An agent-based automatic management and control method for mining engineering design process

Abstract

The application discloses a kind of based on agent's mining engineering design process automatic control method, it is related to process management technical field, including: the source data of mining engineering project is analyzed and collated, forms project basic data;According to project basic data, project task decomposition and execution resource matching are carried out, and project execution plan is formed;Project execution plan is verified with condition, and calls text processing agent to carry out task document flow transfer processing, and forms project text achievement;According to project execution plan and project text achievement, calls drawing processing agent to carry out drawing flow transfer processing, and forms project drawing achievement;Project basic data, project text achievement and project drawing achievement are audited and monitored with state in whole process, and knowledge is stored in the processing, and forms project closed-loop knowledge base.The application is by to the task of mining engineering project constraint aggregation cutting, process constraint projection and resource interlocking resolution, improves the accuracy of task decomposition and the rationality of resource matching.

Description

Technical Field

[0001] This invention relates to the field of process management technology, and in particular to an automatic control method for mining engineering design process based on intelligent agents. Background Technology

[0002] With the development of digital construction and intelligent design platforms in mining enterprises, mining engineering design has gradually shifted from a discrete operation mode driven by human experience to a digital operation mode characterized by multi-source data integration, reuse of standard constraints, and collaborative output of design results. Existing related technologies typically revolve around aspects such as design data management, standard retrieval, text-assisted compilation, drawing-assisted generation, and review and verification. Some solutions have already achieved unified collection of design commissioning data, geological exploration data, experimental data, equipment data, and historical cases. On this basis, knowledge bases, rule bases, large models, or intelligent agent technologies are introduced to assist in the processing of feasibility study documents, design specifications, attached drawings, and review comments.

[0003] However, existing technologies still have significant shortcomings. First, current solutions typically focus on processing single documents or drawings, leading to issues such as insufficient consistency between textual and drawing deliverables, loose connections between different processes, and insufficient reverse constraints from the review process on preceding tasks. Second, while existing solutions can achieve knowledge storage or design deliverable archiving, they mostly remain at the result-level storage level, limiting the depth of application of intelligent mining engineering design systems in project-level process control and cross-project experience inheritance. Summary of the Invention

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

[0005] Therefore, this invention provides an automatic control method for the design process of mining engineering based on intelligent agents to solve the problems of insufficient collaboration between textual and drawing results and the difficulty in supporting full-process control after knowledge is stored in the database.

[0006] To solve the above-mentioned technical problems, the present invention provides the following technical solution: This invention provides an automated management and control method for the design process of mining engineering based on intelligent agents. The method includes: parsing and organizing source data of the mining engineering project to form basic project data; decomposing project tasks and matching execution resources based on the basic project data to form a project execution plan; performing condition verification on the project execution plan and calling a text processing intelligent agent to process task documents, forming project text deliverables; calling a drawing processing intelligent agent to process drawings based on the project execution plan and project text deliverables, forming project drawing deliverables; and conducting full-process review and status monitoring of the basic project data, project text deliverables, and project drawing deliverables, and performing knowledge entry processing to form a closed-loop project knowledge base.

[0007] As a preferred embodiment of the automatic control method for the mining engineering design process based on intelligent agents described in this invention, the specific steps for parsing and organizing the source data of the mining engineering project to form basic project data are as follows: Perform content identification and constraint transformation on the source data of mining engineering projects, extract task restriction information and generate project task control baselines; The project task control baseline is adjusted, linked, and organized to form the project's basic data. The project's basic data includes task constraint information, specification constraint information, equipment boundary information, and drawing acceptance information.

[0008] As a preferred embodiment of the automatic control method for mining engineering design process based on intelligent agents described in this invention, the specific steps of decomposing project tasks and matching execution resources based on project basic data are as follows: Perform constraint aggregation and segmentation processing on the task constraint information, construct a task constraint coupling graph, and perform process constraint projection on the specification constraint information based on the task constraint coupling graph to generate process execution conditions; The process execution conditions are processed to construct the process control state, and boundary mapping is performed in combination with equipment boundary information to generate resource constraint domains; Based on the process control status and resource constraint domain, perform resource interlock resolution and obtain execution resource configuration information.

[0009] As a preferred embodiment of the automatic control method for mining engineering design process based on intelligent agents described in this invention, the project execution plan is formed by processing the transmission of process status and solidifying the plan based on process control status, execution resource configuration information and process execution conditions.

[0010] As a preferred embodiment of the automatic control method for the mining engineering design process based on intelligent agents described in this invention, the condition verification of the project execution plan refers to locking task restriction information and specification constraint information based on the project execution plan through a constraint cross-locking algorithm and performing flow qualification determination to form text access verification information.

[0011] As a preferred embodiment of the automatic control method for mining engineering design process based on intelligent agents described in this invention, the specific steps of calling a text processing intelligent agent to perform task document flow processing and form project text results are as follows: Based on the text access verification information, the text processing agent is invoked to perform targeted attachment of the project's basic data, forming a task document draft. A process path locking and push algorithm is used to perform path locking and flow of task document drafts to obtain text flow information; The text flow information is organized into slots to form project text deliverables.

