CN105717887A - Oil gas processing station intelligent early warning system based on Internet of things sensing data - Google Patents

Oil gas processing station intelligent early warning system based on Internet of things sensing data Download PDF

Info

Publication number
CN105717887A
CN105717887A CN201410738580.4A CN201410738580A CN105717887A CN 105717887 A CN105717887 A CN 105717887A CN 201410738580 A CN201410738580 A CN 201410738580A CN 105717887 A CN105717887 A CN 105717887A
Authority
CN
China
Prior art keywords
early warning
warning
fault
module
flow process
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.)
Granted
Application number
CN201410738580.4A
Other languages
Chinese (zh)
Other versions
CN105717887B (en
Inventor
薛肖
陈红星
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
HONGYOU SOFTWARE Co.,Ltd.
Original Assignee
KARAMAY HONGYOU SOFTWARE CO Ltd
Priority date (The priority date 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 date listed.)
Filing date
Publication date
Application filed by KARAMAY HONGYOU SOFTWARE CO Ltd filed Critical KARAMAY HONGYOU SOFTWARE CO Ltd
Priority to CN201410738580.4A priority Critical patent/CN105717887B/en
Publication of CN105717887A publication Critical patent/CN105717887A/en
Application granted granted Critical
Publication of CN105717887B publication Critical patent/CN105717887B/en
Active legal-status Critical Current
Anticipated expiration legal-status Critical

Links

Classifications

    • YGENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
    • Y02TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
    • Y02PCLIMATE CHANGE MITIGATION TECHNOLOGIES IN THE PRODUCTION OR PROCESSING OF GOODS
    • Y02P90/00Enabling technologies with a potential contribution to greenhouse gas [GHG] emissions mitigation
    • Y02P90/02Total factory control, e.g. smart factories, flexible manufacturing systems [FMS] or integrated manufacturing systems [IMS]

Abstract

The invention discloses an oil gas processing station intelligent early warning system based on Internet of things sensing data, comprising a collection tool, a real time database, a data storage module, an early warning rule setting module, a state identification module, an early warning mode selection module, an early warning expert knowledge database, a forward deduction machine, an early warning module, a correction module, an early warning recording module and an updating module. The oil gas processing station intelligent early warning system based on the Internet of things sensing data can effectively solve the problem that monitoring parameters or instrument indicationsof a DCS (Distributed Control System) are manually observed and the healthy operation state can be determined according to the experience in the conventional oil gas processing station,can help the DCS to promptly know the production state, find the abnormity, know the potential hazard and the corresponding processing measures, and know the adjustment direction of the yield and the quality.

