CN113256160B - Comprehensive pipe rack operation and maintenance dynamic risk evaluation method driven by monitoring data - Google Patents
Comprehensive pipe rack operation and maintenance dynamic risk evaluation method driven by monitoring data Download PDFInfo
- Publication number
- CN113256160B CN113256160B CN202110682404.3A CN202110682404A CN113256160B CN 113256160 B CN113256160 B CN 113256160B CN 202110682404 A CN202110682404 A CN 202110682404A CN 113256160 B CN113256160 B CN 113256160B
- Authority
- CN
- China
- Prior art keywords
- risk
- cloud
- data
- evaluation
- entropy
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
- 238000011156 evaluation Methods 0.000 title claims abstract description 143
- 238000012544 monitoring process Methods 0.000 title claims abstract description 112
- 238000012423 maintenance Methods 0.000 title claims abstract description 111
- 238000006243 chemical reaction Methods 0.000 claims abstract description 23
- 230000007246 mechanism Effects 0.000 claims abstract description 18
- 238000004422 calculation algorithm Methods 0.000 claims abstract description 15
- 238000012502 risk assessment Methods 0.000 claims abstract description 10
- 238000012163 sequencing technique Methods 0.000 claims abstract description 9
- 230000001953 sensory effect Effects 0.000 claims description 12
- 238000013507 mapping Methods 0.000 claims description 10
- 238000004364 calculation method Methods 0.000 claims description 8
- 230000000694 effects Effects 0.000 claims description 3
- 238000005259 measurement Methods 0.000 claims description 3
- 230000005540 biological transmission Effects 0.000 claims 1
- 238000000034 method Methods 0.000 abstract description 23
- 238000011058 failure modes and effects analysis Methods 0.000 abstract description 20
- 238000012545 processing Methods 0.000 abstract description 5
- 238000004458 analytical method Methods 0.000 description 10
- 238000001514 detection method Methods 0.000 description 8
- 238000007726 management method Methods 0.000 description 6
- 201000010099 disease Diseases 0.000 description 5
- 208000037265 diseases, disorders, signs and symptoms Diseases 0.000 description 5
- 230000007613 environmental effect Effects 0.000 description 5
- 238000010276 construction Methods 0.000 description 4
- COCAUCFPFHUGAA-MGNBDDOMSA-N n-[3-[(1s,7s)-5-amino-4-thia-6-azabicyclo[5.1.0]oct-5-en-7-yl]-4-fluorophenyl]-5-chloropyridine-2-carboxamide Chemical compound C=1C=C(F)C([C@@]23N=C(SCC[C@@H]2C3)N)=CC=1NC(=O)C1=CC=C(Cl)C=N1 COCAUCFPFHUGAA-MGNBDDOMSA-N 0.000 description 4
- 229920000915 polyvinyl chloride Polymers 0.000 description 4
- 239000004800 polyvinyl chloride Substances 0.000 description 4
- 230000008569 process Effects 0.000 description 4
- 238000005336 cracking Methods 0.000 description 3
- 238000013461 design Methods 0.000 description 3
- 238000011160 research Methods 0.000 description 3
- 230000006378 damage Effects 0.000 description 2
- 238000013499 data model Methods 0.000 description 2
- 238000010586 diagram Methods 0.000 description 2
- 238000005516 engineering process Methods 0.000 description 2
- 230000008676 import Effects 0.000 description 2
- 230000003068 static effect Effects 0.000 description 2
- 101150081525 LIMK1 gene Proteins 0.000 description 1
- 230000004931 aggregating effect Effects 0.000 description 1
- 230000032683 aging Effects 0.000 description 1
- 230000015572 biosynthetic process Effects 0.000 description 1
- 230000008859 change Effects 0.000 description 1
- 230000008878 coupling Effects 0.000 description 1
- 238000010168 coupling process Methods 0.000 description 1
- 238000005859 coupling reaction Methods 0.000 description 1
- 230000007547 defect Effects 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 230000018109 developmental process Effects 0.000 description 1
- 230000002349 favourable effect Effects 0.000 description 1
- 238000007667 floating Methods 0.000 description 1
- 230000004927 fusion Effects 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 230000008439 repair process Effects 0.000 description 1
- 238000013179 statistical model Methods 0.000 description 1
- 230000035882 stress Effects 0.000 description 1
- 230000001502 supplementing effect Effects 0.000 description 1
- 238000003786 synthesis reaction Methods 0.000 description 1
- 230000001960 triggered effect Effects 0.000 description 1
- 238000012800 visualization Methods 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q10/00—Administration; Management
- G06Q10/06—Resources, workflows, human or project management; Enterprise or organisation planning; Enterprise or organisation modelling
- G06Q10/063—Operations research, analysis or management
- G06Q10/0635—Risk analysis of enterprise or organisation activities
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/30—Information retrieval; Database structures therefor; File system structures therefor of unstructured textual data
- G06F16/36—Creation of semantic tools, e.g. ontology or thesauri
- G06F16/367—Ontology
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/24—Classification techniques
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N5/00—Computing arrangements using knowledge-based models
- G06N5/02—Knowledge representation; Symbolic representation
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q10/00—Administration; Management
- G06Q10/06—Resources, workflows, human or project management; Enterprise or organisation planning; Enterprise or organisation modelling
- G06Q10/063—Operations research, analysis or management
- G06Q10/0639—Performance analysis of employees; Performance analysis of enterprise or organisation operations
- G06Q10/06393—Score-carding, benchmarking or key performance indicator [KPI] analysis
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q10/00—Administration; Management
- G06Q10/20—Administration of product repair or maintenance
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q50/00—Information and communication technology [ICT] specially adapted for implementation of business processes of specific business sectors, e.g. utilities or tourism
- G06Q50/10—Services
- G06Q50/26—Government or public services
Landscapes
- Engineering & Computer Science (AREA)
- Business, Economics & Management (AREA)
- Human Resources & Organizations (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Strategic Management (AREA)
- Economics (AREA)
- Tourism & Hospitality (AREA)
- Entrepreneurship & Innovation (AREA)
- Data Mining & Analysis (AREA)
- General Business, Economics & Management (AREA)
- Marketing (AREA)
- Educational Administration (AREA)
- Development Economics (AREA)
- Operations Research (AREA)
- Quality & Reliability (AREA)
- General Engineering & Computer Science (AREA)
- Game Theory and Decision Science (AREA)
- Life Sciences & Earth Sciences (AREA)
- Evolutionary Computation (AREA)
- Computational Linguistics (AREA)
- Artificial Intelligence (AREA)
- Bioinformatics & Computational Biology (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Computing Systems (AREA)
- Software Systems (AREA)
- Mathematical Physics (AREA)
- Evolutionary Biology (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Animal Behavior & Ethology (AREA)
- Databases & Information Systems (AREA)
- Health & Medical Sciences (AREA)
- General Health & Medical Sciences (AREA)
- Primary Health Care (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The invention discloses a monitoring data-driven comprehensive pipe rack operation and maintenance dynamic risk evaluation method which comprises the steps of (1) establishing a risk evaluation index according to pipe rack operation and maintenance risk classification, classifying pipe rack components and equipment according to a historical drawing, and establishing a risk evaluation mechanism through FMEA risk analysis; (2) analyzing the sensing monitoring data to realize the conversion between the data and the SOD scoring standard; (3) generating an expert evaluation cloud and a sensing monitoring cloud by a reverse cloud generator; (4) iteratively updating the dynamic risk cloud through the risk data, and iteratively correcting the evaluation data; (5) acquiring risk sequencing by using a cloud model gray correlation algorithm by taking the dynamic risk cloud as input; (6) establishing a comprehensive pipe rack operation and maintenance risk knowledge map; (7) and risk inquiry and corresponding maintenance knowledge reasoning are carried out through the operation and maintenance risk knowledge map of the comprehensive pipe rack. The method helps to discover potential risks, leads risks to control intervention time, improves operation and maintenance risk management capacity, and further reduces risk processing cost.