[0012] As a preferred embodiment of the automatic control method for mining engineering design process based on intelligent agents described in this invention, the specific steps of calling a drawing processing intelligent agent to perform drawing flow processing according to the project execution plan and project text results are as follows: Based on the project execution plan and project text deliverables, the project text deliverables are mapped to text-drawing constraints to obtain drawing access verification information; Based on the drawing access verification information, combined with the equipment boundary information and the drawing acceptance information, the acceptance boundary is locked to form the drawing processing draft. Based on the original drawing, the drawing processing agent is invoked to push the planned execution process flow, thus generating drawing flow information.

[0013] As a preferred embodiment of the automatic control method for the mining engineering design process based on intelligent agents described in this invention, the project drawing results are formed by performing element slot convergence organization on the drawing flow information.

[0014] As a preferred embodiment of the intelligent agent-based automatic control method for mining engineering design process described in this invention, the specific steps for conducting full-process review and status monitoring of project basic data, project text deliverables, and project drawing deliverables are as follows: The result constraint back-checking algorithm is used to back-check the project text results and project drawing results against the project basic data to obtain result review information; Extract the design process flow relationship in the project execution plan, track the current acceptance status of project text deliverables and project drawing deliverables, and obtain deliverable status information.

[0015] As a preferred embodiment of the automatic control method for mining engineering design process based on intelligent agents described in this invention, the project closed-loop knowledge base is formed by merging result review information and result status information and performing associated indexing to enter the database.

[0016] The beneficial effects of this invention are as follows: By constraining and aggregating mining project tasks, projecting process constraints, resolving resource interlocks, and transferring status, the originally scattered basic project data can be transformed into a constrained project execution plan, thereby improving the accuracy of task decomposition and the rationality of resource matching; by implementing constraint cross-locking, targeted attachment, process path locking and pushing, and slot convergence organization on text tasks, the text content flows in an orderly manner along the predetermined process path, effectively reducing the probability of subsequent review, modification, and rework. 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 an automatic control method for the design process of mining engineering based on intelligent agents.

[0019] Figure 2 A flowchart for the execution plan.

[0020] Figure 3 This is a flowchart of text flow.

[0021] Figure 4 This is a flowchart for the results review.

[0022] Figure 5 This is a comparison chart of waiting times for different processes.

[0023] Figure 6 This is a comparison chart of text flow review. Detailed Implementation

[0024] 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.

[0025] 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.

[0026] 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.

[0027] Reference Figures 1-6 As one embodiment of the present invention, this embodiment provides an automatic control method for mining engineering design process based on intelligent agents, including the following steps: S1: Analyze and organize the source data of mining engineering projects to form basic project data.

[0028] S1.1: Perform content identification and constraint transformation on the source data of mining engineering projects, extract task restriction information, and generate project task control baselines.

[0029] Specifically, the design commission, geological survey report, test report, historical design case, standard documents, equipment documents and existing drawings in the source data of mining engineering projects are read separately, and the stage requirements, professional requirements, delivery requirements, sequence requirements, acceptance requirements and boundary requirements are extracted and broken down into independent semantic entries to form source data identification entries.

[0030] For each item in the source data, the restriction attributes are identified, and semantic items containing restrictions on project task initiation, completion, acceptance, delivery, and scope of application are identified as task restriction information.

[0031] Organize the task restriction information that applies to the same project task into a single restriction record, and then register the task restriction information that can be passed on and the task restriction information that only applies to the current project task according to the acceptance requirements.

[0032] All constraint records were collected and linked together according to the project phase sequence, project task sequence, and the transmission order determined by the acceptance requirements, so that all constraint records corresponding to the same project task are centrally formed to form the project task control baseline.

[0033] It should be noted that task constraint information refers to the constraints identified from the source data of mining engineering projects (e.g., limiting the conditions for project task initiation, the conditions for project task completion, the direction of project task acceptance, the scope of project task delivery, and the scope of project task application). Task constraint information is used to generate the project task control baseline and serves as the basis for subsequent task constraint coupling diagram construction, process execution condition generation, and text flow qualification determination.

[0034] S1.2: The project task control baseline is adjusted, linked, and organized to form the basic project data.

[0035] Specifically, all constraint records in the project task control baseline are read in the order of project phases. The constraint records corresponding to the previous project task are compared with the constraint records corresponding to the next project task in the order of project tasks to identify the continuity, interruption and conflict relationships between the previous project tasks and the next project tasks in terms of task constraint information.

[0036] Register the task restriction information from the previous project task that can continue to be applied to the next project task as the inherited content; register the task restriction information from the previous project task that is no longer applicable to the next project task as the terminated content; register the task restriction information from the previous project task that is inconsistent in content or in the application boundary between the previous project task and the next project task as the corrected content; and mark the positions of the inherited content, terminated content and corrected content according to the project task order.

[0037] Add the accepted content to the restriction record corresponding to the next project task, retain the terminated content in the restriction record corresponding to the previous project task, and replace the corrected content in the corresponding project task position; organize the task restriction information, specification constraint information, equipment boundary information, and drawing acceptance information that have been completed into the corresponding project task name, and merge and register the task restriction information, specification constraint information, equipment boundary information, and drawing acceptance information according to the project task name and project stage name to obtain the basic project data.