Description

Oil vapor treatment station intelligent early-warning system based on Internet of Things sensing data
Technical field
The invention belongs to oil gas technology of Internet of things field, be specifically related to a kind of oil vapor treatment station intelligent early-warning system based on Internet of Things sensing data.
Background technology
At present, in oil vapor treatment technological process, oil vapor treatment station adopts DCS (DistributedControlSystem, dcs) monitor in real time oil vapor treatment process.In prior art, when judging the operation health status of oil vapor treatment process, mainly take the manual observation DCS monitoring parameter obtained or instrument indicated value, judge again through experience.The subject matter of this kind of method existence is: owing to everyone experience is different, same phenomenon is judged that the result of understanding is incomplete same by different people, cause same phenomenon can be produced different results, it not the people enriched very much particularly with experience, it is difficult to accurately draw the operation health status of oil vapor treatment flow process, thus being unfavorable for the exception finding in oil vapor treatment flow process in time.
Summary of the invention
For the defect that prior art exists, the present invention provides a kind of oil vapor treatment station intelligent early-warning system based on Internet of Things sensing data, can effectively solve the problems referred to above.
The technical solution used in the present invention is as follows:
The present invention provides a kind of oil vapor treatment station intelligent early-warning system based on Internet of Things sensing data, including:
Sampling instrument, for gathering the Real-time Monitoring Data in the science and engineering skill everywhere of oil vapor treatment station;
Real-time data base, for the described Real-time Monitoring Data received and sampling instrument described in real-time storage collects;
Data memory module, for storing the Real-time Monitoring Data relevant to described early warning object that described real-time data base receives;
Early warning rule settings module, is used for setting early warning rule, and wherein, described early warning rule includes key parameter early warning rule, equipment early warning rule and flow process early warning rule;
State recognition module, described Real-time Monitoring Data for being stored by described data memory module is analyzed with the early warning rule set by described early warning rule settings module, judge whether described Real-time Monitoring Data exception occurs, if there is exception, then draw the fault that abnormal described Real-time Monitoring Data reacts;
Modes of warning selects module, for when described state recognition module identifies described fault, selecting this to need the modes of warning of early warning;Wherein, described modes of warning includes key parameter modes of warning, equipment modes of warning and flow process modes of warning;
Early warning expert knowledge library, for the corresponding relation with the form Stored Procedure of knowledge, equipment, key parameter, fault, failure cause and troubleshooting measure;
Direct Reason-Machine, selecting the modes of warning selected by module for receiving described modes of warning, based on described modes of warning, the fault identified with described state recognition module is for ergodic condition, travel through described early warning expert knowledge library, obtain the failure cause corresponding with described fault and troubleshooting measure;
Warning module, for described fault, described failure cause and described troubleshooting measure are integrated into early warning information, and carries out early warning;
Correcting module, for being modified the early warning of described warning module;
Early warning logging modle, for recording the early warning information being corrected, described early warning information includes the corresponding relation between fault, failure cause and troubleshooting measure;
More new module, for adding the early warning information that described early warning logging modle records to described early warning expert knowledge library.
Preferably, described flow process modes of warning refers to: the process flow operation state of each production system flow process that oil vapor treatment station is comprised carries out early warning;
Described equipment modes of warning refers to: the running status of the key equipment comprised in each production system flow process that oil vapor treatment station is comprised carries out early warning;
Described key parameter modes of warning refers to: the state of each crucial detection parameter of the key equipment in each production system flow process that oil vapor treatment station is comprised carries out early warning.
Preferably, described early warning expert knowledge library adopts the expression-form of multiway tree, stores described knowledge.
Preferably, described early warning expert knowledge library adopts the expression-form of multiway tree, stores described knowledge, particularly as follows:
It is referred to as root node with Oilfield Operation District name;
The process name comprised using described Oilfield Operation District title is as the 1st node layer;
The device name comprised using described process name is as the 2nd node layer;
It is referred to as the 3rd node layer with the key parameter name that described device name comprises;
It is referred to as the 4th node layer with the fault name that described key parameter title relates to;
The fault reason information related to using described fault title is as the 5th node layer;
The troubleshooting measure related to using described fault reason information is as the 6th node layer.
Preferably, described direct Reason-Machine specifically includes:
First spider module, for based on described modes of warning, the fault identified with described state recognition module, for ergodic condition, travels through described early warning expert knowledge library, obtains all possible failure cause corresponding with described fault;
Wherein, if being flow process modes of warning, the each production system flow process then oil vapor treatment station comprised carries out running status tracking, for each described production system flow process, any one production system flow process is designated as production system flow process B, obtain the key parameter values broken down of each equipment in production system flow process B, and obtain this key parameter values a plurality of failure cause when fault
If being key parameter modes of warning, by searching described early warning expert knowledge library, it is thus achieved that the described key parameter all possible breakdown reasons under described fault;
If being equipment modes of warning, then obtain the multinomial key parameter values of the device A needing early warning, to key parameter each described, all obtain this key parameter a plurality of failure cause when fault;
If being flow process modes of warning, the each production system flow process then oil vapor treatment station comprised carries out running status tracking, for each described production system flow process, any one production system flow process is designated as production system flow process B, obtain the key parameter values broken down of each equipment in production system flow process B, and obtain this key parameter values a plurality of failure cause when fault;
Logic decision condition memory module, for storing logic decision condition corresponding with every failure cause in described early warning expert knowledge library;Wherein, there is specific fault in the key parameter that described logic decision condition is associate device, and, there is certain logic relation in the fault between associate device, the process logic relation between this certain logic relation with associate device in technological process is consistent;
Second spider module, for all possible failure cause that described first spider module is obtained, is performed both by following operation:
The failure cause that any one is possible is designated as failure cause W, searches described logic decision condition memory module, it is thus achieved that the logic decision condition corresponding with described failure cause W, carry out obtaining the key parameter of associate device;Then, with the key parameter of described associate device for ergodic condition, travel through described early warning expert knowledge library, judge whether the key parameter of described associate device described specific fault occurs, and, whether the fault between associate device meets certain logic relation, if all met, then described failure cause W is physical fault reason;Otherwise, described failure cause W is not the fault actually occurred, and is rejected by failure cause W;
Thus obtain at least one physical fault reason;
