate parameters of the subsequent process can be uniformly quantified, thus forming a direct basis for subsequent parameter screening. After obtaining the degree of transmission influence corresponding to each candidate parameter, the degree of transmission influence corresponding to each candidate parameter is compared with the preset influence threshold to screen out the subsequent process parameters that have a significant impact on the preceding process results, and form a set of affected parameters. In specific implementation, when the degree of transmission influence corresponding to a candidate parameter is greater than the preset influence threshold, it can be determined that the candidate parameter is significantly affected by the continuous transmission influence of the preceding process results and should be included in the set of affected parameters. When the degree of transmission influence corresponding to the candidate parameter does not exceed the preset influence threshold, it indicates that the candidate parameter is less affected by the preceding process results and can be temporarily not regarded as the preferred object for subsequent linkage optimization. Thus, parameters that are weakly related to the preceding process can be excluded from all candidate parameters first, narrowing the processing scope of subsequent linkage optimization. After forming the set of affected parameters, the correlation between the influence of each subsequent process parameter within the set is further analyzed to select a subset of parameters suitable for coordinated adjustment. In specific implementation, the correlation between parameters can be calculated based on the common influence of each subsequent process parameter on each other and on the stage switching boundary during the adjustment process. The correlation can be calculated according to the following formula: ; in, Indicates the first The subsequent process parameters and the first The correlation between the influence of each subsequent process parameter Indicates the first The change in parameter response caused by the adjustment of subsequent process parameters Indicates the first The change in parameter response caused by the adjustment of subsequent process parameters and This indicates the amount of time change caused by adjusting the corresponding parameter. express and The combined influence of parameters on the stage switching boundary. , , This represents the stage coefficient in the correlation calculation. , , These represent daily baseline quantities representing parameter response changes, stage time changes, and boundary co-influence quantities, respectively. Through the above calculations, the degree of coupling and co-effect between different subsequent process parameters can be quantified, thereby determining which parameters are more suitable to be treated simultaneously as linkage optimization objects. After obtaining the impact correlation, the impact correlation between each subsequent process parameter in the impact parameter set can be compared with a preset correlation threshold, and subsequent process parameters with an impact correlation greater than the preset correlation threshold can be selected as the affected parameter subset. Specifically, when a certain subsequent process parameter has a high impact correlation with other parameters in the set, it indicates that there is a strong coupling relationship between this parameter and other parameters in the subsequent adjustment process. If it is adjusted alone, it is easy to cause changes in the response of other parameters or cause the stage switching boundary to shift. Therefore, it is advisable to include it in the affected parameter subset. Conversely, if a parameter is affected by the preceding process but has a low correlation with other parameters, it can not be regarded as a priority linkage optimization object. Through the above two-level screening method, the affected parameter set is first determined based on the degree of transmission of impact, and then the affected parameter subset is further formed based on the impact correlation. This enables the subsequent linkage optimization to maintain the sensitivity to the impact of the preceding process while taking into account the coupling relationship between subsequent process parameters. Through the above processing method, the multi-layered inherited state formed in the preceding process stage can be further mapped to the parameter level of the subsequent process, forming a transition process from "state influence" to "parameter selection". On the one hand, the calculation of the degree of influence means that the influence of the preceding process results on the subsequent process is no longer limited to qualitative judgment, but can be reflected in each candidate parameter in a quantitative form. On the other hand, the introduction of influence correlation means that the parameters selected later are not just "affected parameters", but "parameters suitable for participating in linkage optimization". The resulting subset of affected parameters can provide a clear parameter object basis for determining the subsequent linkage constraint relationship and executing local linkage rapid optimization.

