CN-122022278-A - Intelligent distribution and AI acceptance method and system for maintenance work orders of sewage plants

CN122022278ACN 122022278 ACN122022278 ACN 122022278ACN-122022278-A

Abstract

The invention provides a method and a system for intelligent dispatch and AI acceptance of a maintenance work order of a sewage plant, which relate to the technical field of equipment maintenance, and the method comprises the following steps: constructing a maintenance element knowledge graph model which defines the association relationship among equipment types, maintenance contents, machine-readable acceptance criteria and maintenance periods; and constructing three elliptical assessment spaces including a skill compliance elliptical assessment space, a man-hour capacity elliptical assessment space and a response situation elliptical assessment space aiming at each maintenance work order in the maintenance plan. The invention can accurately match the maintenance resources, automatically complete maintenance acceptance, and improve the professionality, efficiency and quality of equipment maintenance.

Inventors

REN GUOHUI
WANG LI
WEI BIN
LIU XIAOMEI
LI KUO
SUN YAN
ZHOU MINGLIANG
WANG JUNJIE

Assignees

杭州北水云服科技有限公司

Dates

Publication Date: 20260512
Application Date: 20260109

Claims (9)

1. The utility model provides a sewage plant dimension insurance work order intelligence dispatch and AI acceptance method which characterized in that, the method includes: Constructing a maintenance element knowledge graph model which defines the association relationship among equipment types, maintenance contents, machine-readable acceptance criteria and maintenance periods; Automatically generating a maintenance plan for the specific equipment based on the maintenance period in the maintenance element knowledge graph model; Constructing three oval evaluation spaces aiming at each maintenance work order in a maintenance plan, namely constructing a skill compliance oval evaluation space by taking the skill requirement level of each maintenance work order, the skill level label of a candidate maintenance person and a preset skill level compatibility rule as rule anchor points, constructing a man-hour capacity oval evaluation space by taking the current residual available standard man-hour of the candidate maintenance person, the standard man-hour requirement of each maintenance work order and a preset daily man-hour saturation threshold value as capacity anchor points, and constructing a response situation oval evaluation space by taking the emergency level label of each maintenance work order, the real-time position information of the candidate maintenance person and the fixed position information of target equipment as situation anchor points; Calculating the relative position relation and intersection state of the three elliptical evaluation spaces to select a final maintenance person; Receiving maintenance process records uploaded by the distributed final maintenance personnel; inputting the maintenance process record and the corresponding machine-readable acceptance standard in the maintenance element knowledge graph model into an AI model, judging the compliance by the AI model and calculating the overall acceptance rate of the maintenance work order; and comparing the overall acceptance rate with a preset threshold value, and automatically judging the work order acceptance result to form closed-loop management.
2. The method for intelligently distributing and AI acceptance of a maintenance work order for a sewage plant according to claim 1, wherein constructing a maintenance element knowledge graph model, the maintenance element knowledge graph model defining an association relationship among a device type, maintenance content, machine-readable acceptance criteria and maintenance period, comprises: Generating a multi-layer element structure containing equipment types and standardized maintenance contents through unstructured knowledge in the structured definition maintenance field; Generating a machine-readable acceptance criterion corresponding to and visually interpreted by the AI model based on standardized guaranty content defined by the multi-layered element structure, wherein the machine-readable acceptance criterion is a structured text comprising a description of the device state visually identifiable feature; dynamically associating and regularly binding the multi-layer element structure with a machine-readable acceptance criterion to establish a forced logic driving association relationship from the equipment type to the maintenance content and then to the acceptance criterion and the maintenance period; After the establishment of the forced logic driving association relation is completed, in the operation process of the maintenance element knowledge graph model, iterative optimization is carried out on key parameters or standard descriptions in the model through a preset interface according to the acceptance result and feedback data generated by the maintenance element knowledge graph model so as to realize the continuous evolution of the maintenance element knowledge graph model.
3. The method for intelligently distributing and checking AI for a maintenance work order of a sewage plant according to claim 2, wherein automatically generating a maintenance plan for a specific device based on a maintenance period in a knowledge graph model of maintenance factors comprises: inquiring and acquiring the equipment type of the target equipment, the corresponding standardized maintenance content and the associated maintenance period from the maintenance element knowledge graph model; Based on the maintenance period and the current plan generation time point, automatically calculating and generating a plan execution time point corresponding to each maintenance content; And integrating and generating a structured maintenance plan according to the equipment type, the standardized maintenance content, the associated machine-readable acceptance standard, the standard working hours and the plan execution time point.
4. The intelligent dispatching and AI acceptance method for the maintenance work orders of the sewage plant according to claim 3, wherein skill requirement levels of the maintenance work orders, skill level labels of candidate maintenance persons and preset skill level compatibility rules are used as rule anchor points to construct a skill compliance oval assessment space, current residual available standard working hours of the candidate maintenance persons, standard working hour requirements of the maintenance work orders and preset daily working hour saturation threshold values are used as capacity anchor points to construct a working hour capacity oval assessment space, and urgency degree labels of the maintenance work orders, real-time position information of the candidate maintenance persons and fixed position information of target equipment are used as situation anchor points to construct a response situation oval assessment space, comprising: Initializing an evaluation environment in a virtual multidimensional decision coordinate system for a current maintenance work order to be distributed in a structured maintenance plan; Acquiring the skill requirement level of the maintenance work order, the skill level label of the candidate maintenance personnel and a preset compatibility rule as a group of rule anchor points, and constructing a skill compliance oval evaluation space in the evaluation environment; Acquiring the current residual available standard working hours of the candidate maintenance personnel, the standard working hour requirement of the maintenance work order and a preset daily saturation threshold value as a group of capacity anchor points, and constructing a working hour capacity ellipse evaluation space in the evaluation environment; And acquiring an emergency degree label of the maintenance work order, real-time position information of candidate maintenance personnel and fixed position information of target equipment to serve as a configuration potential anchor point, and constructing a response situation ellipse evaluation space in the evaluation environment.
