CN107057749B - The online soft sensor system and measurement method of gasoline stabilizer separation accuracy - Google Patents
The online soft sensor system and measurement method of gasoline stabilizer separation accuracy Download PDFInfo
- Publication number
- CN107057749B CN107057749B CN201610633279.6A CN201610633279A CN107057749B CN 107057749 B CN107057749 B CN 107057749B CN 201610633279 A CN201610633279 A CN 201610633279A CN 107057749 B CN107057749 B CN 107057749B
- Authority
- CN
- China
- Prior art keywords
- pressure
- data
- rvp
- gasoline
- stabilizer
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
- 239000003381 stabilizer Substances 0.000 title claims abstract description 128
- 238000000926 separation method Methods 0.000 title claims abstract description 76
- 238000000691 measurement method Methods 0.000 title claims abstract description 14
- 238000000034 method Methods 0.000 claims abstract description 167
- 230000008569 process Effects 0.000 claims abstract description 67
- 238000005259 measurement Methods 0.000 claims abstract description 45
- 239000007789 gas Substances 0.000 claims description 121
- 239000007788 liquid Substances 0.000 claims description 58
- 239000000463 material Substances 0.000 claims description 43
- 238000004364 calculation method Methods 0.000 claims description 37
- 238000005070 sampling Methods 0.000 claims description 31
- 238000004458 analytical method Methods 0.000 claims description 27
- 102100022002 CD59 glycoprotein Human genes 0.000 claims description 24
- 101000897400 Homo sapiens CD59 glycoprotein Proteins 0.000 claims description 24
- 229920006395 saturated elastomer Polymers 0.000 claims description 21
- 238000001914 filtration Methods 0.000 claims description 19
- 238000010992 reflux Methods 0.000 claims description 15
- 238000003556 assay Methods 0.000 claims description 14
- 239000000126 substance Substances 0.000 claims description 14
- 101100382854 Arabidopsis thaliana CCD7 gene Proteins 0.000 claims description 12
- 101100129496 Arabidopsis thaliana CYP711A1 gene Proteins 0.000 claims description 12
- 101100129499 Arabidopsis thaliana MAX2 gene Proteins 0.000 claims description 12
- 101100083446 Danio rerio plekhh1 gene Proteins 0.000 claims description 12
- 230000008859 change Effects 0.000 claims description 12
- 239000007792 gaseous phase Substances 0.000 claims description 12
- 238000009835 boiling Methods 0.000 claims description 11
- 239000000203 mixture Substances 0.000 claims description 11
- UDYXMTORTDACTG-UHFFFAOYSA-N 1,1,3-tributylthiourea Chemical compound CCCCNC(=S)N(CCCC)CCCC UDYXMTORTDACTG-UHFFFAOYSA-N 0.000 claims description 9
- 239000007791 liquid phase Substances 0.000 claims description 8
- 238000012937 correction Methods 0.000 claims description 7
- 238000004519 manufacturing process Methods 0.000 claims description 7
- -1 AIC5 Proteins 0.000 claims description 6
- 238000013459 approach Methods 0.000 claims description 6
- 230000033228 biological regulation Effects 0.000 claims description 6
- 230000000052 comparative effect Effects 0.000 claims description 6
- 238000009833 condensation Methods 0.000 claims description 5
- 230000005494 condensation Effects 0.000 claims description 5
- 239000012321 sodium triacetoxyborohydride Substances 0.000 claims description 5
- 238000003860 storage Methods 0.000 claims description 5
- 230000001174 ascending effect Effects 0.000 claims description 4
- 238000012512 characterization method Methods 0.000 claims description 3
- 238000001816 cooling Methods 0.000 claims description 3
- 238000010238 partial least squares regression Methods 0.000 claims description 3
- 239000000284 extract Substances 0.000 claims description 2
- 238000007670 refining Methods 0.000 abstract description 8
- 230000008901 benefit Effects 0.000 abstract description 5
- 238000005516 engineering process Methods 0.000 abstract description 5
- 238000003908 quality control method Methods 0.000 abstract description 5
- 238000001311 chemical methods and process Methods 0.000 abstract description 2
- 238000009530 blood pressure measurement Methods 0.000 abstract 1
- 239000011159 matrix material Substances 0.000 abstract 1
- 239000003921 oil Substances 0.000 description 11
- KAKZBPTYRLMSJV-UHFFFAOYSA-N Butadiene Chemical compound C=CC=C KAKZBPTYRLMSJV-UHFFFAOYSA-N 0.000 description 10
- 238000005336 cracking Methods 0.000 description 10
- 238000010586 diagram Methods 0.000 description 9
- 239000000047 product Substances 0.000 description 8
- 238000011084 recovery Methods 0.000 description 6
- 238000010521 absorption reaction Methods 0.000 description 5
- 238000004523 catalytic cracking Methods 0.000 description 5
- 238000004821 distillation Methods 0.000 description 5
- 238000004939 coking Methods 0.000 description 4
- 230000003111 delayed effect Effects 0.000 description 4
- 238000000605 extraction Methods 0.000 description 4
- QWTDNUCVQCZILF-UHFFFAOYSA-N isopentane Chemical compound CCC(C)C QWTDNUCVQCZILF-UHFFFAOYSA-N 0.000 description 4
- 238000012423 maintenance Methods 0.000 description 4
- VNWKTOKETHGBQD-UHFFFAOYSA-N methane Chemical compound C VNWKTOKETHGBQD-UHFFFAOYSA-N 0.000 description 4
- 238000004148 unit process Methods 0.000 description 4
- 230000007246 mechanism Effects 0.000 description 3
- OFBQJSOFQDEBGM-UHFFFAOYSA-N n-pentane Natural products CCCCC OFBQJSOFQDEBGM-UHFFFAOYSA-N 0.000 description 3
- 239000012071 phase Substances 0.000 description 3
- 150000001335 aliphatic alkanes Chemical class 0.000 description 2
- 238000003889 chemical engineering Methods 0.000 description 2
- 238000006243 chemical reaction Methods 0.000 description 2
- 239000007795 chemical reaction product Substances 0.000 description 2
- 238000003795 desorption Methods 0.000 description 2
- 238000001514 detection method Methods 0.000 description 2
- AFABGHUZZDYHJO-UHFFFAOYSA-N dimethyl butane Natural products CCCC(C)C AFABGHUZZDYHJO-UHFFFAOYSA-N 0.000 description 2
- 238000009826 distribution Methods 0.000 description 2
- 230000008676 import Effects 0.000 description 2
- 238000012545 processing Methods 0.000 description 2
- 238000012360 testing method Methods 0.000 description 2
- 238000012795 verification Methods 0.000 description 2
- 241001269238 Data Species 0.000 description 1
- OTMSDBZUPAUEDD-UHFFFAOYSA-N Ethane Chemical compound CC OTMSDBZUPAUEDD-UHFFFAOYSA-N 0.000 description 1
- VGGSQFUCUMXWEO-UHFFFAOYSA-N Ethene Chemical compound C=C VGGSQFUCUMXWEO-UHFFFAOYSA-N 0.000 description 1
- 239000005977 Ethylene Substances 0.000 description 1
- 230000002159 abnormal effect Effects 0.000 description 1
- 239000002199 base oil Substances 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 239000001273 butane Substances 0.000 description 1
- 230000000295 complement effect Effects 0.000 description 1
- 239000000470 constituent Substances 0.000 description 1
- 238000010276 construction Methods 0.000 description 1
- 230000007547 defect Effects 0.000 description 1
- 230000007812 deficiency Effects 0.000 description 1
- 238000013461 design Methods 0.000 description 1
- 239000002283 diesel fuel Substances 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 238000001704 evaporation Methods 0.000 description 1
- 230000008020 evaporation Effects 0.000 description 1
- 230000004907 flux Effects 0.000 description 1
- 239000000295 fuel oil Substances 0.000 description 1
- 238000003306 harvesting Methods 0.000 description 1
- 229930195733 hydrocarbon Natural products 0.000 description 1
- 150000002430 hydrocarbons Chemical class 0.000 description 1
- 125000001972 isopentyl group Chemical group [H]C([H])([H])C([H])(C([H])([H])[H])C([H])([H])C([H])([H])* 0.000 description 1
- 238000007726 management method Methods 0.000 description 1
- 230000000737 periodic effect Effects 0.000 description 1
- 239000003209 petroleum derivative Substances 0.000 description 1
- 238000007781 pre-processing Methods 0.000 description 1
- 238000011897 real-time detection Methods 0.000 description 1
- 238000000611 regression analysis Methods 0.000 description 1
- 238000011160 research Methods 0.000 description 1
- 230000001568 sexual effect Effects 0.000 description 1
- 230000035939 shock Effects 0.000 description 1
- 239000002002 slurry Substances 0.000 description 1
- 230000006641 stabilisation Effects 0.000 description 1
- 238000011105 stabilization Methods 0.000 description 1
- 230000009466 transformation Effects 0.000 description 1
- 238000005292 vacuum distillation Methods 0.000 description 1
Classifications
-
- C—CHEMISTRY; METALLURGY
- C10—PETROLEUM, GAS OR COKE INDUSTRIES; TECHNICAL GASES CONTAINING CARBON MONOXIDE; FUELS; LUBRICANTS; PEAT
- C10G—CRACKING HYDROCARBON OILS; PRODUCTION OF LIQUID HYDROCARBON MIXTURES, e.g. BY DESTRUCTIVE HYDROGENATION, OLIGOMERISATION, POLYMERISATION; RECOVERY OF HYDROCARBON OILS FROM OIL-SHALE, OIL-SAND, OR GASES; REFINING MIXTURES MAINLY CONSISTING OF HYDROCARBONS; REFORMING OF NAPHTHA; MINERAL WAXES
- C10G7/00—Distillation of hydrocarbon oils
- C10G7/12—Controlling or regulating
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01N—INVESTIGATING OR ANALYSING MATERIALS BY DETERMINING THEIR CHEMICAL OR PHYSICAL PROPERTIES
- G01N33/00—Investigating or analysing materials by specific methods not covered by groups G01N1/00 - G01N31/00
- G01N33/22—Fuels; Explosives
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01N—INVESTIGATING OR ANALYSING MATERIALS BY DETERMINING THEIR CHEMICAL OR PHYSICAL PROPERTIES
- G01N33/00—Investigating or analysing materials by specific methods not covered by groups G01N1/00 - G01N31/00
- G01N33/26—Oils; Viscous liquids; Paints; Inks
- G01N33/28—Oils, i.e. hydrocarbon liquids
- G01N33/2829—Mixtures of fuels
Landscapes
- Chemical & Material Sciences (AREA)
- Health & Medical Sciences (AREA)
- Engineering & Computer Science (AREA)
- Life Sciences & Earth Sciences (AREA)
- Oil, Petroleum & Natural Gas (AREA)
- Food Science & Technology (AREA)
- General Chemical & Material Sciences (AREA)
- Chemical Kinetics & Catalysis (AREA)
- Medicinal Chemistry (AREA)
- Physics & Mathematics (AREA)
- Analytical Chemistry (AREA)
- Biochemistry (AREA)
- General Health & Medical Sciences (AREA)
- General Physics & Mathematics (AREA)
- Immunology (AREA)
- Pathology (AREA)
- Organic Chemistry (AREA)
- Production Of Liquid Hydrocarbon Mixture For Refining Petroleum (AREA)
Abstract
The present invention relates to a kind of online soft sensor system of gasoline stabilizer separation accuracy and measurement methods, belong to oil-refining chemical process automatic control technology field, comprising: liquefied gas C5+ content measuring method and stable gasoline Lei Shi vapour pressure measurement method.The existing conventional instrument measurement process parameters such as tower top temperature, tower top pressure, column bottom temperature, tower reactor pressure of the present invention according to gasoline stabilizer, determine the online computation model of hard measurement, the separation accuracy of on-line real-time measuremen gasoline stabilizer, that is two liquefied gas C5+ content, stable gasoline Lei Shi vapour pressure quality indicator values, wherein soft-sensing model coefficient, the step of providing by the method for the present invention and rule determine, the quality index numerical value calculated in real time according to the present invention both can be used as the reference that operation adjusts in real time;The controlled variable that can also be used as Dynamic matrix control device realizes direct closed loop quality control;To propose the stationarity and accuracy of high quality control, device economical operation benefit is improved.