Description
Technical Field
The invention belongs to the field of operation and maintenance risk assessment of pipe corridors, and particularly relates to a comprehensive pipe corridor operation and maintenance dynamic risk evaluation method driven by monitoring data.
Background
The underground comprehensive pipe gallery is easy to suffer from pipeline aging, leakage, gallery body settlement deformation and other diseases in the operation process, so that serious operation and economic consequences are caused. Risks are identified and evaluated, decision basis is provided for operation and maintenance risk management, and risk occurrence probability can be effectively reduced. At present, static data such as expert scores and the like are mostly used as data sources for building operation and maintenance risk evaluation, and methods such as fault trees, multi-attribute decisions, analytic hierarchy process and the like are combined. However, unlike the traditional building form, the pipe gallery is composed of a plurality of pipeline systems which operate independently, the internal risk relationship is complicated, and then a plurality of types of pipelines are often arranged in the same cabin, and coupling risks exist among the pipelines, so that consideration for risk source analysis needs to be taken into consideration at the beginning of risk evaluation. In addition, the running state of the pipe gallery changes frequently, and the expert evaluation is not favorable for timely capturing the risk state and lacks the timeliness.
The Failure Mode and Effects Analysis (FMEA) method describes risks from three dimensions of Severity (Severity), Occurrence (occupancy) and Detection (Detection) (hereinafter referred to as SOD), and has better risk source Analysis capability compared with other evaluation methods. In recent years, some scholars have combined cloud models and FMEA to deal with uncertainty in objective risk and subjective evaluation, further extending the risk analysis capabilities of FMEA. Hu-Chen Liu and the like adopt cloud models to deal with ambiguity and randomness existing in FMEA language evaluation, and make up for the inherent uncertainty defect of FMEA. The Xinlong Li and the like convert the attribute evaluation result of the FMEA into a cloud model, and a risk evaluation method based on the cloud model is provided to effectively analyze the failure risk of the numerical control machine. However, FMEA, while able to describe complex risk sources, has difficulty capturing dynamic risk changes.
In recent years, with the advent of the big data era, dynamic changes reflecting risks by dynamic data have been applied in many fields. Shabtai Isaac and Tsah Edrei propose a statistical model to actively provide an alarm in the event of an elevated worker construction safety risk by calculating the construction safety risk using real-time worker behavior, risk area, etc. data. Xianguo Wu et al propose a dynamic Bayes-based system decision method, aiming at providing guidance for dynamic risk safety analysis of tunnel pavement damage changing along with time. And the Tang-super class converts the monitoring data into corresponding risk probability according to the alarm value, and establishes a risk evaluation method in the construction stage of the subway deep foundation pit by adopting deformation and stress monitoring data in the construction stage as risk evaluation parameters. Therefore, the dynamic risk is captured through dynamic data, time-efficient risk evaluation can be achieved, and related research is lacked in the field of comprehensive pipe galleries at present.
In addition, the subsequent application of the evaluation result is not considered in the existing research, and the risk management method based on risk evaluation still needs to be perfected. In recent years, with the progress of building informatization, some scholars apply knowledge management technology to the building field to improve the value of building information data. The knowledge map describes knowledge in the form of map data, is simple and easy to read, has good algorithm expandability, and greatly improves the read/write operation of a disk and the data access efficiency compared with the traditional database, so that the development of related application systems based on the knowledge map technology is gradually a research hotspot in recent years. Matern Hannah et al use a knowledge-graph-based graph data model for describing and managing information derived from the BIM model and develop systems with the help of this graph database to assist in earlier design. Sabri Qamer Uddin et al, in conjunction with graph theory and CBR, exact graph matching and non-exact graph matching techniques, developed a system that helps architects retrieve similar planar graphs at an early design stage to aid in design. Ali Ismail et al propose a workflow for automatically converting an IFC model into an IFC metagraph and an object graph database, and study and demonstrate the potential of using graph theory concepts and graph databases to manage, visualize and analyze the huge information and complex relationships of BIM models based on IFC standards. The risk relationship of the pipe gallery is complex, the related knowledge and data sources of the operation and maintenance of the pipe gallery are wide, the knowledge map capable of well expressing the semantic relationship inside and among different types of knowledge is introduced, and a more accurate and three-dimensional risk decision auxiliary and countermeasure measure scheme can be provided by combining the risk analysis result.
Disclosure of Invention
The invention aims to provide a comprehensive pipe rack operation and maintenance dynamic risk evaluation method driven by monitoring data, which realizes the analysis of a complex risk source of a pipe rack by combining an expert evaluation through a pipe rack operation and maintenance risk evaluation mechanism; the method has the advantages that sensing monitoring data are introduced to serve as evaluation data, dynamic risks of the pipe gallery are captured, the cloud model is adopted to process data uncertainty, dynamic risk evaluation is achieved, the operation and maintenance risk knowledge graph of the comprehensive pipe gallery is established, risk visualization analysis and maintenance knowledge reasoning are provided, potential risks are found in an auxiliary mode, intervention time is controlled in a front-mounted mode, operation and maintenance risk management capacity is improved, and risk processing cost is reduced.
