CN112257974A - Gas lock well risk prediction model data set, model training method and application - Google Patents
Gas lock well risk prediction model data set, model training method and application Download PDFInfo
- Publication number
- CN112257974A CN112257974A CN202010938869.6A CN202010938869A CN112257974A CN 112257974 A CN112257974 A CN 112257974A CN 202010938869 A CN202010938869 A CN 202010938869A CN 112257974 A CN112257974 A CN 112257974A
- Authority
- CN
- China
- Prior art keywords
- well
- factors
- gate
- data
- risk
- 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.)
- Pending
Links
- 238000013058 risk prediction model Methods 0.000 title claims abstract description 47
- 238000000034 method Methods 0.000 title claims abstract description 39
- 238000012549 training Methods 0.000 title claims abstract description 32
- 238000004140 cleaning Methods 0.000 claims abstract description 17
- 238000012545 processing Methods 0.000 claims abstract description 14
- 230000002159 abnormal effect Effects 0.000 claims abstract description 9
- 238000005516 engineering process Methods 0.000 claims abstract description 7
- 238000011002 quantification Methods 0.000 claims abstract description 7
- 238000012216 screening Methods 0.000 claims abstract description 5
- 238000007619 statistical method Methods 0.000 claims abstract description 5
- 230000008030 elimination Effects 0.000 claims abstract description 4
- 238000003379 elimination reaction Methods 0.000 claims abstract description 4
- 238000000926 separation method Methods 0.000 claims abstract description 4
- 238000003860 storage Methods 0.000 claims description 17
- 238000012360 testing method Methods 0.000 claims description 15
- 238000011156 evaluation Methods 0.000 claims description 14
- 230000007613 environmental effect Effects 0.000 claims description 13
- 238000010801 machine learning Methods 0.000 claims description 11
- 238000010276 construction Methods 0.000 claims description 10
- 238000003915 air pollution Methods 0.000 claims description 9
- 230000003203 everyday effect Effects 0.000 claims description 7
- 239000007788 liquid Substances 0.000 claims description 7
- VNWKTOKETHGBQD-UHFFFAOYSA-N methane Chemical compound C VNWKTOKETHGBQD-UHFFFAOYSA-N 0.000 claims description 6
- 238000004590 computer program Methods 0.000 claims description 4
- 230000000694 effects Effects 0.000 claims description 4
- 238000013139 quantization Methods 0.000 claims description 4
- 230000009193 crawling Effects 0.000 claims description 3
- 238000013480 data collection Methods 0.000 claims description 3
- 238000011835 investigation Methods 0.000 claims description 3
- 239000010865 sewage Substances 0.000 claims description 3
- 238000009423 ventilation Methods 0.000 claims description 3
- 238000004364 calculation method Methods 0.000 claims 1
- 238000012544 monitoring process Methods 0.000 description 7
- 230000003287 optical effect Effects 0.000 description 6
- 238000010586 diagram Methods 0.000 description 4
- 238000004891 communication Methods 0.000 description 3
- 230000006870 function Effects 0.000 description 3
- 238000004519 manufacturing process Methods 0.000 description 3
- 238000013461 design Methods 0.000 description 2
- 238000012986 modification Methods 0.000 description 2
- 230000004048 modification Effects 0.000 description 2
- 239000013307 optical fiber Substances 0.000 description 2
- 230000002093 peripheral effect Effects 0.000 description 2
- 230000000644 propagated effect Effects 0.000 description 2
- 206010003497 Asphyxia Diseases 0.000 description 1
- 208000025174 PANDAS Diseases 0.000 description 1
- 208000021155 Paediatric autoimmune neuropsychiatric disorders associated with streptococcal infection Diseases 0.000 description 1
- 240000000220 Panda oleosa Species 0.000 description 1
- 235000016496 Panda oleosa Nutrition 0.000 description 1
- 231100000570 acute poisoning Toxicity 0.000 description 1
- 230000032683 aging Effects 0.000 description 1
- 238000003491 array Methods 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 230000005540 biological transmission Effects 0.000 description 1
- 238000007405 data analysis Methods 0.000 description 1
- 238000013136 deep learning model Methods 0.000 description 1
- 239000000835 fiber Substances 0.000 description 1
- 238000012423 maintenance Methods 0.000 description 1
- 238000005065 mining Methods 0.000 description 1
- 238000011158 quantitative evaluation Methods 0.000 description 1
- 238000012502 risk assessment Methods 0.000 description 1
- 239000004065 semiconductor Substances 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
- G06Q10/04—Forecasting or optimisation specially adapted for administrative or management purposes, e.g. linear programming or "cutting stock problem"
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/90—Details of database functions independent of the retrieved data types
- G06F16/95—Retrieval from the web
- G06F16/951—Indexing; Web crawling techniques
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/04—Architecture, e.g. interconnection topology
- G06N3/044—Recurrent networks, e.g. Hopfield networks
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/04—Architecture, e.g. interconnection topology
- G06N3/049—Temporal neural networks, e.g. delay elements, oscillating neurons or pulsed inputs
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/08—Learning methods
-
- 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
- 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
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)
- Entrepreneurship & Innovation (AREA)
- General Engineering & Computer Science (AREA)
- Development Economics (AREA)
- Data Mining & Analysis (AREA)
- Software Systems (AREA)
- Life Sciences & Earth Sciences (AREA)
- Molecular Biology (AREA)
- Mathematical Physics (AREA)
- General Health & Medical Sciences (AREA)
- Evolutionary Computation (AREA)
- Computational Linguistics (AREA)
- Biophysics (AREA)
- Biomedical Technology (AREA)
- Educational Administration (AREA)
- Artificial Intelligence (AREA)
- Computing Systems (AREA)
- Game Theory and Decision Science (AREA)
- Health & Medical Sciences (AREA)
- Marketing (AREA)
- Operations Research (AREA)
- Quality & Reliability (AREA)
- Tourism & Hospitality (AREA)
- General Business, Economics & Management (AREA)
- Databases & Information Systems (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The embodiment of the invention discloses a gas lock well risk prediction model data set, a model training method and application. The method comprises the following steps: obtaining internal data of the gate well safety accident emergency event within N years from an urban gas company; carrying out structuralization processing on internal data by using an unstructured data processing technology to obtain a structuralized sample; screening and deleting the items with missing fields and abnormal field values from the structured samples to serve as cleaning samples; performing statistical analysis on the cleaning sample, determining main risk factors of the gas lock well by combining separation mixing, combination approximation and elimination of irrelevant factors, and extracting 35 secondary indexes; quantitatively evaluating the safety state of the gate well in the cleaning sample to obtain 1 safety state quantitative score; and taking 35 risk factor index values and 1 safety state quantification score in the cleaning sample as a data set.