Description
Technical field
The present invention relates to oil-refining chemical process automatic control technology fields, in particular to a kind of gasoline stabilizer
The online soft sensor method of separation accuracy, it may be assumed that liquefied gas C5+ content, two quality index of stable gasoline Lei Shi vapour pressure
Line flexible measurement method.
Background technique
In oil-refining chemical enterprise, the heavy oil conversions such as residual oil, wax oil are light petroleum gas, are the main economic effects for refining oil part
Beneficial source.It the cracking units such as executes the catalytic cracking of conversion reaction task, delayed coking, be hydrocracked, being the pass for refining oil part
Key core apparatus.
Cracked reaction product mainly has gasoline component, liquefied gas component, dry gas component etc..Cracking unit reactor outlet oil
Gas, by main fractionating tower, reaction product after unreacted diesel oil, slurry oil (or wax oil) component are separated, then pass through suction
Stabilization process is received, the products such as stable gasoline, liquefied gas, dry gas are separated under 1MPa or so operating pressure.Vapor recovery unit process
It is widely used important production process in Petrochemical Enterprises.
The major process unit of Vapor recovery unit process is absorption tower, desorber, stabilizer;The Vapor recovery unit mistake of most devices
Journey is also equipped with dry gas reabsorber after absorption tower;Also the absorption tower and desorber for having partial devices are combined into one, i.e. top
For absorption process, lower part is desorption process.Desorber, by thermal distillation, by condensed oil, (reaction oil gas pressurizes condensation after cooling
Oil and absorbing tower base oil mixture) in C2 (ethane, ethylene) component, methane and other low-boiling point gas substances desorb into
Tower top returns to absorption tower charging, and tower reactor liquid phase (commonly referred to as deethanization gasoline) is further separated into stabilizer.Most refinery enterprises
The desorber of industry Vapor recovery unit process, is intended to strong desorption operations, and methane is free of in the liquefied gas for isolating subsequent stabilizer
With other low-boiling point gas substances.
Stabilizer, by the distillation process of tower bottom thermal distillation and tower top cold reflux, by gasoline component and liquefied gas component point
From.The yield of gasoline of cracking unit is usually more much larger than yield of liquefied gas, and the main productive target of stabilizer is further to remove
Liquefied gas component in " deethanization gasoline ", is processed into stable gasoline, therefore, stabilizer, also commonly referred to as gasoline stabilizer.
Because liquefied gas is exactly C3 and C4 component, the stabilizer of some cracking units is also referred to as the de- C4 tower of gasoline or gasoline is de-
Butane tower.
The separation accuracy of gasoline stabilizer, theoretically can be used: " C5 and more heavy constituent contains in the liquefied gas of tower top extraction
Amount (usually write a Chinese character in simplified form are as follows: liquefied gas C5+ content) ", " content of C4 and more light component in the stable gasoline of tower bottom extraction " characterize,
For stable gasoline, an important quality index is " volatility ", practical in the examination and test of products in production, generallys use " thunder
Family name's vapour pressure " index, and the gasoline Lei Shi vapour pressure of petrochemical industry standard, the pass with " content of C4 and more light component in gasoline "
System, is height linear approximate relationship.
In conclusion in oil-refining chemical enterprise, the separation accuracy of gasoline stabilizer generallys use that " liquefied gas C5+ contains
Amount ", " stable gasoline Lei Shi vapour pressure " characterize.
It is produced in practice in Petrochemical Enterprises, liquefied gas C5+ content, stable gasoline Lei Shi vapour pressure generally use timing and adopt
The analysis of sample laboratory obtains.Because manpower limits, the time interval of assay is usually long, and a few hours are primary, even one day
Primary or longer, for Operating Guideline, real-time is poor.For this purpose, having many Petrochemical Enterprises configuration in-line analyzers, real-time detection
The data such as liquefied gas composition.
As Advanced Control Techniques are in the popularity of refining and petrochemical processing units, with multivariable prediction coordinated control and product matter
The Advanced Control Techniques that direct closed-loop control is notable feature are measured, the detection to product quality proposes online real-time continuous survey
The requirement of amount.Timing sampling assay is unable to reach the requirement of online real-time continuous measurement;On-line chromatograph analyzer, but
Have the defects that certain time lag, periodic calibration pulse-type disturbance, and because the analyzers such as on-line chromatograph must configure specimen preprocessing
Reason system, compared with general measure instrument, failure rate is relatively high, and maintenance workload is bigger.
Liquefied gas C5+ content based on general measure instrument, stable gasoline Lei Shi vapour pressure online soft sensor model are established,
On-line real-time measuremen goes out its numerical value, can greatly save hardware investment, reduce maintenance workload, and time lag can be overcome to control
The influence of system, to improve control of product quality precision.
It is corresponding online even if gasoline stabilizer is configured with the in-line analyzer of liquefied gas component, stable gasoline vapour pressure
Hard measurement can also play in analyzer failure and calibration and filter out impulse disturbances, prevent from causing substantially closed loop quality control
Shock vibration, and in-line analyzer can be substituted during failure and continue on-line checking quality index, it is to prevent closed loop quality control
The necessary complement measure of abnormal disturbances.
Since Advanced Control Techniques promote and apply, the flexible measurement method being widely used is " data-driven regression modeling " side
Method, it may be assumed that according to production process experience, choose related process variable (general measure meter parameter) work that may influence tested index
For input, by a large amount of sample data, soft-sensing model is established in recurrence, realizes online soft sensor.This data-driven returns
Modeling, input variable it is more and with the relationship of tested index be it is uncertain, model quality influenced by sample quality it is too big,
Hard measurement reliability and adaptability are usually undesirable.
Summary of the invention
In view of the deficiency of the prior art, the present invention provides a kind of easy reliable gasoline stabilizer separation accuracy
Online soft sensor method.
The technical solution used in the present invention is:
A kind of online soft sensor system of gasoline stabilizer separation accuracy, including stabilizer, stabilizer include tower top dome,
The space that the dome and top first layer column plate of tower surround, the interface of liquid and vault indoor gas on first layer column plate are
The last one separation of ascending vapor logistics composition variation in tower, after top gaseous phase logistics is cooling, condensation enters liquid gas storage tank,
Composition no longer changes, and therefore, liquefied gas C5+ content depends on the indoor vapor liquid equilibrium of tower top vault;
The sensor of thermometric instrument and the sensor of pressure gauge are arranged in tower top dome;
Thermometric instrument measures tower top dome liquidus temperature, and pressure gauge measures tower top pressure;
Liquid gas storage tank extracts logistics out, and a part goes out device as liquefied gas product, measures its flow by flow instrument;It is another
Part is used as tower top cold reflux, is squeezed at the top of stabilizer after flow instrument measures;
Stabilizer includes tower reactor, the space that tower lower part the last layer column plate and tower bottom surround, bottom liquid and gas reactor
Point interface, be flow downward in tower liquid phase stream composition variation the last one separation, the vapour pressure of stable gasoline,
Depending on the vapor liquid equilibrium in tower reactor;
The sensor of thermometric instrument and the sensor of pressure gauge are arranged in tower reactor;
Thermometric instrument measures tower bottom liquidus temperature, and pressure gauge measures tower reactor pressure
A kind of online soft sensor method of gasoline stabilizer separation accuracy, comprising: liquefied gas C5+ content soft measuring method and
Stable gasoline Lei Shi vapour pressure flexible measurement method.
The step of liquefied gas C5+ content soft measuring method are as follows:
Step (1.a) determines that the normal minimum pure material of boiling point is the pass for characterizing liquefied gas C5+ content in liquefied gas C5 component
Key component pure material;
Step (2.a) determines the liquidus temperature measuring instrumentss T of gas-liquid separation interface in gasoline stabilizer tower top domeTOP
With gaseous pressure measuring instrumentss PTOP, what industrial process instrument measurement went out is celsius temperature scale (DEG C) and relative pressure (MPa);
If gas-liquid separation interface does not have temperature measuring instrument in the dome of gasoline stabilizer top, tower top first layer tower can be used
Temperature or top gaseous phase temperature replace under plate;Gaseous pressure P at tower top gas-liquid separating interfaceTOP, it is commonly configured with tower top pressure
Measuring instrumentss numerical value, directlys adopt;
Step (3.a), liquefied gas C5+ content key component pure material gas-liquid separation interface liquid phase temperature in tower top dome
Spend TTOPComparison temperature Tr,C5,TOPIt is calculated as follows:
Tr,C5,TOP=(TTOP+273.15)/Tc,C5
Wherein:
Tc,C5For the critical-temperature of liquefied gas C5+ content key component pure material, unit K
273.15 being 0 DEG C of Celsius temperature of kelvin rating
Step (4.a) calculates key component pure material saturated vapor pressure under liquidus temperature at tower top dome gas-liquid separation
's;
Using distinct methods, the saturated vapor calculated is pressed with pressure value or reduced pressure force value.Vapour pressure pressure value and
Reduced pressure force value can be calculated by simple linear and be converted, and convenient for the statement of following model algorithm, which does not repartition
Pressure value and reduced pressure force value;
Saturated vapor of the liquefied gas C5+ content key component pure material in tower top dome gas-liquid separation interface liquidus temperature
Pressure is expressed as PVP,C5,TOP;
Step (5.a) determines liquefied gas C5+ content soft sensor model:
Wherein:
C5LPGFor liquefied gas C5+ content, unit %
A0,C5、A1.C5For model coefficient
0.101325 is the difference of absolute pressure and relative pressure, units MPa
Step (6.a), off-line calculation determines model coefficient according to the following steps:
Step (6.1a), from the gasoline stabilizer Process History of acquisition certain period of time in dcs (DCS)
Data;
The stabilizer procedure parameter for needing to acquire has:
1)TTOP: tower top dome liquidus temperature, DEG C
2)PTOP: tower top dome gaseous pressure, MPa
If 3) be configured with in-line analyzer, increase liquefied gas C5+ content AIC5Data;
Such as without configuration in-line analyzer, then the liquefied gas C5+ of corresponding period is collected from device analysis data library again
Content analysis data (also uses AIC5Indicate) and its sampling instant data;
Data sampling interval 1,2 or 5 minute;
Data segment time span is established rules then really are as follows:
1) model coefficient is determined using analysis data, then includes continuous 10 or more liquefied gas C5+ contents in the period
Analysis data sample;The amplitude of variation of liquefied gas C5+ content value >=high limit of process chart regulation 1/2;
2) it is configured with in-line analyzer, then includes the significant fluctuating change of quality index: liquefied gas in the period
The fluctuating range of C5+ content >=high limit of process chart regulation 1/3.
