A group standard-based intelligent manufacturing process optimization method and system

Abstract

The application discloses an intelligent manufacturing process optimization method and system based on group standards, relates to the technical field of manufacturing process collaborative optimization, and comprises the following steps: collecting standard process information corresponding to an intelligent manufacturing production line, performing semantic mapping on the standard process information, and generating process mapping information; through a constraint conflict analysis method, comparing the execution conditions and control ranges of different standard constraints according to the process mapping information, identifying conflict relationships and separating inconsistent constraints, obtaining different standard constraints, correcting the constraint sequence of the different standard constraints, and generating a standard constraint structure; performing process optimization matching on deviation source positioning information, obtaining an affected process, performing time sequence correction and adjustment on an execution path of the affected process, and generating an optimized execution scheme. The multi-standard adaptation capability, the deviation source positioning accuracy and the process optimization pertinence are improved.

Description

Technical Field

[0001] This invention relates to the field of collaborative optimization technology for manufacturing processes, and in particular to a method and system for optimizing intelligent manufacturing processes based on group standards. Background Technology

[0002] As intelligent manufacturing production lines develop towards flexibility, collaboration, and standardization, production processes no longer rely solely on single process procedures for control. Instead, they gradually incorporate process constraint information to unify and dynamically manage the production line's operation. Existing intelligent manufacturing process management methods based on standards and specifications typically first organize standard clauses, process requirements, and control conditions from different sources into executable process rules. Then, they combine these rules with process status, cycle time status, and quality inspection information during manufacturing execution to match and verify the actual production process, thereby determining whether the process execution meets the established process requirements.

[0003] While existing process management methods can organize standard clauses and routinely check operational status, they still find it difficult to establish a continuous correlation between standard semantics, constraint conflicts, real-time process trajectories, and sources of deviation when multiple standard constraints act on the same production line process. Summary of the Invention

[0004] In view of the aforementioned existing problems, the present invention is proposed.

[0005] Therefore, this invention provides a smart manufacturing process optimization method based on group standards, which solves the problems of difficulty in correcting conflicts of multiple standard constraints and difficulty in tracing the source of execution deviations.

[0006] To solve the above-mentioned technical problems, the present invention provides the following technical solution: In a first aspect, the present invention provides a method for optimizing intelligent manufacturing processes based on group standards, comprising: collecting standard process information corresponding to the intelligent manufacturing production line; semantically mapping the standard process information to generate process mapping information; using a constraint conflict resolution method, comparing the execution conditions and control ranges of different standard constraints according to the process mapping information, identifying conflict relationships and separating inconsistent constraints, obtaining different standard constraints, performing constraint sequence correction on different standard constraints, and generating a standard constraint structure; reading the real-time operating status according to the production line cycle time, decomposing the real-time operating status into process trajectory segments, embedding the corresponding constraint segments of the standard constraint structure, checking the constraint conformity degree segment by segment, identifying deviations, and generating an execution deviation sequence; using a deviation chain tracing method, performing causal tracing analysis on the execution deviation sequence, obtaining process succession relationships, and tracing back the source of deviations level by level according to the process succession relationships to determine the initial position of the deviation, generating deviation source location information; performing process optimization matching on the deviation source location information, obtaining the affected processes, performing time sequence correction and adjustment on the execution paths of the affected processes, and generating an optimized execution plan.

[0007] As a preferred embodiment of the intelligent manufacturing process optimization method based on group standards described in this invention, the specific steps of collecting standard process information corresponding to the intelligent manufacturing production line, semantically mapping the standard process information, and generating process mapping information are as follows: The process element information in the standard process information is broken down line by line. Semantic equivalence recognition is performed on the process element information to obtain semantic equivalence groups. The semantic equivalence groups are then uniformly replaced and re-concatenated to generate semantic standard expression data. Based on the semantic standard expression data, the process relationships and constraints between processes are compared item by item to obtain the process nodes and their corresponding relationships. The process nodes and their corresponding relationships are then bound and mapped one by one to generate process mapping information.

[0008] As a preferred embodiment of the intelligent manufacturing process optimization method based on group standards described in this invention, the specific steps for obtaining different standard constraints are as follows: Extract process constraint elements from process mapping information, expand and organize process constraint elements to obtain multi-standard constraints, merge the sources of multi-standard constraints, and generate constraint expansion sequence. The constraint expansion sequence is compared pairwise to obtain constraint difference features. The constraint difference features are combined with the differences in execution conditions and the overlap of control scope to determine conflicts and generate a conflict measurement sequence. By using the constraint conflict resolution method, conflict candidates in the conflict measurement sequence are read, and the constraint conflict features corresponding to each conflict candidate are checked one by one. Constraint pairs whose constraint conflict features meet the substantive conflict conditions are retained, and constraint pairs whose constraint conflict features only show object mismatch are eliminated, thus generating a set of conflict relationships. Based on the set of conflict relationships, the control scope, execution conditions and corresponding process nodes of each constraint pair are read. The constraint pair is split into a non-overlapping segment in the front, an overlapping segment in the middle and a non-overlapping segment in the back. The mutually exclusive execution conditions in the overlapping segment in the middle are extracted from the original constraint to generate heterogeneous constraint segments. Heterogeneous constraint fragments and constraints that do not conflict are uniformly merged to obtain a constraint integration sequence. The constraint integration sequence is then classified, organized, and its order is checked to generate constraints of different standards.

[0009] As a preferred embodiment of the intelligent manufacturing process optimization method based on group standards described in this invention, the standard constraint structure is obtained by correcting the execution order of each constraint segment according to different standard constraints, and by rearranging and filling the boundaries of constraints with overlapping conflicts.

[0010] As a preferred embodiment of the intelligent manufacturing process optimization method based on group standards described in this invention, the specific steps for generating the execution deviation sequence are as follows: Read the real-time running status according to the production line cycle time, connect the continuous running content of the same process node in the cycle time sequence to obtain the cycle time running chain, identify the running breakpoint information according to the cycle time running chain, and divide it by the interruption point before and after to generate process trajectory segments. Read the process node and cycle time position based on the process trajectory fragment, insert the process trajectory fragment into the corresponding constraint segment of the standard constraint structure to obtain the trajectory embedding chain, check the cycle time offset, sequence misalignment and range out-of-bounds content segment by segment based on the trajectory embedding chain, and string them together according to the continuous deviation position to generate the execution deviation sequence.