5. The intelligent dispatch and AI acceptance method of a maintenance work order for a sewage plant of claim 4, wherein calculating the relative positional relationship and intersection state of three elliptical assessment spaces to each other to select a final maintenance person comprises: Carrying out standardized mapping processing on the constructed skill compliance oval evaluation space, the man-hour capacity oval evaluation space and the response situation oval evaluation space so as to unify the three oval evaluation spaces to a comparable scale under the same multi-dimensional decision coordinate system and generate a standardized mapping processing result; Based on the standardized mapping processing result, calculating overlapping areas or center distances between every two of three elliptic evaluation spaces to be used as a spatial relationship index for representing matching compactness of candidate maintenance personnel and the maintenance work order in a single dimension; according to a preset weight rule, weighting and fusing each spatial relationship index corresponding to each candidate maintenance person to generate a comprehensive score representing the overall matching degree of the candidate maintenance person and the maintenance work order; And sorting the comprehensive scores of all candidate maintenance personnel from high to low to select the final maintenance personnel and complete work order dispatch.
6. The intelligent dispatching and AI acceptance method for maintenance work orders of sewage plants according to claim 5, wherein receiving the maintenance process records uploaded by the dispatched final maintenance personnel comprises: after the work order dispatch is completed, pushing dispatch notifications containing details of the work order to a mobile execution end used by the selected final maintenance personnel; the mobile execution terminal generates and presents a standardized evidence acquisition interface according to the dispatch notification, guides a final maintenance personnel to execute maintenance operation through the standardized evidence acquisition interface, and uploads maintenance process records according to preset items; and receiving the maintenance process record data packet which is uploaded by the mobile execution terminal and bound with the current work order, and simultaneously, automatically checking the integrity and format of the maintenance process record data packet to obtain the maintenance process record data packet passing the check.
7. The intelligent dispatching and AI acceptance method for maintenance work orders of sewage plants according to claim 6, wherein the machine-readable acceptance criteria corresponding to the maintenance process record and the maintenance element knowledge graph model are input into an AI model, the AI model judges the compliance and calculates the overall acceptance rate of the maintenance work orders, comprising: extracting image evidence before and after the maintenance corresponding to each item of maintenance content from the maintenance process record data packet passing the verification, and simultaneously obtaining corresponding machine-readable acceptance criteria from the maintenance element knowledge graph model in an associated manner; formatting and combining each group of extracted image evidence with a corresponding machine-readable acceptance standard to construct a standardized input unit for the AI model to perform integrated vision and text analysis; invoking an application program interface of the AI model, submitting the standardized input units to the AI model in batches for analysis, and receiving a judging result output by the AI model on whether each item of maintenance content accords with the acceptance standard of the maintenance content; and analyzing the judging result output by the AI model to count the number of the accepted maintenance content items, and calculating the overall acceptance rate of the maintenance work order based on the ratio of the number of the accepted maintenance content items to the total maintenance content items.
8. The intelligent dispatch and AI acceptance method of maintenance work orders for sewage plants of claim 7, wherein comparing the overall acceptance rate with a preset threshold, automatically determining the acceptance result of the work orders, forming closed-loop management, comprises: comparing the overall acceptance rate with a preset threshold, wherein the preset threshold is set based on the equipment type or the importance level, a comparison result is obtained, and the preliminary acceptance conclusion of the maintenance work order is automatically judged based on the comparison result; According to the preliminary acceptance conclusion, automatically updating the current state of the maintenance work order, if the overall acceptance passing rate is greater than or equal to a preset threshold value, updating the work order state to be accepted, and if the overall acceptance passing rate is less than the preset threshold value, updating the work order state to be changed to generate an updated work order state; According to the updated work order state, corresponding closed loop management operation is automatically triggered, namely, for the work order with the checked state, a check report is generated and filed; in the closed-loop management operation process, a traceable closed-loop management record is formed by recording a complete operation log of the acceptance determination, and finally the closed-loop management of the maintenance work order is completed.
9. A system for intelligent dispatch and AI acceptance of maintenance work orders for sewage plants, which implements the method according to any one of claims 1 to 8, comprising: The model construction module is used for constructing a maintenance element knowledge graph model which defines the association relation among equipment types, maintenance contents, machine-readable acceptance standards and maintenance periods; The assessment space definition module is used for constructing three oval assessment spaces for each maintenance work order in the maintenance plan, namely, constructing a skill compliance oval assessment space by taking the skill requirement level of each maintenance work order, the skill level label of a candidate maintenance person and a preset skill level compatibility rule as rule anchor points, constructing a man-hour capacity oval assessment space by taking the current residual available standard man-hour of the candidate maintenance person, the standard man-hour requirement of each maintenance work order and a preset daily man-hour saturation threshold as capacity anchor points, and constructing a response situation oval assessment space by taking the urgency label of each maintenance work order, the real-time position information of the candidate maintenance person and the fixed position information of target equipment as situation anchor points; The calculation and selection module is used for calculating the relative position relation and intersection state of the three elliptical evaluation spaces so as to select final maintenance personnel; The evidence receiving module is used for receiving the distributed maintenance process records uploaded by the final maintenance personnel; the judging and analyzing module is used for inputting the maintenance process record and the corresponding machine-readable acceptance standard in the maintenance element knowledge graph model into an AI model, judging the compliance by the AI model and calculating the overall acceptance rate of the maintenance work order; And the closed-loop management module is used for comparing the overall acceptance rate with a preset threshold value, and automatically judging the work order acceptance result to form closed-loop management.