Description
Technical Field
The invention relates to the field of risk prediction, in particular to a gas lock well risk prediction model data set, a model training method and application.
Background
The gas transmission pipeline is an important infrastructure of cities and towns, and the number of the gas lock wells is distributed more and more along with the wider coverage area of a gas pipe network. The gas lock well belongs to a semi-closed limited space, and the operation in the limited space belongs to high-risk operation, so that the risk of acute poisoning and suffocation accidents is increased. And many gas lock wells are built for a long time, and certain dangerous hidden dangers exist in the working environment and the geographic environment. The pipeline and the valve in the gas lock well have long service time, poor working conditions and aging of instrument equipment; the safety distance between the gas pipeline and other pipelines with different pressures is insufficient, and safety measures are lacked; in the aspect of layout design, some old gas lock wells do not meet the safety design requirements. Therefore, it is necessary to establish a complete gas lock well safety evaluation system, to effectively predict the lock well risk, and to ensure the safety of the personnel in the gas lock well.
The gas lock well system is used as a system with uncertainty, multilevel and openness, a plurality of influencing factors are provided, and the relation among the factors is complicated. Traditional risk assessment methods such as an analytic hierarchy process, a fuzzy evaluation method and a fault tree method all depend too much on experience of researchers, and are strong in subjectivity and poor in prediction effect.
Disclosure of Invention
In recent years, various urban gas groups start to build and perfect a lock well management system in succession, electronic markers and various sensing terminals are installed in massive gas lock wells, real-time monitoring is carried out on the well covers, noise, liquid level, pressure, temperature and the like of the lock wells, and a lock well internet of things is formed by means of a narrow-band internet of things technology, so that real-time monitoring of the lock wells is realized. The gate well management system continuously accumulates the collected gate well related basic data and state data for a long time, and big data of gate well construction, operation and maintenance and management are formed, so that objective, scientific and effective basic data can be provided for risk prediction of the gas gate well.
The invention aims to provide a gas lock well risk prediction model data set, a lock well risk prediction model and an application model for early warning the safety state of a lock well to be evaluated every day so as to solve at least one of the problems in the prior art.
In order to achieve the purpose, the invention adopts the following technical scheme:
the invention provides a manufacturing method of a gas lock well risk prediction model data set, which comprises the following steps:
s10, obtaining internal data of the gate well safety accident emergency events within N years from the urban gas company;
s20, carrying out structuring processing on the internal data by using an unstructured data processing technology, and saving the internal data as a structured sample to a database;
s30, screening and deleting the items with missing fields and abnormal field values from the structured samples, and storing the items as cleaning samples in a database;
s40, performing statistical analysis on the cleaning sample, determining main risk factors of the gas lock well by combining separation mixed factors, combination approximate factors and elimination irrelevant factors, extracting 3 primary indexes of essential factors, environmental factors and state factors, and subdividing the 3 primary indexes into 35 secondary indexes;
s50, quantitatively evaluating the safety state of the gate well in the cleaning sample data to obtain 1 safety state quantitative score;
and S60, taking 35 index values of risk factors and 1 safety state quantization score in the cleaning sample data as a data set.
In one embodiment, the internal data of the emergency event records event base information, lockwell equipment information, and the direct cause of the event.
In a specific embodiment, the 3 primary indicators are subdivided into 35 secondary indicators:
the essential factors include: the depth of the gate well, the service life of the gate well, the complexity of facilities in the well, the pressure level of a pipe network to which the gate well belongs, whether the safe distance between a gas pipeline in the well and other pipelines with different pressures is sufficient, whether lighting facilities in the well meet the requirements, whether ventilation facilities in the well meet the requirements, whether a drainage well/a sewage well/a heat well exist within 10 meters of the periphery of the gate well, whether a septic tank/a methane tank exist within 10 meters of the periphery of the gate well, the accumulated settlement amount of the ground of the area where the gate well exists and the settlement rate of the ground of the area where the gate well exists;
environmental factors include: rainfall at the current day, the highest temperature at the current day, the lowest temperature at the current day, the air pollution index at the current day, rainfall before 1 day, the highest temperature before 1 day, the lowest temperature before 1 day, the air pollution index before 1 day, rainfall before 2 days, the highest temperature before 2 days, the lowest temperature before 2 days, the air pollution index before 2 days, holidays, strategic events and whether the area where the lock well is located is under construction;
the state factors include: whether the well lid is abnormal at the current day, the maximum value of the liquid level in the well at the current day, the minimum value of the liquid level in the well at the current day, the maximum value of the pressure in the well at the current day, the minimum value of the pressure in the well at the current day, the maximum value of the temperature in the well at the current day, the maximum value of the gas concentration in the well at the current day and the minimum value of the gas concentration in the,
the values of 11 essential factors and 9 state factors are extracted from the emergency event structured data, the values of 15 environmental factors are associated with external data according to the event occurrence time, and the external data is obtained by crawling from a website.