Troubleshooting measure acquisition module, for according to described early warning expert knowledge library, it is thus achieved that the troubleshooting measure corresponding with described physical fault reason.
Beneficial effects of the present invention is as follows:
Oil vapor treatment station intelligent early-warning system based on Internet of Things sensing data provided by the invention, can effectively solve current oil vapor treatment station DCS manual observation monitoring parameter or instrument instruction, the problem that micro-judgment runs health status, can be used for assisting DCS to control system, it is made to understand production status much sooner, note abnormalities, know hidden danger and corresponding treatment measures, grasp the adjustment direction of output and quality.
Accompanying drawing explanation
Fig. 1 is the structural representation of the oil vapor treatment station intelligent early-warning system based on Internet of Things sensing data provided by the invention;
Fig. 2 is the big tank liquid level Rate Models figure of early warning provided by the invention;
Fig. 3 is for adopting the calculated liquid level changing trend diagram of method of least square provided by the invention;
Fig. 4 adopts the general calculated liquid level changing trend diagram of Multipoint movable average algorithm;
Fig. 5 is that in early warning expert knowledge library provided by the invention, the figure naturally described of the knowledge of storage shows schematic diagram;
Fig. 6 is early warning representation of knowledge class figure provided by the invention;
Fig. 7 is failure cause decision tree diagram provided by the invention;
Fig. 8 is that failure cause provided by the invention judges class figure.
Detailed description of the invention
Below in conjunction with accompanying drawing, the present invention is described in detail:
As it is shown in figure 1, the present invention provides a kind of oil vapor treatment station intelligent early-warning system based on Internet of Things sensing data, including:
(1) sampling instrument:
For gathering the Real-time Monitoring Data in the science and engineering skill everywhere of oil vapor treatment station;
Concrete acquisition mode is:
The key equipment of oil vapor treatment station science and engineering skill everywhere is mounted on sensor, by the parameter that sensor acquisition is relevant.
Such as, oil vapor treatment station includes multiple production system flow process, e.g., and crude system flow process, steam generator system flow process, flood pattern flow process etc.;For often overlapping production system flow process, all include multiple equipment, for instance, for crude system flow process, including settling tank, separator and purification oil tank etc.;And for each equipment, include again multiple key parameter, for instance, for separator, by installing liquid level sensor and pressure transducer, the key parameter of detection is liquid level and pressure.
Therefore, in the science and engineering skill everywhere of oil vapor treatment station, by Internet of Things sensing equipment, the Real-time Monitoring Data of each monitoring point of Real-time Collection, and the Real-time Monitoring Data collected is transferred in Central Control Room SCADA system;Then, early warning system of the present invention is provided with sampling instrument, and the Real-time Monitoring Data in Central Control Room SCADA system is collected in real-time data base by this sampling instrument by OPC agreement.
(2) real-time data base, for storing the described Real-time Monitoring Data that described sampling instrument collects;
(3) data memory module, for storing the Real-time Monitoring Data relevant to described early warning object that described real-time data base receives;
Herein, it is emphasized that, for improving the real-time of early warning and monitoring, real-time data base only stores real time data, for instance, the life cycle arranging every Real-time Monitoring Data is 1 second, real-time data base is when, after certain Real-time Monitoring Data receiving sampling instrument transmission, the holding time is only 1 second;When reaching the holding time, namely delete this Real-time Monitoring Data.
Owing to real-time data base is all process techniques in whole oil vapor treatment station, early warning system provided by the invention, can concurrently to the multiple system flows in this oil vapor treatment station, or multiple equipment, or multiple key parameters carry out early warning, therefore, for certain early warning object, need from real-time data base, obtain the Real-time Monitoring Data relevant to this early warning object, such as, if oil vapor treatment station includes 3 production system flow processs, respectively crude system flow process, steam generator system flow process, flood pattern flow process, if user has only to this crude system flow process is carried out early warning and monitoring, all Real-time Monitoring Datas relevant to crude system flow process are stored data memory module by having only to, Real-time Monitoring Data without storage other system flow process.
(4) early warning rule settings module, is used for setting early warning rule, and wherein, described early warning rule includes key parameter early warning rule, equipment early warning rule and flow process early warning rule;
(5) state recognition module, described Real-time Monitoring Data for being stored by described data memory module is analyzed with the early warning rule set by described early warning rule settings module, judge whether described Real-time Monitoring Data exception occurs, if there is exception, then draw the fault that abnormal described Real-time Monitoring Data reacts;
Such as, for key parameter early warning rule, for key parameter for liquid level, the liquid level early warning rule of setting is for including liquid level limit (threshold value) early warning and liquid level variation tendency early warning.
With reference to Fig. 2, for the big tank liquid level Rate Models figure of early warning, figure it is seen that pre-set the normal rate of climb scope of liquid level, the highest liquid level and minimum liquid level;Then: the real-time level of the big tank liquid level received is identified, if real-time level exceeds the highest liquid level, then identify liquid level and superelevation fault occurs;If real-time level is lower than minimum liquid level, then identifies liquid level and occur super for fault;Again the rate of climb of the real-time level received is calculated, it is judged that whether the liquid level rate of climb belongs to the normal rate of climb scope of liquid level, if lower than the normal rate of climb scope of liquid level, then identifies liquid level and constant fault occurs.
Wherein, variation tendency early warning is one of relatively conventional computational methods of early warning system, owing to amplitude and the frequency of real time data change do not have rule, so, in prior art, the slope that the multi-point average algorithm generally adopted calculates there will be serious unstability.After tested, method of least square can efficiently solve this problem.
Referring now still to Fig. 2, it can be seen that LINEST function can pass through the straight line using method of least square to calculate with available data best fit, calculates the statistical value of certain straight line, is then back to describe the array of this straight line.LINEST can also be combined with other functions use, then calculate the statistical value of other kinds of linear model in unknown parameter, including multinomial, logarithm, index and power series.Because this function returns numerical value array, so must input with the form of Array Formula.
The formula of straight line is:
Y=mx+b
-or-
Y=m1x1+m2x2+...+b (if having the x value in multiple region)
Wherein, dependent variable y is the functional value of independent variable x.M value is the coefficient corresponding with each x value, and b is constant.Noting, y, x and m can be vectors.LINEST function return array be mn, mn-1 ..., m1, b}.LINEST function also can return to additional regression calculation value.
Grammer is:
LINEST(known_y's,[known_x's],[const],[stats])
LINEST function grammer has following parameter:
Known_y's is required.Y value set known in relational expression y=mx+b.
If cell region corresponding for known_y's is in independent string, then every string of known_x's is considered an independent variable.
If cell region corresponding for known_y's is in independent a line, then every a line of known_x's is considered an independent variable.
Known_x's is optional.X value set known in relational expression y=mx+b.
Cell region corresponding for known_x's can comprise one or more groups variable.If only using a variable, then as long as known_y's and known_x's has identical dimension, then they can be the region of any shape.If using multiple variable, then known_y's is necessary for vector (being namely necessary for a row or column).
If omit known_x's, it assumes that this array be 1,2,3 ... }, its size is identical with known_y's.
Const is optional.One logical value, is used for designating whether to force to be set to 0 by constant b.