The above process is lasted data, imported in Excel table by step (6.2a),
For each procedure parameter, the data of the seclected time period are expressed as an array, such as TTOP,j, j=1,2 ...,
N, n is equal to seclected time period data sample number here;
Step (6.4a), it is first in the Excel table that the above process lasts data and laboratory values or in-line analyzer data
Smothing filtering operation first is carried out to each process parameter data:
(Z)j+1=(1-FL)*(Z)j+FL*(Z)j+1, j=1,2 ..., n
Wherein:
(Z) T is representedTOP、PTOPAnd analyzer data
FLFor smothing filtering coefficient, value is 0.70~0.85
Then in Excel table, using later process parameter data is filtered, array column are calculated:
Step (6.5a) determines soft-sensing model coefficient using one of following two methods:
Step (6.5.1a) calculates soft-sensing model coefficient A referring to the scale calculation method of instrumentation engineering1.C5:
Sampling instant, the numerical error of analysis data are considered, for the computational accuracy for improving model coefficient, A1.C5Using three times
Calculate the method averaged;And A0,C5It is the zero-bit of hard measurement, when off-line calculation finds out a provisional value, and hard measurement is online
It can constantly be calibrated in operation;
Observation comparative approach is shown using curve, respectively:
I. maximum three sample values of liquefied gas C5+ content laboratory values (or in-line analyzer data) are selected, from big to small
It is expressed as AIC5,MAX1、AIC5,MAX2、AIC5,MAX3, the X of three assay samples sampling instantsC5,jIt is expressed as XC5,MAX1、
XC5,MAX2、XC5,MAX3;
Ii. the smallest three sample values of liquefied gas C5+ content laboratory values (or in-line analyzer data) are selected, from big to small
It is expressed as AIC5,MIN1、AIC5,MIN2、AIC5,MIN3, the X of three assay samples sampling instantsC5,jIt is expressed as XC5,MIN1、
XC5,MIN2、XC5,MIN3;
It calculates and determines A1.C5And A0.C5Value:
A0.C5=AIC5,MIN1-A1.C5*XC5,MIN1
Step (6.5.2a), with XC5For input variable, with AIC5To be detected variable, is returned using PLS and obtain hard measurement
Model coefficient;Regression result needs to verify: A1.C5For positive value.Otherwise data should be chosen again to return again;
Step (7.a), it is determined that after model coefficient, the online soft sensor of liquefied gas C5+ content is realized, host computer can be used
Soft instrument software, host computer calculation procedure are realized or are realized using DCS Function Block Configuration;
Liquefied gas C5+ content soft instrument or calculation procedure, by OPC obtain DCS in gasoline stabilizer tower top temperature,
Liquefied gas C5+ content data is sent into DCS after the completion of calculating and shown or for used in closed-loop control by tower top pressure process data.Tower
Push up liquefied gas C5+ content chemical examination/analyzer data, be sent into host computer in, by correction module periodically or real time calibration liquefied gas
Zero-bit of the C5+ containing numerical quantity.
The step of stable gasoline Lei Shi vapour pressure flexible measurement method are as follows:
Step (1.b) determines that the highest pure material of normal boiling point is characterization stable gasoline Lei Shi in C4 component in stable gasoline
The key component pure material of vapour pressure;
Step (2.b) determines the liquidus temperature measuring instrumentss T of gas-liquid separation interface in gasoline stabilizer tower reactorBTMIt is gentle
Phase pressure measuring instrumentss PBTM, what industrial process instrument measurement went out is celsius temperature scale (DEG C) and relative pressure (MPa).
If there is no tower reactor gaseous pressure measuring instrumentss data in DCS, it is determined that gasoline stabilizer top gaseous phase pressure is surveyed
Measure instrument PTOP, liquefied gas output flow measurement instrument FLPGWith overhead reflux flow measurement instrument FRE, gasoline stabilizer tower reactor pressure
The calculation method of power is provided in subsequent step (6.3b);
Step (3.b), stable gasoline Lei Shi vapour pressure key component pure material gas-liquid separation interface liquid phase in tower reactor
Temperature TBTMComparison temperature Tr,RVP,BTMIt is calculated as follows
Tr,RVP,BTM=(TBTM+273.15)/Tc,RVP
Wherein:
Tc,RVPFor the critical-temperature of stable gasoline Lei Shi vapour pressure key component pure material,
Unit K
273.15 be 0 DEG C of Celsius temperature of kelvin rating
Step (4.b) determines stable gasoline Lei Shi vapour pressure key component pure material liquid phase temperature at tower reactor gas-liquid separation
Spend lower saturated vapor pressure;
Using distinct methods, the saturated vapor calculated is pressed with pressure value or reduced pressure force value.Vapour pressure pressure value and
Reduced pressure force value can be calculated by simple linear and be converted, and convenient for the statement of following model algorithm, which does not repartition
Pressure value and reduced pressure force value;
Saturated vapor pressure table of the stable gasoline Lei Shi vapour pressure key component in tower reactor gas-liquid separation interface liquidus temperature
It is shown as PVP,RVP,BTM。
Step (5.b) determines stable gasoline Lei Shi vapour pressure soft-sensing model:
Wherein:
RVPSTABFor stable gasoline Lei Shi vapour pressure, unit kPa
A0,RVP、A1,RVPFor model coefficient
0.101325 is the difference of absolute pressure and relative pressure, units MPa
Step (6.b), off-line calculation obtains soft-sensing model coefficient according to the following steps:
Step (6.1b) acquires the gasoline stabilizer Process History data of certain period of time from dcs;
The stabilizer procedure parameter for needing to acquire has:
1)TBTM: tower reactor liquidus temperature, DEG C
2)PBTM: tower reactor gaseous pressure (if there is measuring instrumentss DCS data are shown), MPa
3) such as tower reactor gaseous pressure does not have measuring instrumentss DCS data to show, then acquires top gaseous phase pressure (MPa), liquefies
Gas output flow FLPGWith overhead reflux flow FRE, two of them data on flows, unit is consistent
4) it is such as configured with in-line analyzer, then is further added by stable gasoline vapour pressure AIRVPData;
Such as without configuration in-line analyzer, then the stable gasoline for collecting the corresponding period from device analysis data library again steams
Vapour pressure analysis data (also uses AIRVPIndicate) and its sampling instant data;
Sampling interval 1,2 or 5 minute;
Data segment time span is established rules then really:
Model coefficient is determined using analysis data, then includes continuous 10 or more stable gasoline vapour pressures in the period
Test data sample;Amplitude of variation >=6kPa of stable gasoline vapour pressure;
It is configured with in-line analyzer, then includes the significant fluctuating change of quality index: stable gasoline in the period
Fluctuating range >=4kPa of vapour pressure;
The above process is lasted data, imported in Excel table by step (6.2b);
For each procedure parameter, the data of the seclected time period are expressed as an array, such as TBTM,j, j=1,2 ...,
N, n is equal to seclected time period data sample number here;
Step (6.3b) jumps to step (6.4b) if having collected tower reactor gaseous pressure data.This step by step,
Gasoline stabilizer tower reactor pressure is calculated using following algorithm:
Stabilizer pressure drop is calculated first estimates coefficient ADP:
Wherein:
ADPFor constant, using mean value computation as a result, not considering unit;
PBTM,0For constant, value is equal to the tower reactor pressure mean value obtained from operating experience or site instructions pressure gauge;
Then tower reactor pressure value array is calculated
PBTM,j=PTOP,j+ADP*(FLPG,j+FRE,j)
Step (6.4b), it is first in the Excel table that the above process lasts data and laboratory values or in-line analyzer data
Smothing filtering operation first is carried out to each process parameter data:
(Z)j+1=(1-FL)*(Z)j+FL*(Z)j+1, j=1,2 ..., n
Wherein:
(Z) T is representedBTM、PBTMAnd analyzer data
FLFor smothing filtering coefficient: for observed temperature, pressure and in-line analyzer data,
FL=0.70~0.85;
For the tower reactor pressure data being calculated, FL=0.85~0.97;
Then in Excel table, using later process parameter data is filtered, array column are calculated:
Step (6.5b) determines soft-sensing model coefficient using one of following two methods;
Step (6.5.1b) calculates soft-sensing model coefficient referring to the scale calculation method of instrumentation engineering;
Sampling instant, the numerical error of analysis data are considered, for the computational accuracy for improving model coefficient, A1,RVPUsing three times
Calculate the method averaged;And A0,RVPIt is the zero-bit of hard measurement, when off-line calculation finds out a provisional value, and hard measurement is online
It can constantly be calibrated in operation;
Observation comparative approach is shown using curve, respectively:
I. maximum three sample values of stable gasoline vapour pressure laboratory values (or in-line analyzer data) are selected, from big to small
It is expressed as AIRVP,MAX1、AIRVP,MAX2、AIRVP,MAX3, the X of three assay samples sampling instantsRVP,jIt is expressed as XRVP,MAX1、
XRVP,MAX2、XRVP,MAX3;
Ii. the smallest three sample values of stable gasoline vapour pressure laboratory values (or in-line analyzer data) are selected, from greatly to
It is small to be expressed as AIRVP,MIN1、AIRVP,MIN2、AIRVP,MIN3, the X of three assay samples sampling instantsRVP,jIt is expressed as
XRVP,MIN1、XRVP,MIN2、XRVP,MIN3。
It calculates and determines A1,RVPAnd A0,RVPValue:
A0.RVP=AIRVP,MIN1-A1.RVP*XRVP,MIN1
Step (6.5.2b), with XRVPFor input variable, with AIRVPTo be detected variable, is returned using PLS and obtain hard measurement
Model coefficient;Regression result needs to verify: A1.RVPFor positive value.Otherwise data should be chosen again to return again;
When DCS does not have tower reactor pressure measured data, stable gasoline Lei Shi vapour pressure hard measurement mould that step (5.b) provides
Type, pattern also can be changed are as follows:
At this point, not executing step (6.3b) and step (6.4b), the mistake that step (6.2b) imports Excel table is directlyed adopt
Journey lasts data and laboratory values or in-line analyzer data, seeks model coefficient using the PLS Return Law:
Smothing filtering operation is carried out to each process parameter data first:
(Z)j+1=(1-FL)*(Z)j+FL*(Z)j+1, j=1,2 ..., n
Wherein:
(Z) T is representedBTM、PTOP、And analyzer data
FLFor smothing filtering coefficient:
For temperature, pressure and in-line analyzer data, FL=0.70~0.85;
For data on flows: FL=0.85~0.97;
Then in Excel table, using later process parameter data is filtered, two array column are calculated:
Finally with X1RVP、X2RVPFor input variable, with AIRVPTo be detected variable, using PLS, that is, Partial Least Squares Regression
Method obtains soft-sensing model coefficient;
Regression result should meet:
(1)A1,RVP、A2,RVPIt is positive value;
(2)Value agreement be equal to stabilizer voltage drop value.F hereinLPGAnd FRE, above-mentioned
It filters and chooses its mean value in later process parameter data.