[0011] As a preferred embodiment of the intelligent manufacturing process optimization method based on group standards described in this invention, the step of performing causal tracing analysis on the execution deviation sequence through the deviation chain tracing method to obtain the process succession relationship includes the following specific steps: Based on the execution deviation sequence, the deviation node elements are read item by item, and the consecutive deviation records are linked together into a deviation transmission chain according to the order of execution time to generate deviation chain data. Using the deviation chain tracing method, adjacent process nodes are read back item by item in reverse chronological order of the deviation chain data. The execution order, deviation direction changes, and deviation persistence position changes of adjacent process nodes are compared, and the comparison content is organized into node readback segments. Based on the node readback segments, adjacent process nodes that are sequentially connected, have continuously extending deviation directions, and whose deviation positions can be transmitted and connected are selected. The upstream process node is used as the candidate source point and the downstream process node is used as the deviation receiving point. The deviation back transmission path is recorded, and the continuously connected adjacent process nodes are merged to generate the process receiving relationship.

[0012] As a preferred embodiment of the intelligent manufacturing process optimization method based on group standards described in this invention, the specific steps for determining the initial location of the deviation and generating deviation source location information by tracing back the source of deviation step by step according to the process succession relationship are as follows: Based on the process succession relationship, each candidate process node is read in reverse, and the deviation tracing characteristics of each candidate process node are jointly analyzed to generate a source point pointing sequence. Based on the source point pointing sequence, candidate process nodes are arranged according to the strength of the source point pointing, and the process succession relationship is checked step by step to check whether there are still deviation records of the same type and in the same direction before the candidate process nodes, thus generating a reverse backtracking sequence. The process node that first shows a deviation and can continuously trigger subsequent deviations is selected as the initial position of the deviation. The initial position, the corresponding deviation type, the corresponding deviation direction, and the corresponding constraint segment are bound together to generate deviation source location information.

[0013] As a preferred embodiment of the intelligent manufacturing process optimization method based on group standards described in this invention, the specific steps for optimizing and matching the deviation source location information to obtain the affected processes are as follows: Extract deviation source elements from deviation source location information, and find process nodes that are connected to the initial position from the deviation source elements. Expand the connected process nodes according to the execution order to generate a deviation influence chain. Based on the deviation impact chain, check the connection status between each process node and the corresponding constraint segment, mark the process nodes that meet the deviation propagation conditions as affected processes, and generate affected processes.

[0014] As a preferred embodiment of the intelligent manufacturing process optimization method based on group standards described in this invention, the generation of the optimized execution scheme is achieved by performing time-series correction and adjustment on the execution time and sequence of the affected processes, obtaining correction path information, and binding the correction path information with the corresponding constraint segments.

[0015] Secondly, the present invention provides an intelligent manufacturing process optimization system based on group standards, comprising: The process mapping module is used to collect standard process information corresponding to the intelligent manufacturing production line, perform semantic mapping on the standard process information, and generate process mapping information. The constraint correction module is used to identify conflicting relationships and separate inconsistent constraints by comparing the execution conditions and control ranges of different standard constraints based on process mapping information through constraint conflict resolution methods, obtaining different standard constraints, correcting the constraint order of different standard constraints, and generating a standard constraint structure. The deviation identification module is used to perform real-time running status alignment and matching verification of the standard constraint structure, identify deviations from the standard constraint structure, and generate an execution deviation sequence. The deviation tracing module is used to perform causal tracing analysis on the execution deviation sequence through the deviation chain tracing method, obtain the process succession relationship, and trace back the deviation source level by level according to the process succession relationship to determine the initial position of the deviation and generate deviation source location information. The process optimization module is used to optimize and match the deviation source location information, obtain the affected processes, perform timing correction and adjustment on the execution path of the affected processes, and generate an optimized execution plan.

[0016] The beneficial effects of this invention are as follows: by using the constraint conflict resolution method, multi-source standard constraints are transformed into an executable standard constraint structure, completing the coordinated organization of standard semantics, execution conditions, and control scope; by using the deviation chain tracing method, the execution deviation sequence is read back in reverse and the initial position of the deviation is located, and then the affected process is matched and the timing correction is adjusted, which improves the multi-standard adaptation capability, the accuracy of deviation source location, and the targeting of process optimization. Attached Figure Description

[0017] To more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings used in the following description of the embodiments will be briefly introduced. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0018] Figure 1 This is a flowchart of a smart manufacturing process optimization method based on group standards.

[0019] Figure 2 This is a schematic diagram of an intelligent manufacturing process optimization system based on group standards.

[0020] Figure 3 Generate flowcharts for different standard constraints.

[0021] Figure 4 Create a flowchart for the process succession relationship. Detailed Implementation

[0022] To make the above-mentioned objects, features and advantages of the present invention more apparent and understandable, the specific embodiments of the present invention will be described in detail below with reference to the accompanying drawings.

[0023] Many specific details are set forth in the following description in order to provide a full understanding of the invention. However, the invention may also be practiced in other ways different from those described herein, and those skilled in the art can make similar extensions without departing from the spirit of the invention. Therefore, the invention is not limited to the specific embodiments disclosed below.

[0024] Secondly, the term "one embodiment" or "embodiment" as used herein refers to a specific feature, structure, or characteristic that may be included in at least one implementation of the present invention. The phrase "in one embodiment" appearing in different places in this specification does not necessarily refer to the same embodiment, nor is it a single or selective embodiment that is mutually exclusive with other embodiments.

[0025] Reference Figures 1-4 This is one embodiment of the present invention, which provides a smart manufacturing process optimization method based on group standards, including the following steps: S1. Collect standard process information corresponding to the intelligent manufacturing production line, perform semantic mapping on the standard process information, and generate process mapping information.

[0026] S1.1. Decompose the process element information in the standard process information one by one, perform semantic equivalence recognition on the process element information, obtain semantic equivalence groups, and uniformly replace and re-concatenate the semantic equivalence groups to generate semantic standard expression data.