If const is TRUE or is omitted, b will calculate in usual manner.
If const is FALSE, b will be set as 0, and adjustment m value makes y=mx simultaneously.
Stats is optional.One logical value, is used for designating whether to return additional regression calculation value.
If stats is TRUE, then LINEST function return additional regression calculation value, the array at this moment returned for mn, mn-1 ..., m1, b;Sen, sen-1 ..., se1, seb;R2, sey;F, df;Ssreg, ssresid}.
If stats is FALSE or is omitted, LINEST function drop-off to pick-up radio m and constant b.
With reference to Fig. 3, for adopting the calculated liquid level changing trend diagram of method of least square provided by the invention;With reference to Fig. 4, for adopting the calculated liquid level changing trend diagram of general Multipoint movable average algorithm;Comparison diagram 3 and Fig. 4, it will be evident that the liquid level changing trend diagram that the present invention obtains more can react the overall trend of liquid level change.
Again such as, in practical application, following early warning rule can be set:
1) separator pressure < 0.22
2) liquid level superelevation
3) liquid level does not change more than 3 minute datas
4) in sewerage flow process, 3#, 4#, 5# force (forcing) pump can not stop, and its effusion meter FIC302 not can exceed that one hour downtime, and otherwise pipeline can coagulate.Know whether pipeline stops by effusion meter and Pressure gauge
5) stator temperature not can exceed that 107 DEG C and lubricating oil temperature not can exceed that 70 DEG C
6) if the liquid level rate of climb is less than 0.2 m/h (variation tendency warning), such as 0.1 m/h, early warning.
Visible, most early warning rules can be attributed to most basic multilevel iudge, namely, decision content whether more than, less than, equal to, be more than or equal to, less than or equal to etc. comparison operation, for 3) data for change, typically represent liquid level can one only small interval in fluctuation, it is possible to regard superelevation and ultralow warning as, for 4) in judge that force (forcing) pump stopping can be changed to flow less than certain value and pressure less than certain value.For 6) warning of liquid level climbing speed, wherein comprise the calculating to speed and the comparison to result of calculation, function using rate calculations as a function, can be regarded as a master data source by us.
(6) modes of warning selects module, for when described state recognition module identifies described fault, selecting this to need the modes of warning of early warning;Wherein, described modes of warning includes key parameter modes of warning, equipment modes of warning and flow process modes of warning;
Wherein, described flow process modes of warning refers to: the process flow operation state of each production system flow process that oil vapor treatment station is comprised carries out early warning;
Described equipment modes of warning refers to: the running status of the key equipment comprised in each production system flow process that oil vapor treatment station is comprised carries out early warning;
Described key parameter modes of warning refers to: the state of each crucial detection parameter of the key equipment in each production system flow process that oil vapor treatment station is comprised carries out early warning.
(7) early warning expert knowledge library, for the corresponding relation with the form Stored Procedure of knowledge, equipment, key parameter, fault, failure cause and troubleshooting measure;
With reference to Fig. 5, show schematic diagram for the figure naturally described of the knowledge of storage in early warning expert knowledge library;In the present invention, early warning expert knowledge library adopts the expression-form of multiway tree, stores described knowledge, particularly as follows:
It is referred to as root node with Oilfield Operation District name;
The process name comprised using described Oilfield Operation District title is as the 1st node layer;
The device name comprised using described process name is as the 2nd node layer;
It is referred to as the 3rd node layer with the key parameter name that described device name comprises;
It is referred to as the 4th node layer with the fault name that described key parameter title relates to;
The fault reason information related to using described fault title is as the 5th node layer;
The troubleshooting measure related to using described fault reason information is as the 6th node layer.
Such as Fig. 6, for early warning representation of knowledge class figure of the present invention;From Fig. 6, it can be seen that the degree of depth and the range of early warning expert knowledge library have uncertainty, from tree root to leaves, every node layer all corresponding 1 arrives multiple child nodes.The reasoning process of inference machine, is converted to the Traversal Problem to knowledge tree.
(8) direct Reason-Machine, the modes of warning selected by module is selected for receiving described modes of warning, based on described modes of warning, the fault identified with described state recognition module is for ergodic condition, travel through described early warning expert knowledge library, obtain the failure cause corresponding with described fault and troubleshooting measure;
Described direct Reason-Machine specifically includes:
First spider module, for based on described modes of warning, the fault identified with described state recognition module, for ergodic condition, travels through described early warning expert knowledge library, obtains all possible failure cause corresponding with described fault;
Wherein, if being flow process modes of warning, the each production system flow process then oil vapor treatment station comprised carries out running status tracking, for each described production system flow process, any one production system flow process is designated as production system flow process B, obtain the key parameter values broken down of each equipment in production system flow process B, and obtain this key parameter values a plurality of failure cause when fault
If being key parameter modes of warning, by searching described early warning expert knowledge library, it is thus achieved that the described key parameter all possible breakdown reasons under described fault;
If being equipment modes of warning, then obtain the multinomial key parameter values of the device A needing early warning, to key parameter each described, all obtain this key parameter a plurality of failure cause when fault;
If being flow process modes of warning, the each production system flow process then oil vapor treatment station comprised carries out running status tracking, for each described production system flow process, any one production system flow process is designated as production system flow process B, obtain the key parameter values broken down of each equipment in production system flow process B, and obtain this key parameter values a plurality of failure cause when fault;
Logic decision condition memory module, for storing logic decision condition corresponding with every failure cause in described early warning expert knowledge library;Wherein, there is specific fault in the key parameter that described logic decision condition is associate device, and, there is certain logic relation in the fault between associate device, the process logic relation between this certain logic relation with associate device in technological process is consistent;
Second spider module, for all possible failure cause that described first spider module is obtained, is performed both by following operation:
The failure cause that any one is possible is designated as failure cause W, searches described logic decision condition memory module, it is thus achieved that the logic decision condition corresponding with described failure cause W, carry out obtaining the key parameter of associate device;Then, with the key parameter of described associate device for ergodic condition, travel through described early warning expert knowledge library, judge whether the key parameter of described associate device described specific fault occurs, and, whether the fault between associate device meets certain logic relation, if all met, then described failure cause W is physical fault reason;Otherwise, described failure cause W is not the fault actually occurred, and is rejected by failure cause W;
Thus obtain at least one physical fault reason;
Troubleshooting measure acquisition module, for according to described early warning expert knowledge library, it is thus achieved that the troubleshooting measure corresponding with described physical fault reason.