In order to achieve the purpose, the invention adopts the following technical scheme:
a comprehensive pipe rack operation and maintenance dynamic risk evaluation method driven by monitoring data comprises the following steps:
(1) establishing a pipe gallery FMEA risk evaluation mechanism: establishing a risk evaluation index according to the pipe gallery operation and maintenance risk classification, classifying pipe gallery components and equipment according to historical drawings, and establishing a risk evaluation mechanism based on an SOD scoring standard through FMEA risk analysis according to the risk evaluation index, the pipe gallery components and the equipment classification;
(2) accessing the sensing monitoring data into a risk evaluation mechanism: analyzing the sensing monitoring data to realize the conversion between the sensing monitoring data and the SOD scoring standard;
(3) evaluation cloud generation: realizing uncertainty conversion from evaluation data to a cloud model through a reverse cloud generator, and generating an expert evaluation cloud and a sensing monitoring cloud;
a reverse cloud generator:
；
in the formula, Ex is expectation, a typical value representing risk degree, En is entropy, representing risk discrete degree, He is super entropy, representing risk uncertainty measurement, when input data is expert evaluation value, N is expert number, x is expert evaluation valueiIs an evaluation value of the i-th expert,
the average value of expert evaluation is s, and the standard deviation of expert evaluation is s; when the input data is the sensing monitoring alarm times, N is the number of monitoring points, xiThe evaluation value of the corresponding occurrence degree of the alarm times of the ith monitoring point,
the average value of the evaluation values of the incidence degrees corresponding to the alarm times of the monitoring points is obtained, and s is the standard deviation of the evaluation values of the incidence degrees corresponding to the alarm times of the monitoring points;
and (3) evaluating cloud by experts: cp=(Exp O,Enp O,Hep O,Exp S,Enp SHep S,Exp D,Enp D,Hep D)，CpEvaluation of cloud, Ex for expertsp OFor experts to evaluate the expectation of the O score, Enp OEntropy of O-rating for expert evaluation, Hep OHyper-entropy, Ex, for expert evaluation of O-scoresp SFor experts to evaluate the expectation of the S score, Enp SEntropy of S-score for expert evaluation, Hep SFor experts to evaluate the hyper-entropy, Ex, of the S scorep DFor experts to evaluate the expectation of a D score, Enp DEvaluation of entropy of D-score for experts, Hep DEvaluating the super entropy of the D score for the expert;
sensing and monitoring cloud: cm=(Exm O,Enm O,Hem O,Exm S,Enm SHem S,Exm D,Enm D,Hem D)，CmMonitoring clouds, Ex for sensingm OFor the expectation of converting the sensing monitoring data into O score, Enm OEntropy, He, for conversion of sensory monitoring data into O-scoresm OFor the conversion of the sensed monitoring data into O-scored hyper-entropy, Exm SFor the expectation of converting the sensing monitoring data into S score, Enm SEntropy, He, for conversion of sensory monitoring data into S-scoresm SFor converting sensing monitoring data into super entropy, Ex of S scorem DFor the conversion of the sensory monitoring data into an expectation of D-score, Enm DEntropy, He, for conversion of sensory monitoring data into D-scoresm DConverting the sensing monitoring data into D-grade super entropy;
(4) and (3) risk data iteration: accessing the continuously updated sensing monitoring cloud and the original data through risk data iteration, updating the dynamic risk cloud, and iteratively correcting the evaluation data;
wherein iterating the tth dynamic risk cloud:
in the formula, Exd (t)To iterate the expectation of the tth dynamic risk cloud, End (t)Entropy of the dynamic risk cloud for the t-th iteration, Hed (t)To iterate the hyper-entropy, Ex, of the tth dynamic risk cloudd (t-1)To iterate the expectation of the dynamic risk cloud for the t-1 st time, End (t-1)Entropy of dynamic risk clouds for iteration t-1, Hed (t-1)To iterate the hyper entropy, Ex, of the dynamic risk cloud for the t-1 st times (t)For the expectation of the sensing monitoring cloud for the iteration t, Ens (t)Monitoring entropy of cloud, He, for sensing of t-th iterations (t)Monitoring the hyper-entropy of the cloud for the sensing of the t-th iteration;
(5) risk ranking: acquiring risk sequencing by using a cloud model gray correlation algorithm by taking the dynamic risk cloud as input;
(6) establishing a comprehensive pipe rack operation and maintenance risk knowledge map: the comprehensive pipe rack operation and maintenance risk knowledge graph comprises a mode layer and a data layer, wherein the mode layer of the comprehensive pipe rack operation and maintenance risk knowledge graph is established according to the mapping relation between components and risks by taking a pipe rack FMEA (failure mode and effects evaluation) mechanism as a core; collecting basic triple data according to a data framework of a mode layer and establishing a data layer, wherein the basic triple data comprise nodes, node attribute data, relations and relation attribute data; storing actual data files in a Neo4j database to establish a comprehensive pipe rack operation and maintenance risk knowledge map;
(7) risk querying and knowledge reasoning: and risk query and corresponding maintenance knowledge reasoning are carried out through the established operation and maintenance risk knowledge map of the comprehensive pipe rack.
Further, the risk data iteration in the step (4) is performed according to the following steps:
step 1: defining evaluation times t =0, sensing monitoring cloud input state values s =0, and initial dynamic risk cloud Cd=(Exd O,End O,Hed O,Exd S,End SHed S,Exd D,End D,Hed D)，CdTo initiate a dynamic risk cloud, Exd OExpectation of O scoring for initial state data, End OEntropy, He, of O scores for initial State datad OHyper-entropy, Ex, scoring O for initial state datad SExpectation of scoring S for initial state data, End SEntropy, He, of S score for initial State datad SSuper entropy, Ex, scoring S for initial State datad DExpectation of scoring D for initial state data, End DEntropy, He, scoring D for initial state datad DIs in an initial stateState data is the super entropy of the D score;
step 2: assigning the expert evaluation cloud Cp to the dynamic risk cloud C when t =0d (t)Taking expert evaluation as initial evaluation data; when t is more than or equal to 1, using the cloud comprehensive algorithm result as the current dynamic risk cloud input value Cd (t)(ii) a If sensing monitoring cloud input exists, changing the state value s =1, and inputting the sensing monitoring cloud Cm (t)；
And step 3: if s =1, outputting C from step 2d (t)、Cm (t)Substituting into a cloud comprehensive algorithm;
and 4, step 4: assigning t = t +1, and outputting dynamic risk cloud Cd (t)Changing the state value s =0, and completing 1 dynamic risk cloud updating calculation;
and 5: and when a new sensing monitoring cloud is input, substituting the output result of the step 4, and circularly executing the step 2-4 iterative computation to obtain the continuously updated dynamic risk cloud.
The method provided by the invention has the advantages that:
(1) introducing FMEA, carrying out the complicated risk source analysis of utility tunnel, establishing the evaluation standard based on the risk description dimension, and then establishing a pipe gallery FMEA risk evaluation mechanism, can effectively describe the complicated risk of pipe gallery.
(2) And introducing sensing monitoring data as evaluation data to capture the dynamic risk of the pipe gallery. And establishing a risk data iteration method, and performing data iteration on continuously input sensing monitoring data to obtain continuously updated evaluation data while realizing good fusion of the expert evaluation static data and the sensing monitoring dynamic data so as to reflect the dynamic change of the risk.
(3) And processing the evaluation data by adopting a cloud model so as to solve the problem of uncertainty caused by subjective evaluation and objective environment of experts.
(4) The risk information is visualized through the knowledge map, the algorithm is linked, the maintenance knowledge is linked, a quantitative basis is provided for risk disposal knowledge reasoning and operation and maintenance risk management decision making in the operation and maintenance process, operation and maintenance management personnel are assisted in decision making, and the operation and maintenance efficiency of the comprehensive pipe rack is improved.
Drawings
FIG. 1 is a schematic view of the structure of the present invention.
Fig. 2 is a schematic diagram of the correspondence between the alarm times and the occurrence degree levels of the present invention.
FIG. 3 is a schematic diagram of a risk data iteration method according to the present invention.
Fig. 4 is a utility tunnel operation and maintenance risk knowledge map (risk-free path) of the present invention.
FIG. 5 is a mapping relationship between the dynamic evaluation method of the present invention and a knowledge graph containing risk paths.
Detailed Description
As shown in fig. 1, the method for evaluating the operation and maintenance dynamic risk of the utility tunnel driven by monitoring data provided by this embodiment includes the following steps:
(1) establishing a pipe gallery FMEA risk evaluation mechanism: establishing a risk evaluation index according to the operation and maintenance risk classification of the pipe gallery, and classifying pipe gallery components and equipment according to a historical drawing; in the embodiment, the risk evaluation indexes are established by adopting the classification of the operation and maintenance risks of the pipe gallery established by Guojiaqi and the like, as shown in a table 1; the tube lane components and equipment classifications are shown in table 2 and can be adjusted according to the specific tube lane. And establishing a SOD-based risk evaluation mechanism by FMEA risk analysis according to the classification and risk evaluation indexes of the pipe gallery components and the equipment, as shown in Table 3. To obtain initial baseline data, the expert was invited to perform SOD evaluation based on table 3.