[0038] S1.3: Project basic data includes task constraint information, specification constraint information, equipment boundary information, and drawing acceptance information.

[0039] Specifically, after completing the acceptance, correction, and association processing, all restriction records are collected item by item according to the project task name. Then, the content limiting the project task initiation conditions, project task completion conditions, project task acceptance direction, project task delivery scope, and project task applicable scope are extracted from all restriction records and organized into task restriction information.

[0040] Extract the content that defines the scope of application of the standard, the boundary of standard implementation, and the corresponding relationship of the standard from all the restriction records, and organize them into standard constraint information.

[0041] Extract the information on the scope of use of the equipment, the boundaries of the equipment's capabilities, and the scope of the equipment's configuration from all the restriction records, and organize them into equipment boundary information.

[0042] Extract the information on the connection between drawings, the continuity of drawings, and the order of drawing succession from all the restriction records, and organize it into drawing succession information.

[0043] The task restriction information, specification constraint information, equipment boundary information, and drawing assignment information are merged and registered according to the project task name and project phase name to obtain the basic project data.

[0044] S2: Based on the project's basic data, decompose project tasks and match execution resources to form a project execution plan.

[0045] S2.1: Perform constraint aggregation and segmentation processing on the task constraint information, construct a task constraint coupling graph, and perform process constraint projection on the specification constraint information according to the task constraint coupling graph to generate process execution conditions.

[0046] Specifically, all task restriction information in the project basic data is read in the order of project phases, and the project task initiation conditions, project task completion conditions, project task undertaking direction, project task delivery scope, and project task applicable scope are extracted from each task restriction information.

[0047] Group task restriction information that corresponds to the project task initiation conditions and the project task completion conditions into the same task restriction group; group task restriction information that has the same project task acceptance direction and consecutive project task delivery scope into the same task restriction group; and group task restriction information that has overlapping project task application scope into adjacent task restriction groups.

[0048] According to the project phase sequence and project task sequence, the succession relationship between the previous task constraint group and the next task constraint group is registered item by item. Task constraint groups whose applicable scope of project tasks does not overlap and whose project task delivery scope is independent are registered as parallel relationships; task constraint groups whose applicable scope of project tasks overlap and whose project task delivery scope is mutually restrictive are registered as conflict relationships; and all the registered relationships are connected in series according to the task constraint group name to form a task constraint coupling diagram.

[0049] Based on the task constraint coupling diagram, the standard constraint information in the basic project data is retrieved item by item. Standard constraint information with consistent application scope, consistent execution boundary, and consistent correspondence is projected onto the corresponding task constraint group. This makes each task constraint group form standard constraint content corresponding to the project task initiation conditions, project task completion conditions, project task acceptance direction, project task delivery scope, and project task application scope, and generates process execution conditions.

[0050] It should be noted that the task constraint coupling diagram is a relational graph structure formed by dividing the task constraint information in the project basic data according to the project stage sequence, project task sequence, and task succession relationship. It is used to represent the succession relationship, parallel relationship, and conflict relationship between different project tasks. The task constraint coupling diagram can transform the scattered task constraint information into an association basis that can be used for process constraint projection and process execution condition generation.

[0051] S2.2: Perform process control state construction processing on the process execution conditions, and perform boundary mapping in combination with equipment boundary information to generate resource constraint domains.

[0052] Specifically, all process execution conditions under the same task restriction group name are organized into the same process control record in chronological order; the allowed entry range of the process is determined according to the applicable scope of the specifications in the process control record; the allowed continuation range of the process is determined according to the execution boundary of the specifications in the process control record; and the allowed acceptance range of the process is determined according to the corresponding relationship of the specifications in the process control record, thus forming the process control status. The process control status includes the allowed entry range of the process, the allowed continuation range of the process, and the allowed acceptance range of the process.

[0053] Read the equipment boundary information in the project's basic data item by item according to the task restriction group name, match the equipment usage scope with the process's allowed entry scope, match the equipment capability boundary with the process's allowed continuation scope, and match the equipment configuration scope with the process's allowed acceptance scope. Equipment boundary information whose equipment usage scope falls within the process's allowed entry scope, equipment capability boundary falls within the process's allowed continuation scope, and equipment configuration scope falls within the process's allowed acceptance scope is determined to be consistent. Equipment boundary information that does not fall within the corresponding scope is determined to be inconsistent.

[0054] Register the corresponding consistent equipment boundary information under the corresponding process control status name, exclude inconsistent equipment boundary information from the corresponding process control status, and merge the registered process control statuses according to the project stage sequence and project task sequence to generate resource constraint domains.

[0055] It should be noted that a resource constraint domain refers to the set of available resource ranges formed by the correspondence between process control status and equipment boundary information. It is used to limit which equipment usage range, equipment capacity boundary, and equipment configuration range the current process can occupy. The resource constraint domain is essentially the resource boundary content corresponding to the current process. It does not refer to all resources in general, but only represents the information range that is consistent with the current process control status and can enter the subsequent resource interlock resolution and execution resource configuration.

[0056] S2.3: Based on the process control status and resource constraint domain, perform resource interlock resolution to obtain execution resource configuration information.