[0027] It should be explained that the standard process information is read line by line, and the content describing the same processing step, inspection step, or transfer step in the standard process information is divided into process element information according to its position. Each process element information is read in the order of the preceding and following statements in the standard process information. The current process element information and the adjacent process element information are placed in the same verification position. First, it is verified whether the two process element information points to the same process step. Then, it is verified whether the two process element information play the same role in the preceding and following connection. It is verified whether the preceding and following statements can still be maintained after the two process element information are replaced. Process element information with consistent process steps, consistent connection roles, and no change in the continuity of preceding and following statements after replacement is identified as semantically equivalent content and grouped into the same semantic equivalence group. From the semantic equivalence group, the expression that can cover the meaning of the same group and is convenient for subsequent connection comparison is selected as the unified expression. The remaining process element information in the semantic equivalence group is replaced with the unified expression. The replaced process element information is re-connected in the order of the standard process information to generate semantic standard expression data.

[0028] It should also be noted that the significance and role of semantic equivalence identification lies in the unified identification and merging of content in standard process information that is expressed differently but points to the same processing, inspection, or transfer stage. This avoids confusion in subsequent process comparisons caused by different group standards, enterprise standards, or operating specifications using different wording. By judging whether the process stages are consistent, whether the connecting functions are consistent, and whether the statements before and after the replacement are continuous, equivalent content can be unified into a standard expression that is easy to compare. This provides a consistent data foundation for subsequent extraction of process connection relationships, constraint conflict resolution, and generation of standard constraint structures, thereby improving the accuracy and continuity of multi-standard process mapping.

[0029] S1.2. Based on the semantic standard, compare the connection and constraint descriptions between the processes item by item to obtain the process nodes and their corresponding relationships. Bind and map the process nodes and their corresponding relationships one by one to generate process mapping information.

[0030] It should be explained that, following the semantic standard, adjacent process contents are read item by item in sequence according to the order in the data. The completion status of the previous process and the entry condition of the next process are placed in the same comparison position to check whether the completion status can continue with the entry condition. Adjacent process contents that can continue are marked as succession relationships. The constraint descriptions corresponding to the succession relationships are then read to check whether the constraint descriptions limit the same process content, whether they act on the same succession position, and whether they are consistent with the order. Constraint descriptions with consistent limiting content, consistent succession position, and consistent order are marked as correspondence relationships. Each process node is determined based on the succession relationships and correspondence relationships, and process nodes and correspondence relationships are bound item by item according to the order in the semantic standard data. The bound process nodes and correspondence relationships are mapped and organized according to the succession order of the process nodes to generate process mapping information.

[0031] It should also be noted that, in the semantic standard expression data, the correspondence is established by first identifying whether the content of the preceding process can be continuously connected with the content of the following process to determine the process node; then, the constraint descriptions related to the process node are read to verify whether the constraint descriptions limit the same process content, whether they act on the same connecting position, and whether they conform to the execution order. Constraint descriptions that meet these conditions are then associated with the process node.

[0032] S2. Using the constraint conflict resolution method, the execution conditions and control scope of different standard constraints are compared based on the process mapping information to identify conflict relationships and separate inconsistent constraints, obtain different standard constraints, perform constraint order correction on different standard constraints, and generate a standard constraint structure.

[0033] S2.1 Extract process constraint elements from process mapping information, expand and organize process constraint elements to obtain multi-standard constraints, merge the sources of multi-standard constraints, and generate constraint expansion sequence.

[0034] It should be explained that the process involves reading each process node and its corresponding relationship from the process mapping information, extracting and organizing the limiting content, execution boundaries, and preceding and following requirements corresponding to the process nodes into process constraint elements; expanding the process constraint elements according to the sequence of process nodes, listing process constraint elements from different standard sources under the same process node in their corresponding expansion positions, and maintaining a continuous correspondence between the process constraint elements and process nodes and their corresponding relationships to obtain multi-standard constraints; merging the sources of the multi-standard constraints based on their standard source, applicable process, and following position, grouping multi-standard constraints with the same standard source and applicable process into the same source segment, and placing multi-standard constraints with different standard sources but the same following position into the same comparison segment; and continuously organizing the merged multi-standard constraints according to the sequence of process nodes and the arrangement of source segments to generate a constraint expansion sequence.

[0035] S2.2. Perform pairwise comparisons on the constraint expansion sequence to obtain constraint difference features. Combine the constraint difference features with the differences in execution conditions and the overlap of control ranges, and perform conflict determination to generate a conflict measurement sequence.

[0036] It should be noted that, following the order of the process nodes in the constraint expansion sequence, the multi-standard constraints within the same comparison segment are read item by item. Any two multi-standard constraints are placed in the same comparison position. First, the process content defined by the two multi-standard constraints is compared to see if they are consistent. Then, the execution order, allowed state, and prohibited state requirements of the two multi-standard constraints are compared to see if they differ. Finally, the process segments affected by the two multi-standard constraints are compared to see if they overlap or partially overlap, thus obtaining constraint difference characteristics. Based on the constraint difference characteristics, the differences in execution conditions and the overlap of control scope are read. Multi-standard constraints with opposite execution conditions, overlapping or partial overlap of control scope, and consistent corresponding process content are marked as conflict candidates. Multi-standard constraints with only different wording but consistent execution requirements are marked as non-conflicting content. The constraint difference characteristics, differences in execution conditions, and overlap of control scope corresponding to the conflict candidates are recorded according to the order of the process nodes. Non-conflicting content is retained in its original position, generating a conflict measurement sequence.

[0037] S2.3. Using the constraint conflict resolution method, read the conflict candidates in the conflict measurement sequence, check the constraint conflict features corresponding to each conflict candidate, retain the constraint pairs whose constraint conflict features meet the substantive conflict conditions, and remove the constraint pairs whose constraint conflict features only show object mismatch, thus generating a set of conflict relationships.