In table 3, the risk evaluation of the member C11 power cable and the risk of U11 settlement is taken as an example, and the evaluation criteria are described as follows: (1) for the power cable, the severity evaluation of consequences when the settlement risk occurs is divided into ten grades of 0-9, and the severity of the corresponding risk is serious harm from no later result to no warning; (2) the evaluation of the occurrence degree corresponding to the severity of different grades is divided into ten grades of 0-9, and the occurrence frequency of corresponding risks is from never occurring to always existing; (3) the detection degree evaluation that the settlement risk of the power cable can be detected in daily operation and maintenance is divided into ten grades of 0-9, and the detection degree corresponding to the settlement risk can be always found to be undetectable.
TABLE 1 pipe gallery operation and maintenance Risk Classification
TABLE 2 piping lane Member and Equipment Classification
TABLE 3 SOD-based risk evaluation Table
(2) Accessing the sensing monitoring data into a risk evaluation mechanism: analyzing the sensing monitoring data to realize the conversion between the sensing monitoring data and the SOD scoring standard; the conversion relation between different types of sensing monitoring data and the SOD scoring standard is different, and analysis processing is required to be carried out during access. Taking environmental quality monitoring as an example, the technical specification of urban comprehensive pipe gallery engineering (GB50838-2015) specifies monitoring parameters of different components, and this embodiment establishes an environmental quality monitoring value analysis table, as shown in table 4,
TABLE 4 analysis of environmental quality monitoring values
As shown in table 4, when the monitoring value is located normal interval, the operation of the pipe gallery is not affected, and when the monitoring value is located alarm interval, the normal operation of the pipe gallery is affected.
Because the severity is related to the risk consequence, the sensing monitoring data cannot reflect the risk consequence, and thus, severity evaluation data does not exist; the detection degree is related to the detectability, the sensing monitoring parameters are related to the environmental safety risk indexes, and the evaluation value of the risk corresponding to the detection degree is 0, so that the risk can be always found; the occurrence degree is the occurrence probability of different risk degrees, the increase of the monitoring alarm times represents the increase of the corresponding environmental safety risk occurrence probability, the occurrence degree is obviously influenced, and the more the alarm times, the higher the corresponding occurrence degree grade is. According to the expert opinion, 30 days are taken as a monitoring period, the corresponding relation between the alarm frequency and the occurrence degree grade is appointed as shown in figure 2, and the alarm frequency is converted into an occurrence degree evaluation value. In addition, the piping lane usually comprises a plurality of pipeline sections, and the risk occurrence probability of different pipeline sections has the difference, need set up a plurality of monitoring point record alarm times according to piping lane actual conditions.
(3) Evaluation cloud generation: realizing uncertainty conversion from evaluation data to a cloud model through a reverse cloud generator, and generating an expert evaluation cloud and a sensing monitoring cloud; the method is characterized in that a reverse cloud generator is adopted to describe quantitative data of expert evaluation and alarm frequency occurrence degree evaluation values by cloud model characteristic value expectation (Ex), entropy (En) and super entropy (He).
A reverse cloud generator:
；
in the formula, Ex is expectation, a typical value representing risk degree, En is entropy, representing risk discrete degree, He is super entropy, representing risk uncertainty measurement, when input data is expert evaluation value, N is expert number, x is expert evaluation valueiIs an evaluation value of the i-th expert,
the average value of expert evaluation is s, and the standard deviation of expert evaluation is s; when the input data is the sensing monitoring alarm times, N is the number of monitoring points, xiThe evaluation value of the corresponding occurrence degree of the alarm times of the ith monitoring point,
the average value of the evaluation values of the incidence degrees corresponding to the alarm times of the monitoring points is obtained, and s is the standard deviation of the evaluation values of the incidence degrees corresponding to the alarm times of the monitoring points;
and (3) evaluating cloud by experts: cp=(Exp O,Enp O,Hep O,Exp S,Enp SHep S,Exp D,Enp D,Hep D)，CpEvaluation of cloud, Ex for expertsp OFor experts to evaluate the expectation of the O score, Enp OEntropy of O-rating for expert evaluation, Hep OHyper-entropy, Ex, for expert evaluation of O-scoresp SFor experts to evaluate the expectation of the S score, Enp SEntropy of S-score for expert evaluation, Hep SFor experts to evaluate the hyper-entropy, Ex, of the S scorep DFor experts to evaluate the expectation of a D score, Enp DEvaluation of entropy of D-score for experts, Hep DEvaluating the super entropy of the D score for the expert;
sensing and monitoring cloud: cm=(Exm O,Enm O,Hem O,Exm S,Enm SHem S,Exm D,Enm D,Hem D)，CmMonitoring clouds, Ex for sensingm OFor the expectation of converting the sensing monitoring data into O score, Enm OEntropy, He, for conversion of sensory monitoring data into O-scoresm OFor the conversion of the sensed monitoring data into O-scored hyper-entropy, Exm SFor the expectation of converting the sensing monitoring data into S score, Enm SEntropy, He, for conversion of sensory monitoring data into S-scoresm SFor converting sensing monitoring data into super entropy, Ex of S scorem DFor the conversion of the sensory monitoring data into an expectation of D-score, Enm DEntropy, He, for conversion of sensory monitoring data into D-scoresm DAnd converting the sensing monitoring data into D-grade super entropy.
(4) And (3) risk data iteration: and (4) accessing the continuously updated sensing monitoring cloud and the original data through risk data iteration, updating the dynamic risk cloud, and iteratively correcting the evaluation data.
The iteration step is shown in fig. 3, and specifically includes the following steps:
step 1: defining evaluation times t =0, and monitoring cloud input stateState value s =0, initial dynamic risk cloud Cd=(Exd O,End O,Hed O,Exd S,End SHed S,Exd D,End D,Hed D)，CdTo initiate a dynamic risk cloud, Exd OExpectation of O scoring for initial state data, End OEntropy, He, of O scores for initial State datad OHyper-entropy, Ex, scoring O for initial state datad SExpectation of scoring S for initial state data, End SEntropy, He, of S score for initial State datad SSuper entropy, Ex, scoring S for initial State datad DExpectation of scoring D for initial state data, End DEntropy, He, scoring D for initial state datad DThe initial state data is the hyper-entropy of the D score.
Step 2: when t =0, the expert evaluates cloud CpAssign value to dynamic Risk cloud Cd (t)Taking expert evaluation as initial evaluation data; when t is more than or equal to 1, using the last cloud comprehensive algorithm result as the current dynamic risk cloud input value Cd (t)(ii) a If sensing monitoring cloud input exists, changing the state value s =1, and inputting the sensing monitoring cloud Cm (t)；
And step 3: if s =1, outputting C from step 2d (t)、Cm (t)Substituting into a cloud comprehensive algorithm;
and 4, step 4: assigning t = t +1, and outputting dynamic risk cloud Cd (t)Changing the state value s =0, and completing 1 dynamic risk cloud updating calculation;
and 5: when a new sensing monitoring cloud is input, the step is triggered, the output result of the step 4 is substituted, and the steps 2-4 are executed in a circulating mode, so that the continuously updated dynamic risk cloud is obtained.