[0057] Specifically, the process control status under the same task constraint group name is matched with the resource constraint domain item by item to form a process resource correspondence record; the process resource correspondence record is compared item by item, and the case where the process control status simultaneously corresponds to the same equipment usage range, the same equipment capacity boundary, or the same equipment configuration range is identified as a resource interlock record.

[0058] For each resource interlock record, the resource constraint fields corresponding to the process control status that have met the requirements of the process's allowed entry range, the process's allowed continuation range, and the process's allowed acceptance range have been completed with the previous process are retained as the resource configuration content for the current process. The resource constraint fields corresponding to the process control status that have not met the requirements of the process's allowed entry range, the process's allowed continuation range, or the process's allowed acceptance range have not been completed with the previous process are removed from the resource interlock record and converted into continuation occupancy content. This releases the simultaneous occupancy relationship of multiple process control statuses on the same resource constraint field.

[0059] Organize the current process resource configuration content and the subsequent occupied content into the corresponding process control status name to form execution resource configuration information.

[0060] S2.4: Based on the process control status, execution resource allocation information, and process execution conditions, process status transmission processing and plan solidification are performed to form a project execution plan.

[0061] Specifically, the process control status, execution resource configuration information, and process execution conditions under the same task restriction group name are matched one by one to form a process status correspondence record; the correspondence between the process status correspondence record of the previous process and the process status correspondence record of the next process is compared item by item according to the project task sequence to see the process's allowed entry range, process's allowed continuation range, and process's allowed acceptance range.

[0062] The following relationships are defined as follows: An entry connection is defined as the situation where the allowed entry range of the subsequent process status record corresponds to the allowed acceptance range of the process after the completion of the previous process status record; a continuation connection is defined as the situation where the allowed continuation range of the subsequent process status record is continuous with the allowed continuation range of the previous process status record; and an acceptance connection is defined as the situation where the allowed acceptance range of the subsequent process status record matches the transmitted content of the previous process status record.

[0063] The content in the previous process status record that has met the allowed continuation range of the process and whose current process resource configuration content has been fully occupied is transferred to the next process status record. The content in the previous process status record that has not yet met the allowed continuation range of the process is retained in the previous process status record, forming the process status transfer result.

[0064] The sequence of preceding and following processes, the order of resource usage, and the normative restrictions in the process execution conditions are uniformly registered according to the project stage sequence and the project task sequence in the process status transmission results, forming a project execution plan.

[0065] like Figure 5This section describes the changes in waiting times for processes during the project execution plan formation phase. In the control group, project task names, project phase names, and equipment registration information were extracted from the source data of the mining engineering project. A basic task progress table was formed according to the project phase sequence and project task sequence. Each task was directly mapped to the preset equipment usage scope, equipment capacity boundary, and equipment configuration range. When project tasks corresponded to the same equipment registration information, they were arranged in the order of registration, with the later project task remaining in a waiting state until the earlier project task was completed. The time difference between the moment each project task met the entry conditions and the moment it actually entered processing was recorded, and the average value was calculated. In the experimental group, task constraint information, specification constraint information, equipment boundary information, and drawing acceptance information were first extracted from the source data of the mining engineering project to form the basic project data. Task constraint information that corresponds to the project task initiation conditions and completion conditions was grouped into the same task constraint group to construct a task constraint coupling diagram. The scope of application of specifications, the execution boundaries of specifications, and the correspondence between specifications were projected onto the corresponding task constraint groups to form the process execution conditions. The equipment usage scope, equipment capacity boundaries, and equipment configuration scope were then matched item by item with the process's allowed entry scope, allowed continuation scope, and allowed acceptance scope. Equipment registration content that did not meet the correspondence was removed from the current process, while equipment registration content that met the correspondence was retained as the resource configuration content for the current process. Finally, the waiting time for each process was calculated. The figure shows that the experimental group's overall performance was lower than the control group, indicating a reduction in resource conflicts and delayed occupancy.

[0066] S3: Perform condition validation on the project execution plan and call the text processing agent to process the task documents and generate project text deliverables.

[0067] S3.1: Based on the project execution plan, the task restriction information and specification constraint information are locked through the constraint cross-locking algorithm and the qualification for circulation is determined to form text access verification information.

[0068] Specifically, the constraint cross-locking algorithm reads the sequence of preceding and following processes, resource usage order, and specification constraints in the project execution plan according to the project phase order and project task order. Then, it reads the task constraint information and specification constraint information in the project basic data according to the same project task name, and organizes the sequence of preceding and following processes, resource usage order, specification constraints, task constraint information, and specification constraint information into the same locking comparison grid.

[0069] Within each locking comparison grid, a constraint cross-locking algorithm is used to compare the sequence of preceding and following processes with the project task initiation conditions, project task completion conditions, and project task acceptance direction in the task constraint information item by item; the specification constraints are compared with the specification scope of application, specification execution boundaries, and specification correspondence in the specification constraint information item by item, and the flow matching degree is calculated, expressed as: ; in, For flow matching degree, For the number of sequential matches, To standardize the number of matches, This represents the number of sequence deviations. To standardize the number of deviations, To standardize the constraint inhibition coefficient.