[0029] Furthermore, S3 determines the stage switching boundary offset trend of subsequent processes, specifically including: Obtain the target switching conditions corresponding to the subsequent process stage, and extract the boundary state parameters related to the target switching conditions at the end of the preceding process stage. The boundary state parameters include the end pressure deviation, end displacement deviation, and end velocity deviation. The boundary state parameters are compared with the target switching conditions to obtain the current boundary state deviation value; Based on the magnitude and direction of the current boundary state deviation, the boundary offset evaluation value is calculated according to the preset evaluation function, and the boundary offset evaluation value is compared with the preset boundary trend judgment threshold range. Based on the comparison result, the offset trend of the stage switching boundary of the subsequent process is determined to move forward, backward, or remain unchanged. Specifically, after determining the subset of affected parameters, the switching boundary offset trend of subsequent process stages is further identified. Since the switching boundary of subsequent process stages usually corresponds to preset pressure, displacement, or velocity conditions, when there is a deviation between the end state of the preceding process stage and the target switching condition, the actual switching timing of the subsequent process stage may change. Therefore, by analyzing the end state of the preceding process stage, a basis can be provided for adjusting the switching boundary of subsequent stages. In practice, the target switching conditions corresponding to the subsequent process stage can be obtained first, and the boundary state parameters related to the target switching conditions at the end of the preceding process stage can be extracted. The boundary state parameters include the end pressure deviation, end displacement deviation, and end velocity deviation, which are used to characterize the deviation of the actual state at the end of the preceding process stage from the target switching conditions. Then, the boundary state parameters are compared with the target switching conditions to obtain the current boundary state deviation value, so as to quantify the difference between the end state of the preceding process stage and the ideal switching conditions of the subsequent process stage. After obtaining the current boundary state deviation value, the boundary offset evaluation value can be further calculated based on the magnitude and direction of the deviation. In specific implementation, a preset evaluation function can be used to comprehensively calculate the end pressure deviation, end displacement deviation, and end velocity deviation to obtain the boundary offset evaluation value. The calculation method can be expressed as follows: ; in, This represents the boundary offset evaluation value. Indicates the deviation of the end pressure. Indicates the displacement deviation at the end point. Indicates the terminal velocity deviation. , , These represent the evaluation weights of the corresponding deviation items. The end pressure deviation, end displacement deviation, and end velocity deviation can be determined by the difference between the actual value at the end of the preceding process stage and the corresponding target switching condition. The evaluation weights can be preset according to the degree of influence of different deviation items on the stage switching boundary. After obtaining the boundary offset evaluation value, it can be compared with the preset boundary trend judgment interval to determine the switching boundary offset trend of subsequent process stages. In specific implementation, when the boundary offset evaluation value is greater than the preset upper threshold, it can be determined that the stage switching boundary has a backward trend. When the boundary offset evaluation value is less than the preset lower threshold, it can be determined that the stage switching boundary has a forward trend. When the boundary offset evaluation value is within the preset boundary trend judgment threshold interval, it can be determined that the stage switching boundary remains unchanged. Thus, the influence of the end state of the preceding process stage on the switching conditions of the subsequent process stage can be transformed into a clear boundary offset trend, providing a basis for subsequent stage switching boundary correction and linkage optimization.

[0030] S4. Based on the multi-level inheritance state and the phase switching boundary offset trend, determine the linkage constraint relationship between each parameter in the affected parameter subset and the collaborative adjustment relationship between the affected parameter subset and the phase switching boundary. Specifically, after completing the construction of the multi-layer inheritance state, determining the affected parameter subset, and identifying the stage switching boundary offset trend, it is further necessary to clarify the interaction relationship between subsequent process parameters and between subsequent process parameters and stage switching boundaries, so as to provide a constraint basis for subsequent local linkage rapid optimization. Since the parameters in the affected parameter subset have mutual influence during the adjustment process, and the adjustment of some parameters will also affect the stage switching boundary, it is necessary to form the corresponding linkage constraint relationship and collaborative adjustment relationship before subsequent optimization. In practice, the influence direction of the preceding process reflected by the multi-layer inheritance state can be combined to conduct correlation analysis on each subsequent process parameter in the affected parameter subset to determine the adjustment priority, adjustment range and mutual constraint relationship of each parameter in the subsequent adjustment process, thereby forming a linkage constraint relationship between each parameter in the affected parameter subset. In this way, it is possible to avoid adjusting each parameter independently and improve the consistency of subsequent parameter linkage optimization.

[0031] Based on this, the collaborative adjustment relationship between the affected parameter subset and the stage switching boundary can be further determined by combining the stage switching boundary offset trend. In specific implementation, the adjustment direction and adjustment magnitude of subsequent process parameters related to the stage switching boundary can be coordinated according to the forward movement trend, backward movement trend, or unchanged state of the stage switching boundary, so that the parameter adjustment and boundary correction are consistent. In this way, the adjustment of subsequent process parameters and the adjustment of stage switching boundary can be carried out in synergy, thereby providing a foundation for subsequent local linkage and rapid optimization.