Concrete, in early warning expert knowledge library, for a fault, it is possible to there is a plurality of failure cause, for instance, for separator this fault of liquid level superelevation, the possible breakdown reason existed includes: separator pressure is low, plugged filter and pneumatic operated valve fault;At this time, it may be necessary to by logic decision condition, the possible failure cause obtained is screened, obtains physical fault reason.Such as, for separator pressure this failure cause low, the step carrying out failure cause investigation is: logic decision condition memory module exists the logic decision condition of each failure cause, and logic decision condition as corresponding in: separator pressure this failure cause low can be: separator liquid level is ultralow more than and separator pressure;Therefore, after judging this logic decision condition, traversal early warning expert knowledge library again, find separator current pressure values, and judge whether separator current pressure values now pressure this fault ultralow occurs, if there is, then show the physical fault reason that separator pressure this failure cause low is separator liquid level this fault of superelevation.
In practical application, the difference according to modes of warning, set logic decision condition can be extremely complex, more complicated logic decision condition is the logic decision of multiple basic comparison combination, e.g., and (X > 1) AND (y=0)), such as Fig. 7, for failure cause decision tree diagram;Such as Fig. 8, judge class figure for failure cause;For the ease of the more complicated failure cause logic decision condition being likely to occur in the future, adopt decision tree representation to represent failure cause logic decision condition, will substantially compare as leaf node.Such as expression formula (TI101>50) AND ((LI101>100) OR (LI101<10)), it is possible to represent with decision tree.
For equipment modes of warning and flow process modes of warning, can be generally designated as:
If being equipment modes of warning, then obtain the multinomial key parameter values of the device A needing early warning, to key parameter each described, all obtain this key parameter a plurality of failure cause when fault;Then, based on process logic relation between every key parameter of this device A in technological process, by failure cause logic decision condition, the failure cause obtained is screened, remove irrational failure cause, obtain the physical fault reason of accurate described device A, then find the troubleshooting measure of physical fault reason;
If being flow process modes of warning, the each production system flow process then oil vapor treatment station comprised carries out running status tracking, for each described production system flow process, any one production system flow process is designated as production system flow process B, obtain the key parameter values broken down of each equipment in production system flow process B, and obtain this key parameter values a plurality of failure cause when fault;Then, based on the logical relation between equipment each in production system flow process and, according to logical relation between each key parameter in same equipment, by failure cause logic decision condition, the failure cause obtained is screened, remove irrational failure cause, obtain the physical fault reason of accurate production system flow process B, then find the troubleshooting measure of physical fault reason.
(9) warning module, for described fault, described failure cause and described troubleshooting measure are integrated into early warning information, and carries out early warning;
(10) correcting module, for being modified the early warning of described warning module;
(11) early warning logging modle, for recording the early warning information being corrected, described early warning information includes the corresponding relation between fault, failure cause and troubleshooting measure;
(12) more new module, for adding the early warning information that described early warning logging modle records to described early warning expert knowledge library.
At present, DCS control system automatically controls industry in the world and belongs to dominant systems, oil vapor treatment station intelligent early-warning system based on Internet of Things sensing data provided by the invention is used for assisting DCS to control system, it is made to understand production status much sooner, note abnormalities, know hidden danger and corresponding treatment measures, grasp the adjustment direction of output and quality.
For certain oil vapor treatment station, crude system flow process is that oily entrance 1#~4# vapour liquid separator separation oil and gas is carried out in five oil collecting zones such as A, B, C, D, E, gas enters natural gas processing system, liquid mixes heat through one section, enter one section of settling tank (7#, 8#), two-stage nitration settling tank (5#, 6#) is entered, subsequently into purifying oil tank (1#, 2#, 3#, 4#, 9#, 10#) then through two-stage nitration heating and agent-feeding treatment.If crude oil purification is not up to standard, by pump, the liquid purifying oil tank is squeezed into two-stage nitration, again settle, purify oil up to standard outer defeated.
For above-mentioned oil vapor treatment station, to the process of crude system flow process early warning it is:
(1) gather separator, purify the Real-time Monitoring Data such as oil tank, a section of settling tank, two-stage nitration settling tank;
(2) crude system flow process health degree is calculated based on following formula:
Crude system flow process health degree=Σ function (equipment i) * weight i;Σ weight i=1
(3) set early warning rule, to purify oil tank, set following early warning rule:
Liquid level limit (threshold value) early warning, liquid level variation tendency early warning, level imbalance early warning.
For liquid level variation tendency early warning, adopting method of least square to calculate and purify oil tank level rate of change, judge to purify oil tank state (oil-feed, outer defeated, draining, standing) according to rate setting scope, following table is to purify oil tank parameter attribute information:
(4) based on early warning rule, intelligent early-warning is carried out:
Purifying oil tank level and allow the highest 10 meters, the low liquid level of run duration is generally 1.5 meters (out-of-limit early warning), average oil-feed liquid level rate of climb 0.2-0.4 m/h (variation tendency early warning, out-of-limit early warning), and slack tank enters completely about needs 25 hours.In actual motion, if oil-feed liquid level rising speed oil frequent fluctuation (deviation pre-alert) between 0.2 m/h and 0.4 m/h, analyzing possible origin liquid pipeline and infusion pump causes extremely, Field Force carries out investigation process;If the liquid level rate of climb is more than 0.4 m/h (variation tendency early warning, out-of-limit early warning), such as 0.6 m/h, it was predicted that 12.5 hours canfuls, early warning, analyzing the possible reasons such as origin liquid measure is excessive, settling tank water outlet is not smooth to cause, Field Force carries out investigation process;If liquid level rising speed oil is less than 0.2 m/h (variation tendency early warning, out-of-limit early warning), such as 0.1 m/h, early warning, analyze reasons such as being likely to too small, the device pipeline breakage of origin liquid measure and cause, Field Force carries out investigation process.
The present invention also provides for a kind of oil vapor treatment station intelligent early-warning method, comprises the following steps:
After step 1, warning module start, monitoring object is carried out on-line monitoring, obtains Real-time Monitoring Data;Then, for the Real-time Monitoring Data received, all early warning rules of pre-stored are scanned for, it is judged that whether Real-time Monitoring Data meets certain early warning rule;If it is satisfied, then generation early warning, and early warning information is pushed to early warning subscription response module;
Step 2, early warning subscription response module judges whether currently this monitoring object to be reported to the police, if it is, release early warning, and records the releasing time;
Step 3, warning module is in running, if receiving the message that early warning expert knowledge library is modified, then updates early warning expert knowledge library, and proceeds early warning and monitoring.
In a word, oil vapor treatment station intelligent early-warning system based on Internet of Things sensing data provided by the invention, can effectively solve current oil vapor treatment station DCS manual observation monitoring parameter or instrument instruction, the problem that micro-judgment runs health status, can be used for assisting DCS to control system so that it is understand production status much sooner, it has been found that abnormal, know hidden danger and corresponding treatment measures, grasp the adjustment direction of output and quality.
The above is only the preferred embodiment of the present invention; it should be pointed out that, for those skilled in the art, under the premise without departing from the principles of the invention; can also making some improvements and modifications, these improvements and modifications also should look protection scope of the present invention.