In one embodiment, the quantitative evaluation of the gate well safety state of the internal data of the gate well safety accident emergency event within N years, and the obtaining of the 1 safety state quantitative score comprises:
and (4) employing 10 experts related to the gas system, referring to the quantitative standard of the safety state level of the lock well by the experts, quantitatively scoring the safety state of each sample data, and taking the average value of the scoring results of the 10 experts as 1 quantitative score of the safety state of the corresponding sample.
In one embodiment, the lockwell security status level quantification criteria include:
based on the on-site investigation of a gas lock well system and the professional consultation of relevant personnel at all levels such as experts, technicians, constructors and safety personnel, the safety state of the lock well is divided into 5 levels of normal, slight fault, major fault, serious fault and catastrophic fault by combining the current gas pipeline management and operation system in China, and each level corresponds to a quantitative score.
In a second aspect, the present invention provides a method for performing gate risk prediction model training using the data set of the first aspect of the present invention, comprising:
s12, dividing the data set into a training set and a testing set;
s22, training the training set by using an LSTM machine learning model, taking 35 risk factor indexes as input and 1 safety state quantitative score as output, generating a gas lock well safety risk prediction model, and obtaining the relation between the risk factors and the safety state scores;
s32, carrying out model evaluation by using the test set, and if the evaluation effect is not ideal, subdividing the training set and the test set; and training and testing the model based on the new training set and the new testing set until the model is evaluated to be qualified.
In a third aspect, the present invention provides a method for applying a gate risk prediction model generated by the training method of the second aspect, including:
performing data collection and pretreatment on 35 risk factor indexes of the gate to be evaluated every day, using the risk factor indexes as the input of the gate risk prediction model, calculating to obtain corresponding safety state quantitative scores so as to perform early warning,
wherein, the environmental factors trigger the web crawler to crawl new external data every day by using the cunning device.
A fourth aspect of the present invention provides a computer device comprising:
one or more processors;
storage means for storing one or more programs;
when executed by the one or more processors, cause the one or more processors to implement a method as described in any one of the first, second and third aspects of the invention.
A fifth aspect of the invention provides a storage medium having stored thereon a computer program which, when executed by a processor, performs a method according to any one of the first, second and third aspects of the invention.
The invention has the following beneficial effects:
the invention provides a gas lock well risk prediction model data set, a model training method and application, which make full use of historical monitoring big data and a machine learning method of a lock well system, take the complexity of an urban gas lock well and the complexity of accident reasons into consideration, and combine with an expert fuzzy comprehensive evaluation method based on a fuzzy mathematical theory to provide a more objective, scientific and easy-to-operate gas lock well risk prediction model construction method combining the big data and the expert experience advantages. The gate well risk prediction model is used for quantifying the accident occurrence possibility and the consequence severity of a gate well, calculating and evaluating the risk level of the gate well, giving a risk prediction in advance and improving the accuracy and the effectiveness of the gate well risk prediction to a great extent.
Drawings
The following describes embodiments of the present invention in further detail with reference to the accompanying drawings.
Fig. 1 shows a diagram of a hardware architecture for implementing the method of the invention according to an embodiment of the present application.
FIG. 2 illustrates a flow chart of a method of making a gas lock risk prediction model dataset according to an embodiment of the present application.
FIG. 3 shows a schematic diagram of a lockwell risk prediction model based on the LSTM machine learning method according to one embodiment of the present application.
Fig. 4 shows a schematic structural diagram of a computer device for implementing the gas lock well risk prediction method of the present invention according to an embodiment of the present invention.
Detailed Description
In order to more clearly illustrate the invention, the invention is further described below with reference to preferred embodiments and the accompanying drawings. Similar parts in the figures are denoted by the same reference numerals. It is to be understood by persons skilled in the art that the following detailed description is illustrative and not restrictive, and is not to be taken as limiting the scope of the invention.
Fig. 1 illustrates an exemplary system architecture 100 to which embodiments of the gas lock risk prediction model dataset production method, the lock risk prediction model training method, or the prediction method using the lock risk prediction model of the present application may be applied.
As shown in fig. 1, the system architecture 100 may include a user terminal 101, a network 104, and a server 107. The network 104 is used to provide a medium for communication links between the user terminals 101 and the server 107. Network 104 may include various connection types, such as wired, wireless communication links, or fiber optic cables, among others.
A user may use the user terminal 101 to interact with the server 107 via the network 104 to create a gas lock risk prediction model dataset, train a lock risk prediction model, or predict using a lock risk prediction model, etc.