[0032] S5. Based on the linkage constraint relationship and the collaborative adjustment relationship, perform local linkage rapid optimization on the affected parameter subset and stage switching boundary to obtain the optimized control result of the subsequent process; Furthermore, the linkage constraint relationship and the collaborative adjustment relationship are determined, and local linkage rapid optimization is performed, specifically including: Based on the influence correlation between each subsequent process parameter in the affected parameter subset and the stage switching boundary offset trend, calculate the target adjustment amount, adjustment sequence and adjustment range for each subsequent process parameter, and form the linkage constraint relationship between each parameter in the affected parameter subset. Based on the offset trend of the stage switching boundary, the target adjustment amount of the subsequent process parameters corresponding to the stage switching boundary is associated and matched with the boundary correction amount of the stage switching boundary to form a coordinated adjustment relationship between the affected parameter subset and the stage switching boundary. Based on the linkage constraint relationship and the collaborative adjustment relationship, the subsequent process parameters and stage switching boundaries within the affected parameter subset are synchronously adjusted, and local linkage rapid optimization is performed within the parameter range corresponding to the affected parameter subset and the boundary range corresponding to the stage switching boundary to obtain the optimized control results of the subsequent process.

[0033] Specifically, after determining the linkage constraint relationship and the collaborative adjustment relationship, the next step is to enter the local linkage rapid optimization stage of the subsequent process. Unlike the traditional method of overall debugging of all subsequent process parameters, this implementation method only focuses on the optimization of the affected parameter subset and the stage switching boundary, and incorporates the coupling relationship between parameters and the collaborative relationship between parameters and boundaries into the adjustment process. This makes the optimization process more focused, reduces the number of debugging times caused by irrelevant parameters, and improves the optimization efficiency of subsequent processes. In the process of forming parameter linkage constraints, the influence correlation between each subsequent process parameter in the affected parameter subset and the stage switching boundary offset trend can be combined to calculate the target adjustment amount, adjustment sequence, and adjustment range corresponding to each subsequent process parameter, i.e., the first... The target adjustment amount for each subsequent process parameter can be determined in the following way: ; in, Indicates the first The target adjustment amount for each subsequent process parameter. Indicates the first The degree of transmission of influence corresponding to each subsequent process parameter This represents the boundary offset evaluation value. Indicates the first The correlation between the overall impact of each subsequent process parameter and the other affected parameters. , , These represent the corresponding adjustment weights. According to the first The influence correlation between each subsequent process parameter and the remaining parameters is calculated comprehensively. Based on the above calculation results, the parameter adjustment priority can be further determined according to the size of the target adjustment amount, and the corresponding adjustment range can be determined in combination with the allowable change range of each parameter, thereby forming a linkage constraint relationship between each parameter in the affected parameter subset. In this way, the subsequent parameter adjustment is no longer carried out in isolation, but is a linkage adjustment based on the mutual constraints and sequential coordination between parameters. During the coordinated adjustment of parameters and stage switching boundaries, the target adjustment amount of subsequent process parameters corresponding to the stage switching boundary can be further correlated and matched with the boundary correction amount of the stage switching boundary based on the offset trend of the stage switching boundary, so as to form a coordinated adjustment relationship. In specific implementation, the boundary correction amount of the stage switching boundary can be determined in the following way: ; in, This represents the boundary correction amount for the phase switching boundary. This represents the boundary offset evaluation value. Indicates the first The target adjustment amount for each subsequent process parameter. Indicates the first The weight of the influence of each subsequent process parameter on the stage switching boundary , Indicates the boundary correction factor. This indicates the number of parameters in the affected parameter subset. Through the above processing, the boundary correction amount can not only reflect the offset trend of the stage switching boundary itself, but also reflect the joint effect of related parameter adjustments on boundary changes, thereby ensuring consistency between parameter adjustments and boundary correction. After obtaining the linkage constraint relationship and the collaborative adjustment relationship, the subsequent process parameters and stage switching boundaries in the affected parameter subset can be adjusted synchronously. Local linkage rapid optimization can be performed within the corresponding parameter range and boundary range. The set values ​​of each subsequent process parameter can be updated sequentially according to the determined adjustment order, and the stage switching boundary can be corrected synchronously to obtain the updated combination of subsequent process control parameters. Since the optimization range is limited to the range corresponding to the affected parameter subset and stage switching boundary, compared with the full parameter readjustment method, parameter convergence can be completed within a smaller search range, thereby obtaining the optimized control result of the subsequent process.

[0034] S6. Obtain the execution results of subsequent processes, and determine the mismatch type based on the deviation between the execution results and the expected results and the dominant deviation item corresponding to the deviation; Furthermore, S6 determines the mismatch type based on the deviation between the execution result and the expected result, specifically including: The actual execution trajectory, actual stage switching result, and actual parameter response result of the subsequent process are compared with the corresponding expected execution trajectory, expected stage switching result, and expected parameter response result to obtain the trajectory deviation value, boundary deviation value, and parameter deviation value. The trajectory deviation value, boundary deviation value, and parameter deviation value are compared with the corresponding mismatch judgment thresholds, and the deviation item that exceeds the corresponding mismatch judgment threshold and has the largest deviation amplitude is determined as the dominant deviation item. When the dominant deviation term corresponds to the state determination result of the multi-level inheritance state, it is determined to be an inheritance state mismatch; When the dominant deviation term corresponds to the correction result of the stage switching boundary, it is determined to be a stage switching boundary mismatch; when the dominant deviation term corresponds to the screening result of the affected parameter subset, it is determined to be a parameter subset mismatch.