[0038] It should be noted that the constraint conflict resolution method reads conflict candidates according to the order of the conflict metric sequence. The two multi-standard constraints corresponding to the conflict candidates are placed in the same verification position. First, it is verified whether the two multi-standard constraints limit the same process content. Then, it is verified whether the execution conditions of the two multi-standard constraints have opposite requirements, and whether the control scope of the two multi-standard constraints overlaps or partially overlaps. The process-limited content, opposite execution content, and scope coverage content formed by the verification are organized into constraint conflict features. When a constraint conflict feature simultaneously has process-limited content, opposite execution content, and scope coverage content, it is determined that the constraint conflict feature meets the substantive conflict condition, and the corresponding constraint pair is retained. When a constraint conflict feature only shows inconsistency in process content or mismatched objects, and the difference in execution conditions cannot affect the same control scope, it is determined that the constraint conflict feature only shows object mismatch, and the corresponding constraint pair is removed. Finally, the retained constraint pairs are summarized according to the process node order in the conflict metric sequence to generate a conflict relationship set.

[0039] It should also be noted that the constraint conflict resolution method reads the conflict measurement sequence according to the process node order, and puts the conflict candidates under the same process node into the verification position one by one. First, it checks whether the conflict candidate points to the same process content, then checks whether the execution requirement has the opposite direction, and finally checks whether the control scope has coverage or partial coverage. When the same process content, opposite execution requirement and control scope coverage are all true at the same time, the conflict candidate is judged as a substantial conflict and retained as a constraint pair. When the conflict candidate only shows that the object of action is inconsistent, the receiving position is not corresponding or the control scope cannot overlap, the conflict candidate is judged as object mismatch and eliminated, thereby realizing the screening of conflict candidates into the set of conflict relationships.

[0040] S2.4. Based on the set of conflict relationships, read the control range, execution conditions and corresponding process nodes of each constraint pair, split the constraint pair into a front non-overlapping segment, a middle overlapping segment and a rear non-overlapping segment, and extract the mutually exclusive execution conditions in the middle overlapping segment from the original constraint to generate heterogeneous constraint segments.

[0041] It should be noted that, based on the conflict relationship set, constraint pairs are read group by group. The control ranges in the constraint pairs are aligned according to the sequential positions of the corresponding process nodes. First, the starting part belonging only to the previous control range is marked as the first non-overlapping segment. Then, the part covered by both control ranges is marked as the middle overlapping segment. Finally, the ending part belonging only to the next control range is marked as the next non-overlapping segment. The execution conditions corresponding to the middle overlapping segments are read. Contents within the same middle overlapping segment that require opposite directions, whose execution states cannot be simultaneously established, or whose succession order is mutually exclusive are identified as mutually exclusive execution conditions. The mutually exclusive execution conditions are extracted from the original constraints, and the succession position relationship between the first non-overlapping segment, the middle overlapping segment, and the next non-overlapping segment and the corresponding process node is retained respectively. The fragments after being split are merged according to the arrangement order in the conflict relationship set to generate heterogeneous constraint fragments.

[0042] S2.5. Merge heterogeneous constraint fragments and constraints that do not conflict into a unified sequence to obtain a constraint integration sequence. Then, classify, organize, and check the order of the constraint integration sequence to generate constraints of different standards.

[0043] It should be explained that, according to the sequential order of process nodes, heterogeneous constraint fragments and non-conflicting constraints are read. Heterogeneous constraint fragments are returned to their original receiving positions in the corresponding process nodes, and non-conflicting constraints are arranged continuously in their original positions. For heterogeneous constraint fragments and non-conflicting constraints under the same process node, each item is checked to see if the receiving position is consistent, whether the execution conditions are continuously applicable, and whether there are any omissions or repetitions in the control scope. Contents with consistent receiving positions and continuous applicability are grouped into the same constraint segment. Contents with missing control scopes are added to adjacent constraint segments, and content with repeated control scopes is retained once according to the process sequence, thus obtaining a constraint integration sequence. Then, the constraint integration sequence is classified and organized according to the standard source, applicable process, and execution order. Contents with the same source and consistent applicable processes are grouped into the same category, and content with different sources but acting on the same process node is arranged side by side according to the sequential relationship. Finally, the classified and organized constraint content is checked item by item to see if it conforms to the process node order and execution sequence, and content with misaligned order is adjusted to the corresponding receiving position to generate different standard constraints.

[0044] S2.6. Correct the execution order of each constraint fragment according to different standard constraints, and rearrange and fill the boundaries of constraints with coverage conflicts to generate a standard constraint structure.

[0045] It should be noted that, according to different standard constraints, each constraint fragment is read item by item, and each constraint fragment is placed in the same correction sequence according to the order of the corresponding process nodes. First, it is checked whether the execution start point of the adjacent constraint fragment is later than the completion position of the previous constraint fragment. Then, it is checked whether the execution content of the adjacent constraint fragments can be continuously inherited. Constraint fragments with advanced, delayed, or broken sequences are adjusted to the correct execution position of the corresponding process node. The constraints with overlapping conflicts are read, and the constraints with overlapping conflicts are rearranged according to common coverage positions, individual coverage positions, and missing positions. Mutually exclusive contents in common coverage positions are separated and arranged, individual coverage positions are retained under the corresponding process node, and missing positions are filled between adjacent constraint fragments to form a continuous boundary. The standard constraint structure is generated by continuously organizing according to the corrected execution order, rearranged control range, and filled boundary relationship.

[0046] S3. Read the real-time operating status according to the production line cycle time, decompose the real-time operating status into process trajectory segments, embed the corresponding constraint segments of the standard constraint structure, check the constraint matching degree segment by segment, identify deviations, and generate an execution deviation sequence.

[0047] S3.1 Read the real-time running status according to the production line cycle time, connect the continuous running content of the same process node in sequence according to the cycle time to obtain the cycle time running chain, identify the running breakpoint information according to the cycle time running chain, and divide it by the interruption point before and after to generate process trajectory segments.

[0048] It should be noted that the real-time operating status is read step by step according to the production line cycle time. Operational content belonging to the same process node and whose preceding and following cycles can be continuously connected are sequentially linked according to the cycle time order to obtain the cycle time operation chain. The cycle time operation chain is checked item by item along the cycle time operation chain for the cycle time interval, process node continuity, and operational status connection relationship of adjacent operational content. When there is a missing cycle time between adjacent operational content, a process node jump that cannot be reconnected, or the operational status changes from continuous execution to a pause, the corresponding position is identified as operational breakpoint information. Based on the operational breakpoint information, the points of interruption in the preceding and following cycles in the cycle time operation chain are marked. Using these points of interruption as the dividing points, the continuous operational content before the dividing point is grouped into the same segment, and the continuous operational content after the dividing point is grouped into the next segment, generating a process trajectory segment.