The user terminal 101 may be a variety of electronic devices with a display screen including, but not limited to, smart phones, tablets, laptop portable computers, desktop computers, and the like.
The server 107 may be a server providing various services, and the server 107 may analyze and process the received internal data of the emergency event of the gate well safety accident within N years, and feed back the processing result to the user terminal 101.
It should be noted that the number of the user terminals, the networks and the servers in fig. 1 is only schematic. There may be any number of terminal devices, networks, and servers, as desired for implementation.
Example one
As shown in fig. 2, an embodiment of the present invention provides a method for creating a gas lock risk prediction model data set, which in one example includes:
s10, obtaining internal data of the gate well safety accident emergency events within N years from the urban gas company;
in one particular embodiment, internal data for a lockwell safety accident emergency within N years is obtained from a city gas company. Generally, the data is better and better, for example, the safety accident data of various gas lock wells in nearly 3 years is selected.
In one embodiment, the internal data of the emergency event records the detailed information of the emergency event, including the basic information of the event, the information of the well-locking equipment and the direct reason of the event, and one emergency event report corresponds to the emergency event.
S20, carrying out structuring processing on the internal data by using an unstructured data processing technology, and saving the internal data as a structured sample to a database;
in a specific embodiment, the internal data is unstructured data stored in the form of word and pdf documents, the unstructured data is structured by using an unstructured data processing technology, and the unstructured data is stored in a database as a structured sample;
s30, screening and deleting the items with missing fields and abnormal field values from the structured samples, and storing the items as cleaning samples in a database;
in a specific embodiment, screening out entries with field missing and field value abnormal from the sample, deleting the entries with field missing and field value abnormal, and saving the entries in a database as cleaned samples;
s40, performing statistical analysis on the cleaning sample, determining main risk factors of the gas lock well by combining separation mixed factors, combination approximate factors and elimination irrelevant factors, extracting 3 primary indexes of essential factors, environmental factors and state factors, and subdividing the 3 primary indexes into 35 secondary indexes;
in a specific embodiment, a Pandas and StatsModels data analysis mining tool in Python is used for carrying out statistical analysis on a structured and cleaned sample, main risk factors of a gas lock well are determined by separating mixed factors, combining approximate factors and eliminating irrelevant factors, 3 primary indexes are extracted, and the 3 primary indexes can be subdivided into 35 secondary indexes;
in a specific embodiment, the 3 primary indicators are an essential factor, an environmental factor and a status factor, respectively.
The 3 primary indexes are subdivided into 35 secondary indexes:
the essential factors include: the depth of the gate well, the service life of the gate well, the complexity of facilities in the well, the pressure level of a pipe network to which the gate well belongs, whether the safe distance between a gas pipeline in the well and other pipelines with different pressures is sufficient, whether lighting facilities in the well meet the requirements, whether ventilation facilities in the well meet the requirements, whether a drainage well/a sewage well/a heat well exist within 10 meters of the periphery of the gate well, whether a septic tank/a methane tank exist within 10 meters of the periphery of the gate well, the accumulated settlement amount of the ground of the area where the gate well exists and the settlement rate of the ground of the area where the gate well exists;
environmental factors include: rainfall at the current day, the highest temperature at the current day, the lowest temperature at the current day, the air pollution index at the current day, rainfall before 1 day, the highest temperature before 1 day, the lowest temperature before 1 day, the air pollution index before 1 day, rainfall before 2 days, the highest temperature before 2 days, the lowest temperature before 2 days, the air pollution index before 2 days, holidays, strategic events and whether the area where the lock well is located is under construction;
the state factors include: whether the well lid is abnormal at the current day, the maximum value of the liquid level in the well at the current day, the minimum value of the liquid level in the well at the current day, the maximum value of the pressure in the well at the current day, the minimum value of the pressure in the well at the current day, the maximum value of the temperature in the well at the current day, the maximum value of the gas concentration in the well at the current day and the minimum value of the gas concentration in the,
the values of 11 essential factors and 9 state factors are extracted from the emergency event structured data, the values of 15 environmental factors are associated with external data according to the event occurrence time, and the external data can be obtained by crawling from a website by using a web crawler technology. Table 1 shows 3 primary indexes of the gate well risk prediction model subdivided into 35 secondary index risk factor field tables.
TABLE 1 Gate Risk prediction model 3 Primary index subdivision 35 Secondary index Risk factor field Table
As can be seen from table 1, the combination of the codes, types, and values corresponding to the 35 secondary indexes enables us to more intuitively know the internal data recording event basic information, the well locking equipment information, and the direct reason of the event of the emergency event.
S50, quantitatively evaluating the safety state of the gate well in the cleaning sample data to obtain 1 safety state quantitative score;
in a specific embodiment, the S50 includes:
and (4) employing 10 experts related to the gas system, referring to the quantitative standard of the safety state level of the lock well by the experts, quantitatively scoring the safety state of each sample data, and taking the average value of the scoring results of the 10 experts as 1 quantitative score of the safety state of the corresponding sample.
In one embodiment, the lockwell security status level quantification criteria include:
based on the on-site investigation of a gas lock well system and the professional consultation of relevant personnel at all levels such as experts, technicians, constructors and safety personnel, the safety state of the lock well is divided into 5 levels of normal, slight fault, major fault, serious fault and catastrophic fault by combining the current gas pipeline management and operation system in China, and each level corresponds to a quantitative score. Table 2 shows the lock well safety status level quantification criteria.