[0035] Specifically, after completing the local linkage rapid optimization of the subsequent process, the execution results of the subsequent process are further obtained and compared with the corresponding expected results to determine whether the current optimization result is consistent with the expectation. The execution results here may include the actual execution trajectory of the subsequent process, the actual stage switching results, and the actual parameter response results. The expected results are the expected execution trajectory, the expected stage switching results, and the expected parameter response results obtained based on the aforementioned linkage optimization process. By comparing the two, it is possible to identify whether there is a deviation in the current subsequent process optimization process and provide a basis for subsequent mismatch type determination. In practice, the actual execution trajectory is compared with the expected execution trajectory to obtain the trajectory deviation value; the actual stage switching result is compared with the expected stage switching result to obtain the boundary deviation value; and the actual parameter response result is compared with the expected parameter response result to obtain the parameter deviation value. In this way, the deviations of different aspects in the subsequent process can be separated and quantified, avoiding the judgment of mismatch based on a single deviation result and improving the accuracy of mismatch identification. After obtaining the trajectory deviation value, boundary deviation value, and parameter deviation value, each deviation value is further compared with the corresponding mismatch judgment threshold. The deviation item that exceeds the corresponding mismatch judgment threshold and has the largest deviation amplitude is identified as the dominant deviation item. In other words, when a certain type of deviation not only exceeds its corresponding allowable range, but also has the largest amplitude among all deviations exceeding the threshold, it can be identified that the deviation item is the main source of the mismatch of the current optimization result. When multiple deviations exist at the same time, the dominant mismatch factor is identified first to avoid the lack of targeting in the subsequent correction process. After identifying the dominant deviation item, the mismatch type is further determined based on its corresponding source. When the dominant deviation item corresponds to the state determination result of the multi-level inheritance state, it can be identified as an inheritance state mismatch; when the dominant deviation item corresponds to the correction result of the stage switching boundary, it can be identified as a stage switching boundary mismatch; when the dominant deviation item corresponds to the screening result of the affected parameter subset, it can be identified as a parameter subset mismatch. This provides a clear basis for subsequent sub-path correction and improves the pertinence and effectiveness of the subsequent correction process.

[0036] S7. Based on the mismatch type, perform sub-path correction on the multi-layer inheritance state, affected parameter subset, stage switching boundary offset trend, linkage constraint relationship or collaborative adjustment relationship, and use the correction results for subsequent optimization of the current process cycle or rapid optimization of linkage process in the next process cycle.

[0037] Furthermore, S7 performs path correction based on the mismatch type, specifically including: When the mismatch type is inherited state mismatch, the state evaluation results corresponding to each state characteristic parameter are recalculated, and the multi-level inherited state is regenerated based on the recalculated state evaluation results. When the mismatch type is stage switching boundary mismatch, the current boundary state deviation value is recalculated, and the stage switching boundary offset trend is redetermined based on the recalculated current boundary state deviation value. When the mismatch type is parameter subset mismatch, the degree of transmission influence and the degree of influence correlation of each candidate parameter in the subsequent process are re-determined, and the affected parameter subset is re-selected based on the re-determined degree of transmission influence and degree of influence correlation.