[0070] It should be noted that the normative constraint suppression coefficient is obtained by mapping the degree of difference between the number of normative deviations and the number of normative matches, and its value range is [value missing]. .

[0071] Based on the workflow matching degree, the corresponding project task names are judged for workflow eligibility. Tasks meeting the workflow matching degree threshold are selected (obtained by verifying the sequence of preceding and following processes, resource usage order, and specification restrictions in the project execution plan against the task restriction information and specification constraint information in the project's basic data; the value range is:). The project task names are registered as approved content, while those that do not meet the flow matching threshold are registered as unapproved content. The registration is then merged and completed according to the project stage order and the project task order to form text access verification information.

[0072] It should be noted that the constraint cross-locking algorithm first reads the sequence of preceding and following processes, the order of resource usage, and the content of specification restrictions in the project execution plan. Then, it reads the task restriction information and specification constraint information in the project basic data and organizes them into a locking comparison grid according to the same project task name. Subsequently, it compares the sequence of preceding and following processes with the project task start conditions, project task completion conditions, and project task acceptance direction item by item. It also compares the content of specification restrictions with the scope of application of the specification, the execution boundary of the specification, and the correspondence of the specification item by item. Based on the comparison results, it calculates the flow matching degree and judges the flow eligibility of the corresponding project task name according to the flow matching degree, forming text access verification information.

[0073] S3.2: Based on the text access verification information, call the text processing agent to perform targeted attachment of the project's basic data to form the task document draft.

[0074] Specifically, the approved content in the text access verification information is read according to the project stage sequence and project task sequence. The task restriction information, specification constraint information, equipment boundary information and drawing acceptance information in the project basic data are read according to the same project task name. The approved content and the task restriction information, specification constraint information, equipment boundary information and drawing acceptance information corresponding to the same project task are organized into the corresponding grid.

[0075] The text processing agent is invoked to classify the task constraint information into the project task description for each attached grid, the specification constraint information into the specification basis, the equipment boundary information into the equipment condition, and the drawing acceptance information into the drawing association. All the content that has been attached under the same project task name is concatenated into continuous document content according to the project stage order and the project task order, and all continuous document content is registered separately according to the project task name to form the task document draft.

[0076] It should be noted that the text processing agent is used to receive text-related content from the project's basic data, and in the subsequent path-locking process, it calls the sequence of preceding and following procedures, resource occupancy order, and specification restrictions in the project execution plan. It performs targeted attachment, path-locking, and slot-based organization on the task document drafts. The text processing agent can organize and arrange the project task descriptions, specification basis, equipment conditions, and drawing-related content according to the project task sequence, thereby forming project text deliverables corresponding to the project execution plan.

[0077] S3.3: The process path locking and push algorithm is used to perform path locking and flow of the task document draft and obtain text flow information.

[0078] Specifically, the continuous document content corresponding to each project task name is organized into a path-corresponding grid along with the preceding and following process sequences and resource usage sequences of the same project task. Within each path-corresponding grid, the content corresponding to the preceding process in the preceding and following process sequences is determined as the flow starting point, and the content corresponding to the following process in the preceding and following process sequences is determined as the flow ending point. The flow order of the continuous document content is determined according to the resource usage sequence.

[0079] The process path locking push algorithm is used to push the continuous document content at the starting point of the flow to the end point of the flow item by item along the sequence of the preceding and following processes. The continuous document content that has not reached the resource occupation order is retained in the current position, and the continuous document content that has reached the resource occupation order is transferred to the continuous document content corresponding to the next project task.

[0080] After the push is completed, all continuous document content will be re-registered according to the project stage order and project task order, and the document content corresponding to each project task name will be organized into text flow information.

[0081] It should be noted that the process path locking push algorithm is a processing method used to pass the content of consecutive documents in the task document draft backward along the sequence of previous and subsequent processes in the project execution plan. The process path locking push algorithm can determine the timing of content push based on the resource occupancy order, so that the content of consecutive documents that have reached the resource occupancy order enters the corresponding content of the next project task, and the content of consecutive documents that have not reached the resource occupancy order is retained in the corresponding content of the current project task, thereby ensuring that the text flow process is consistent with the project execution plan.

[0082] S3.4: Organize the text flow information into slots to form project text deliverables.

[0083] Specifically, the system reads all the document content after the transfer in the text transfer information according to the project stage order and the project task order. It then groups the document content of the same project task name into the same document slot and groups the document content of the previous project task name that has been transferred to the next project task name into the document slot corresponding to the next project task name.

[0084] For each document slot, the content of the transferred document is corrected according to the sequence of preceding and following procedures and the order of resource usage in the project execution plan. The transferred document content that is consistent with the sequence of preceding and following procedures in the project execution plan and has reached the resource usage order is retained in the current document slot, while the transferred document content that is inconsistent with the sequence of preceding and following procedures in the project execution plan or has not yet reached the resource usage order is removed from the current document slot.

[0085] After the documents are transferred and stored in the same document slot, the content is divided into sections and merged according to the project task description, standard basis, equipment conditions and drawing-related content. Then, all the content after being divided and merged is continuously connected according to the project stage order and project task order. The content of the continuously connected document slots is registered according to the project task name to form the project text deliverables.