Claims (5)

1. the oil vapor treatment station intelligent early-warning system based on Internet of Things sensing data, it is characterised in that including:
Sampling instrument, for gathering the Real-time Monitoring Data in the science and engineering skill everywhere of oil vapor treatment station;
Real-time data base, for the described Real-time Monitoring Data received and sampling instrument described in real-time storage collects;
Data memory module, for storing the Real-time Monitoring Data relevant to described early warning object that described real-time data base receives;
Early warning rule settings module, is used for setting early warning rule, and wherein, described early warning rule includes key parameter early warning rule, equipment early warning rule and flow process early warning rule;
State recognition module, described Real-time Monitoring Data for being stored by described data memory module is analyzed with the early warning rule set by described early warning rule settings module, judge whether described Real-time Monitoring Data exception occurs, if there is exception, then draw the fault that abnormal described Real-time Monitoring Data reacts;
Modes of warning selects module, for when described state recognition module identifies described fault, selecting this to need the modes of warning of early warning;Wherein, described modes of warning includes key parameter modes of warning, equipment modes of warning and flow process modes of warning;
Early warning expert knowledge library, for the corresponding relation with the form Stored Procedure of knowledge, equipment, key parameter, fault, failure cause and troubleshooting measure;
Direct Reason-Machine, selecting the modes of warning selected by module for receiving described modes of warning, based on described modes of warning, the fault identified with described state recognition module is for ergodic condition, travel through described early warning expert knowledge library, obtain the failure cause corresponding with described fault and troubleshooting measure;
Warning module, for described fault, described failure cause and described troubleshooting measure are integrated into early warning information, and carries out early warning;
Correcting module, for being modified the early warning of described warning module;
Early warning logging modle, for recording the early warning information being corrected, described early warning information includes the corresponding relation between fault, failure cause and troubleshooting measure;
More new module, for adding the early warning information that described early warning logging modle records to described early warning expert knowledge library.
2. the oil vapor treatment station intelligent early-warning system based on Internet of Things sensing data according to claim 1, it is characterised in that described flow process modes of warning refers to: the process flow operation state of each production system flow process that oil vapor treatment station is comprised carries out early warning;
Described equipment modes of warning refers to: the running status of the key equipment comprised in each production system flow process that oil vapor treatment station is comprised carries out early warning;
Described key parameter modes of warning refers to: the state of each crucial detection parameter of the key equipment in each production system flow process that oil vapor treatment station is comprised carries out early warning.
3. the oil vapor treatment station intelligent early-warning system based on Internet of Things sensing data according to claim 1, it is characterised in that described early warning expert knowledge library adopts the expression-form of multiway tree, stores described knowledge.
4. the oil vapor treatment station intelligent early-warning system based on Internet of Things sensing data according to claim 3, it is characterised in that described early warning expert knowledge library adopts the expression-form of multiway tree, stores described knowledge, particularly as follows:
It is referred to as root node with Oilfield Operation District name;
The process name comprised using described Oilfield Operation District title is as the 1st node layer;
The device name comprised using described process name is as the 2nd node layer;
It is referred to as the 3rd node layer with the key parameter name that described device name comprises;
It is referred to as the 4th node layer with the fault name that described key parameter title relates to;
The fault reason information related to using described fault title is as the 5th node layer;
The troubleshooting measure related to using described fault reason information is as the 6th node layer.
5. the oil vapor treatment station intelligent early-warning system based on Internet of Things sensing data according to claim 1, it is characterised in that described direct Reason-Machine specifically includes:
First spider module, for based on described modes of warning, the fault identified with described state recognition module, for ergodic condition, travels through described early warning expert knowledge library, obtains all possible failure cause corresponding with described fault;
Wherein, if being flow process modes of warning, the each production system flow process then oil vapor treatment station comprised carries out running status tracking, for each described production system flow process, any one production system flow process is designated as production system flow process B, obtain the key parameter values broken down of each equipment in production system flow process B, and obtain this key parameter values a plurality of failure cause when fault
If being key parameter modes of warning, by searching described early warning expert knowledge library, it is thus achieved that the described key parameter all possible breakdown reasons under described fault;
If being equipment modes of warning, then obtain the multinomial key parameter values of the device A needing early warning, to key parameter each described, all obtain this key parameter a plurality of failure cause when fault;
If being flow process modes of warning, the each production system flow process then oil vapor treatment station comprised carries out running status tracking, for each described production system flow process, any one production system flow process is designated as production system flow process B, obtain the key parameter values broken down of each equipment in production system flow process B, and obtain this key parameter values a plurality of failure cause when fault;
Logic decision condition memory module, for storing logic decision condition corresponding with every failure cause in described early warning expert knowledge library;Wherein, there is specific fault in the key parameter that described logic decision condition is associate device, and, there is certain logic relation in the fault between associate device, the process logic relation between this certain logic relation with associate device in technological process is consistent;
Second spider module, for all possible failure cause that described first spider module is obtained, is performed both by following operation:
The failure cause that any one is possible is designated as failure cause W, searches described logic decision condition memory module, it is thus achieved that the logic decision condition corresponding with described failure cause W, carry out obtaining the key parameter of associate device;Then, with the key parameter of described associate device for ergodic condition, travel through described early warning expert knowledge library, judge whether the key parameter of described associate device described specific fault occurs, and, whether the fault between associate device meets certain logic relation, if all met, then described failure cause W is physical fault reason;Otherwise, described failure cause W is not the fault actually occurred, and is rejected by failure cause W;
Thus obtain at least one physical fault reason;
Troubleshooting measure acquisition module, for according to described early warning expert knowledge library, it is thus achieved that the troubleshooting measure corresponding with described physical fault reason.
CN201410738580.4A 2014-12-05 2014-12-05 Oil vapor treatment station intelligent early-warning system based on Internet of Things sensing data Active CN105717887B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN201410738580.4A CN105717887B (en) 2014-12-05 2014-12-05 Oil vapor treatment station intelligent early-warning system based on Internet of Things sensing data