Otherwise, data should be chosen again to return again;
Step (7b), it is determined that after model coefficient, the canbe used on line of stable gasoline Lei Shi vapour pressure hard measurement be can be used
Position machine soft instrument software, host computer calculation procedure use DCS Function Block Configuration;
Stable gasoline Lei Shi vapour pressure soft instrument or calculation procedure obtain the tower bottom temperature of gasoline stabilizer in DCS by OPC
Degree, tower reactor pressure (/ tower top pressure, flow of lpg and overhead reflux flow) process data, by stable gasoline after the completion of calculating
Lei Shi Vapor pressure data is sent into DCS and is shown or for used in closed-loop control;Stable gasoline Lei Shi vapour pressure chemical examination/analyzer data,
Be sent into host computer in, by correction module periodically or real time calibration stable gasoline Lei Shi vapour pressure numerical value zero-bit.
Step (4.a) and step (4.b) described in the online soft sensor method of gasoline stabilizer separation accuracy of the present invention: pure
The saturated vapor pressure calculation method of substance, can refer to related chemistry engineering manual, and following two method usually can be used:
1) pure material hold-up vapour pressure is calculated using Riedel method:
Wherein:
For the reduced pressure force value of pure material vapour pressure, no unit;
TrTemperature, no unit are compared for pure material;
A, B, C, D are Riedel method vapour pressure algorithm coefficient;
2) pure material hold-up vapour pressure is calculated using Lee-Kesler method:
Wherein:
For the reduced pressure force value of pure material vapour pressure, no unit;
TrTemperature, no unit are compared for pure material;
ω is the eccentric factor of pure material, no unit.
Step (6.1a) and (6.1b) described in the online soft sensor method of gasoline stabilizer separation accuracy of the present invention: to make
Quality index has significant change between data Harvest time, can carry out operation adjustment appropriate to stabilizer.
Step (7.a) described in the online soft sensor method of gasoline stabilizer separation accuracy of the present invention and step (7.b) institute
Be related to: hard measurement zero adjustment, correction module are the basic skills or software module of hard measurement engineering technology, this patent side
Method is not described further.
The present invention compared with the prior art the utility model has the advantages that
The online soft sensor method of gasoline stabilizer separation accuracy of the present invention, from distillation process vapor liquid equilibrium mechanism and correlation
Physical chemistry engineering principles set out, and by the approximation under actual production conditions and simplify processing, establish deterministic simple line
The soft-sensing model of sexual intercourse realizes that two the liquefied gas C5+ content of gasoline stabilizer, stable gasoline Lei Shi vapour pressure quality refer to
Target online soft sensor substantially increases its reliability and adaptability.
The online soft sensor method of gasoline stabilizer separation accuracy of the present invention is surveyed using the ordinary temperature of stabilizer, pressure
Meter information is measured, is calculated by soft-sensing model, on-line real-time measuremen liquefied gas C5+ content, stable gasoline Lei Shi vapour pressure,
Middle soft instrument model algorithm is based on distillation process vapor liquid equilibrium mechanism and relevant physicochemical engineering principles.The hard measurement side
Method is easy to be reliable, and lag time is not present, and compared with configuring in-line analyzer detection method, has with low investment, maintenance workload
Advantage small, without lag time;Compared with using data-driven regression modeling method, have relationship model simple clear, reliable
The advantage that property is high, adaptable, modeling data sample requirement amount is small.Hard measurement is established using the method for the present invention, resettles closed loop
Quality control system can be further improved the control of product quality stationarity and accuracy of gasoline stabilizer.
Detailed description of the invention
Fig. 1 is that the liquefied gas C5+ content of the online soft sensor embodiment of the method for gasoline stabilizer separation accuracy of the present invention is soft
The flow chart of measurement method;
Fig. 2 is the stable gasoline Lei Shi steam of the online soft sensor embodiment of the method for gasoline stabilizer separation accuracy of the present invention
Press the flow chart of flexible measurement method;
Fig. 3 is the liquefied gas C5+ content hard measurement of the online soft sensor method of gasoline stabilizer separation accuracy of the present invention
Canbe used on line schematic diagram;
Fig. 4 is the soft survey of stable gasoline Lei Shi vapour pressure of the online soft sensor method of gasoline stabilizer separation accuracy of the present invention
The canbe used on line schematic diagram of amount;
Fig. 5 is gasoline stabilizer process flow and the instrument of the online soft sensor method of gasoline stabilizer separation accuracy of the present invention
Table distribution schematic diagram (has tower reactor pressure DCS data);
Fig. 6 is gasoline stabilizer process flow and the instrument of the online soft sensor method of gasoline stabilizer separation accuracy of the present invention
Table distribution schematic diagram (does not have tower reactor pressure DCS data);
Fig. 7 is the catalytic cracking dress of certain Petrochemical Enterprises of the online soft sensor method of gasoline stabilizer separation accuracy of the present invention
Set the online soft sensor of the liquefied gas C5+ content of stabilizer and the comparison figure of in-line analyzer operation data;
Fig. 8 is certain Petrochemical Enterprises delayed coking unit of the online soft sensor method of gasoline stabilizer separation accuracy of the present invention
The online soft sensor of the liquefied gas C5+ content of stabilizer and the comparison figure of in-line analyzer operation data;
Fig. 9 is Petrochemical Enterprises (one) the catalytic cracking dress of the online soft sensor method of gasoline stabilizer separation accuracy of the present invention
Set the stable gasoline Lei Shi vapour pressure online soft sensor of stabilizer and the comparison figure of in-line analyzer operation data;
Figure 10 is Petrochemical Enterprises (two) catalytic cracking of the online soft sensor method of gasoline stabilizer separation accuracy of the present invention
The stable gasoline Lei Shi vapour pressure online soft sensor of device stabilizer and the comparison figure of in-line analyzer operation data.
Specific embodiment
Referring to the drawings and embodiment the present invention will be described in detail:
The online soft sensor method of gasoline stabilizer separation accuracy of the present invention, the totality according to attached drawing 1, attached drawing 2 is real respectively
Process is applied, realizes the on-line real-time measuremen of liquefied gas C5+ content, stable gasoline Lei Shi vapour pressure.
For the online soft sensor method of present invention be described in more detail gasoline stabilizer separation accuracy, first in conjunction with attached drawing
5, attached drawing 6 illustrates the process flow and instrument measurement situation of gasoline stabilizer.
Referring to attached drawing 5 or attached drawing 6.Deethanization gasoline from the extraction of desorber bottom enters from gasoline stabilizer middle and upper part,
Stablize bottom reboiler and heat source is provided, the gaseous stream gone out from liquid evaporation on tower reactor, each layer column plate, across the sky of each column plate
Gap flows up;Overhead reflux provides condensation power, and the high component of each layer gaseous stream mid-boiling point is just condensed into the liquid on column plate
Body, to underflow stream;When device operates normally, tower top dome, each layer column plate, tower reactor are in respective vapor liquid equilibrium state.
The space that tower top dome, the i.e. dome of tower and top first layer column plate 1 surround, liquid on first layer column plate and
The interface 2 of vault indoor gas, is the last one separation of ascending vapor logistics composition variation in tower, and top gaseous phase logistics is cold
But, after condensation enters liquid gas storage tank, composition no longer changes, and therefore, liquefied gas C5+ content depends on the indoor gas-liquid of tower top vault
Balance.Thermometric instrument 3 measures tower top dome liquidus temperature, and pressure gauge 4 measures tower top pressure.Liquid gas storage tank extraction logistics, one
Part goes out device as liquefied gas product, measures its flow by flow instrument 5;Another part is as tower top cold reflux, through overcurrent
Amount instrument 6 is squeezed at the top of stabilizer after measuring.
The boundary that the space that tower reactor, i.e. tower lower part the last layer column plate 7 and tower bottom surround, bottom liquid and gas reactor divide
Face 8 is the last one separation of the liquid phase stream composition variation to flow downward in tower, and the vapour pressure of stable gasoline is (with C4 group
Divide content height linear approximate relationship), depending on the vapor liquid equilibrium in tower reactor.Thermometric instrument 9 measures tower bottom liquidus temperature, pressure
Power instrument 10 measures tower reactor pressure.
Above-mentioned temperature, pressure, flow measurement instrument are that the standard of gasoline stabilizer instrument is matched in addition to pressure gauge 10
It sets.
Cracking unit newly-built in recent years or that key technological transformation was carried out into several years, gasoline stabilizer tower reactor usually configure
Pressure gauge 10 (DCS data are shown), referring to attached drawing 5.
The old device of past construction, typically no configuration tower reactor pressure gauge (DCS data are shown), referring to attached drawing 6.This
When, tower reactor pressure can be estimated by tower top pressure and Ta Nei ascending vapor logistics flux by simple operation.
The online soft sensor method of gasoline stabilizer separation accuracy of the present invention, comprising:
Liquefied gas C5+ content soft measuring method and stable gasoline Lei Shi vapour pressure flexible measurement method;
The step of liquefied gas C5+ content soft measuring method:
Step (1.a) determines that the normal minimum pure material of boiling point is the pass for characterizing liquefied gas C5+ content in liquefied gas C5 component
Key component pure material;
It is checked in from the cracking unit analysis data library of place, the normal minimum pure material of boiling point is isoamyl in liquefied gas C5 component
Alkane;
Step (2.a) determines the liquidus temperature T of gas-liquid separation interface in the dome of gasoline stabilizer topTOPAnd PTOP;
Liquid phase is replaced sometimes without temperature measuring instrument with top gaseous phase temperature at this.The instrument measurement goes out Celsius temperature,
Unit DEG C;
Gas-liquid separation interface gaseous pressure PTOP, i.e. tower top pressure measuring instrumentss directly adopt.The instrument measurement goes out phase
To pressure, units MPa;
Step (3.a) is checked in from common chemical engineering/databook: the critical-temperature of isopentane is 462.926K, different
Pentane gas-liquid separation interface liquidus temperature T in tower top domeTOPComparison temperature Tr,C5,TOPAre as follows:
Tr,C5,TOP=(TTOP+273.15)/462.926
Wherein:
273.15 be 0 DEG C of Celsius temperature of kelvin rating
Step (4.a) determines key component pure material saturated vapor pressure under liquidus temperature at tower top dome gas-liquid separation.