[0086] like Figure 6The diagram illustrates the changes in the flow and review effectiveness of project text deliverables. In the control group, based on the basic task progress table, the project task description, standard basis, equipment conditions, and drawing-related content were assembled into a task document according to the order of project task names and passed forward in the order of preceding and following processes. During the transfer, the task restriction information and standard constraint information corresponding to the current project task were not checked for satisfaction, nor was the resource occupancy order checked. Therefore, after the document arrived at the project task location, the consistency between the project task description and task restriction information, the standard basis and standard constraint information, and the equipment conditions and equipment boundary information were checked item by item. Based on this, the text flow pass rate, path consistency rate, deliverable constraint matching degree, and first-round review pass rate were statistically analyzed. In the experimental group, the sequence of preceding and following processes, resource occupancy order, and specification restrictions in the project execution plan were first read. These were then compared with the task restriction information and specification constraints corresponding to the same-named project task, forming a locking comparison grid. The sequence of preceding and following processes was compared item by item with the project task initiation conditions, project task completion conditions, and project task assignment direction. The specification restrictions were also compared item by item with the specification's scope of application, specification execution boundaries, and specification correspondence. Project tasks meeting the comparison results were included in the text access verification information. The content was then linked with the task restriction information, specification constraints, equipment boundary information, and drawing assignment information to form the task document draft. This draft was only pushed to the next project task when the resource occupancy order was reached, and then assigned to the corresponding document slot according to the project task name. The experimental group outperformed the control group in all indicators, indicating more orderly text flow and less rework during review.

[0087] S4: Based on the project execution plan and project text deliverables, call the drawing processing agent to process the drawings and generate project drawing deliverables.

[0088] S4.1: Based on the project execution plan and project text deliverables, perform text-drawing constraint mapping on the project text deliverables to obtain drawing access verification information.

[0089] Specifically, based on the project phase sequence and project task sequence, the sequential order of preceding and following processes, resource usage order, and specification restrictions in the project execution plan are read. Then, the project task descriptions, specification basis, equipment conditions, and drawing associations in the project text deliverables are read according to the same project task name, and organized into a corresponding mapped grid.

[0090] Map the project task description to process succession constraints, the standard basis to standard constraints, the equipment conditions to equipment constraints, and the drawing associations to drawing succession constraints. Then, map the process succession constraints to the sequence of preceding and following processes, the standard constraints to the standard restrictions, the equipment constraints to the resource occupancy sequence, and the drawing succession constraints to the connection positions of preceding and following processes.

[0091] Register the corresponding valid content as approved content and the corresponding invalid content as unapproved content, and complete the merging and registration according to the project stage sequence and project task sequence to form drawing access verification information.

[0092] S4.2: Based on the drawing access verification information, combined with the equipment boundary information and drawing acceptance information, the acceptance boundary is locked to form the drawing processing draft.

[0093] Specifically, the approved content in the drawing access verification information is read according to the project stage sequence and project task sequence, and the equipment boundary information and drawing acceptance information in the project basic data are read according to the same project task name; the approved content, equipment boundary information and drawing acceptance information are organized into a boundary locking grid.

[0094] Within each boundary-locked grid, the equipment usage range, equipment capability boundary, and equipment configuration range in the equipment boundary information are matched one by one with the equipment constraints in the passing content. The drawing connection relationship, drawing inheritance relationship, and drawing acceptance order in the drawing acceptance information are matched one by one with the drawing acceptance constraints in the passing content. The content where both the equipment constraint and the drawing acceptance constraint are matched is retained as locked content, while the content where the equipment constraint or the drawing acceptance constraint is not matched is excluded from the locked content.

[0095] The locked content is consolidated according to the project phase sequence and the project task sequence, and then registered separately according to the project task name to form the drawing processing draft.

[0096] S4.3: Based on the drawing processing draft, call the drawing processing intelligent agent to push the planned execution process flow, forming drawing flow information.

[0097] Specifically, based on the drawing processing draft, the locked content corresponding to each project task name is organized into a push corresponding grid with the preceding and following process sequence and resource occupation sequence of the same project task; within each push corresponding grid, the content corresponding to the preceding process in the preceding and following process sequence is determined as the push start point, and the content corresponding to the following process in the preceding and following process sequence is determined as the push end point, and then the push order of the locked content is determined according to the resource occupation sequence.

[0098] The drawing processing agent is invoked to push the locked content at the starting point to the end point in the order of the preceding and following processes. Locked content whose resource occupation order has been reached is transferred to the content corresponding to the next project task name, while locked content whose resource occupation order has not been reached is retained in the content corresponding to the current project task name.

[0099] All locked content after push notifications are re-registered according to project phase and task order. The push notification drawing content corresponding to each project task name is organized to form drawing flow information.