Wherein iterating the tth dynamic risk cloud:
in the formula, Exd (t)To iterate the expectation of the tth dynamic risk cloud, End (t)Entropy of the dynamic risk cloud for the t-th iteration, Hed (t)To iterate the hyper-entropy, Ex, of the tth dynamic risk cloudd (t-1)To iterate the expectation of the dynamic risk cloud for the t-1 st time, End (t-1)Entropy of dynamic risk clouds for iteration t-1, Hed (t-1)To iterate the hyper entropy, Ex, of the dynamic risk cloud for the t-1 st times (t)For the expectation of the sensing monitoring cloud for the iteration t, Ens (t)Monitoring entropy of cloud, He, for sensing of t-th iterations (t)The cloud's hyper-entropy is monitored for the sensing of the t-th iteration.
It should be noted that when the input cloud model is null, the cloud synthesis algorithm iteration is not executed, and the non-null cloud model is directly output, for example, when the input value is a severity-based expert evaluation cloud and a null sensing monitoring cloud, the expert evaluation cloud is directly output as a final result.
(5) Risk ranking: and taking the dynamic risk cloud as input, and obtaining quantitative risk information through risk sequencing. In the embodiment, the risk evaluation data is analyzed by using a cloud model-grey correlation algorithm proposed by LEE and the like, and risk sequencing is calculated.
Assuming that there are m components and n risks, taking the risk ranking of the i (i = (1,2, …, m) th component as an example, the calculation steps are as follows:
step 1: defining a grey correlation maximum risk reference sequence based on a cloud model:
step 2: calculating a maximum risk reference sequence value, taking the degree of occurrence as an example:
；
wherein j = (1,2, …, n) characterizes the jth risk of the ith component;
and step 3: calculating the distance between each risk sequence and the maximum risk reference sequence based on SOD indexes:
；
wherein d isijDenotes the distance between the jth risk of the ith member and the maximum reference sequence, k =1 denotes the degree of occurrence, k =2 denotes the degree of severity, k =3 denotes the degree of detection, w denotes the number of componentskThe corresponding weight of the three indexes is determined by the actual condition of the project.
And 4, step 4: definition of dmax=maxdijAnd dmin=mindijAnd further calculating the gray relevance g of the jth risk of the ith componentijComprises the following steps:
；
and 5: and (4) aggregating the risk ranking calculation results of the m components, and representing the risk degree by grey relevance, so that the risk ranking of any component can be obtained.
In practical application, part of the risks are high in occurrence degree and severity and are easy to detect, and part of the risks are low in occurrence degree and severity and are difficult to detect; the FMEA evaluation system compensates the influence of the self-attribute of the risk on the evaluation result through the detection degree, and the SOD evaluation is integrated to be used as a final risk sequencing result. Meanwhile, based on the evaluation of severity and occurrence degree, crisis degree sequencing is introduced to represent short-term risk conditions so as to assist risk decision; and the crisis degree ranking evaluation data are dynamic risk clouds based on severity and incidence, and the calculation steps are consistent with those of risk ranking.
(6) Establishing a comprehensive pipe rack operation and maintenance risk knowledge map: the comprehensive pipe rack operation and maintenance risk knowledge graph comprises a mode layer and a data layer, wherein the mode layer of the comprehensive pipe rack operation and maintenance risk knowledge graph is established according to the mapping relation between components and risks by taking a pipe rack FMEA (failure mode and effects evaluation) mechanism as a core; collecting basic triple group data according to a model layer data framework, and establishing a comprehensive pipe rack operation and maintenance risk knowledge map data layer in a Neo4j map database in a mode of importing data files.
The mode layer provides decision support for operation and maintenance risk analysis and maintenance, so that the data model can comprehensively and simply describe knowledge elements related to operation and maintenance risks of the comprehensive pipe rack and logic relations of the knowledge elements, and disease position positioning and accurate risk information can be provided according to a risk analysis result. The mode layer is established according to a mapping relation between a component and a risk by taking a pipe gallery FMEA evaluation mechanism as a core.
Firstly, a class hierarchy structure is constructed, corresponding risks occur to the components according to a pipe gallery risk occurrence mechanism, corresponding risk disposal is further executed, the risk disposal is summarized into component-risk disposal operation and maintenance decision logic, and the bottom end of the logic is maintenance knowledge required by risk disposal, so that the class hierarchy comprises the component, the risks and the maintenance knowledge.
Establishing a class relation, dividing a pipe gallery system according to BIM model drawing paper, establishing a 'has _ part _ of' hierarchical relation in a component class, assisting in positioning the position of the component, establishing a 'potential _ link' connection relation between the component class and a maintenance knowledge class to represent the potential association of the component and the maintenance knowledge, obtaining a component-risk mapping relation according to a FMEA (failure mode analysis) scoring mechanism, and establishing a 'connected _ with' connection relation between the component class and the risk class to realize the risk information query based on the component; and further expanding the risk maintenance knowledge, establishing a 'be _ handled _ with' connection relation between the risk class and the maintenance knowledge class, and providing maintenance knowledge retrieval.
Finally, establishing attributes which comprise class attributes and class relation attributes; attributes are used to describe internal information of a class, and attribute constraints are used to describe values, types, domains, ranges, etc. of attributes. For the convenience of retrieval, the coding property "code: xx" is established for the component class, risk class and maintenance knowledge class. In order to improve the readability of the knowledge graph, a Chinese name attribute 'name: xx' is further established for component classes, risk classes and maintenance knowledge classes. To store risk maintenance knowledge, a content attribute "content: xx" is established for the maintenance knowledge class. And finally, establishing a risk degree attribute R _ value: xx and a risk level attribute R _ sort: xx for the linked _ with relation so as to provide risk information query based on the knowledge graph for visually presenting a dynamic risk evaluation algorithm result.
According to the mode layer, the operation and maintenance risk knowledge graph data layer of the comprehensive pipe gallery comprises three types of triples, namely (entity) - [ relationship ] - (entity), (entity) - { attribute: attribute value } and (relationship) - { attribute: attribute value }, and specific basic triples are as follows: (entity) - [ relationship ] - (entity): (component) - [ has _ part _ of ] - (component) - [ patent _ limk ] - (maintenance knowledge), (component) - [ connected _ with ] - (risk) - [ be _ handled _ with ] - (maintenance knowledge); (entity) - { Attribute: Attribute value }: (means) - { code: xx }, (risk) - { code: xx }, (maintenance knowledge) - { code: xx }, (means) - { name: xx }, (risk) - { name: xx }, (maintenance knowledge) - { content: xx }; (relationship) - { attribute: attribute value }: (connected _ with) - { R _ value: xx }, (connected _ with) - { R _ sort: xx }. Collecting basic ternary group data, including nodes, node attribute data, relations and relation attribute data; the data sources include relevant laws and regulations, expert knowledge, historical maintenance information, and the like.
Wherein, the node and node attribute data: building block entity: and (component) - { Code: xx } and (component) - { Name: xx } attribute triad data corresponding to the Code attribute value and the Name attribute value, such as 'Code: C11', 'Name: power cable', are obtained according to the classification of the pipe gallery component and the equipment in the table 2, and finally, a component entity is established.