[0038] Specifically, after determining the mismatch type, corresponding sub-path corrections are performed according to different mismatch types to avoid uniform readjustment of all states and parameters. In this way, the correction process can be directly targeted at the current main source of mismatch, thereby improving correction efficiency and providing updated basis for subsequent optimization of the current process cycle or rapid optimization of the linkage process of the next process cycle. When the mismatch type is inherited state mismatch, it means that the multi-layered inherited state formed by the results of the preceding process cannot accurately reflect the actual impact of the subsequent process. At this time, the state evaluation results corresponding to each state characteristic parameter can be recalculated, and the multi-layered inherited state can be regenerated based on the recalculated state evaluation results. This allows the characterization results of the output state of the preceding process to be rematched with the current actual working conditions, thereby providing an updated state basis for subsequent parameter screening and boundary offset trend judgment. When the mismatch type is stage switching boundary mismatch, it means that the judgment result of the current stage switching boundary offset trend is inconsistent with the actual switching situation of the subsequent process. At this time, the current boundary state deviation value can be recalculated, and the stage switching boundary offset trend can be re-determined based on the recalculated current boundary state deviation value. The forward, backward or unchanged state of the stage switching boundary can be re-identified, so as to provide a more accurate basis for subsequent boundary correction and parameter coordination adjustment. When the mismatch type is parameter subset mismatch, it indicates that the currently selected affected parameter subset cannot accurately cover the parameter range that plays a major role in the results of subsequent processes. The degree of propagation and correlation of each candidate parameter in the subsequent process can be re-determined, and the affected parameter subset can be re-selected based on the re-determined degree of propagation and correlation. This method allows the parameters participating in subsequent linkage optimization to re-correspond to the current actual operating conditions, avoiding the involvement of irrelevant parameters in adjustments or the omission of key parameters.

[0039] Furthermore, in S7, the corrected results will be used for subsequent optimization of the current process cycle or for rapid optimization of the linked process in the next process cycle, specifically including: The corrected multi-level inheritance state, affected parameter subset, stage switching boundary offset trend, linkage constraint relationship and collaborative adjustment relationship are stored to form a template for subsequent optimization calls; When there are unexecuted subsequent process stages in the current process cycle, the subsequent optimization call template is invoked to perform rapid optimization of the unexecuted subsequent process stages in the current process cycle. At the start of the next process cycle, the subsequent optimization call template is invoked as the initial optimization basis for constructing multi-layer inheritance states, determining the affected parameter subset, determining the stage switching boundary offset trend, and performing local linkage rapid optimization in the next process cycle.

[0040] Specifically, after completing the sub-path correction, the corrected results are further retained so that they can continue to be used for subsequent optimization of the current process cycle or for rapid optimization of the linked process in the next process cycle. This ensures that the state results, parameter results, and boundary results after the sub-path correction do not only affect the current correction process but continue to participate in subsequent optimization calls, thereby improving the continuity and stability of rapid optimization of the linked process. In practice, the corrected multi-layer inheritance state, affected parameter subset, stage switching boundary offset trend, linkage constraint relationship and collaborative adjustment relationship can be stored to form a subsequent optimization call template. This subsequent optimization call template can be understood as the optimization basis updated for the current working condition. It can reflect the latest correspondence between the previous process results, the subsequent process parameter screening results and the boundary correction results, thereby providing a unified call basis for the subsequent optimization process. When there are still unexecuted subsequent process stages in the current process cycle, the subsequent optimization call template can be directly called to perform linkage process rapid optimization on the unexecuted subsequent process stages in the current process cycle. In this way, the corrected results can be applied to the remaining process stages of the current cycle in a timely manner, avoiding the continuation of the identified mismatch problems to the subsequent execution process of the current cycle, thereby improving the timeliness of optimization within the current process cycle. When entering the next process cycle, the subsequent optimization call template can also be called as the initial optimization basis for building multi-level inheritance states, determining the affected parameter subset, determining the stage switching boundary offset trend, and performing local linkage rapid optimization in the next process cycle. The correction results in the current process cycle are continued to be used in the next process cycle, thereby realizing cross-cycle optimization inheritance, reducing the process of repeated identification and repeated adjustment in the next process cycle, and improving the continuity and efficiency of linkage process rapid optimization. Example 1

[0041] First, the operational result data of the plasticizing stage is obtained, including pressure data, velocity data, displacement data, stage duration data, and change data before and after switching. Then, feature extraction is performed on the operational result data to obtain state feature parameters that characterize the output state of the plasticizing stage.

[0042] After obtaining the state characteristic parameters, the different transmission influence dimensions of the subsequent injection and holding pressure stages are grouped according to the results of the plasticizing stage, forming a stability parameter group, a boundary offset parameter group, and a risk transmission parameter group; then, based on the judgment results corresponding to each parameter group, a stable output state, a stage boundary offset state, and an abnormal risk transmission state are constructed, thus forming a multi-layered inherited state.

[0043] Subsequently, based on the multi-layered inheritance state, the degree of transmission influence corresponding to candidate parameters such as injection pressure, injection speed, holding pressure, and holding time is calculated, and subsequent process parameters with a transmission influence degree higher than a preset influence threshold are selected to form an affected parameter set. Then, based on the influence correlation between each subsequent process parameter in the affected parameter set, subsequent process parameters with an influence correlation degree higher than a preset correlation threshold are selected as the affected parameter subset. In this embodiment, injection speed, injection pressure, and holding pressure are determined as the affected parameter subset.