[0100] It should be noted that the drawing processing agent takes the sequence of preceding and following processes, resource occupancy order, and specification restrictions in the project execution plan as input for flow constraints, and the project task description, specification basis, equipment conditions, and drawing association content in the project text deliverables as input for drawing constraints; it takes the equipment boundary information and drawing acceptance information in the project basic data as input for boundary basis; and it takes the drawing access verification information as input for entry conditions, and establishes a corresponding relationship according to the project task name. The working principle of the drawing processing agent is as follows: by locking the acceptance boundary through the corresponding equipment boundary information, drawing acceptance information, and project text deliverables, a drawing processing draft is formed. Then, according to the sequence of preceding and following processes and resource occupancy order in the project execution plan, the execution plan execution process flow is pushed to the drawing processing draft, and the element slots are bundled and organized on the flowed drawing content to form the project drawing deliverables.

[0101] S4.4: Perform element slot convergence organization on the drawing flow information to form the project drawing deliverables.

[0102] Specifically, drawings belonging to the same project task name are assigned to the same element slot, and drawings belonging to the previous project task name and already transferred to the next project task name are assigned to the element slot corresponding to the next project task name.

[0103] For each element slot, the pushed drawing content is corrected according to the sequence of preceding and following processes and the order of resource usage in the project execution plan. Pushed drawing content that is consistent with the sequence of preceding and following processes in the project execution plan and has reached the resource usage order is retained in the current element slot, while pushed drawing content that is inconsistent with the sequence of preceding and following processes in the project execution plan or has not yet reached the resource usage order is moved out of the current element slot.

[0104] After the content of the pushed drawings is retained in the same element slot, it is divided into segments and merged according to the element receiving content, equipment layout content, boundary annotation content, and drawing association identification content. Then, all the segmented and merged element content is continuously connected according to the project stage sequence and project task sequence, and registered separately according to the project task name to form the project drawing deliverables.

[0105] It should be noted that project drawing deliverables refer to the collection of drawing content formed under the constraints of the project execution plan. This collection is formed by mapping the project task descriptions, specifications, equipment conditions, and drawing-related content in the project text deliverables through text-drawing constraint mapping, boundary locking, planned execution process flow and push, and element slot aggregation organization. Project drawing deliverables can reflect the process inheritance relationships, specification restriction relationships, equipment boundary relationships, and drawing inheritance relationships of the current project tasks.

[0106] S5: Conduct full-process review and status monitoring of project basic data, project text deliverables, and project drawing deliverables, and perform knowledge entry processing to form a closed-loop project knowledge base.

[0107] S5.1: The result constraint back-checking algorithm is used to back-check the project text results and project drawing results into the project basic data to obtain the result review information.

[0108] Specifically, the project task description is compared item by item with the task constraint information, and the number of matching tasks and the number of deviations are calculated; the standard basis content is compared item by item with the standard constraint information, and the number of matching standards and the number of deviations are calculated; the equipment condition content is compared item by item with the equipment boundary information, and the number of matching equipment items and the number of deviations are calculated; the drawing association content, the pushed drawing content, and the element acceptance content are compared item by item with the drawing acceptance information, and the number of matching drawings and the number of deviations are calculated.

[0109] The result constraint matching degree is calculated based on the number of task matches, specification matches, equipment matches, drawing matches, and the number of task deviations, specification deviations, equipment deviations, and drawing deviations. The threshold for meeting the result constraint matching degree (obtained by normalizing the item-by-item correspondence between project text deliverables and project drawing deliverables and project basic data, with a value range of [value missing]) is then set. Project task names that do not meet the requirements for matching the results constraints are registered as approved content, while project task names that do not meet the requirements for matching the results constraints are registered as deviation content.

[0110] The expression for calculating the constraint matching degree of the calculation result is: ; in, To constrain the matching degree of the results, Indicates the number of tasks matched. This represents the number of task deviations. To standardize the number of matches, To standardize the number of deviations, Match the number of devices. This represents the number of equipment deviations. Match the quantity to the drawings. This represents the amount of deviation from the drawing specifications.

[0111] The approved and deviated items will be merged and registered according to the project phase and task order to form the results review information.

[0112] It should be noted that the outcome constraint backcheck algorithm is a processing method used to match project text deliverables and project drawing deliverables back to the task constraint information, specification constraint information, equipment boundary information, and drawing acceptance information in the project basic data. The outcome constraint backcheck algorithm can determine the matching and deviation between project text deliverables and project drawing deliverables and project basic data through item-by-item comparison, thereby providing a basis for the formation of outcome review information.

[0113] S5.2: Extract the design process flow relationship in the project execution plan, track the current acceptance position of the project text deliverables and project drawing deliverables, and obtain the deliverable status information.

[0114] Specifically, the contents of the preceding and following processes are linked one by one to form a design process flow relationship. The contents of the project task description, the specifications, the equipment conditions, the drawing associations, the pushed drawings, and the elements are respectively mapped to the contents of the preceding and following processes in the design process flow relationship.

[0115] The project's textual and drawing deliverables are traced item by item along the design process flow, and their current position is checked according to the project task name. The content that has entered the next process is identified as the arrived content in the current receiving position, the content that is still in the previous process is identified as the untransmitted content in the current receiving position, and the content that has been assigned to the corresponding resource occupancy sequence and has completed the connection between the preceding and following processes is identified as the completed content in the current receiving position.

[0116] Arrived content, undelivered content, and completed content are grouped and registered according to the project phase sequence and project task sequence to obtain the result status information.