TABLE 2 Gate safety status level quantification Standard
| State classification | Is normal | Minor fault | Major failure | Major failure | Catastrophic failure |
| Quantifying a score | 1 | 2 | 3 | 4 | 5 |
As can be seen from table 2, the gate safety status is specified by 5 levels of quantified scores.
And S60, taking 35 risk factor index values and 1 safety state quantization score corresponding to the cleaning sample data as a data set.
In one embodiment, each sample datum corresponds to 35 risk factor indicator values and 1 security status quantification score. And (3) taking the 35 risk factor indexes as model input, and taking 1 safety state quantization score as model output to establish a data set.
Aiming at the existing problems, the invention provides a gas lock well risk prediction model data set, which considers the complexity of urban gas lock wells and the complexity of accident reasons, combines an expert fuzzy comprehensive evaluation method based on a fuzzy mathematical theory, quantifies the accident occurrence possibility and the consequence severity of the lock wells, calculates and evaluates the risk level of the lock wells, gives risk prediction in advance, and can improve the accuracy and the effectiveness of the lock well risk prediction to a great extent.
Example two
The method for performing gate risk prediction model training by using the data set of the first embodiment includes:
s12, dividing the data set into a training set and a testing set;
s22, training the training set by using an LSTM machine learning model, taking 35 risk factor indexes as input and 1 safety state quantitative score as output, generating a gas lock well safety risk prediction model, and obtaining the relation between the risk factors and the safety state scores;
s32, carrying out model evaluation by using the test set, and if the evaluation effect is not ideal, subdividing the training set and the test set; and training and testing the model based on the new training set and the new testing set until the model is evaluated to be qualified.
In a specific embodiment, as shown in fig. 3, the gate risk prediction model based on the LSTM machine learning method according to an embodiment of the present invention, only the influence of weather, construction, holidays, policy events, etc. on the gas gate risk of the current day can be learned, and the influence of these factors on the gas gate risk of the future days is difficult to learn, using the conventional machine learning and deep learning models. The LSTM model can well solve the problem, 11 indexes of the lock well intrinsic risk factors, 15 indexes of the lock well environment risk factors and 9 indexes of the lock well state risk factors are used as input, and the lock well safety risk score of the day, namely 1 safety state quantitative score, is used as output to obtain the gas lock well safety risk prediction model.
Aiming at the existing problems, the invention provides a method for carrying out gate well risk prediction model training by using the data set of the first embodiment, fully utilizes historical monitoring big data and a machine learning method of a gate well system, considers the complexity of an urban gas gate well and the complexity of accident reasons, and combines an expert fuzzy comprehensive evaluation method based on a fuzzy mathematical theory to provide a more objective, scientific and easy-to-operate gas gate well risk prediction model construction method combining big data and expert experience advantages. The gate well risk prediction model is used for quantifying the accident occurrence possibility and the consequence severity of a gate well, calculating and evaluating the risk level of the gate well, giving a risk prediction in advance and improving the accuracy of the gate well risk prediction to a great extent.
EXAMPLE III
The method for applying the gate risk prediction model generated by the training method of the second embodiment comprises the following steps:
performing data collection and pretreatment on 35 risk factor indexes of the gate to be evaluated every day, using the risk factor indexes as the input of the gate risk prediction model, calculating to obtain corresponding safety state quantitative scores so as to perform early warning,
wherein, the environmental factors trigger the web crawler to crawl new external data every day by using the cunning device.
Aiming at the existing problems, the invention provides a method for applying a lock well risk prediction model generated by using the training method of the second embodiment, fully utilizes historical monitoring big data and a machine learning method of a lock well system, considers the complexity of urban gas lock wells and the complexity of accident reasons, and combines an expert fuzzy comprehensive evaluation method based on a fuzzy mathematical theory to provide a more objective, scientific and easy-to-operate gas lock well risk prediction model construction method combining big data and expert experience advantages. The gate well risk prediction model is used for quantifying the accident occurrence possibility and the consequence severity of a gate well, calculating and evaluating the risk level of the gate well, giving a risk prediction in advance and improving the accuracy of the gate well risk prediction to a great extent.
Example four
As shown in fig. 4, an embodiment of the present invention provides a schematic structural diagram of a computer device, and the computer device 12 shown in fig. 4 is only an example and should not bring any limitation to the functions and the scope of the embodiment of the present invention.
As shown in FIG. 4, computer device 12 is in the form of a general purpose computing device. The components of computer device 12 may include, but are not limited to: one or more processors or processing units 16, a system memory 28, and a bus 18 that couples various system components including the system memory 28 and the processing unit 16.
The system memory 28 may include computer system readable media in the form of volatile memory, such as Random Access Memory (RAM)30 and/or cache memory 32. Computer device 12 may further include other removable/non-removable, volatile/nonvolatile computer system storage media. By way of example only, storage system 34 may be used to read from and write to non-removable, nonvolatile magnetic media (not shown in FIG. 4, and commonly referred to as a "hard drive"). Although not shown in FIG. 4, a magnetic disk drive for reading from and writing to a removable, nonvolatile magnetic disk (e.g., a "floppy disk") and an optical disk drive for reading from or writing to a removable, nonvolatile optical disk (e.g., a CD-ROM, DVD-ROM, or other optical media) may be provided. In these cases, each drive may be connected to bus 18 by one or more data media interfaces. Memory 28 may include at least one program product having a set (e.g., at least one) of program modules that are configured to carry out the functions of embodiments of the invention.