[0044] Simultaneously, the target switching conditions corresponding to the injection stage are obtained, and the end pressure deviation, end displacement deviation, and end velocity deviation related to the target switching conditions at the end of the plasticizing stage are extracted to determine the offset trend of the current stage switching boundary. In this embodiment, it is determined that the switching boundary of the injection stage has a tendency to shift backward.

[0045] Based on this, the linkage constraint relationship between each parameter within the affected parameter subset is further determined, and the collaborative adjustment relationship between the affected parameter subset and the stage switching boundary is determined. Then, based on the linkage constraint relationship and the collaborative adjustment relationship, local linkage rapid optimization is performed on the injection speed, injection pressure, holding pressure and the corresponding stage switching boundary to obtain the optimized control results of the subsequent process.

[0046] During subsequent execution, the actual execution trajectory, actual stage switching results, and actual parameter response results are further acquired and compared with the corresponding expected results. If the boundary deviation value exceeds the corresponding mismatch judgment threshold and is the dominant deviation item, the mismatch type is determined to be stage switching boundary mismatch; then, the current boundary state deviation value is recalculated, and the stage switching boundary offset trend is redefined. The corrected multi-layer inheritance state, stage switching boundary offset trend, linkage constraint relationship, and collaborative adjustment relationship are stored for subsequent optimization of the current process cycle or rapid optimization of the linkage process in the next process cycle.

[0047] This embodiment enables the identification of the transfer effect of plasticizing stage results on injection and holding stages, and allows for the linkage adjustment of subsequent process parameters and stage switching boundaries within a relatively small optimization range. Example 2

[0048] In this embodiment, the preceding process stage is the injection stage, and the subsequent process stage is the holding pressure stage.

[0049] First, the operation result data at the end of the injection stage is obtained. The operation result data includes pressure data, velocity data, displacement data, stage duration data, and change data before and after switching. Then, feature extraction is performed on the operation result data to obtain state feature parameters that characterize the output state of the injection stage.

[0050] After obtaining the state characteristic parameters, the different transmission influence dimensions of the subsequent holding pressure stage are grouped according to the injection stage results to form a stability parameter group, a boundary offset parameter group, and a risk transmission parameter group; then, based on the judgment results corresponding to each parameter group, a stable output state, a stage boundary offset state, and an abnormal risk transmission state are constructed, thus forming a multi-layered inherited state.

[0051] Subsequently, based on the multi-layered inheritance state, the degree of transmission influence corresponding to candidate parameters such as holding pressure, holding time, and holding switching position is calculated, and subsequent process parameters with a transmission influence degree higher than a preset influence threshold are selected to form an affected parameter set. Then, based on the influence correlation between each subsequent process parameter in the affected parameter set, subsequent process parameters with an influence correlation degree higher than a preset correlation threshold are selected as the affected parameter subset. In this embodiment, holding pressure and holding time are determined as the affected parameter subset.

[0052] Simultaneously, the target switching conditions corresponding to the pressure holding stage are obtained, and the end pressure deviation, end displacement deviation, and end velocity deviation related to the target switching conditions at the end of the injection stage are extracted to determine the offset trend of the current stage switching boundary. In this embodiment, it is determined that the switching boundary of the pressure holding stage has a forward shifting trend.

[0053] Based on this, the linkage constraint relationship between the holding pressure and the holding time is further determined, and the coordinated adjustment relationship between them and the stage switching boundary is determined. Then, based on the linkage constraint relationship and the coordinated adjustment relationship, local linkage fast optimization is performed on the affected parameter subset and the stage switching boundary to obtain the optimized control result of the holding stage.

[0054] After the pressure holding phase is completed, the actual execution trajectory, actual phase switching results, and actual parameter response results are further obtained and compared with the corresponding expected results. If the parameter deviation value exceeds the corresponding mismatch judgment threshold and is the dominant deviation item, the mismatch type is determined to be parameter subset mismatch. Subsequently, the degree of transmission influence and influence correlation of subsequent candidate parameters are re-determined, and the affected parameter subset is re-screened. The corrected multi-level inheritance state, affected parameter subset, phase switching boundary offset trend, linkage constraint relationship, and collaborative adjustment relationship are stored and used as the initial basis for rapid optimization of the linkage process in the next process cycle.

[0055] This embodiment enables the identification of the impact of injection stage results on the holding pressure stage, and allows the continued use of corrected optimization results in subsequent cycles, thereby improving the continuity and stability of rapid optimization of the linkage process.