Using distinct methods, the saturated vapor calculated is pressed with pressure value or reduced pressure force value.Vapour pressure pressure value and
Reduced pressure force value can be converted by simple computation;It is convenient to state for following model algorithm, which does not repartition pressure
Value and reduced pressure force value;
Liquefied gas C5+ content key component is indicated in the saturated vapor pressure of tower top dome gas-liquid separation interface liquidus temperature
For PVP,C5,TOP;
The saturated vapor pressure of temperature is (right at gas-liquid separation in tower top dome using Riedel method calculating isopentane for embodiment
Than pressure value):
Step (5.a) determines liquefied gas C5+ content soft sensor model:
Wherein:
C5LPGFor liquefied gas C5+ content, unit %;
A0,C5、A1.C5;For model coefficient;
0.101325 is the difference of absolute pressure and relative pressure;
Step (6.a), model coefficient determines according to the following steps:
Step (6.1a), the gasoline stabilizer Process History data that certain period of time is acquired from DCS determine;
The stabilizer procedure parameter for needing to acquire has:
1)TTOP: tower top dome liquidus temperature, DEG C
2)PTOP: tower top dome gaseous pressure, MPa
3) embodiment is configured with in-line analyzer, increases liquefied gas C5+ content AIC5Data;
Data sampling interval 5 minutes;
Data segment time span is established rules then really are as follows:
1) model coefficient is determined using analysis data, then includes continuous 10 or more liquefied gas C5+ contents in the period
Analysis data sample;The amplitude of variation of liquefied gas C5+ content value >=high limit of process chart regulation 1/2;
2) it is configured with in-line analyzer, then includes the significant fluctuating change of quality index: liquefied gas in the period
The fluctuating range of C5+ content >=high limit of process chart regulation 1/3.
The above process is lasted data, imported in Excel table by step (6.2a);
For each procedure parameter, the data of the seclected time period are expressed as an array, such asJ=1,2 ...,
N, n is equal to seclected time period data sample number here;
Step (6.4a), it is first in the Excel table that the above process lasts data and laboratory values or in-line analyzer data
Smothing filtering operation first is carried out to each process parameter data:
(Z)j+1=(1-FL)*(Z)j+FL*(Z)j+1, j=1,2 ..., n
Wherein:
(Z) T is representedTOP、PTOPAnd analyzer data AIC5
FLFor smothing filtering coefficient, value is 0.70~0.85
Then in Excel table, using later process parameter data is filtered, array column are calculated:
Step (6.5a) determines soft-sensing model coefficient using one of following two methods;
Step (6.5.1a) calculates soft-sensing model coefficient referring to the scale calculation method of instrumentation engineering;
For the computational accuracy for improving model coefficient, A1.C5Using calculating the method averaged three times;And A0,C5It is hard measurement
Zero-bit, when off-line calculation, find out a provisional value, and online soft sensor has zero adjustment;
Observation comparative approach is shown using curve, respectively:
I. maximum three sample values of liquefied gas C5+ content laboratory values (or in-line analyzer data) are selected, from big to small
It is expressed as AIC5,MAX1、AIC5,MAX2、AIC5,MAX3, the X of three assay samples sampling instantsC5,jIt is expressed as XC5,MAX1、
XC5,MAX2、XC5,MAX3;
Ii. the smallest three sample values of liquefied gas C5+ content laboratory values (or in-line analyzer data) are selected, from big to small
It is expressed as AIC5,MIN1、AIC5,MIN2、AIC5,MIN3, the X of three assay samples sampling instantsC5,jIt is expressed as XC5,MIN1、
XC5,MIN2、XC5,MIN3;
It calculates and determines A1.C5And A0,C5Value:
A0.C5=AIC5,MIN1-A1.C5*XC5,MIN1
Step (6.5.2a), with XC5For input variable, with AIC5To be detected variable, is returned using PLS and obtain hard measurement
Model coefficient;Regression result needs to verify: A1.C5For positive value.Otherwise data should be chosen again to return again;
Step (7.a), it is determined that after model coefficient, host computer soft instrument is can be used in the canbe used on line of liquefied gas C5+ content
Software, host computer calculation procedure use DCS Function Block Configuration;
Using host computer realize liquefied gas C5+ content online soft sensor, referring to attached drawing 3: host computer soft instrument software or
Calculation procedure obtains gasoline stabilizer top temperature, tower top pressure process data in DCS by OPC, based on process shown in block diagram
After the completion of calculation, liquefied gas C5+ content data feeding DCS is shown or for used in closed-loop control;
Gasoline stabilizer top liquefied gas C5+ content chemical examination/analyzer data is sent into host computer, fixed by correction module
The zero-bit of phase or real time calibration liquefied gas C5+ containing numerical quantity.
It is realized using DCS Function Block Configuration, also can refer to the implementation of 3 configuration of attached drawing.
The step of stable gasoline Lei Shi vapour pressure flexible measurement method:
Step (1.b) determines that the highest pure material of normal boiling point is characterization stable gasoline Lei Shi in C4 component in stable gasoline
The key component pure material of vapour pressure;
It is checked in from the cracking unit analysis data library of place, the highest pure material of normal boiling point is in C4 component in stable gasoline
1,3- butadiene;
Step (2.b) determines the liquidus temperature T of gas-liquid separation interface in gasoline stabilizer tower reactorBTMAnd gaseous pressure
PBTM;
If there is no tower reactor gaseous pressure measuring instrumentss data in DCS, it is determined that gasoline stabilizer top gaseous phase pressure is surveyed
Measure instrument PTOP, liquefied gas output flow measurement instrument FLPGWith overhead reflux flow measurement instrument FRE;Gasoline stabilizer tower reactor pressure
Power is calculated using the method that subsequent step (6.3b) provides;
Step (3.b), checks in from common chemical engineering/databook: the critical-temperature of 1,3-butadiene is
446.223K then 1,3-butadiene gas-liquid separation interface liquidus temperature T in tower reactorBTMComparison temperature Tr,RVP,BTMAre as follows:
Tr,RVP,BTM=(TBTM+273.15)/446.223
Wherein:
273.15 be the kelvin rating of Celsius temperature DEG C
Step (4.b) determines 1,3-butadiene saturated vapor pressure under liquidus temperature at the gas-liquid separation of tower reactor;
Using distinct methods, the saturated vapor calculated is pressed with pressure value or reduced pressure force value.Vapour pressure pressure value and
Reduced pressure force value can be converted by simple computation;It is convenient to state for following model algorithm, which does not repartition pressure
Value and reduced pressure force value:
Saturated vapor pressure table of the stable gasoline Lei Shi vapour pressure key component in tower reactor gas-liquid separation interface liquidus temperature
It is shown as PVP,RVP,BTM。
Embodiment calculates 1,3- butadiene using Riedel method and steams in the saturation of tower reactor gas-liquid separation interface liquidus temperature
Vapour pressure (reduced pressure force value):
Step (5.b) determines stable gasoline Lei Shi vapour pressure soft-sensing model:
Wherein:
RVPSTABFor stable gasoline Lei Shi vapour pressure, unit kPa
A0,RVP、A1,RVPFor model coefficient;
Step (6.b), off-line calculation obtains soft-sensing model coefficient according to the following steps:
Step (6.1b) acquires the gasoline stabilizer Process History data of certain period of time from dcs;
The stabilizer procedure parameter for needing to acquire has:
1)TBTM: tower reactor liquidus temperature, DEG C
2)PBTM: tower reactor gaseous pressure (if there is measuring instrumentss DCS data are shown), MPa
3) such as old device does not have tower reactor gaseous pressure measuring instrumentss DCS data to show, then acquires top gaseous phase pressure
(MPa), liquefied gas output flow FLPGWith overhead reflux flow FRE, two datas on flows, unit is consistent
4) embodiment is configured with in-line analyzer, is further added by stable gasoline vapour pressure AIRVPData;
Sampling interval 2 minutes;
Data segment time span is established rules then really:
1) model coefficient is determined using analysis data, then includes continuous 10 or more stable gasoline vapour pressures in the period
Analysis data sample;Amplitude of variation >=6kPa of stable gasoline vapour pressure;
2) it is configured with in-line analyzer, includes the significant fluctuating change of quality index: stable gasoline in the period
Fluctuating range >=4kPa of vapour pressure;
The above process is lasted data, imported in Excel table by step (6.2b);
For each procedure parameter, the data of the seclected time period are expressed as an array, such as TBTM,j, j=1,2 ...,
N, n is equal to seclected time period data sample number here;
Step (6.3b) jumps to step (6.4b) if having collected tower reactor gaseous pressure data.This step by step,
Gasoline stabilizer tower reactor pressure is calculated using following algorithm:
Stabilizer pressure drop is calculated first estimates coefficient ADP:
Wherein:
ADPFor constant, using mean value computation as a result, not considering unit;
PBTM,0For constant, value is equal to the tower reactor pressure mean value obtained from operating experience or site instructions pressure gauge;
Then tower reactor pressure value array is calculated
PBTM,j=PTOP,j+ADP*(FLPG,j+FRE,j)
Step (6.4b), it is first in the Excel table that the above process lasts data and laboratory values or in-line analyzer data
Smothing filtering operation first is carried out to each process parameter data:
(Z)j+1=(1-FL)*(Z)j+FL*(Z)j+1, j=1,2 ..., n
Wherein:
(Z) T is representedBTM、PBTMAnd analyzer data
FLFor smothing filtering coefficient: for observed temperature, pressure and in-line analyzer data,
FL=0.70~0.85;
For the tower reactor pressure data being calculated, FL=0.85~0.97;
Then in Excel table, using later process parameter data is filtered, array column are calculated:
Step (6.5b) determines soft-sensing model coefficient using one of following two methods;
Step (6.5.1b) calculates soft-sensing model coefficient referring to the scale calculation method of instrumentation engineering;
For the computational accuracy for improving model coefficient, A1,RVPUsing calculating the method averaged three times;And A0,RVPIt is hard measurement
Zero-bit, when off-line calculation find out a provisional value can positive online soft sensor have zero adjustment:
Observation comparative approach is shown using curve, respectively:
I. maximum three sample values of stable gasoline vapour pressure laboratory values (or in-line analyzer data) are selected, from big to small
It is expressed as AIRVP,MAX1、AIRVP,MAX2、AIRVP,MAX3, the X of three assay samples sampling instantsRVP,jIt is expressed as XRVP,MAX1、
XRVP,MAX2、XRVP,MAX3;
Ii. the smallest three sample values of stable gasoline vapour pressure laboratory values (or in-line analyzer data) are selected, from greatly to
It is small to be expressed as AIRVP,MIN1、AIRVP,MIN2、AIRVP,MIN3, the X of three assay samples sampling instantsRVP,jIt is expressed as
XRVP,MIN1、XRVP,MIN2、XRVP,MIN3。
It calculates and determines A1,RVPAnd A0.RVPValue:
A0.RVP=AIRVP,MIN1-A1.RVP*XRVP,MIN1
Step (6.5.2b), with XRVPFor input variable, with AIRVPTo be detected variable, is returned using PLS and obtain hard measurement
Model coefficient;Regression result needs to verify: A1.RVPFor positive value.Otherwise data should be chosen again to return again;
When DCS does not have tower reactor pressure measured data, stable gasoline Lei Shi vapour pressure hard measurement mould that step (5.b) provides
Type, pattern change are as follows:
Wherein:
RVPSTABFor stable gasoline Lei Shi vapour pressure, unit kPa
A0,RVP、A1,RVP、A2,RVPFor model coefficient;
At this point, not executing step (6.3b) and step (6.4b), the mistake that step (6.2b) imports Excel table is directlyed adopt
Journey lasts data and laboratory values or in-line analyzer data, seeks model coefficient using PLS Return Law method:
Smothing filtering operation is carried out to each process parameter data first:
(Z)j+1=(1-FL)*(Z)j+FL*(Z)j+1, j=1,2 ..., n
Wherein:
(Z) T is representedBTM、PTOP、And analyzer data
FLFor smothing filtering coefficient:
For temperature, pressure and in-line analyzer data, FL=0.70~0.85;
For data on flows: FL=0.85~0.97;
Then in Excel table, using later process parameter data is filtered, two array column are calculated:
Finally with X1RVP、X2RVPFor input variable, with AIRVPTo be detected variable, using PLS, that is, Partial Least Squares Regression
Method obtains soft-sensing model coefficient;
Regression result should meet:
(1)A1,RVP、A2,RVPIt is positive value;
(2)Value agreement be equal to stabilizer voltage drop value.F hereinLPGAnd FRE, above-mentioned
It filters and chooses its mean value in later process parameter data.