[0049] It should also be noted that the breakpoint information includes the location of missing cycle time, the location where the process node cannot be reconnected after jumping, and the location where the running status changes from continuous execution to pause and interruption.

[0050] S3.2. Read the process node and cycle position according to the process trajectory fragment, insert the process trajectory fragment into the corresponding constraint segment of the standard constraint structure, obtain the trajectory embedding chain, check the cycle offset, sequence misalignment and range out-of-bounds content segment by segment according to the trajectory embedding chain, and string them together according to the continuous deviation position to generate the execution deviation sequence.

[0051] It should be explained that, based on the process trajectory fragments, process nodes and cycle times are read item by item. The process nodes are matched with the corresponding process nodes in the standard constraint structure. The cycle times are aligned sequentially with the start and end positions of the corresponding constraint segments in the standard constraint structure. The matched process trajectory fragments are then inserted into the corresponding constraint segments to obtain the trajectory embedding chain. The process trajectory fragments and corresponding constraint segments are read segment by segment along the trajectory embedding chain. First, it is checked whether the cycle times are earlier or later than the allowed position of the corresponding constraint segment to identify cycle time offsets. Then, it is checked whether the arrangement of process nodes in the trajectory embedding chain deviates from the sequential order in the standard constraint structure to identify sequence misalignments. Next, it is checked whether the execution range of the process trajectory fragments exceeds the control boundary of the corresponding constraint segment to identify out-of-bounds content. The cycle time offsets, sequence misalignments, and out-of-bounds content are written into the corresponding deviation positions according to the process node order. Adjacent deviation positions with the same deviation direction and consecutive deviation types are then concatenated to generate an execution deviation sequence.

[0052] S4. Using the deviation chain tracing method, perform causal tracing analysis on the execution deviation sequence, obtain the process succession relationship, and trace back the deviation source level by level according to the process succession relationship to determine the initial position of the deviation and generate deviation source location information.

[0053] S4.1. Based on the execution deviation sequence, read the deviation node elements one by one, and connect the consecutive deviation records into a deviation transmission chain according to the order of execution time to generate deviation chain data.

[0054] It should be noted that, based on the execution deviation sequence, the deviation node elements are read item by item, arranged in chronological order of execution time, and adjacent deviation node elements are placed in the same concatenation position. First, it is checked whether the execution times corresponding to adjacent deviation node elements are consecutive, then it is checked whether the deviation types corresponding to adjacent deviation node elements can be inherited and extended, and then it is checked whether the deviation directions corresponding to adjacent deviation node elements are consistent or progressively change. Deviation records with consecutive execution times, inheritable deviation types, and consistent or progressively changing deviation directions are sequentially concatenated into a deviation transmission chain. Deviation records with broken execution times, uninheritable deviation types, or reversed deviation directions are used as the segment positions of the deviation transmission chain, and subsequent consecutive deviation records are re-concatenated according to the segment positions to generate deviation chain data.

[0055] S4.2. Using the deviation chain tracing method, read back adjacent process nodes one by one in reverse chronological order along the deviation chain data, compare the execution order, deviation direction changes, and deviation persistence position changes of adjacent process nodes, and organize the comparison content into node readback segments.

[0056] It should be noted that the deviation record is read item by item in reverse chronological order along the deviation chain data using the deviation chain tracing method. The deviation record with a later time position is taken as the readback starting point, and the deviation record with an adjacent time position and an earlier time position is taken as the readback object. The process node corresponding to the readback starting point and the process node corresponding to the readback object are placed in the same readback position. The sequential relationship of adjacent process nodes is checked according to the time mark in the deviation chain data. If the process node corresponding to the readback object is located before the process node corresponding to the readback starting point, and the end position of the process node corresponding to the readback object can be connected to the entry position of the process node corresponding to the readback starting point, then it is recorded that the execution order is valid. Continue to read the deviation direction change of adjacent process nodes. If the deviation direction of the process node corresponding to the readback starting point can be restored to the process node corresponding to the readback object along the reverse path, or the deviation direction of the process node corresponding to the readback object can be continuously extended to the process node corresponding to the readback starting point, then it is recorded that the deviation direction change can be backtracked. Next, read the continuous position change of the deviation of adjacent process nodes. If the continuous position of the deviation of the process node corresponding to the readback starting point is adjacent, coincident, or continuously moving along the process path with the continuous position of the deviation of the process node corresponding to the readback object, it is recorded as the continuous position change of deviation can be inherited. Finally, write the execution order before and after, the deviation direction change can be backtracked, and the continuous position change of deviation can be inherited into the same readback position, and arrange the comparison content of each readback position continuously in reverse time order to generate node readback fragments.

[0057] It should also be noted that the deviation chain tracing method first concatenates continuous deviation records in the order of execution time of the deviation sequence to form deviation chain data; then, it reads back adjacent process nodes in reverse chronological order along the deviation chain data to check whether the execution order, deviation direction changes, and deviation continuity position changes can be continuously inherited; the adjacent process nodes that can be continuously inherited are connected to form a deviation feedback path, and the process node that first appeared with a deviation and can continuously trigger subsequent deviations is searched level by level in the deviation feedback path, thereby realizing the deviation source location.

[0058] The change in deviation direction is extracted from the deviation records of adjacent process nodes in the execution deviation sequence. It mainly depends on factors such as whether the cycle position is advanced or delayed, whether the process sequence is moved forward or backward, and whether the execution range exceeds the boundary. It determines whether the deviation continues to extend in the same direction and expands progressively between the preceding and following processes, or whether a reverse jump occurs, thus forming a change in deviation direction.

[0059] The change in the position of deviation is extracted from the positional relationship of continuous occurrence of deviations in the execution deviation sequence. It mainly depends on the occurrence position, duration segment and continuation position of the deviation in adjacent process nodes to determine whether the deviation position is adjacent, coincident or moves continuously along the process path, thus forming the change in the position of deviation.