[0056] Although embodiments of the invention have been shown and described, it will be understood by those skilled in the art that various changes, modifications, substitutions and alterations can be made to these embodiments without departing from the principles and spirit of the invention, the scope of which is defined by the appended claims and their equivalents.

Claims

1. A method for rapid optimization of the linkage process in a hydraulic injection molding equipment, characterized in that, include: S1. Obtain the operation result data of the preceding process stage and extract features to obtain the state feature parameters that characterize the output state of the preceding process. S2. Based on the state feature parameters, group the different transmission influence dimensions of the subsequent process according to the results of the preceding process, and construct a multi-layer inheritance state according to the state determination rules corresponding to the state feature parameters of each group. S3. Through the multi-layer inheritance state, determine the degree of transmission influence of each candidate parameter in the subsequent process, and determine the subset of affected parameters based on the degree of transmission influence, while determining the stage switching boundary offset trend. S4. Based on the multi-layer inheritance state and the stage switching boundary offset trend, determine the linkage constraint relationship between each parameter in the affected parameter subset and the collaborative adjustment relationship between the affected parameter subset and the stage switching boundary; S5. Based on the linkage constraint relationship and the collaborative adjustment relationship, perform local linkage rapid optimization on the affected parameter subset and stage switching boundary to obtain the optimized control result of the subsequent process; S6. Obtain the execution results of subsequent processes, and determine the mismatch type based on the deviation between the execution results and the expected results and the dominant deviation item corresponding to the deviation; S7. Based on the mismatch type, perform sub-path correction on the multi-layer inheritance state, affected parameter subset, stage switching boundary offset trend, linkage constraint relationship or collaborative adjustment relationship, and use the correction results for subsequent optimization of the current process cycle or rapid optimization of the linkage process in the next process cycle.

2. The method for rapid optimization of the linkage process of a hydraulic injection molding equipment according to claim 1, characterized in that, The feature extraction of the running result data in S1 specifically includes: Fluctuation amplitude analysis and fluctuation frequency analysis were performed on the pressure data, velocity data and displacement data of the preceding process stage to obtain the corresponding fluctuation characteristic parameters; The rate of change and direction of change of the stage duration data of the preceding process stage and the change data before and after the switch are analyzed to obtain the corresponding trend characteristic parameters. Based on the fluctuation characteristic parameters and the trend characteristic parameters, state characteristic parameters are extracted to characterize the output state of the preceding process.

3. The method for rapid optimization of the linkage process of a hydraulic injection molding equipment according to claim 1, characterized in that, In step S2, the different dimensions of the impact of the preceding process on the subsequent process are grouped according to the results of the preceding process. Specifically, this includes: Based on the influence of each state characteristic parameter on the stability of subsequent processes, the corresponding state characteristic parameters are divided into stability parameter groups. Based on the influence of each state characteristic parameter on the switching conditions of subsequent process stages, the corresponding state characteristic parameters are divided into boundary offset parameter groups. Based on the influence of each state characteristic parameter on the probability of subsequent process anomalies, the corresponding state characteristic parameters are divided into risk transmission parameter groups.

4. The method for rapid optimization of the linkage process of a hydraulic injection molding equipment according to claim 3, characterized in that, The construction of a multi-level inheritance state in S2 specifically includes: Each state characteristic parameter in the stability parameter group is compared with the corresponding stability judgment threshold, and the output stability of the preceding process stage is determined based on the comparison result to generate a stable output state. Each state feature parameter in the boundary offset parameter group is compared with the corresponding boundary offset judgment threshold, and the boundary offset direction and degree at the end of the preceding process stage are determined based on the comparison result to generate the stage boundary offset state. The state characteristic parameters in the risk transmission parameter group are weighted according to the preset parameter weights to obtain the abnormal risk evaluation value. The abnormal risk evaluation value is then compared with the corresponding risk judgment threshold, and an abnormal risk transmission state is generated based on the comparison result.

5. The method for rapid optimization of the linkage process of a hydraulic injection molding equipment according to claim 4, characterized in that, The determination of the affected parameter subset in subsequent processes in S3 specifically includes: Based on the stable output state, stage boundary offset state, and abnormal risk transmission state, calculate the degree of transmission influence of each candidate parameter in the subsequent process. The degree of transmission influence corresponding to each candidate parameter is compared with a preset influence threshold, and subsequent process parameters with a degree of transmission influence greater than the preset influence threshold are selected to form a set of affected parameters; Based on the degree of influence correlation between each subsequent process parameter in the affected parameter set, subsequent process parameters with an influence correlation greater than a preset correlation threshold are selected as a subset of affected parameters.