Otherwise, data should be chosen again to return again;
Step (7.b), it is determined that after model coefficient, the canbe used on line of stable gasoline Lei Shi vapour pressure hard measurement be can be used
Host computer soft instrument software, host computer calculation procedure use DCS Function Block Configuration;
The online soft sensor of stable gasoline Lei Shi vapour pressure is realized using host computer, referring to attached drawing 4: host computer soft instrument is soft
Part or calculation procedure obtain column bottom temperature, tower reactor the pressure (/ tower top pressure, liquid gas flow of gasoline stabilizer in DCS by OPC
Amount and overhead reflux flow) process data, after the completion of workflow management shown in block diagram, by stable gasoline Lei Shi Vapor pressure data
DCS is sent into show or for used in closed-loop control;
Stable gasoline Lei Shi vapour pressure chemical examination/analyzer data is sent into host computer, periodically or real-time by correction module
Calibrate the zero-bit of stable gasoline Lei Shi vapour pressure numerical value.
It is realized using DCS Function Block Configuration, also can refer to the implementation of 4 configuration of attached drawing.
Case verification illustrates
Attached drawing 7-10 is the liquefied gas C5+ content of gasoline stabilizer separation accuracy online soft sensor method, stable gasoline thunder
Family name's vapour pressure hard measurement case verification diagram.
The data of in-line analyzer are relatively accurate, strong real-time.For verify online soft sensor operation result, following four
The example that application example, selection are furnished with corresponding in-line analyzer.
In diagram, hard measurement operation data curve and in-line analyzer operation data curve are placed among the same coordinate system
It is compared.
Example illustrates:
Attached drawing 7, be certain Petrochemical Enterprises Stabilizer In Fccu liquefied gas C5+ content online soft sensor and
The comparison of line analysis instrument operation data, abscissa: the time, the sampling interval 1 minute;Ordinate: liquefied gas C5+ content, unit %
(volume).
Attached drawing 8 is the online soft sensor and in-line analyzer of the delayed coking unit liquefied gas C5+ content of certain Petrochemical Enterprises
The comparison of operation data, abscissa: the time, the sampling interval 1 minute;Ordinate: liquefied gas C5+ content, unit % (volume).
The online soft sensor operation data it can be seen that stabilizer liquefied gas C5+ content is illustrated from attached drawing 7 and 8 liang of attached drawing,
With its in-line analyzer operation data, variation tendency is consistent.
Attached drawing 9 is the online soft survey of the Stabilizer In Fccu stable gasoline Lei Shi vapour pressure of Petrochemical Enterprises (one)
The comparison of amount and in-line analyzer operation data, abscissa: the time, the sampling interval 1 minute;Ordinate: stable gasoline vapour pressure,
Unit kPa.
The Stabilizer In Fccu tower reactor is configured with pressure measuring instruments.
Attached drawing 10 is online soft sensor and the in-line analyzer operation of the stable gasoline Lei Shi vapour pressure of Petrochemical Enterprises (two)
The comparison of data, abscissa: the time, the sampling interval 1 minute;Ordinate: stable gasoline vapour pressure, unit kPa.
The Stabilizer In Fccu tower reactor does not configure pressure measuring instruments.Tower reactor pressure uses tower top pressure, liquid
Change throughput, overhead reflux flow rate calculation.
From attached drawing 9 and 10 liang of attached drawing diagrams it can be seen that the online soft sensor of stabilizer stable gasoline Lei Shi vapour pressure is transported
Row data, with its in-line analyzer operation data, variation tendency is consistent.
The method of the present invention passes through soft survey using based on the existing ordinary temperature of gasoline stabilizer, pressure measuring instruments signal
Online liquefied gas C5+ content, stable gasoline Lei Shi vapour pressure index value in real time is calculated in amount.With the online quality analysis of configuration
Instrument is compared, have it is with low investment, without the small advantage of lag time, maintenance workload.It is promoted and applied with Advanced Control Techniques at this stage
In " data-driven regression analysis " modeling flexible measurement method for being widely used compare, have that input variable is few, relationship model is simple
Clearly, the characteristics of model coefficient is easily obtained;For the gasoline stabilizer that temperature, pressure measuring instruments are assorted, separation accuracy
Soft measuring instrument, as long as determining gain coefficient and zero-bit.The soft-sensing model of Kernel-based methods Analysis on Mechanism, adaptability for working condition
It is greatly improved with reliability.
The method of the present invention with catalytic cracking, delayed coking, be hydrocracked etc. cracking units Vapor recovery unit process gasoline
Stabilizer is research object, for the de- fourth of liquefied gas-gasoline separation tower similar in other oil refining apparatus, such as reformer
The stabilizer of alkane tower, atmospheric and vacuum distillation unit lighter hydrocarbons recovery process, also can refer to application.
The method of the present invention is studied from Advanced Control Techniques application work and is obtained, since its modeling method is based on chemical industry machine
Reason, thus in the engineering technology such as the flowsheeting of cracking unit gasoline stabilizer process, technology Calculation, these can also be applied
Calculation method and computation model.
Embodiment described above only carries out the present invention in the preferred embodiment of cracking unit gasoline stabilizer
Description, is not intended to limit the scope of the present invention, without departing from the spirit of the design of the present invention, ordinary skill
The various changes and improvements that personnel make technical solution of the present invention, or applied to similar other liquefied gas-gasoline separation
Tower should all be placed in the protection scope that claims of the present invention determines.