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN201410738580.4A CN105717887B (en) 2014-12-05 2014-12-05 Oil vapor treatment station intelligent early-warning system based on Internet of Things sensing data

Publications (2)

Publication Number Publication Date
CN105717887A true CN105717887A (en) 2016-06-29
CN105717887B CN105717887B (en) 2018-03-02

Family

ID=56144405

Family Applications (1)

Application Number Title Priority Date Filing Date
CN201410738580.4A Active CN105717887B (en) 2014-12-05 2014-12-05 Oil vapor treatment station intelligent early-warning system based on Internet of Things sensing data

Country Status (1)

Country Link
CN (1) CN105717887B (en)

Cited By (9)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN108764761A (en) * 2018-07-11 2018-11-06 华电电力科学研究院有限公司 A kind of thermal power plant's high temperature steam pipeline whole-life cycle fee system and its operation method based on 3D pipeline models
CN109299143A (en) * 2018-11-28 2019-02-01 重庆邮电大学 The knowledge fast indexing method in the data interoperation knowledge on testing library based on Redis caching
CN109508880A (en) * 2018-11-12 2019-03-22 西南石油大学 Cementing quality appraisal procedure based on ontology
CN110531742A (en) * 2019-09-16 2019-12-03 重庆华能水电设备制造有限公司 A kind of generator current collecting equipment real time monitoring and method for diagnosing faults
CN110853744A (en) * 2019-09-04 2020-02-28 太原市中心医院 Hospital quality control management system under big data
CN110868338A (en) * 2019-11-21 2020-03-06 北京高因科技有限公司 Early warning monitoring method
CN111639779A (en) * 2019-03-01 2020-09-08 北京国双科技有限公司 Information processing method and related equipment
CN111994137A (en) * 2020-09-04 2020-11-27 深圳科安达电子科技股份有限公司 Alarm analysis method based on railway signal centralized monitoring
CN112700343A (en) * 2019-10-23 2021-04-23 中国石油天然气股份有限公司 Operation monitoring method and system based on oil gas Internet of things

Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20070028706A1 (en) * 2005-07-29 2007-02-08 Mandziuk Michael W Abrasion-resistant pig, and materials and methods for making same
CN101388085A (en) * 2007-09-14 2009-03-18 李清东 Rapid failure diagnosis reasoning machine
CN102163300A (en) * 2011-04-20 2011-08-24 南京航空航天大学 Method for optimizing fault diagnosis rules based on ant colony optimization algorithm
CN102434388A (en) * 2011-11-17 2012-05-02 高丙团 Health status online monitoring device of wind generating set and monitoring method of monitoring device
CN102637019A (en) * 2011-02-10 2012-08-15 武汉科技大学 Intelligent integrated fault diagnosis method and device in industrial production process