6. The method for rapid optimization of the linkage process of a hydraulic injection molding equipment according to claim 1, characterized in that, The step-switching boundary offset trend for subsequent processes, as determined in S3, specifically includes: Obtain the target switching conditions corresponding to the subsequent process stage, and extract the boundary state parameters related to the target switching conditions at the end of the preceding process stage. The boundary state parameters include the end pressure deviation, the end displacement deviation, and the end velocity deviation. The boundary state parameters are compared with the target switching conditions to obtain the current boundary state deviation value; Based on the magnitude and direction of the current boundary state deviation, a boundary offset evaluation value is calculated according to a preset evaluation function. The boundary offset evaluation value is then compared with a preset boundary trend determination threshold range. Based on the comparison result, the offset trend of the subsequent process stage switching boundary—whether it moves forward, backward, or remains unchanged—is determined.

7. The method for rapid optimization of the linkage process of a hydraulic injection molding equipment according to claim 1, characterized in that, The process of determining the linkage constraint relationship, the collaborative adjustment relationship, and performing local linkage rapid optimization specifically includes: Based on the influence correlation between each subsequent process parameter in the affected parameter subset and the offset trend of the stage switching boundary, the target adjustment amount, adjustment order and adjustment range of each subsequent process parameter are calculated to form the linkage constraint relationship between each parameter in the affected parameter subset. Based on the offset trend of the stage switching boundary, the target adjustment amount of the subsequent process parameters corresponding to the stage switching boundary is associated and matched with the boundary correction amount of the stage switching boundary to form a collaborative adjustment relationship between the affected parameter subset and the stage switching boundary. According to the linkage constraint relationship and the collaborative adjustment relationship, the subsequent process parameters and the stage switching boundary within the affected parameter subset are synchronously adjusted, and local linkage rapid optimization is performed within the parameter range corresponding to the affected parameter subset and the boundary range corresponding to the stage switching boundary to obtain the optimized control result of the subsequent process.

8. The method for rapid optimization of the linkage process of a hydraulic injection molding equipment according to claim 1, characterized in that, In step S6, the mismatch type is determined based on the deviation between the execution result and the expected result, specifically including: The actual execution trajectory, actual stage switching result, and actual parameter response result of the subsequent process are compared with the corresponding expected execution trajectory, expected stage switching result, and expected parameter response result to obtain the trajectory deviation value, boundary deviation value, and parameter deviation value. The trajectory deviation value, boundary deviation value, and parameter deviation value are compared with the corresponding mismatch judgment thresholds, and the deviation item that exceeds the corresponding mismatch judgment threshold and has the largest deviation amplitude is determined as the dominant deviation item. When the dominant deviation term corresponds to the state determination result of the multi-level inheritance state, it is determined to be an inheritance state mismatch; When the dominant deviation term corresponds to the correction result of the stage switching boundary, it is determined to be a stage switching boundary mismatch; When the dominant deviation term corresponds to the screening result of the affected parameter subset, it is determined to be a parameter subset mismatch.

9. The method for rapid optimization of the linkage process of a hydraulic injection molding equipment according to claim 1, characterized in that, The path correction in step S7 based on the mismatch type specifically includes: When the mismatch type is inherited state mismatch, the state evaluation results corresponding to each state feature parameter are recalculated, and the multi-layer inherited state is regenerated based on the recalculated state evaluation results. When the mismatch type is stage switching boundary mismatch, the current boundary state deviation value is recalculated, and the stage switching boundary offset trend is redetermined based on the recalculated current boundary state deviation value. When the mismatch type is parameter subset mismatch, the degree of transmission influence and the degree of influence correlation of each candidate parameter in the subsequent process are re-determined, and the affected parameter subset is re-selected based on the re-determined degree of transmission influence and degree of influence correlation.

10. A method for rapid optimization of the linkage process of a hydraulic injection molding equipment according to claim 9, characterized in that, In step S7, the correction results are used for subsequent optimization of the current process cycle or for rapid optimization of the linked process in the next process cycle, specifically including: The corrected multi-level inheritance state, affected parameter subset, stage switching boundary offset trend, linkage constraint relationship and collaborative adjustment relationship are stored to form a template for subsequent optimization calls; When there is an unexecuted subsequent process stage in the current process cycle, the subsequent optimization call template is invoked to perform rapid optimization of the unexecuted subsequent process stage in the current process cycle. At the start of the next process cycle, the subsequent optimization call template is invoked as the initial optimization basis for constructing multi-layer inheritance states, determining the affected parameter subset, determining the stage switching boundary offset trend, and performing local linkage rapid optimization in the next process cycle.

Images