Risk entity: namely, the specific risk data of the operation and maintenance of the pipe gallery, the corresponding encoding attribute values and Name attribute values of the (risk) - { Code: xx } and (risk) - { Name: xx } attribute ternary group data, such as Code: U11 and Name: settlement, are obtained according to the operation and maintenance risk classification of the pipe gallery in table 1, and finally a risk entity is established.
And (3) maintaining a knowledge entity: namely, the maintenance method of the pipe gallery diseases, the existing related standards of the operation and maintenance of the pipe gallery do not describe the concrete maintenance method of the pipe gallery diseases in detail, and only the maintenance content and the basic maintenance principle are specified. Therefore, in this embodiment, according to the configuration of the pipe gallery system, the disease maintenance method is queried in the corresponding system field to obtain the maintenance knowledge entity data, and finally the maintenance knowledge entity is established. The maintenance knowledge entity attribute triplets include (maintenance knowledge) - { code: xx }, (maintenance knowledge) - { name: xx } and (maintenance knowledge) - { content: xx }, the code encoding attribute reference member entity and the risk entity encoding mode, denoted by "M + arabic numeral", for example "M1". The name attribute stores the name of the maintenance method, such as "name: PE polyvinyl chloride lining repair method ". content maintenance method the content attribute stores a specific maintenance method, such as "content: the method is characterized in that a to-be-repaired reinforced pipeline is used as a carrier, polyvinyl chloride (PE) short pipes subjected to special pipe orifice cutting processing are connected section by section in a hydraulic mode and synchronously pulled into the to-be-repaired pipeline, so that the pipeline is repaired or reinforced. ", finally establishing the maintenance knowledge.
Relationship and relationship attribute data: (means) - [ has _ part _ of ] - (means): and (4) obtaining the connection relation between the systems in the pipe gallery and the components in the systems according to the tree classification in the table 2.
(means) - [ potential _ link ] - (maintenance knowledge): defining its relationships such that each component is linked to each maintenance knowledge item characterizes the uncertainty of the potential relationship between the component and the maintenance knowledge item without knowledge of the risk relationship.
(risk) - [ be _ handled _ with ] - (maintenance knowledge): and according to the system field and risk characteristics of the maintenance knowledge entity, one-to-one correspondence is carried out, and one risk node corresponds to one maintenance knowledge node.
The operation and maintenance knowledge graph of the utility tunnel without the risk path is obtained, for example, as shown in fig. 4, a partial graph of the drainage pipeline is used, and the relationship between the component and the risk is lacking, so that the component-risk mapping relationship is obtained based on the dynamic risk evaluation method for the operation and maintenance of the utility tunnel, the evaluation result is used as the risk relationship attribute, and the risk information is stored.
(means) - [ connected _ with ] - (risk): the knowledge graph risk path based on the dynamic risk evaluation method firstly judges whether a mapping relation exists according to operation and maintenance dynamic risk evaluation results, if the SOD evaluation values are all 0, the mapping relation does not exist, and if not, the relation is defined as connected _ with. The "connected _ with" relationship includes two relationship attribute triplets, namely (connected _ with) - { R _ value: xx } and (connected _ with) - { R _ sort: xx }. The R _ Value attribute Value is calculated by the risk ranking to obtain a gray degree gij as a risk Value attribute, for example, "R _ Value: 1.00". The R _ sort attribute value is obtained from the result of the dynamic risk ranking calculation for the operation and maintenance of the pipe rack, for example, "R _ sort: 1". Supplementing the risk relationship mapping and the risk information on the basis of fig. 5, and finally establishing a comprehensive pipe gallery operation and maintenance knowledge map, taking a drainage pipeline part map as an example for display, as shown in fig. 5, forming the relation among the nodes of the internal components of the pipe gallery through 'has _ part _ of', and being capable of quickly positioning the position of the drainage pipeline. The drainage pipeline nodes are connected with all the maintenance knowledge nodes through the 'potential _ link', and the drainage pipeline is represented to have potential connection with the existing maintenance knowledge. In addition, according to the result of the risk sorting algorithm, the drainage pipeline is connected with the pipeline cracking risk node through a connected _ with relation, the risk node is further connected with the corresponding maintenance knowledge node through a be _ handled _ with relation, a component-risk-maintenance knowledge risk path is finally formed, and the real relation of the potential _ link is confirmed. Therefore, when any component meets the path query condition, the maintenance knowledge corresponding to the risk of the component can be automatically judged in a plurality of maintenance knowledge.
And establishing 4 entity nodes, csv files and 3 relation csv files according to the knowledge graph architecture and risk sequencing calculation result. The csv file stores one attribute per column and one entity per row. The csv file fromname and toname fields store related entity names for relationship matching, the rel field stores a relationship type, the other fields store relationship attributes, and each row stores a relationship. Import local files to Neo4j external repository: the 8 csv files are stored to the Neo4j/import folder for subsequent node and relationship importation.
Establishing a node: the nodes are imported and established by combining the LOAD CSV statement and the CREATE statement, taking a Component entity node AS an example, the node comprises a code attribute and a name attribute, a Cypher command [ LOAD CSV WITH HEADERS FROM "file:// R2tom. CSV" AS relation1MATCH (entry 1: Risk { name: relation1. relation }), (entry 2: Component { name: relation1. relation }) CREATE (entry 1) - [: connected _ with { R _ ue: relations.R _ value, R _ socket: relations.R _ socket } ] - > (entry 2) ] is input to CREATE the Component entity node.
Matching and establishing the relation: a node is imported, matched and established by combining a LOAD CSV statement, a MATCH statement and a CREATE statement, and a connected _ with relation is taken as an example, the connected _ with relation comprises an 'R _ sort' attribute and an 'R _ value' attribute, a Cypher command [ LOAD CSV WITH HEADERS FROM 'file:// R2tom. CSV' relation1MATCH (entry 1: Risk { name: relation1. front }), entry 2: Component { name: relation1. name }) CREATE (entry 1) -, connected _ with { R _ value: relations.R _ value, R _ sort: relations.R _ sort } ] - (entry 2) ].
In addition, when the relation attribute is established by the Csv file data read by the Neo4j, the default attribute value type is a string type, so that the SET statement is adopted to convert the attribute values of the string type of "R _ sort" and "R _ value" into an integer type and a floating point type respectively, and a Cypher command [ MATCH (: Component) - [ R: connected _ with ] - (: Risk) SET r.R _ sort = Integer (r.R _ sort) is input
MATCH (: Component) - [ r: connected _ with ] - (: Risk) SET r.R _ sort = tofoat (r.R _ sort) ]. And finally, building a comprehensive pipe gallery dynamic operation and maintenance risk knowledge graph, wherein the comprehensive pipe gallery dynamic operation and maintenance risk knowledge graph comprises 68 nodes and 824 relations.
(7) Risk querying and knowledge reasoning: and risk query and corresponding maintenance knowledge inference based on the comprehensive pipe gallery dynamic operation and maintenance risk knowledge graph are realized through a Cypher language.