[0060] S4.3. Based on the node readback segment, filter adjacent process nodes that are sequentially connected, have continuously extended deviation directions, and whose deviation positions can be transmitted and connected. Use the upstream process node as the candidate source point and the downstream process node as the deviation receiving point, record the deviation back transmission path, merge the continuously connected adjacent process nodes, and generate the process receiving relationship.

[0061] It should be explained that, based on the node readback segments, the execution order, deviation direction changes, and deviation persistence position changes of adjacent process nodes are read item by item, and each group of adjacent process nodes is placed in the same filtering position; first, according to the execution order, it is checked whether the upstream process node is located before the downstream process node, and whether the completion position of the upstream process node can continue the entry position of the downstream process node; then, according to the deviation direction change, it is checked whether the deviation direction of the upstream process node is consistent with the deviation direction of the downstream process node, or whether a progressive expansion relationship can be formed from the upstream process node to the downstream process node; finally, according to the deviation persistence position change, it is checked whether the deviation persistence position of the upstream process node is consistent with the deviation persistence position of the downstream process node. The nodes are located adjacently, overlappingly, or continuously moving along the process path. Adjacent process nodes that can be inherited in the execution sequence, whose deviation direction can be continuously extended, and whose deviation position can be transmitted and connected are selected. Among the selected adjacent process nodes, the upstream process node in the preceding position is used as the candidate source point, and the downstream process node in the following position is used as the deviation receiving point. The position through which the deviation is transmitted from the deviation receiving point to the candidate source point is recorded in the reverse reading order of the node back reading segment, and each back transmission position is written into the deviation back transmission path. Finally, adjacent process nodes in the deviation back transmission path that have continuous execution sequence, continuous deviation direction change, and continuous deviation position change are merged in sequence to generate the process inheritance relationship.

[0062] S4.4. Based on the process succession relationship, read each candidate process node in reverse and perform joint analysis on the deviation tracing characteristics of each candidate process node to generate the source point pointing sequence.

[0063] It should be noted that, according to the process succession relationship, each candidate process node is read in reverse order along the deviation feedback path. Each candidate process node is placed in the traceability verification position, and the deviation occurrence time, deviation direction extension, deviation continuity position connection, and subsequent deviation succession number corresponding to each candidate process node are extracted from the node backread segment to form deviation traceability features. When performing joint analysis on deviation traceability features, the deviation occurrence time of each candidate process node is first compared, and the candidate process node with the earlier occurrence time and still able to connect to the subsequent deviation succession point is taken as the priority target. Then, the deviation direction extension is checked, and the candidate process node whose deviation direction can continuously point to the downstream process node is retained. The deviation continuity position connection is further checked, and the candidate process node whose deviation continuity position can be transmitted to the deviation succession point along the process succession relationship is retained. Finally, the subsequent deviation succession number corresponding to the retained candidate process nodes is counted, and each candidate process node is arranged in the order of earlier occurrence time, continuous direction extension, complete position connection, and more succession number to generate the source point pointing sequence.

[0064] The expression for the joint analysis of the deviation tracing characteristics of each candidate process node is as follows: ; in, For the first The source point pointing strength of each candidate process node; For the first Each candidate process node can accept the number of times it can accept subsequent deviation records; For the first The continuity of the deviation direction of each candidate process node is determined by reading the deviation direction from the candidate process node to the subsequent deviation connection point along the process connection relationship, and judging whether the deviation direction is consistent, whether it extends progressively, and whether there is a reverse jump. For the first The degree of continuity of deviation position of each candidate process node is obtained by reading the position of deviation occurrence between the candidate process node and the subsequent deviation connection point along the process connection relationship, and judging whether the deviation position is adjacent, coincident, or continuously moved along the process path. For the first The interval between the time when the deviation occurs in each candidate process node and the time when the earliest deviation occurs; This refers to the sequence number of the candidate process node.

[0065] S4.5 Based on the source point pointing sequence, arrange the candidate process nodes according to the strength of the source point pointing, and check step by step along the process succession relationship to see if there are still deviation records of the same type and in the same direction before the candidate process nodes, and generate a reverse backtracking sequence.

[0066] It should be noted that, based on the source point pointing sequence, candidate process nodes are read item by item. These nodes are then ranked according to the strength of the source point pointing, based on the order of the deviation occurrence time, the continuity of the deviation direction, the complete connection of the deviation's duration, and the number of subsequent deviations. Starting from the first-ranked candidate process node, the process is read back level by level along the process connection relationship, placing the read-backed ...

[0067] S4.6. Select the process node that first shows a deviation and can continuously trigger subsequent deviations based on the reverse backtracking sequence as the initial position of the deviation. Bind the initial position, the corresponding deviation type, the corresponding deviation direction and the corresponding constraint segment to generate deviation source location information.

[0068] It should be noted that, based on the reverse backtracking sequence, each process node and its corresponding deviation record are read item by item. The process nodes are arranged from earliest to latest according to the time of deviation occurrence, and each process node is checked to see if the process node at the beginning of the sequence can be connected to the subsequent deviation connection point through the process succession relationship. If the deviation record of the process node at the beginning of the sequence can be continuously transmitted to the subsequent process node along the deviation direction, and the deviation continuity position can be connected with the deviation continuity position of the subsequent process node, then the process node is selected as the initial position of the deviation. The deviation type and corresponding deviation direction corresponding to the initial position are then read from the execution deviation sequence, and the constraint fragment corresponding to the initial position is read from the standard constraint structure. Finally, the initial position, the corresponding deviation type, the corresponding deviation direction, and the corresponding constraint fragment are bound according to the correspondence of the same process node to generate deviation source location information.

[0069] S5. Perform process optimization and matching on the deviation source location information, obtain the affected processes, perform timing correction and adjustment on the execution path of the affected processes, and generate an optimized execution plan.

[0070] S5.1 Extract deviation source elements from deviation source location information, and find process nodes that are connected to the initial position from the deviation source elements. Expand the connected process nodes according to the execution order to generate a deviation influence chain.