Description

Intelligent distribution and AI acceptance method and system for maintenance work orders of sewage plants Technical Field The invention relates to the technical field of equipment maintenance, in particular to an intelligent distribution and AI acceptance method and system for a maintenance work order of a sewage plant. Background In the field of equipment maintenance of sewage treatment plants, the equipment type is complicated, the standardization degree of maintenance operation flow is high, and the requirements on the accuracy of work order distribution and objectivity of acceptance are strict. In the prior art, the distribution of the maintenance work order is mostly dependent on manual matching of maintenance personnel according to experience, and the acceptance work is also mainly dependent on subjective judgment after on-site checking or checking of maintenance photos by management personnel, so that the technical defects of lack of a set of intelligent matching and acceptance system based on structural knowledge association and multidimensional quantitative evaluation are overcome. Specifically, in the traditional work order dispatching process, only basic skills or duty arrangement of maintenance personnel are simply referred, a systematic association evaluation mechanism of multidimensional factors such as equipment maintenance requirements, personnel capacity, workload, response efficiency and the like is not established, so that the problems of lack of reasonable quantitative basis for matching the work order with the maintenance personnel, skill mismatching, uneven personnel load or untimely response and the like are easily caused, meanwhile, the acceptance criteria are mostly unstructured text description, standardized definition which can be directly interpreted by a machine is lacking, the intellectualization and automation of the acceptance process cannot be realized, the acceptance efficiency is affected, and the condition that the acceptance result is inconsistent is easily caused by artificial subjective judgment is also easily caused. In sum, the defects can cause the problems that the precision and the efficiency of work order dispatch are low, timeliness and professionality of maintenance work are difficult to guarantee, and the quality of the maintenance work cannot be objectively and efficiently judged due to non-uniform execution of acceptance criteria, so that the overall quality and the management efficiency of equipment maintenance work of a sewage treatment plant are limited to a certain extent. Disclosure of Invention The technical problem to be solved by the invention is to provide the intelligent dispatching and AI acceptance checking method and system for the maintenance work orders of the sewage treatment plant, which can promote the rationality and objectivity of acceptance of dispatching the maintenance work orders and ensure the quality and efficiency of maintenance work of equipment of the sewage treatment plant. In order to solve the technical problems, the technical scheme of the invention is as follows: In a first aspect, a method for intelligently distributing and AI acceptance of a maintenance work order of a sewage plant, the method comprising: Constructing a maintenance element knowledge graph model which defines the association relationship among equipment types, maintenance contents, machine-readable acceptance criteria and maintenance periods; Automatically generating a maintenance plan for the specific equipment based on the maintenance period in the maintenance element knowledge graph model; Constructing three oval evaluation spaces aiming at each maintenance work order in a maintenance plan, namely constructing a skill compliance oval evaluation space by taking the skill requirement level of each maintenance work order, the skill level label of a candidate maintenance person and a preset skill level compatibility rule as rule anchor points, constructing a man-hour capacity oval evaluation space by taking the current residual available standard man-hour of the candidate maintenance person, the standard man-hour requirement of each maintenance work order and a preset daily man-hour saturation threshold value as capacity anchor points, and constructing a response situation oval evaluation space by taking the emergency level label of each maintenance work order, the real-time position information of the candidate maintenance person and the fixed position information of target equipment as situation anchor points; and calculating the relative position relation and intersection state of the three elliptic evaluation spaces to select the final maintenance personnel. Receiving maintenance process records uploaded by the distributed final maintenance personnel; inputting the maintenance process record and the corresponding machine-readable acceptance standard in the maintenance element knowledge graph model into an AI model, judging the compliance by the AI model