Taking the drainage pipeline as an example, the risk analysis and knowledge reasoning can be realized as follows:
and (3) component-risk query, wherein relevant risk information of the specified component can be obtained by inputting Cypher sentences, and the risk information comprises risk indexes and risk sequences thereof: MATCH p = (: Component { name: 'drainage pipe') - [: connected _ with ] - >) RETURN p; in combination with the WHERE statement, the specified component risk ranking information may be queried based on the risk ranking execution condition: MATCH p = (: Component { name: 'drainage pipe') - [ r: connected _ with ] - >) WHERE r.R _ sort <4 RETURN p.
And (3) risk-component query, wherein risk information of the specified risk related components can be obtained by inputting Cypher sentences, and the risk information comprises component names and risk degree sequencing thereof: MATCH p = (: Risk { name:' pipeline cracking) - [: connected _ with ] - () RETURN p.
And (3) inquiring a component-risk-maintenance knowledge reasoning path, and inputting Cypher sentences to obtain maintenance knowledge of the specified risk of the specified component: MATCH p = (: Component { name: 'drainage pipe') - [: connected _ with ] - (: Risk { name: 'pipeline cracking') - [: be _ handled _ with ] - >) RETURN p.
The above description is only a preferred embodiment of the present invention, but the scope of the present invention is not limited thereto, and any modification and replacement based on the technical solution and inventive concept provided by the present invention should be covered within the scope of the present invention.
Claims (1)
1. A comprehensive pipe rack operation and maintenance dynamic risk evaluation method driven by monitoring data is characterized by comprising the following steps:
(1) establishing a pipe gallery FMEA risk evaluation mechanism: establishing a risk evaluation index according to the pipe gallery operation and maintenance risk classification, classifying pipe gallery components and equipment according to historical drawings, and establishing a risk evaluation mechanism based on an SOD scoring standard through FMEA risk analysis according to the risk evaluation index, the pipe gallery components and the equipment classification;
(2) accessing the sensing monitoring data into a risk evaluation mechanism: analyzing the sensing monitoring data to realize the conversion between the sensing monitoring data and the SOD scoring standard;
(3) evaluation cloud generation: realizing uncertainty conversion from evaluation data to a cloud model through a reverse cloud generator, and generating an expert evaluation cloud and a sensing monitoring cloud;
a reverse cloud generator:
；
in the formula, Ex is expectation, a typical value representing risk degree, En is entropy, representing risk discrete degree, He is super entropy, representing risk uncertainty measurement, when input data is expert evaluation value, N is expert number, x is expert evaluation valueiIs an evaluation value of the i-th expert,
the average value of expert evaluation is s, and the standard deviation of expert evaluation is s; when the input data is the sensing monitoring alarm times, N is the number of monitoring points, xiThe evaluation value of the corresponding occurrence degree of the alarm times of the ith monitoring point,
the average value of the evaluation values of the incidence degrees corresponding to the alarm times of the monitoring points is obtained, and s is the standard deviation of the evaluation values of the incidence degrees corresponding to the alarm times of the monitoring points;
and (3) evaluating cloud by experts: cp=(Exp O,Enp O,Hep O,Exp S,Enp SHep S,Exp D,Enp D,Hep D)，CpEvaluation of cloud, Ex for expertsp OFor experts to evaluate the expectation of the O score, Enp OEntropy of O-rating for expert evaluation, Hep OHyper-entropy, Ex, for expert evaluation of O-scoresp SFor experts to evaluate the expectation of the S score, Enp SEntropy of S-score for expert evaluation, Hep SFor experts to evaluate the hyper-entropy, Ex, of the S scorep DFor experts to evaluate the expectation of a D score, Enp DEvaluation of entropy of D-score for experts, Hep DEvaluating the super entropy of the D score for the expert;
sensing and monitoring cloud: cm=(Exm O,Enm O,Hem O,Exm S,Enm SHem S,Exm D,Enm D,Hem D)，CmMonitoring clouds, Ex for sensingm OFor the expectation of converting the sensing monitoring data into O score, Enm OEntropy, He, for conversion of sensory monitoring data into O-scoresm OFor the conversion of the sensed monitoring data into O-scored hyper-entropy, Exm SFor the expectation of converting the sensing monitoring data into S score, Enm SEntropy, He, for conversion of sensory monitoring data into S-scoresm SFor converting sensing monitoring data into super entropy, Ex of S scorem DFor the conversion of the sensory monitoring data into an expectation of D-score, Enm DEntropy, He, for conversion of sensory monitoring data into D-scoresm DConverting the sensing monitoring data into D-grade super entropy;
(4) and (3) risk data iteration: accessing the continuously updated sensing monitoring cloud and the original data through risk data iteration, updating the dynamic risk cloud, and iteratively correcting the evaluation data;
the risk data iteration is performed according to the following steps:
step 1: defining evaluation times t =0, sensing monitoring cloud input state values s =0, and initial dynamic risk cloud Cd=(Exd O,End O,Hed O,Exd S,End SHed S,Exd D,End D,Hed D) ，CdTo initiate a dynamic risk cloud, Exd OExpectation of O scoring for initial state data, End OEntropy, He, of O scores for initial State datad OHyper-entropy, Ex, scoring O for initial state datad SExpectation of scoring S for initial state data, End SEntropy, He, of S score for initial State datad SSuper entropy, Ex, scoring S for initial State datad DExpectation of scoring D for initial state data, End DEntropy, He, scoring D for initial state datad DThe initial state data is the super entropy of the D score;
step 2: when t =0, the expert evaluates cloud CpAssign value to dynamic Risk cloud Cd (t)Taking expert evaluation as initial evaluation data; when t is more than or equal to 1, using the cloud comprehensive algorithm result as the current dynamic risk cloud input value Cd (t)(ii) a If sensing monitoring cloud transmission existsChanging the state value s =1, and inputting the sensing monitoring cloud Cm (t)；
And step 3: if s =1, outputting C from step 2d (t)、Cm (t)Substituting into a cloud comprehensive algorithm;
and 4, step 4: assigning t = t +1, and outputting dynamic risk cloud Cd (t)Changing the state value s =0, and completing 1 dynamic risk cloud updating calculation;
and 5: when a new sensing monitoring cloud is input, substituting the output result of the step 4, and circularly executing the step 2-4 iterative computation to obtain a continuously updated dynamic risk cloud;
wherein iterating the tth dynamic risk cloud:
in the formula, Exd (t)To iterate the expectation of the tth dynamic risk cloud, End (t)Entropy of the dynamic risk cloud for the t-th iteration, Hed (t)To iterate the hyper-entropy, Ex, of the tth dynamic risk cloudd (t-1)To iterate the expectation of the dynamic risk cloud for the t-1 st time, End (t -1)Entropy of dynamic risk clouds for iteration t-1, Hed (t-1)To iterate the hyper entropy, Ex, of the dynamic risk cloud for the t-1 st times (t)For the expectation of the sensing monitoring cloud for the iteration t, Ens (t)Monitoring entropy of cloud, He, for sensing of t-th iterations (t)Monitoring the hyper-entropy of the cloud for the sensing of the t-th iteration;
(5) risk ranking: acquiring risk sequencing by using a cloud model gray correlation algorithm by taking the dynamic risk cloud as input;