[0117] S5.3: Merge the results review information and results status information, and perform associated indexing to enter the database, forming a closed-loop knowledge base for the project.

[0118] Specifically, the approved content, content with audit deviations, content that has arrived, content that has not been delivered, and completed content are organized into merged content under the same project task name; the approved content and completed content in the merged content are assigned to the same project task index position, the content with audit deviations and the content that has not been delivered are assigned to the same project task index position, and the approved content and the content that has arrived are assigned to the same project task index position, forming associated index content under the same project task name.

[0119] All related index content is merged and registered according to the project phase order and project task order, and then stored in the database according to the project task name to form a closed-loop project knowledge base.

[0120] In summary, this invention improves the accuracy of task decomposition and the rationality of resource matching by: constraining and aggregating mining project tasks, projecting process constraints, resolving resource interlocks, and transferring status. This transforms the originally scattered basic project data into a constrained project execution plan. By implementing constraint cross-locking, targeted attachment, process path locking and pushing, and slot clustering for text tasks, the invention ensures that text content flows orderly along predetermined process paths, effectively reducing the probability of subsequent review, modification, and rework.

[0121] 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. An automatic control method for mining engineering design process based on intelligent agents, characterized in that, include: Step S1: Analyze and organize the source data of the mining project to form the basic data of the project; Step S2: Based on the project's basic data, decompose the project tasks and match execution resources to form a project execution plan; Step S3: Perform condition verification on the project execution plan and call the text processing agent to process the task document flow and form the project text output. Step S4: Based on the project execution plan and project text deliverables, invoke the drawing processing agent to process the drawings and generate project drawing deliverables. Step S5 involves conducting a full-process review and status monitoring of the project's basic data, textual deliverables, and drawing deliverables, and then performing knowledge entry processing to form a closed-loop project knowledge base.

2. The automatic control method for mining engineering design process based on intelligent agents as described in claim 1, characterized in that, The process of analyzing and organizing the source data of the mining project to form the project's basic data involves the following steps: Perform content identification and constraint transformation on the source data of mining engineering projects, extract task restriction information and generate project task control baselines; The project task control baseline is adjusted, linked, and organized to form the project's basic data. The project's basic data includes task constraint information, specification constraint information, equipment boundary information, and drawing acceptance information.

3. The automatic control method for mining engineering design process based on intelligent agents as described in claim 2, characterized in that, The specific steps for decomposing project tasks and matching execution resources based on project basic data are as follows: Perform constraint aggregation and segmentation processing on the task constraint information, construct a task constraint coupling graph, and perform process constraint projection on the specification constraint information based on the task constraint coupling graph to generate process execution conditions; The process execution conditions are processed to construct the process control state, and boundary mapping is performed in combination with equipment boundary information to generate resource constraint domains; Based on the process control status and resource constraint domain, perform resource interlock resolution and obtain execution resource configuration information.

4. The automatic control method for mining engineering design process based on intelligent agents as described in claim 3, characterized in that, The project execution plan is formed by processing the transmission of process status and solidifying the plan based on the process control status, execution resource allocation information, and process execution conditions.

5. The automatic control method for mining engineering design process based on intelligent agents as described in claim 4, characterized in that, The condition verification of the project execution plan refers to locking task restriction information and specification constraint information based on the project execution plan through a constraint cross-locking algorithm, and determining the eligibility for transfer to form text access verification information.

6. The automatic control method for mining engineering design process based on intelligent agents as described in claim 5, characterized in that, The specific steps for invoking the text processing agent to process the task document flow and generate the project text output are as follows: Based on the text access verification information, the text processing agent is invoked to perform targeted attachment of the project's basic data, forming a task document draft. A process path locking and push algorithm is used to perform path locking and flow of task document drafts to obtain text flow information; The text flow information is organized into slots to form project text deliverables.

7. The automatic control method for mining engineering design process based on intelligent agents as described in claim 6, characterized in that, The specific steps for invoking a drawing processing AI to handle drawing workflow based on the project execution plan and project text deliverables are as follows: Based on the project execution plan and project text deliverables, the project text deliverables are mapped to text-drawing constraints to obtain drawing access verification information; Based on the drawing access verification information, combined with the equipment boundary information and the drawing acceptance information, the acceptance boundary is locked to form the drawing processing draft. Based on the original drawing, the drawing processing agent is invoked to push the planned execution process flow, thus generating drawing flow information.

8. The automatic control method for mining engineering design process based on intelligent agents as described in claim 1, characterized in that, The project drawing deliverables are formed by organizing the drawing flow information into graphic element slots.

9. The automatic control method for mining engineering design process based on intelligent agents as described in claim 8, characterized in that, The specific steps for conducting full-process review and status monitoring of project basic data, project text deliverables, and project drawing deliverables are as follows: The result constraint back-checking algorithm is used to back-check the project text results and project drawing results back to the project basic data to obtain result review information; Extract the design process flow relationship in the project execution plan, track the current acceptance position of project text deliverables and project drawing deliverables, and obtain deliverable status information.

10. The automatic control method for mining engineering design process based on intelligent agents as described in claim 1, characterized in that, The project closed-loop knowledge base is formed by merging the result review information and result status information and performing associated indexing to store them in the database.

Images