Patent Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20070028706A1 (en) * 2005-07-29 2007-02-08 Mandziuk Michael W Abrasion-resistant pig, and materials and methods for making same
CN101388085A (en) * 2007-09-14 2009-03-18 李清东 Rapid failure diagnosis reasoning machine
CN102637019A (en) * 2011-02-10 2012-08-15 武汉科技大学 Intelligent integrated fault diagnosis method and device in industrial production process
CN102163300A (en) * 2011-04-20 2011-08-24 南京航空航天大学 Method for optimizing fault diagnosis rules based on ant colony optimization algorithm
CN102434388A (en) * 2011-11-17 2012-05-02 高丙团 Health status online monitoring device of wind generating set and monitoring method of monitoring device

Cited By (13)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN108764761A (en) * 2018-07-11 2018-11-06 华电电力科学研究院有限公司 A kind of thermal power plant's high temperature steam pipeline whole-life cycle fee system and its operation method based on 3D pipeline models
CN109508880A (en) * 2018-11-12 2019-03-22 西南石油大学 Cementing quality appraisal procedure based on ontology
CN109299143B (en) * 2018-11-28 2022-03-22 重庆邮电大学 Knowledge fast indexing method of data interoperation test knowledge base based on Redis cache
CN109299143A (en) * 2018-11-28 2019-02-01 重庆邮电大学 The knowledge fast indexing method in the data interoperation knowledge on testing library based on Redis caching
CN111639779B (en) * 2019-03-01 2023-08-29 北京国双科技有限公司 Information processing method and related equipment
CN111639779A (en) * 2019-03-01 2020-09-08 北京国双科技有限公司 Information processing method and related equipment
CN110853744A (en) * 2019-09-04 2020-02-28 太原市中心医院 Hospital quality control management system under big data
CN110531742A (en) * 2019-09-16 2019-12-03 重庆华能水电设备制造有限公司 A kind of generator current collecting equipment real time monitoring and method for diagnosing faults
CN112700343A (en) * 2019-10-23 2021-04-23 中国石油天然气股份有限公司 Operation monitoring method and system based on oil gas Internet of things
CN110868338B (en) * 2019-11-21 2021-05-25 北京高因科技有限公司 Early warning monitoring method
CN110868338A (en) * 2019-11-21 2020-03-06 北京高因科技有限公司 Early warning monitoring method
CN111994137A (en) * 2020-09-04 2020-11-27 深圳科安达电子科技股份有限公司 Alarm analysis method based on railway signal centralized monitoring
CN111994137B (en) * 2020-09-04 2022-07-12 深圳科安达电子科技股份有限公司 Alarm analysis method based on railway signal centralized monitoring

Also Published As

Publication number Publication date
CN105717887B (en) 2018-03-02

Similar Documents

Publication Publication Date Title
CN105717887A (en) Oil gas processing station intelligent early warning system based on Internet of things sensing data
CN105717888B (en) A kind of intelligent early-warning system
CN105891629B (en) A kind of discrimination method of transformer equipment failure
CN108027594A (en) For detecting the abnormal method in water distribution system
CN104091035B (en) Health monitoring method for effective loads of space station based on data-driven algorithm
CN107701378B (en) A kind of wind-driven generator fault early warning method
CN104360677A (en) Cigarette processing quality evaluation and diagnosis method
CN109753027A (en) It is a kind of industry manufacture in parameter monitoring method and device
Branisavljević et al. Automatic, semi-automatic and manual validation of urban drainage data
CN106405280B (en) A kind of intelligent substation on-line monitoring parameter trend method for early warning
CN105306439A (en) Feature rule detection method based on decision tree self-repairing
CN106875613A (en) A kind of fire alarm Situation analysis method
CN104950873B (en) The online test method of process control loops intermittent oscillation
CN107742883A (en) A kind of power system topology island system for rapidly identifying and method based on Spark
CN109614451A (en) Industrial big data intellectual analysis decision making device
CN106933097B (en) Chemical process fault diagnosis method based on multi-layer optimization PCC-SDG
CN109711428A (en) A kind of saturated gas pipeline internal corrosion speed predicting method and device
CN116451608A (en) Mixed wind power prediction method and device for complex terrain
CN105183612A (en) Server available memory abnormal growth and running state assessment method
CN108388224A (en) A kind of Desalting and Dewatering from Crude Oil process operation method for monitoring state based near infrared spectrum
CN105607477B (en) A kind of process control loops detection method of oscillations decomposed based on improvement local mean value
TW201020806A (en) Product quality fault detection method and real metrology data evaluation method
CN104731056B (en) The method of the quick operation stability for judging chemical production device
CN113723649A (en) Chemical process abnormity monitoring method based on semi-supervised model and model optimization method
CN107977727A (en) A kind of method that probability is blocked based on social development and climatic factor prediction cable network

Legal Events

Date Code Title Description
C06 Publication
PB01 Publication
C10 Entry into substantive examination
SE01 Entry into force of request for substantive examination
GR01 Patent grant
GR01 Patent grant
CP03 Change of name, title or address

Address after: 834000 Jiyun Road, Karamay District, Karamay City, Xinjiang Uygur Autonomous Region, 197-2

Patentee after: HONGYOU SOFTWARE Co.,Ltd.

Address before: 834000 No. 22, long march road, the Xinjiang Uygur Autonomous Region, Karamay

Patentee before: KARAMAY HONGYOU SOFTWARE Co.,Ltd.

CP03 Change of name, title or address