[0071] It should be explained that the initial position, corresponding deviation type, corresponding deviation direction, and corresponding constraint fragment are read from the deviation source location information. The initial position is used as the search starting point, and the process nodes are read one by one along the process succession sequence in the standard constraint structure. The read process nodes are placed in the same succession verification position as the initial position. First, it is checked whether the process node is located after the initial position. Then, it is checked whether the entry position of the process node can continue the initial position or the completion position of the previous process node. It is then checked whether the constraint fragment corresponding to the process node has a continuous interaction relationship with the corresponding constraint fragment. The process nodes that are located later, can be entered at the same position, and have continuous constraint fragments are identified as process nodes that have a succession relationship with the initial position. The succession process nodes are then expanded in sequence according to the execution order. Finally, the initial position, corresponding deviation type, corresponding deviation direction, and the expanded succession process nodes are continuously connected in series to generate a deviation influence chain.

[0072] S5.2. Check the connection status between each process node and the corresponding constraint segment according to the deviation influence chain, mark the process nodes that meet the deviation transmission conditions as affected processes, and generate affected processes.

[0073] It should be explained that, according to the deviation influence chain, each process node and its corresponding constraint segment are read item by item, and adjacent process nodes and their corresponding constraint segments are placed in the same acceptance and verification position; first, it is verified whether the deviation direction of the previous process node can be extended to the next process node, then it is verified whether the action boundary of the previous corresponding constraint segment can be connected to the action boundary of the next corresponding constraint segment, and then it is verified whether the operation content of the next process node is affected by the deviation record of the previous process node; process nodes whose deviation direction can be extended, whose action boundary can be connected, and whose operation content is affected by the deviation record are determined to meet the deviation transmission conditions, and process nodes that meet the deviation transmission conditions are marked as affected processes; process nodes whose deviation direction cannot be extended, whose action boundary is broken, or whose operation content is not affected by the deviation record are excluded from the deviation influence chain; finally, the marked affected processes are connected in series according to the execution order in the deviation influence chain to generate the affected processes.

[0074] S5.3 By performing timing correction and adjustment on the execution time and sequence of the affected processes, correction path information is obtained, and the correction path information and corresponding constraint segments are bound together to generate an optimized execution plan.

[0075] It should be explained that by reading the execution time and sequence of each affected process, affected processes with earlier execution times are moved to the start position allowed by the corresponding constraint segment, and affected processes with later execution times are moved to the finish position allowed by the corresponding constraint segment. The interval between adjacent affected processes is then corrected according to the receiving requirements. The process sequence is then checked to see if there are any premature, delayed, or cross-segment jumps between adjacent affected processes. Affected processes with premature sequences are moved back after the previous affected process, affected processes with delayed sequences are moved to the corresponding receiving position, and affected processes with cross-segment jumps are added to the missing intermediate receiving position. Finally, the corrected execution time, sequence, and receiving position are continuously organized into correction path information, and the correction path information is bound to the corresponding constraint segment item by item according to the same affected process to generate an optimized execution plan.

[0076] This embodiment also provides an intelligent manufacturing process optimization system based on group standards, including: The process mapping module is used to collect standard process information corresponding to the intelligent manufacturing production line, perform semantic mapping on the standard process information, and generate process mapping information. The constraint correction module is used to identify conflicting relationships and separate inconsistent constraints by comparing the execution conditions and control ranges of different standard constraints based on process mapping information through constraint conflict resolution methods, obtaining different standard constraints, correcting the constraint order of different standard constraints, and generating a standard constraint structure. The deviation identification module is used to perform real-time running status alignment and matching verification of the standard constraint structure, identify deviations from the standard constraint structure, and generate an execution deviation sequence. The deviation tracing module is used to perform causal tracing analysis on the execution deviation sequence through the deviation chain tracing method, obtain the process succession relationship, and trace back the deviation source level by level according to the process succession relationship to determine the initial position of the deviation and generate deviation source location information. The process optimization module is used to optimize and match the deviation source location information, obtain the affected processes, perform timing correction and adjustment on the execution path of the affected processes, and generate an optimized execution plan.

[0077] In summary, this invention improves multi-standard adaptability, accuracy of deviation source location, and targeted process optimization by: using a constraint conflict resolution method to transform multi-source standard constraints into an executable standard constraint structure, thereby completing the coordinated organization of standard semantics, execution conditions, and control scope; and using a deviation chain tracing method to reverse-read the execution deviation sequence and locate the initial position of the deviation, then matching the affected processes and performing timing correction adjustments.

[0078] It should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and are not intended to limit it. Although the present invention has been described in detail with reference to preferred embodiments, those skilled in the art should understand that modifications or equivalent substitutions can be made to the technical solutions of the present invention without departing from the spirit and scope of the technical solutions of the present invention, and all such modifications or substitutions should be covered within the scope of the claims of the present invention.