Claims (6)
1. a kind of online soft sensor system of gasoline stabilizer separation accuracy, which is characterized in that including stabilizer, stabilizer includes
Tower top dome, the space that the dome and top first layer column plate of tower surround, liquid and vault indoor gas on first layer column plate
Interface is the last one separation of ascending vapor logistics composition variation in tower, and top gaseous phase logistics is cooling, condensation enters liquid
After LPG cylinder, composition no longer changes, and therefore, liquefied gas C5+ content depends on the indoor vapor liquid equilibrium of tower top vault;
The sensor of thermometric instrument and the sensor of pressure gauge are arranged in tower top dome;
Thermometric instrument measures tower top dome liquidus temperature, and pressure gauge measures tower top pressure;
Liquid gas storage tank extracts logistics out, and a part goes out device as liquefied gas product, measures its flow by flow instrument;Another part
As tower top cold reflux, squeezed into after flow instrument measures at the top of stabilizer;
Stabilizer includes tower reactor, the space that tower lower part the last layer column plate and tower bottom surround, bottom liquid and gas reactor point
Interface, is the last one separation of the liquid phase stream composition variation to flow downward in tower, and the vapour pressure of stable gasoline depends on
In the vapor liquid equilibrium in tower reactor;
The sensor of thermometric instrument and the sensor of pressure gauge are arranged in tower reactor;
Thermometric instrument measures tower bottom liquidus temperature, and pressure gauge measures tower reactor pressure;
Including liquefied gas C5+ content soft measuring method and stable gasoline Lei Shi vapour pressure flexible measurement method;
The step of liquefied gas C5+ content soft measuring method are as follows:
Step (1.a) determines that the normal minimum pure material of boiling point is crucial group for characterizing liquefied gas C5+ content in liquefied gas C5 component
Divide pure material;
Step (2.a) determines the liquidus temperature T of gas-liquid separation interface in the dome of gasoline stabilizer topTOPWith gaseous pressure PTOP,
What industrial process instrument measurement went out is celsius temperature scale (DEG C) and relative pressure (MPa);
Step (3.a), liquefied gas C5+ content key component pure material gas-liquid separation interface liquidus temperature in tower top dome
TTOPComparison temperature Tr,C5,TOPIt is calculated as follows:
Tr,C5,TOP=(TTOP+273.15)/Tc,C5
Wherein:
Tc,C5For the critical-temperature of liquefied gas C5+ content key component pure material,
273.15 be 0 DEG C of Celsius temperature of kelvin rating;
Step (4.a) determines key component pure material saturated vapor pressure under liquidus temperature at tower top dome gas-liquid separation;
Using distinct methods, the saturated vapor calculated is pressed with pressure value or reduced pressure force value;Vapour pressure pressure value and comparison
Pressure value can be calculated by simple linear and be converted;It is convenient to state for following model algorithm, which does not repartition pressure
Value and its reduced pressure force value;
Liquefied gas C5+ content key component is expressed as in the saturated vapor pressure of tower top dome gas-liquid separation interface liquidus temperature
PVP,C5,TOP;
Step (5.a) determines liquefied gas C5+ content soft sensor model:
Wherein:
C5LPGFor liquefied gas C5+ content, unit %;
A0,C5、A1.C5For model coefficient;
0.101325 is the difference of absolute pressure and relative pressure;
Step (6.a), off-line calculation obtains model coefficient according to the following steps:
Step (6.1a), the gasoline stabilizer Process History data that certain period of time is acquired from dcs determine;
The stabilizer procedure parameter for needing to acquire has:
1)TTOP: tower top dome liquidus temperature, DEG C;
2)PTOP: tower top dome gaseous pressure, MPa;
If 3) be configured with in-line analyzer, it is further added by liquefied gas C5+ content AIC5Data;
Such as without configuration in-line analyzer, then from device analysis data library, then the liquefied gas C5+ content of corresponding period is collected
Analysis data and its sampling instant data;
The above process is lasted data, imported in Excel table by step (6.2a);
For each procedure parameter, the data of the seclected time period are expressed as an array, TTOP,j, j=1,2 ..., n, n etc. here
In seclected time period data sample number;
Step (6.4a) is right first in the Excel table that the above process lasts data and laboratory values or in-line analyzer data
Each process parameter data carries out smothing filtering operation:
(Z)j+1=(1-FL)*(Z)j+FL*(Z)j+1, j=1,2 ..., n
Wherein:
(Z) T is representedTOP、PTOPAnd analyzer data AIC5
FLFor smothing filtering coefficient, value is 0.70~0.85
Then in Excel table, using later process parameter data is filtered, array column are calculated:
Step (6.5a) determines soft-sensing model coefficient using one of following two methods;
Step (6.5.1a) calculates soft-sensing model coefficient referring to the scale calculation method of instrumentation engineering;
Sampling instant, the numerical error of analysis data are considered, for the computational accuracy for improving model coefficient, A1.C5Using calculating three times
The method averaged;And A0,C5It is the zero-bit of hard measurement, when off-line calculation finds out a provisional value;
Observation comparative approach is shown using curve, respectively:
I. maximum three sample values of liquefied gas C5+ content laboratory values are selected, are expressed as AI from big to smallC5,MAX1、AIC5,MAX2、
AIC5,MAX3, the X of three assay samples sampling instantsC5,jIt is expressed as XC5,MAX1、XC5,MAX2、XC5,MAX3;
Ii. the smallest three sample values of liquefied gas C5+ content laboratory values are selected, are expressed as AI from big to smallC5,MIN1、AIC5,MIN2、
AIC5,MIN3, the X of three assay samples sampling instantsC5,jIt is expressed as XC5,MIN1、XC5,MIN2、XC5,MIN3;
It calculates and determines A1.C5And A0,C5Value:
A0.C5=AIC5,MIN 1-A1.C5*XC5,MIN 1
Step (6.5.2a), with XC5For input variable, with AIC5To be detected variable, is returned using PLS and obtain soft-sensing model system
Number A1.C5And A0,C5;Regression result needs to verify: A1.C5For positive value, it otherwise should choose again data and return again;
Step (7.a), it is determined that after model coefficient, it is soft that host computer soft instrument can be used in the canbe used on line of liquefied gas C5+ content
Part, host computer calculation procedure use DCS Function Block Configuration;
Liquefied gas C5+ content soft instrument or calculation procedure pass through gasoline stabilizer tower top temperature, tower top pressure in OPC acquisition DCS
Liquefied gas C5+ content data is sent into DCS after the completion of calculating and shown or for used in closed-loop control by process data;Tower top liquefied gas
C5+ content chemical examination/analyzer data, be sent into host computer in, by correction module periodically or real time calibration liquefied gas C5+ content number
The zero-bit of value;
The step of stable gasoline Lei Shi vapour pressure flexible measurement method are as follows:
Step (1.b) determines that the highest pure material of normal boiling point is characterization stable gasoline Lei Shi steam in C4 component in stable gasoline
The key component pure material of pressure;
Step (2.b) determines the liquidus temperature T of gas-liquid separation interface in gasoline stabilizer tower reactorBTMGaseous pressure PBTM, industry
What process instrument was measured is celsius temperature scale (DEG C) and relative pressure (MPa);
If there is no tower reactor gaseous pressure measuring instrumentss data in DCS, it is determined that gasoline stabilizer top gaseous phase pressure-measuring device
Table PTOP, liquefied gas output flow measurement instrument FLPGWith overhead reflux flow measurement instrument FRE, subsequent step (6.3b) provides vapour
The calculation method of oily stabilizer tower reactor pressure;
Step (3.b), stable gasoline Lei Shi vapour pressure key component pure material gas-liquid separation interface liquidus temperature in tower reactor
TBTMComparison temperature Tr,RVP,BTMIt is calculated as follows:
Tr,RVP,BTM=(TBTM+273.15)/Tc,RVP
Wherein:
Tc,RVPFor the critical-temperature of stable gasoline Lei Shi vapour pressure key component pure material,
Unit K
273.15 be 0 DEG C of Celsius temperature of kelvin rating
Step (4.b) determines stable gasoline Lei Shi vapour pressure key component pure material under liquidus temperature at tower reactor gas-liquid separation
Saturated vapor pressure;
Using distinct methods, the saturated vapor calculated is pressed with pressure value or reduced pressure force value;Vapour pressure pressure value and comparison
Pressure value can be calculated by simple linear and be converted, and convenient for the statement of following model algorithm, which does not repartition pressure
Value and reduced pressure force value;
Stable gasoline Lei Shi vapour pressure key component is expressed as in the saturated vapor pressure of tower reactor gas-liquid separation interface liquidus temperature
PVP,RVP,BTM;
Step (5.b) determines stable gasoline Lei Shi vapour pressure soft-sensing model:
Wherein:
RVPSTABFor stable gasoline Lei Shi vapour pressure, unit kPa
A0,RVP、A1,RVPFor model coefficient
0.101325 is the difference of absolute pressure and relative pressure, units MPa
Step (6.b), off-line calculation obtains soft-sensing model coefficient according to the following steps:
Step (6.1b) acquires the gasoline stabilizer Process History data of certain period of time from dcs;
The stabilizer procedure parameter for needing to acquire has:
1)TBTM: tower reactor liquidus temperature, DEG C;
2)PBTM: tower reactor gaseous pressure, MPa;
3) such as tower reactor gaseous pressure does not have measuring instrumentss DCS data to show, then acquires top gaseous phase pressure (MPa), and liquefied gas produces
Outflow FLPGWith overhead reflux flow FRE, two of them data on flows is constant when unit is consistent and hard measurement on-line implement
It can;
4) it is such as configured with in-line analyzer, then is further added by stable gasoline vapour pressure AIRVPData;
Such as without configuration in-line analyzer, then from device analysis data library, then the stable gasoline steam of corresponding period is collected
Press analysis data and its sampling instant data;
The above process is lasted data, imported in Excel table by step (6.2b);
For each procedure parameter, the data of the seclected time period are expressed as an array, TBTM,j, j=1,2 ..., n, n etc. here
In seclected time period data sample number;
Step (6.3b) jumps to step (6.4b) if having collected tower reactor gaseous pressure data;This is used step by step
Following algorithm calculates gasoline stabilizer tower reactor pressure:
Stabilizer pressure drop is calculated first estimates coefficient ADP:
Wherein:
ADPFor constant, using mean value computation as a result, not considering unit;
PBTM,0For constant, value is equal to the tower reactor pressure mean value obtained from operating experience or site instructions pressure gauge;
Then tower reactor pressure value array is calculated
PBTM,j=PTOP,j+ADP*(FLPG,j+FRE,j)
Step (6.4b) is right first in the Excel table that the above process lasts data and laboratory values or in-line analyzer data
Each process parameter data carries out smothing filtering operation:
(Z)j+1=(1-FL)*(Z)j+FL*(Z)j+1, j=1,2 ..., n
Wherein:
(Z) T is representedBTM、PBTMAnd analyzer data
FLFor smothing filtering coefficient: for observed temperature, pressure and in-line analyzer data,
FL=0.70~0.85;
For the tower reactor pressure data being calculated, FL=0.85~0.97;
Then in Excel table, using later process parameter data is filtered, array column are calculated:
Step (6.5b) determines soft-sensing model coefficient using one of following two methods;
Step (6.5.1b) calculates soft-sensing model coefficient referring to the scale calculation method of instrumentation engineering;
Sampling instant, the numerical error of analysis data are considered, for the computational accuracy for improving model coefficient, A1,RVPUsing calculating three times
The method averaged;And A0,RVPIt is the zero-bit of hard measurement, when off-line calculation finds out a provisional value;
Observation comparative approach is shown using curve, respectively:
I. maximum three sample values of stable gasoline vapour pressure laboratory values are selected, are expressed as AI from big to smallRVP,MAX1、AIRVP,MAX2、
AIRVP,MAX3, the X of three assay samples sampling instantsRVP,jIt is expressed as XRVP,MAX1、XRVP,MAX2、XRVP,MAX3;
Ii. the smallest three sample values of stable gasoline vapour pressure laboratory values are selected, are expressed as AI from big to smallRVP,MIN1、
AIRVP,MIN2、AIRVP,MIN3, the X of three assay samples sampling instantsRVP,jIt is expressed as XRVP,MIN1、XRVP,MIN2、XRVP,MIN3;
It calculates and determines A1,RVPAnd A0,RVPValue:
A0.RVP=AIRVP,MIN 1-A1.RVP*XRVP,MIN 1
Step (6.5.2b), with XRVPFor input variable, with AIRVPTo be detected variable, is returned using PLS and obtain soft-sensing model
Coefficient;Regression result needs to verify: A1.RVPFor positive value;Otherwise data should be chosen again to return again;
When DCS does not have tower reactor pressure measured data, the soft survey of stable gasoline Lei Shi vapour pressure that step (5.b) is provided also can be changed
The pattern of model is measured, and model coefficient is sought using the PLS Return Law;
Step (7.b), it is determined that after model coefficient, the canbe used on line of stable gasoline Lei Shi vapour pressure hard measurement be can be used upper
Machine soft instrument software, host computer calculation procedure use DCS Function Block Configuration;
Stable gasoline Lei Shi vapour pressure soft instrument or calculation procedure obtain column bottom temperature, the tower of gasoline stabilizer in DCS by OPC
Stable gasoline Lei Shi Vapor pressure data is sent into DCS after the completion of calculating and shown or for used in closed-loop control by kettle press process data;
Stable gasoline Lei Shi vapour pressure chemical examination/analyzer data is sent into host computer, by the way that correction module is regular or real time calibration is stablized
The zero-bit of gasoline Lei Shi vapour pressure numerical value.
2. the online soft sensor system of gasoline stabilizer separation accuracy according to claim 1, it is characterised in that: step
(2.a) can use tower top first layer column plate if gas-liquid separation interface does not have temperature measuring instrument in the dome of gasoline stabilizer top
Lower temperature or top gaseous phase temperature replace;Gas-liquid separation interface gaseous pressure PTOP, it is configured with tower top pressure measuring instrumentss number
Value, directlys adopt.