A program/utility 40 having a set (at least one) of program modules 42 may be stored, for example, in memory 28, such program modules 42 including, but not limited to, an operating system, one or more application programs, other program modules, and program data, each of which examples or some combination thereof may comprise an implementation of a network environment. Program modules 42 generally carry out the functions and/or methodologies of the described embodiments of the invention.
The processor unit 16 executes various functional applications and data processing, for example, implementing the method provided in the third embodiment, by running a program stored in the system memory 28.
Aiming at the existing problems, the invention provides computer equipment, provides a more objective, scientific and easy-to-operate gas lock well risk prediction model construction method combining big data and expert experience advantages by fully utilizing historical monitoring big data and a machine learning method of a lock well system, considering the complexity of an urban gas lock well and the complexity of accident reasons and combining an expert fuzzy comprehensive evaluation method based on a fuzzy mathematical theory. The gate well risk prediction model is used for quantifying the accident occurrence possibility and the consequence severity of a gate well, calculating and evaluating the risk level of the gate well, giving a risk prediction in advance and improving the accuracy of the gate well risk prediction to a great extent.
EXAMPLE five
Another embodiment of the present invention provides a storage medium having stored thereon a computer program which, when executed by a processor, implements the method as provided in the first, second and third embodiments above.
In practice, the computer-readable storage medium may take any combination of one or more computer-readable media. The computer readable medium may be a computer readable signal medium or a computer readable storage medium. A computer readable storage medium may be, for example, but not limited to, an electronic, magnetic, optical, electromagnetic, infrared, or semiconductor system, apparatus, or device, or any combination of the foregoing. More specific examples (a non-exhaustive list) of the computer readable storage medium would include the following: an electrical connection having one or more wires, a portable computer diskette, a hard disk, a Random Access Memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disc read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the foregoing. In the present embodiment, a computer readable storage medium may be any tangible medium that can contain, or store a program for use by or in connection with an instruction execution system, apparatus, or device.
A computer readable signal medium may include a propagated data signal with computer readable program code embodied therein, for example, in baseband or as part of a carrier wave. Such a propagated data signal may take many forms, including, but not limited to, electro-magnetic, optical, or any suitable combination thereof. A computer readable signal medium may also be any computer readable medium that is not a computer readable storage medium and that can communicate, propagate, or transport a program for use by or in connection with an instruction execution system, apparatus, or device.
Program code embodied on a computer readable medium may be transmitted using any appropriate medium, including but not limited to wireless, wireline, optical fiber cable, RF, etc., or any suitable combination of the foregoing.
Computer program code for carrying out operations for aspects of the present invention may be written in any combination of one or more programming languages, including an object oriented programming language such as Java, Smalltalk, C + + or the like and conventional procedural programming languages, such as the "C" programming language or similar programming languages. The program code may execute entirely on the user's computer, partly on the user's computer, as a stand-alone software package, partly on the user's computer and partly on a remote computer or entirely on the remote computer or server. In the case of a remote computer, the remote computer may be connected to the user's computer through any type of network, including a Local Area Network (LAN) or a Wide Area Network (WAN), or the connection may be made to an external computer (for example, through the Internet using an Internet service provider).
Aiming at the existing problems, the invention provides a storage medium, fully utilizes historical monitoring big data of a lock well system and a machine learning method, considers the complexity of the urban gas lock well and the complexity of accident reasons, combines an expert fuzzy comprehensive evaluation method based on a fuzzy mathematical theory, and provides a more objective, scientific and easy-to-operate gas lock well risk prediction model construction method combining the big data and the expert experience advantages. The gate well risk prediction model is used for quantifying the accident occurrence possibility and the consequence severity of a gate well, calculating and evaluating the risk level of the gate well, giving a risk prediction in advance and improving the accuracy of the gate well risk prediction to a great extent.
It should be understood that the above-mentioned embodiments of the present invention are only examples for clearly illustrating the present invention, and are not intended to limit the embodiments of the present invention, and it will be obvious to those skilled in the art that other variations or modifications may be made on the basis of the above description, and all embodiments may not be exhaustive, and all obvious variations or modifications may be included within the scope of the present invention.
Claims (9)
1. A method for making a gas lock well risk prediction model dataset is characterized by comprising the following steps:
s10, obtaining internal data of the gate well safety accident emergency events within N years from the urban gas company;
s20, carrying out structuring processing on the internal data by using an unstructured data processing technology, and saving the internal data as a structured sample to a database;
s30, screening and deleting the items with missing fields and abnormal field values from the structured samples, and storing the items as cleaning samples in a database;
s40, performing statistical analysis on the cleaning sample, determining main risk factors of the gas lock well by combining separation mixed factors, combination approximate factors and elimination irrelevant factors, extracting 3 primary indexes of essential factors, environmental factors and state factors, and subdividing the 3 primary indexes into 35 secondary indexes;
s50, quantitatively evaluating the safety state of the lock well in the cleaning sample to obtain 1 safety state quantitative score;
and S60, taking 35 index values of the risk factors and 1 safety state quantization score in the cleaning sample as a data set.
2. The method of claim 1, wherein the internal data of the emergency event records event base information, lockwell equipment information, and an event direct cause.