Claims

1. A method for optimizing intelligent manufacturing processes based on group standards, characterized in that, include: Collect standard process information corresponding to intelligent manufacturing production lines, perform semantic mapping on the standard process information, and generate process mapping information; By using the constraint conflict resolution method, the execution conditions and control scope of different standard constraints are compared based on the process mapping information, conflict relationships are identified and inconsistent constraints are separated, different standard constraints are obtained, constraint order correction is performed on different standard constraints, and a standard constraint structure is generated. The real-time operating status is read according to the production line cycle time. The real-time operating status is broken down into process trajectory segments and embedded into the corresponding constraint segments of the standard constraint structure. The constraint matching degree is checked segment by segment, deviations are identified, and an execution deviation sequence is generated. By using the deviation chain tracing method, causal tracing analysis is performed on the execution deviation sequence to obtain the process succession relationship. Based on the process succession relationship, the source of deviation is traced back level by level to determine the initial position of the deviation and generate deviation source location information. The process optimization and matching of deviation source location information is performed to obtain the affected processes, the execution path of the affected processes is adjusted by timing correction, and an optimized execution plan is generated. The steps for collecting standard process information corresponding to the intelligent manufacturing production line, semantically mapping the standard process information, and generating process mapping information are as follows: The process element information in the standard process information is broken down line by line. Semantic equivalence recognition is performed on the process element information to obtain semantic equivalence groups. The semantic equivalence groups are then uniformly replaced and re-concatenated to generate semantic standard expression data. Based on the semantic standard expression data, the process connection and constraint descriptions between the processes are compared item by item to obtain the process nodes and their corresponding relationships. The process nodes and their corresponding relationships are bound and mapped one by one to generate process mapping information. The specific steps for obtaining different standard constraints are as follows: Extract process constraint elements from process mapping information, expand and organize process constraint elements to obtain multi-standard constraints, merge the sources of multi-standard constraints, and generate constraint expansion sequence. The constraint expansion sequence is compared pairwise to obtain constraint difference features. The constraint difference features are combined with the differences in execution conditions and the overlap of control scope to determine conflicts and generate a conflict measurement sequence. By using the constraint conflict resolution method, conflict candidates in the conflict measurement sequence are read, and the constraint conflict features corresponding to each conflict candidate are checked one by one. Constraint pairs whose constraint conflict features meet the substantive conflict conditions are retained, and constraint pairs whose constraint conflict features only show object mismatch are eliminated, thus generating a set of conflict relationships. Based on the set of conflict relationships, the control scope, execution conditions and corresponding process nodes of each constraint pair are read. The constraint pair is split into a non-overlapping segment in the front, an overlapping segment in the middle and a non-overlapping segment in the back. The mutually exclusive execution conditions in the overlapping segment in the middle are extracted from the original constraint to generate heterogeneous constraint segments. Heterogeneous constraint fragments and constraints that do not conflict are uniformly merged to obtain a constraint integration sequence. The constraint integration sequence is then classified, sorted, and its order is checked to generate constraints of different standards. The specific steps for generating the execution deviation sequence are as follows: Read the real-time running status according to the production line cycle time, connect the continuous running content of the same process node in the cycle time sequence to obtain the cycle time running chain, identify the running breakpoint information according to the cycle time running chain, and divide it by the interruption point before and after to generate process trajectory segments. Read the process node and cycle time position based on the process trajectory fragment, insert the process trajectory fragment into the corresponding constraint segment of the standard constraint structure to obtain the trajectory embedding chain, check the cycle time offset, sequence misalignment and range out-of-bounds content segment by segment based on the trajectory embedding chain, and string them together according to the continuous deviation position to generate the execution deviation sequence.

2. The intelligent manufacturing process optimization method based on group standards as described in claim 1, characterized in that, The standard constraint structure is obtained by correcting the execution order of each constraint fragment according to different standard constraints, and by rearranging and filling the boundaries of constraints with coverage conflicts.

3. The intelligent manufacturing process optimization method based on group standards as described in claim 1, characterized in that, The method of deviation chain tracing is used to perform causal tracing analysis on the execution deviation sequence to obtain the process succession relationship. The specific steps are as follows: Based on the execution deviation sequence, the deviation node elements are read item by item, and the consecutive deviation records are linked together into a deviation transmission chain according to the order of execution time to generate deviation chain data. Using the deviation chain tracing method, adjacent process nodes are read back item by item in reverse chronological order of the deviation chain data. The execution order, deviation direction changes, and deviation persistence position changes of adjacent process nodes are compared, and the comparison content is organized into node readback segments. Based on the node readback segments, adjacent process nodes that are sequentially connected, have continuously extending deviation directions, and whose deviation positions can be transmitted and connected are selected. The upstream process node is used as the candidate source point and the downstream process node is used as the deviation receiving point. The deviation back transmission path is recorded, and the continuously connected adjacent process nodes are merged to generate the process receiving relationship.

4. The intelligent manufacturing process optimization method based on group standards as described in claim 3, characterized in that, The process of tracing back the source of deviation step by step according to the process succession relationship to determine the initial location of the deviation and generate deviation source location information is as follows: Based on the process succession relationship, each candidate process node is read in reverse, and the deviation tracing characteristics of each candidate process node are jointly analyzed to generate a source point pointing sequence. Based on the source point pointing sequence, candidate process nodes are arranged according to the strength of the source point pointing, and the process succession relationship is checked step by step to check whether there are still deviation records of the same type and in the same direction before the candidate process nodes, thus generating a reverse backtracking sequence. The process node that first shows a deviation and can continuously trigger subsequent deviations is selected as the initial position of the deviation. The initial position, the corresponding deviation type, the corresponding deviation direction, and the corresponding constraint segment are bound together to generate deviation source location information.

5. The intelligent manufacturing process optimization method based on group standards as described in claim 4, characterized in that, The specific steps for optimizing and matching the deviation source location information to obtain the affected processes are as follows: Extract deviation source elements from deviation source location information, and find process nodes that are connected to the initial position from the deviation source elements. Expand the connected process nodes according to the execution order to generate a deviation influence chain. Based on the deviation impact chain, check the connection status between each process node and the corresponding constraint segment, mark the process nodes that meet the deviation propagation conditions as affected processes, and generate affected processes.

6. The intelligent manufacturing process optimization method based on group standards as described in claim 1, characterized in that, The optimized execution plan is generated by performing time-series correction and adjustment on the execution time and sequence of the affected processes, obtaining correction path information, and binding the correction path information with the corresponding constraint fragments.

7. A smart manufacturing process optimization system based on group standards, based on the smart manufacturing process optimization method based on group standards according to any one of claims 1 to 6, characterized in that, include: The process mapping module is used to collect standard process information corresponding to the intelligent manufacturing production line, perform semantic mapping on the standard process information, and generate process mapping information. The constraint correction module is used to identify conflicting relationships and separate inconsistent constraints by comparing the execution conditions and control ranges of different standard constraints based on process mapping information through constraint conflict resolution methods, obtaining different standard constraints, correcting the constraint order of different standard constraints, and generating a standard constraint structure. The deviation identification module is used to perform real-time running status alignment and matching verification of the standard constraint structure, identify deviations from the standard constraint structure, and generate an execution deviation sequence. The deviation tracing module is used to perform causal tracing analysis on the execution deviation sequence through the deviation chain tracing method, obtain the process succession relationship, and trace back the deviation source level by level according to the process succession relationship to determine the initial position of the deviation and generate deviation source location information. The process optimization module is used to optimize and match the deviation source location information, obtain the affected processes, perform timing correction and adjustment on the execution path of the affected processes, and generate an optimized execution plan.

Images