3. the online soft sensor system of gasoline stabilizer separation accuracy according to claim 2, it is characterised in that: step
(4.a) and step (4.b) determines that key component pure material is satisfied under liquidus temperature at gas-liquid separation in tower top dome or tower reactor
With vapour pressure calculation method are as follows:
1) pure material hold-up vapour pressure is calculated using Riedel method:
Wherein:
For pure material vapour pressure reduced pressure, no unit;
TrTemperature, no unit are compared for pure material;
A, B, C, D are Riedel method vapour pressure algorithm coefficient;
2) pure material hold-up vapour pressure is calculated using Lee-Kesler method:
Wherein:
For pure material vapour pressure reduced pressure, no unit;
TrTemperature, no unit are compared for pure material;
ω is pure material eccentric factor, no unit.
4. the online soft sensor system of gasoline stabilizer separation accuracy according to claim 3, it is characterised in that: step
(6.1a), from dcs acquire certain period of time gasoline stabilizer Process History data, the sampling interval 1,2 or
5 minutes;
Time segment length is established rules then really:
1) model coefficient is determined using analysis data, then includes continuous 10 or more liquefied gas C5+ content chemical examinations in the period
Data sample;The amplitude of variation of liquefied gas C5+ content value >=high limit of process chart regulation 1/2;
2) it is configured with in-line analyzer, then include the significant fluctuating change of quality index in the period: liquefied gas C5+ contains
The fluctuating range of amount >=high limit of process chart regulation 1/3;
To make quality index during acquiring data have significant change, operation adjustment appropriate can be carried out to stabilizer.
5. the online soft sensor system of gasoline stabilizer separation accuracy according to claim 4, it is characterised in that: step
(6.1b), from dcs acquire certain period of time gasoline stabilizer Process History data, the sampling interval 1,2 or
5 minutes;
Time segment length is established rules then really:
1) model coefficient is determined using analysis data, then includes continuous 10 or more stable gasoline vapour pressure chemical examinations in the period
Data sample;Amplitude of variation >=6kPa of stable gasoline vapour pressure;
2) it is configured with in-line analyzer, then include the significant fluctuating change of quality index in the period: stable gasoline steams
Fluctuating range >=4kPa of vapour pressure;
To make quality index during acquiring data have significant change, operation adjustment appropriate can be carried out to stabilizer.
6. the online soft sensor system of gasoline stabilizer separation accuracy according to claim 5, it is characterised in that: step
(6.5.2b), when DCS does not have tower reactor pressure measured data, stable gasoline Lei Shi vapour pressure hard measurement that step (5.b) provides
Model, pattern are changeable are as follows:
Wherein:
RVPSTABFor stable gasoline Lei Shi vapour pressure, unit kPa
A0,RVP、A1,RVP、A2,RVPFor model coefficient
0.101325 is the difference of absolute pressure and relative pressure, units MPa
PTOPFor gasoline stabilizer tower top pressure, relative pressure, MPa
FLPG、FREFor gasoline stabilizer flow of lpg and overhead reflux flow, when unit is consistent and hard measurement on-line implement not
Change
PVP,RVP,BTMFor stable gasoline Lei Shi vapour pressure key component tower reactor gas-liquid separation interface liquidus temperature saturated vapor
Pressure
At this point, not executing step (6.3b) and step (6.4b), the process for directlying adopt step (6.2b) importing Excel table is gone through
When data and laboratory values or in-line analyzer data, model coefficient is sought using the PLS Return Law:
Smothing filtering operation is carried out to each process parameter data first:
(Z)j+1=(1-FL)*(Z)j+FL*(Z)j+1, j=1,2 ..., n
Wherein:
(Z) T is representedBTM、PTOP、And analyzer data
FLFor smothing filtering coefficient:
For temperature, pressure and in-line analyzer data, FL=0.70~0.85;
For data on flows: FL=0.85~0.97;
Then in Excel table, using later process parameter data is filtered, two array column are calculated:
Finally with X1RVP、X2RVPFor input variable, with AIRVPTo be detected variable, obtained using PLS, that is, partial least-squares regression method
Obtain soft-sensing model coefficient;
Regression result should meet:
(1)A1,RVP、A2,RVPIt is positive value;
(2)Value agreement be equal to stabilizer voltage drop value;F hereinLPGAnd FRE, in above-mentioned filtering
Its mean value is chosen in later process parameter data;
Otherwise, data should be chosen again to return again.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201610633279.6A CN107057749B (en) | 2016-08-04 | 2016-08-04 | The online soft sensor system and measurement method of gasoline stabilizer separation accuracy |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201610633279.6A CN107057749B (en) | 2016-08-04 | 2016-08-04 | The online soft sensor system and measurement method of gasoline stabilizer separation accuracy |
Publications (2)
Publication Number | Publication Date |
---|---|
CN107057749A CN107057749A (en) | 2017-08-18 |
CN107057749B true CN107057749B (en) | 2018-12-07 |
Family
ID=59616755
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201610633279.6A Active CN107057749B (en) | 2016-08-04 | 2016-08-04 | The online soft sensor system and measurement method of gasoline stabilizer separation accuracy |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN107057749B (en) |
Families Citing this family (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN110551517B (en) * | 2019-09-12 | 2021-08-31 | 扬子石化-巴斯夫有限责任公司 | Gasoline reflux adjusting method for quenching oil tower |
Citations (6)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US4345463A (en) * | 1980-04-22 | 1982-08-24 | Electronic Associates, Inc. | On-line gas measurement and analysis system |
JPS6457158A (en) * | 1987-08-27 | 1989-03-03 | Nippon Oil Co Ltd | Flash point measuring apparatus for middle cut and flash point control method |
CN1996192A (en) * | 2006-12-28 | 2007-07-11 | 浙江大学 | Industrial soft measuring instrument based on bionic intelligence and soft measuring method therefor |
CN101017373A (en) * | 2006-12-28 | 2007-08-15 | 浙江大学 | Industrial process multiresolution softsensoring instrument and method thereof |
CN204536317U (en) * | 2015-04-27 | 2015-08-05 | 国电长源湖北生物质气化科技有限公司 | A kind of biomass gasification furnace gas humidity on-line measurement system |
CN105403683A (en) * | 2015-12-14 | 2016-03-16 | 石化盈科信息技术有限责任公司 | On-line soft measuring method for refinery enterprise heating furnace fuel gas calorific values |
-
2016
- 2016-08-04 CN CN201610633279.6A patent/CN107057749B/en active Active
Patent Citations (6)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US4345463A (en) * | 1980-04-22 | 1982-08-24 | Electronic Associates, Inc. | On-line gas measurement and analysis system |
JPS6457158A (en) * | 1987-08-27 | 1989-03-03 | Nippon Oil Co Ltd | Flash point measuring apparatus for middle cut and flash point control method |
CN1996192A (en) * | 2006-12-28 | 2007-07-11 | 浙江大学 | Industrial soft measuring instrument based on bionic intelligence and soft measuring method therefor |
CN101017373A (en) * | 2006-12-28 | 2007-08-15 | 浙江大学 | Industrial process multiresolution softsensoring instrument and method thereof |
CN204536317U (en) * | 2015-04-27 | 2015-08-05 | 国电长源湖北生物质气化科技有限公司 | A kind of biomass gasification furnace gas humidity on-line measurement system |
CN105403683A (en) * | 2015-12-14 | 2016-03-16 | 石化盈科信息技术有限责任公司 | On-line soft measuring method for refinery enterprise heating furnace fuel gas calorific values |
Non-Patent Citations (2)
Title |
---|
"催化裂化装置稳定汽油饱和蒸气压的在线软测量与实时控制";周丽君等;《内蒙古石油化工》;20010930;第27卷(第3期);第124-127页 * |
"稳定塔控制汽油饱和蒸汽压和液化气碳五含量的优化操作";蒋定建等;《石油化工应用》;20160229;第35卷(第2期);第107-110,118页 * |
Also Published As
Publication number | Publication date |
---|---|
CN107057749A (en) | 2017-08-18 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CA2397555A1 (en) | Petroleum distillation method and system | |
RU2498274C2 (en) | System and method for alkylation process analysis | |
US20170269559A1 (en) | System and Method for Online Measurement of Vapor Pressure in Hydrocarbon Process Streams | |
CA2577781A1 (en) | Refinery crude unit performance monitoring using advanced analytic techniques for raw material quality prediction | |
RU2678271C1 (en) | Method for obtaining reporting data on gas-condensate characteristics of plastic gas for stables under abnormally high plate pressure | |
CN108760789A (en) | A kind of crude oil fast evaluation method | |
CN107057749B (en) | The online soft sensor system and measurement method of gasoline stabilizer separation accuracy | |
CN103906553B (en) | The high accuracy correlating method of equilibrium and the evaluation method that balances each other | |
CN109978059A (en) | A kind of crude oil distillation process primary distillation tower rushes tower fault early warning method | |
CN107207975A (en) | Apparatus and method for studying naphtha reforming process | |
CN106198405B (en) | System for monitoring hydrogen-oxygen stable isotope ratio of atmospheric water vapor | |
RU2335526C2 (en) | Device and method of hydrogen to carbon ratio determination and adjustment in liquid fraction of pyrolysis product | |
EP3463606B1 (en) | System and method for online measurement of vapor pressure in hydrocarbon process streams | |
WO2017062416A1 (en) | Hybrid analyzer for fluid processing processes | |
US20190257808A1 (en) | Chemo-Metrical Prediction of Methane Index for the Natural Gas | |
Wang et al. | Experimental and predicted vapor–liquid equilibrium for binary systems with diethanolamine, m-cresol and p-cresol at 20.0 kPa | |
CN106118715B (en) | Petroleum fractionator operation conditions observer | |
Friedman et al. | First-principles distillation inference models for product quality prediction | |
Rathbun | Programmed automation of modulator cold jet flow for comprehensive two-dimensional gas chromatographic analysis of vacuum gas oils | |
US3392589A (en) | Specific gravity measurement system | |
CN109603184B (en) | Approximate quick calculation method for rectifying tower section and control method for rectifying tower | |
CN109988595A (en) | Screening method of composite neutralizer for fractionating tower top in crude oil distillation | |
CN104215493B (en) | Vacuum distillation apparatus and test method for determining content of polycyclic aromatic hydrocarbon in light lubricating oil | |
CN101545890B (en) | Method for on-line measurement of sec-butyl alcohol content in butylene hydration product | |
JP2016159208A (en) | Crude petroleum composition estimation method, absorption process simulation method for absorbent liquefying facility, and process simulation method for recovery facility |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
PB01 | Publication | ||
PB01 | Publication | ||
SE01 | Entry into force of request for substantive examination | ||
SE01 | Entry into force of request for substantive examination | ||
GR01 | Patent grant | ||
GR01 | Patent grant |