3. The method of claim 1, wherein the 3 primary indicators are subdivided into 35 secondary indicators:
the essential factors include: the depth of the gate well, the service life of the gate well, the complexity of facilities in the well, the pressure level of a pipe network to which the gate well belongs, whether the safe distance between a gas pipeline in the well and other pipelines with different pressures is sufficient, whether lighting facilities in the well meet the requirements, whether ventilation facilities in the well meet the requirements, whether a drainage well/a sewage well/a heat well exist within 10 meters of the periphery of the gate well, whether a septic tank/a methane tank exist within 10 meters of the periphery of the gate well, the accumulated settlement amount of the ground of the area where the gate well exists and the settlement rate of the ground of the area where the gate well exists;
environmental factors include: rainfall at the current day, the highest temperature at the current day, the lowest temperature at the current day, the air pollution index at the current day, rainfall before 1 day, the highest temperature before 1 day, the lowest temperature before 1 day, the air pollution index before 1 day, rainfall before 2 days, the highest temperature before 2 days, the lowest temperature before 2 days, the air pollution index before 2 days, holidays, strategic events and whether the area where the lock well is located is under construction;
the state factors include: whether the well lid is abnormal at the current day, the maximum value of the liquid level in the well at the current day, the minimum value of the liquid level in the well at the current day, the maximum value of the pressure in the well at the current day, the minimum value of the pressure in the well at the current day, the maximum value of the temperature in the well at the current day, the maximum value of the gas concentration in the well at the current day and the minimum value of the gas concentration in the,
the values of 11 essential factors and 9 state factors are extracted from the emergency event structured data, the values of 15 environmental factors are associated with external data according to the event occurrence time, and the external data is obtained by crawling from a website.
4. The method according to claim 1, wherein the S50 includes:
and based on the gas system related experts, performing quantitative safety state scoring on each cleaning sample data by referring to the safety state grade quantitative standard of the lock well, and taking the average value of scoring results as 1 safety state quantitative score of the corresponding sample.
5. The method of claim 4, wherein the lockwell safety status level quantification criteria comprises:
based on the on-site investigation of a gas lock well system and the professional consultation of relevant personnel at all levels such as experts, technicians, constructors and safety personnel, the safety state of the lock well is divided into 5 levels of normal, slight fault, major fault, serious fault and catastrophic fault by combining the current gas pipeline management and operation system in China, and each level corresponds to a quantitative score.
6. Method for performing a gate risk prediction model training using the data set of any of claims 1-5,
s12, dividing the data set into a training set and a testing set;
s22, training the training set by using an LSTM machine learning model, taking 35 risk factor indexes as input and 1 safety state quantitative score as output, generating a gas lock well safety risk prediction model, and obtaining the relation between the risk factors and the safety state scores;
s32, carrying out model evaluation by using the test set, and if the evaluation effect is not ideal, subdividing the training set and the test set; and training and testing the model based on the new training set and the new testing set until the model is evaluated to be qualified.
7. The method for predicting the gate well risk prediction model generated by the training method of claim 6 is characterized in that 35 risk factor indexes of a gate well to be evaluated are subjected to data collection and pretreatment every day and are used as input of the gate well risk prediction model, corresponding safety state quantitative scores are obtained through calculation, and therefore early warning is carried out,
wherein the environmental factors trigger the web crawler to crawl new external data every day by using a timer.
8. A computer device, comprising:
one or more processors;
storage means for storing one or more programs;
when executed by the one or more processors, cause the one or more processors to implement the method of any one of claims 1-7.
9. A storage medium on which a computer program is stored which, when being executed by a processor, carries out the method according to any one of claims 1-7.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010938869.6A CN112257974A (en) | 2020-09-09 | 2020-09-09 | Gas lock well risk prediction model data set, model training method and application |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010938869.6A CN112257974A (en) | 2020-09-09 | 2020-09-09 | Gas lock well risk prediction model data set, model training method and application |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN112257974A true CN112257974A (en) | 2021-01-22 |
Family
ID=74233087
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202010938869.6A Pending CN112257974A (en) | 2020-09-09 | 2020-09-09 | Gas lock well risk prediction model data set, model training method and application |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN112257974A (en) |
Cited By (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN114331022A (en) * | 2021-12-02 | 2022-04-12 | 国能网信科技(北京)有限公司 | Method and device for monitoring safety information of downhole operation and storage medium |
| CN114386797A (en) * | 2021-12-29 | 2022-04-22 | 天翼物联科技有限公司 | Internet of things card management and control method, system and device and storage medium |
| CN114693193A (en) * | 2022-06-02 | 2022-07-01 | 中国人民解放军海军工程大学 | Equipment scientific research project risk factor evaluation system and method |
| CN115099586A (en) * | 2022-06-10 | 2022-09-23 | 上海异工同智信息科技有限公司 | Method and device for identifying operation risk |
Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20180089389A1 (en) * | 2016-09-26 | 2018-03-29 | International Business Machines Corporation | System, method and computer program product for evaluation and identification of risk factor |
| CN108510194A (en) * | 2018-03-30 | 2018-09-07 | 平安科技（深圳）有限公司 | Air control model training method, Risk Identification Method, device, equipment and medium |
| US20180365229A1 (en) * | 2017-06-19 | 2018-12-20 | Vettd, Inc. | Systems and methods to determine and utilize semantic relatedness between multiple natural language sources to determine strengths and weaknesses |
| CN109523386A (en) * | 2018-10-18 | 2019-03-26 | 广东工业大学 | A kind of investment portfolio risk prediction technique of GMM in conjunction with LSTM |
| CN110967153A (en) * | 2019-12-04 | 2020-04-07 | 北京无线电计量测试研究所 | Automatic inspection system and method for gas lock well |
| US20200117903A1 (en) * | 2018-10-10 | 2020-04-16 | Autodesk, Inc. | Architecture, engineering and construction (aec) construction safety risk analysis system and method for interactive visualization and capture |
| CN111325455A (en) * | 2020-02-13 | 2020-06-23 | 中国安全生产科学研究院 | Limited space operation safety risk assessment system |
-
2020
- 2020-09-09 CN CN202010938869.6A patent/CN112257974A/en active Pending
Patent Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20180089389A1 (en) * | 2016-09-26 | 2018-03-29 | International Business Machines Corporation | System, method and computer program product for evaluation and identification of risk factor |
| US20180365229A1 (en) * | 2017-06-19 | 2018-12-20 | Vettd, Inc. | Systems and methods to determine and utilize semantic relatedness between multiple natural language sources to determine strengths and weaknesses |
| CN108510194A (en) * | 2018-03-30 | 2018-09-07 | 平安科技（深圳）有限公司 | Air control model training method, Risk Identification Method, device, equipment and medium |
| US20200117903A1 (en) * | 2018-10-10 | 2020-04-16 | Autodesk, Inc. | Architecture, engineering and construction (aec) construction safety risk analysis system and method for interactive visualization and capture |
| CN109523386A (en) * | 2018-10-18 | 2019-03-26 | 广东工业大学 | A kind of investment portfolio risk prediction technique of GMM in conjunction with LSTM |
| CN110967153A (en) * | 2019-12-04 | 2020-04-07 | 北京无线电计量测试研究所 | Automatic inspection system and method for gas lock well |
| CN111325455A (en) * | 2020-02-13 | 2020-06-23 | 中国安全生产科学研究院 | Limited space operation safety risk assessment system |
Non-Patent Citations (2)
| Title |
|---|
| 徐征: "燃气闸井安全评价指标体系研究", 《中国优秀硕士学位论文全文数据库 工程科技Ⅰ辑》, no. 1, pages 11 - 53 * |
| 陈毓飞: "燃气管道风险预测方法的研究与实现", 《中国优秀硕士学位论文全文数据库 工程科技Ⅱ辑》, no. 8, pages 15 - 30 * |
Cited By (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN114331022A (en) * | 2021-12-02 | 2022-04-12 | 国能网信科技(北京)有限公司 | Method and device for monitoring safety information of downhole operation and storage medium |
| CN114331022B (en) * | 2021-12-02 | 2024-10-11 | 国能数智科技开发(北京)有限公司 | Method, equipment and storage medium for monitoring downhole operation safety information |
| CN114386797A (en) * | 2021-12-29 | 2022-04-22 | 天翼物联科技有限公司 | Internet of things card management and control method, system and device and storage medium |
| CN114693193A (en) * | 2022-06-02 | 2022-07-01 | 中国人民解放军海军工程大学 | Equipment scientific research project risk factor evaluation system and method |
| CN115099586A (en) * | 2022-06-10 | 2022-09-23 | 上海异工同智信息科技有限公司 | Method and device for identifying operation risk |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN112257974A (en) | Gas lock well risk prediction model data set, model training method and application | |
| Wang et al. | Data-driven risk assessment on urban pipeline network based on a cluster model | |
| CN105678481A (en) | Pipeline health state assessment method based on random forest model | |
| CN116777223B (en) | Urban underground pipe network safety comprehensive risk assessment method and system | |
| CN112529327A (en) | Method for constructing fire risk prediction grade model of buildings in commercial areas | |
| CN111612301A (en) | Weight self-adjustment-based gas buried pipeline leakage risk assessment method and device | |
| CN113240211B (en) | Method and device for predicting wastewater discharge, computer equipment and storage medium | |
| CN113688169A (en) | Mine potential safety hazard identification and early warning system based on big data analysis | |
| CN201229289Y (en) | Corrosion predicting device | |
| CN115660513B (en) | Monitoring method and system based on hydraulic engineering aqueduct deformation | |
| Xu et al. | A framework for statistical analysis of water pipeline field performance data | |
| CN116720324A (en) | Traction substation key equipment fault early warning method and system based on prediction model | |
| CN116756825A (en) | Group structural performance prediction system for middle-small span bridge | |
| Reed et al. | Metrics as tools for bridging climate science and applications | |
| CN117591530B (en) | Data cross section processing method and system | |
| CN111737993B (en) | Method for extracting equipment health state from fault defect text of power distribution network equipment | |
| Holloway et al. | Exploring the use of water resource recovery facility instrument data to visualise dynamic resilience to environmental stressors | |
| CN115689118A (en) | Oil and gas pipeline risk evaluation method and system, storage medium and electronic equipment | |
| Weeraddana et al. | Utilizing machine learning to prevent water main breaks by understanding pipeline failure drivers | |
| CN116739391B (en) | Multi-system rail transit emergency collaborative decision-making method and device | |
| CN113656666B (en) | Relay protection fixed value data interaction method | |
| Kumar et al. | A hybrid framework for mining spatial characteristics of sewer defects from inspection databases | |
| Marinelli et al. | Investigation of earthmoving trucks deterioration using discriminant analysis | |
| CN118427736B (en) | Enterprise environmental health risk classification method and system based on environmental monitoring | |
| CN118626790A (en) | Power equipment fault data mining method and system in complex information environment |
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 |