(6) establishing a comprehensive pipe rack operation and maintenance risk knowledge map: the comprehensive pipe rack operation and maintenance risk knowledge graph comprises a mode layer and a data layer, wherein the mode layer of the comprehensive pipe rack operation and maintenance risk knowledge graph is established according to the mapping relation between components and risks by taking a pipe rack FMEA (failure mode and effects evaluation) mechanism as a core; collecting basic triple data according to a data framework of a mode layer and establishing a data layer, wherein the basic triple data comprise nodes, node attribute data, relations and relation attribute data; storing actual data files in a Neo4j database to establish a comprehensive pipe rack operation and maintenance risk knowledge map;
(7) risk querying and knowledge reasoning: and risk query and corresponding maintenance knowledge reasoning are carried out through the established operation and maintenance risk knowledge map of the comprehensive pipe rack.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110682404.3A CN113256160B (en) | 2021-06-21 | 2021-06-21 | Comprehensive pipe rack operation and maintenance dynamic risk evaluation method driven by monitoring data |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110682404.3A CN113256160B (en) | 2021-06-21 | 2021-06-21 | Comprehensive pipe rack operation and maintenance dynamic risk evaluation method driven by monitoring data |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN113256160A CN113256160A (en) | 2021-08-13 |
| CN113256160B true CN113256160B (en) | 2021-09-28 |
Family
ID=77188744
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202110682404.3A Active CN113256160B (en) | 2021-06-21 | 2021-06-21 | Comprehensive pipe rack operation and maintenance dynamic risk evaluation method driven by monitoring data |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN113256160B (en) |
Families Citing this family (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN117252449B (en) * | 2023-11-20 | 2024-01-30 | 水润天府新材料有限公司 | Full-penetration drainage low-noise pavement construction process and system |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2011150319A2 (en) * | 2010-05-28 | 2011-12-01 | Geostellar, Llc | System and method for geomatic modeling of a diverse resource base across broad landscapes |
| CN105096037A (en) * | 2015-07-16 | 2015-11-25 | 广东产品质量监督检验研究院 | Failure risk determination method of photovoltaic assembly |
| CN106920040A (en) * | 2017-03-01 | 2017-07-04 | 西南交通大学 | Freeway tunnel street accidents risks appraisal procedure based on Fuzzy Level Analytic Approach |
| CN109636110A (en) * | 2018-11-09 | 2019-04-16 | 中国电力科学研究院有限公司 | A kind of method and device obtaining protective relaying device operating status |
| CN112396358A (en) * | 2021-01-18 | 2021-02-23 | 西南交通大学 | Cloud model and fault tree ancient building risk analysis method based on entropy algorithm improvement |
-
2021
- 2021-06-21 CN CN202110682404.3A patent/CN113256160B/en active Active
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2011150319A2 (en) * | 2010-05-28 | 2011-12-01 | Geostellar, Llc | System and method for geomatic modeling of a diverse resource base across broad landscapes |
| CN105096037A (en) * | 2015-07-16 | 2015-11-25 | 广东产品质量监督检验研究院 | Failure risk determination method of photovoltaic assembly |
| CN106920040A (en) * | 2017-03-01 | 2017-07-04 | 西南交通大学 | Freeway tunnel street accidents risks appraisal procedure based on Fuzzy Level Analytic Approach |
| CN109636110A (en) * | 2018-11-09 | 2019-04-16 | 中国电力科学研究院有限公司 | A kind of method and device obtaining protective relaying device operating status |
| CN112396358A (en) * | 2021-01-18 | 2021-02-23 | 西南交通大学 | Cloud model and fault tree ancient building risk analysis method based on entropy algorithm improvement |
Non-Patent Citations (7)
| Title |
|---|
| A novel method of risk assessment based on cloud inference for natural gas pipelines;Guo, Yanbao 等;《JOURNAL OF NATURAL GAS SCIENCE AND ENGINEERING》;20160331;第30卷;第421-429页 * |
| Application of comprehensive relational grade theory in expert system of transformer fault diagnosis;Jianpo Li 等;《 2009 International Workshop on Intelligent Systems and Applications》;20091231;第1-4页 * |
| Flood risk assessment based on entropy weight and cloud model: A case study in Huaihe River Basin of China;Yugang Tian 等;《 2010 The 2nd Conference on Environmental Science and Information Application Technology》;20101231;第80-83页 * |
| 兰州地铁下穿黄河富水砂卵石层盾构掘进关键技术研究;杨成;《中国博士学位论文全文数据库 工程科技Ⅱ辑》;中国学术期刊（光盘版）电子杂志社;20200315(第3期);第C033-37页 * |
| 基于云理论的工程项目风险评价;崔楷;《河南科学》;20120430;第30卷(第4期);第526-528页 * |
| 基于数据挖掘的火灾分析模型及应用研究;谢道文;《中国博士学位论文全文数据库 工程科技Ⅰ辑》;中国学术期刊（光盘版）电子杂志社;20141215(第12期);第B026-3页 * |
| 运营病害隧道健康状态评估;吴剑飞;《中国优秀硕士学位论文全文数据库 工程科技Ⅱ辑》;中国学术期刊（光盘版）电子杂志社;20170715(第7期);第C034-291页 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN113256160A (en) | 2021-08-13 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| WO2021103492A1 (en) | Risk prediction method and system for business operations | |
| Yu et al. | A digital twin-based decision analysis framework for operation and maintenance of tunnels | |
| Karan et al. | BIM and GIS integration and interoperability based on semantic web technology | |
| Yang et al. | Data-driven accident consequence assessment on urban gas pipeline network based on machine learning | |
| Jiang et al. | Multi-ontology fusion and rule development to facilitate automated code compliance checking using BIM and rule-based reasoning | |
| CN113723632A (en) | Industrial equipment fault diagnosis method based on knowledge graph | |
| Yang et al. | Intelligent bridge management via big data knowledge engineering | |
| Li et al. | Dynamic early warning of rockburst using microseismic multi-parameters based on Bayesian network | |
| Zhou et al. | BIM and ontology-based knowledge management for dam safety monitoring | |
| Jiang et al. | Knowledge driven approach for smart bridge maintenance using big data mining | |
| Bloch | Connecting research on semantic enrichment of BIM-review of approaches, methods and possible applications. | |
| CN117273133A (en) | Construction method of multi-source heterogeneous data knowledge graph of power distribution network | |
| Jiang et al. | Semantic enrichment for BIM: Enabling technologies and applications | |
| CN115654381A (en) | Water supply pipeline leakage detection method based on graph neural network | |
| Du et al. | Clustering and ontology-based information integration framework for surface subsidence risk mitigation in underground tunnels | |
| CN116484024A (en) | Multi-level knowledge base construction method based on knowledge graph | |
| CN116644192B (en) | Knowledge graph construction method based on reliability of aircraft parts | |
| CN113256160B (en) | Comprehensive pipe rack operation and maintenance dynamic risk evaluation method driven by monitoring data | |
| Huang et al. | Dynamic seismic damage assessment of distributed infrastructure systems using graph neural networks and semi-supervised machine learning | |
| CN113449120B (en) | Pipeline safety comprehensive supervision method combining spatial information | |
| Yin | Big Data Analysis and Modeling Method of College Student Employment Management Based on Deep Learning Model | |
| CN117687824A (en) | Satellite fault diagnosis system based on quality problem knowledge graph | |
| Liu et al. | Knowledge Graph Construction and Application of Spatial Arrangement of Underground Powerhouse | |
| Zhou et al. | Development and Application of an Industry Foundation Classes‐Based Metro Protection Information Model | |
| CN116776984A (en) | Disaster chain reasoning prediction method and system based on knowledge graph |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |