CN114819193A - Data processing method and system for liquefied natural gas storage tank operation data mining - Google Patents

Data processing method and system for liquefied natural gas storage tank operation data mining Download PDF

Info

Publication number
CN114819193A
CN114819193A CN202210744755.7A CN202210744755A CN114819193A CN 114819193 A CN114819193 A CN 114819193A CN 202210744755 A CN202210744755 A CN 202210744755A CN 114819193 A CN114819193 A CN 114819193A
Authority
CN
China
Prior art keywords
characteristic value
storage tank
temperature
liquefied natural
natural gas
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Granted
Application number
CN202210744755.7A
Other languages
Chinese (zh)
Other versions
CN114819193B (en
Inventor
陈雷
张超
顾佳
刘刚
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
China University of Petroleum East China
Original Assignee
China University of Petroleum East China
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by China University of Petroleum East China filed Critical China University of Petroleum East China
Priority to CN202210744755.7A priority Critical patent/CN114819193B/en
Publication of CN114819193A publication Critical patent/CN114819193A/en
Application granted granted Critical
Publication of CN114819193B publication Critical patent/CN114819193B/en
Active legal-status Critical Current
Anticipated expiration legal-status Critical

Links

Images

Classifications

    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
    • G06N20/00—Machine learning
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
    • G06F16/20—Information retrieval; Database structures therefor; File system structures therefor of structured data, e.g. relational data
    • G06F16/24—Querying
    • G06F16/245—Query processing
    • G06F16/2458—Special types of queries, e.g. statistical queries, fuzzy queries or distributed queries
    • G06F16/2465—Query processing support for facilitating data mining operations in structured databases
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F2216/00—Indexing scheme relating to additional aspects of information retrieval not explicitly covered by G06F16/00 and subgroups
    • G06F2216/03—Data mining

Landscapes

  • Engineering & Computer Science (AREA)
  • Theoretical Computer Science (AREA)
  • Software Systems (AREA)
  • Physics & Mathematics (AREA)
  • General Engineering & Computer Science (AREA)
  • Databases & Information Systems (AREA)
  • Data Mining & Analysis (AREA)
  • Mathematical Physics (AREA)
  • General Physics & Mathematics (AREA)
  • Evolutionary Computation (AREA)
  • Computing Systems (AREA)
  • Medical Informatics (AREA)
  • Computer Vision & Pattern Recognition (AREA)
  • Artificial Intelligence (AREA)
  • Fuzzy Systems (AREA)
  • Probability & Statistics with Applications (AREA)
  • Computational Linguistics (AREA)
  • Filling Or Discharging Of Gas Storage Vessels (AREA)

Abstract

The invention belongs to the technical field of data processing, and provides a data processing method and a data processing system for running data mining of a liquefied natural gas storage tank, based on LNG storage tank heat exchange and phase change physical models, the original operation data of the LNG storage tank is recombined or calculated, a plurality of characteristic values are reconstructed on the basis of considering parameters or relations such as the difference between the temperature inside and outside the storage tank, the temperature difference between the fed liquefied natural gas and the liquefied natural gas in the storage tank, the operation parameters of equipment, the relation between the physical state parameters of the gas phase in the tank and the volume related parameters, the relation between the physical state parameters of the gas phase at the outlet of a compressor and the discharge capacity, and the like, and the fitting is carried out on the plurality of characteristic values, and the result shows that compared with the fitting of the original characteristic values, fitting is carried out on the basis of a plurality of characteristic values constructed in the invention, so that the occurrence of overfitting is greatly reduced, and the reliability of running data mining in digital twin can be ensured.

Description

Data processing method and system for liquefied natural gas storage tank operation data mining
Technical Field
The invention belongs to the technical field of data processing, and particularly relates to a data processing method and a data processing system for liquefied natural gas storage tank operation data mining.
Background
For liquefied natural gas (Liquefied Natural Gas，LNG) When the storage tank carries out digital twinning in the physical operation process, the change process of each parameter in the process flow needs to be accurately described.LNGThe storage tank has complex process flow, multiple operating parameters and evaporated gas (C)Boil-Off Gas，BOG) The volatilization volume of the system is influenced by factors such as environment temperature, liquid level in the tank, heat insulation effect of the pipe wall, liquid receiving and sending of the storage tank and the like, the traditional mechanism model has deviation, and data mining is necessary to be carried out on actual operation data to seek the internal correlation of the operation data.
The inventor finds that the traditional machine learning method is directly utilized for miningLNGThe problem of overfitting of actual operation data of the storage tank exists, the epitaxial simulation of the operation condition is not facilitated, and the precision and the application range of digital twins are limited.
Disclosure of Invention
The invention provides a data processing method and a data processing system for mining operation data of a liquefied natural gas storage tank, which aim to solve the problems and belong to the field of data processingLNGA storage tank operation physical process digital twin operation data preprocessing method; the invention reduces the occurrence of overfitting to the maximum extent and can ensure the reliability of the running data mining in the digital twin.
In order to achieve the purpose, the invention is realized by the following technical scheme:
in a first aspect, the present invention provides a data processing method for liquefied natural gas storage tank operation data mining, including:
obtaining a first characteristic value according to the difference value between the temperature inside the storage tank and the temperature outside the storage tank; obtaining a second characteristic value according to the difference between the temperature of the fed liquefied natural gas and the temperature of the liquefied natural gas in the storage tank; constructing a third characteristic value based on the equipment operation parameters; obtaining a fourth characteristic value according to the derivative of the temperature of the liquefied natural gas in the tank to the time; obtaining a fifth characteristic value according to the derivative of the algebraic operation result of the physical state parameter and the volume related parameter of the gas phase in the tank to the time; obtaining a sixth characteristic value according to algebraic operation of gas-phase physical state parameters and discharge capacity at the outlet of the compressor;
and according to a machine learning algorithm, performing data fitting by taking one or more of the first characteristic value, the second characteristic value, the third characteristic value, the fourth characteristic value, the fifth characteristic value and the sixth characteristic value as a label and taking other characteristic values as input.
Further, the first characteristic value is equal to the difference between the temperature of the outer wall of the storage tank and the temperature of the liquefied natural gas in the storage tank.
Further, the second characteristic value is equal to the difference between the temperature of the fed liquefied natural gas and the temperature of the liquefied natural gas in the storage tank multiplied by the flow of the feed port.
Further, the third characteristic value is equal to a bottom-of-tank pump displacement.
Further, the fourth characteristic value is equal to the derivative of the temperature of the liquefied natural gas in the tank with respect to time multiplied by the liquid level height of the liquefied natural gas.
Further, the fifth characteristic value is equal to a ratio of a difference value between the height of the storage tank and the liquid level height of the liquefied natural gas multiplied by the gas phase pressure to the temperature of the boil-off gas, and then the ratio is subjected to time derivation.
Further, the sixth characteristic value is equal to a ratio of a product of the compressor outlet pressure and the compressor outlet flow to the compressor outlet temperature.
In a second aspect, the present invention also provides a data processing system for liquefied natural gas storage tank operation data mining, comprising:
a feature value construction module configured to: obtaining a first characteristic value according to the difference value between the temperature inside the storage tank and the temperature outside the storage tank; obtaining a second characteristic value according to the difference between the temperature of the fed liquefied natural gas and the temperature of the liquefied natural gas in the storage tank; constructing a third characteristic value based on the equipment operation parameters; obtaining a fourth characteristic value according to the derivative of the temperature of the liquefied natural gas in the tank to the time; obtaining a fifth characteristic value according to the derivative of the algebraic operation result of the physical state parameter and the volume related parameter of the gas phase in the tank to the time; obtaining a sixth characteristic value according to algebraic operation of gas-phase physical state parameters and discharge capacity at the outlet of the compressor;
a data fitting module configured to: and according to a machine learning algorithm, performing data fitting by taking one or more of the first characteristic value, the second characteristic value, the third characteristic value, the fourth characteristic value, the fifth characteristic value and the sixth characteristic value as a label and taking other characteristic values as input.
Compared with the prior art, the invention has the beneficial effects that:
the invention is based onLNGStorage tank heat exchange and phase change physical model, pairLNGThe original operation data of the storage tank is recombined or calculated, a plurality of characteristic values are reconstructed on the basis of considering parameters or relations such as the difference between the temperature inside and outside the storage tank, the temperature difference between the fed liquefied natural gas and the liquefied natural gas in the storage tank, the operation parameters of equipment, the relation between the physical state parameters of the gas phase in the tank and the volume related parameters, the relation between the physical state parameters of the gas phase at the outlet of the compressor and the displacement and the like, and the plurality of characteristic values are fitted.
Drawings
The accompanying drawings, which form a part hereof, are included to provide a further understanding of the present embodiments, and are incorporated in and constitute a part of this specification, illustrate exemplary embodiments of the present embodiments and together with the description serve to explain the present embodiments without unduly limiting the present embodiments.
FIG. 1 shows the fitting effect of original features in example 1 of the present invention;
FIG. 2 is a graph showing the fitting effect after recombining characteristic values according to example 1 of the present invention;
FIG. 3 is a schematic diagram of data preprocessing based on a thermal equilibrium physical model of a storage tank system according to embodiment 1 of the present invention.
The specific implementation mode is as follows:
the invention is further described with reference to the following figures and examples.
It should be noted that the following detailed description is exemplary and is intended to provide further explanation of the disclosure. Unless defined otherwise, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this application belongs.
Example 1:
the embodiment provides a data processing method for liquefied natural gas storage tank operation data mining, which comprises the following steps:
obtaining a first characteristic value according to the difference value between the temperature inside the storage tank and the temperature outside the storage tank; obtaining a second characteristic value according to the difference between the temperature of the fed liquefied natural gas and the temperature of the liquefied natural gas in the storage tank; constructing a third characteristic value based on the equipment operation parameters; obtaining a fourth characteristic value according to the derivative of the temperature of the liquefied natural gas in the tank to the time; obtaining a fifth characteristic value according to the derivative of the algebraic operation result of the physical state parameter and the volume related parameter of the gas phase in the tank to the time; obtaining a sixth characteristic value according to algebraic operation of gas-phase physical state parameters and discharge capacity at the outlet of the compressor;
and according to a machine learning algorithm, performing data fitting by taking one or more of the first characteristic value, the second characteristic value, the third characteristic value, the fourth characteristic value, the fifth characteristic value and the sixth characteristic value as a label and taking other characteristic values as input.
The new characteristic value obtained by the method in the embodiment can be effectively solvedLNGOverfitting problems that are trapped when the tank is running raw data machine learning, makeLNGThe extendibility of the machine learning result of the storage tank operation data is better, and a new characteristic value is provided for the storage tank operation data mining based on a machine learning method; specifically, based onLNGPhysical model of heat exchange and phase change of storage tank, pairLNGThe original operation data of the storage tank is recombined or calculated to obtain a new characteristic value, which comprises a heat input part of the storage tank,LNGA heat storage portion and a phase change heat portion; wherein, the storage tank heat input part: for heat input caused by temperature difference, constructing a new characteristic value based on the original temperature data difference; for heat input caused by a heat source of equipment, constructing a new characteristic value based on main parameters of equipment operation; in canLNGA heat storage part: get jar insideLNGThe derivative of the temperature with time and constructing a new characteristic value according to the derivative; phase change heat portion: to the interior of the tankBOGTaking the derivative of the algebraic operation result of the physical state parameters and the volume related parameters of the gas phase in the tank to the time,constructing a new characteristic value according to the characteristic value; for discharge outside the tankBOGAnd obtaining an algebraic operation result of original data such as gas-phase physical state parameters and discharge capacity of the compressor outlet, and constructing a new characteristic value according to the algebraic operation result.
Constructing a first characteristic value forLNGExtracting first characteristic value from the related original data of the internal and external temperature of the storage tankX 1 Which is aT W AndT L as a result of the recombination in the form ofT W -T L . Wherein the content of the first and second substances,T W representing the temperature of the outer wall of the storage tank,T L representing the inside of a storage tankLNGAnd (3) temperature.
Constructing second characteristic values forLNGExtracting a second characteristic value from the raw data related to the cold/heat input of the storage tankX 2 Which is aQ I 、T I AndT L as a result of the recombination in the form ofQ I （T I -T L ). Wherein the content of the first and second substances,Q I representing the feed inlet flow;T I represents the feedLNG(ii) temperature;T L representing the inside of a storage tankLNGAnd (3) temperature.
Establishing a third characteristic value, and establishing a new characteristic value based on main parameters of equipment operation aiming at heat input caused by a heat source of the equipment; the specific method comprisesLNGThe storage tank generates heat by a tank bottom pump, and a third characteristic value is extracted according to the related original data of the heat source of the equipmentX 3 ：Q P . Wherein the content of the first and second substances,Q P representing the tank bottom pump displacement.
Constructing a fourth characteristic value for the inside of the tankLNG heat storage part, getLNGThe derivative of the temperature with time and constructing a new characteristic value according to the derivative; the specific method is to extract a fourth characteristic valueX 4 Which is aLAndT L as a result of the recombination in the form of
Figure 267552DEST_PATH_IMAGE001
. Wherein the content of the first and second substances,T L representative tankLNG(ii) temperature;LrepresentsLNGThe liquid level height;trepresenting time.
Constructing a fifth characteristic value aiming at the phase change heat part in the tankBOGTaking the derivative of the algebraic operation result of the physical state parameters and the volume related parameters of the gas phase in the tank to the time, and constructing a new characteristic value according to the derivative; the specific method is to extract the fifth characteristic valueX 5 Which is ap、H tank 、LAndT g as a result of the recombination in the form of
Figure 928603DEST_PATH_IMAGE002
. Wherein the content of the first and second substances,prepresents the gas phase pressure;T g representsBOG(ii) temperature;H tank which represents the height of the storage tank,m；LrepresentsLNGThe height of the liquid level.
Constructing a sixth characteristic value for the phase change heat portion discharged outside the tankBOGMeasuring, namely obtaining an algebraic operation result of original data such as gas-phase physical state parameters and discharge capacity of an outlet of the compressor, and constructing a new characteristic value according to the algebraic operation result; the specific method is to extract the sixth characteristic valueX 6 Which is ap dis 、Q dis AndT dis as a result of the recombination in the form of
Figure 433533DEST_PATH_IMAGE003
. Wherein the content of the first and second substances,p dis represents the compressor outlet pressure;Q dis representing compressor outlet flow;T dis representing the compressor outlet temperature.
FIG. 3 is a schematic diagram of data preprocessing based on a physical model of thermal equilibrium of the storage tank system; from the thermal point of view, toLNGThe inner space of the storage tank is a research object, and the balance between heat input and conversion in unit time exists.LNGThe heat input in the storage tank mainly comprisesLNGThe heat conducted into the storage tank from the environment around the storage tank,LNGCold/heat input from tank feed andLNGstorage tank bottom pump operation productHeating; based on the law of conservation of energy, inputLNGThe heat of the tank being increased in addition toLNGThe residual heat is outside the temperatureLNGUndergoes a phase change toBOGOne part of the gas phase is accumulated in the gas phase space of the storage tank to cause the pressure, the temperature and the height of the gas phase space to change, and the other part of the gas phase is pumped out of the tank by the compressor.
The heat input part data preprocessing is to divide all the data collected by the heat input part into three subdata sets for preprocessing, including heat conducting part data preprocessing,LNGThe method comprises the steps of preprocessing data of a storage tank feeding initiation cold/heat input part and preprocessing data of a tank bottom pump operation heat production part, and extracting first characteristic values respectivelyX 1 The second characteristic valueX 2 And a third characteristic valueX 3 。
Partial data of heat conductivity is preprocessed based on heat conductivity calculation equation byLNGTemperature of external monitoring point of storage tank and internal of storage tankLNGMeasuring heat conductivity by temperature difference, and extracting first characteristic valueX 1 ：T W -T L 。
Figure 387583DEST_PATH_IMAGE004
Wherein the content of the first and second substances,q c indicating the heat input caused by the heat conduction of the tank body,W；λ e the equivalent thermal conductivity of the storage tank insulating layer is shown,W/（mK）；A tank the external surface area of the tank is indicated,m 2 ；T W the temperature of a temperature monitoring point near the outer surface of the storage tank heat insulation layer is represented,K；T L indicating storage tanksLNGThe temperature of the mixture is controlled by the temperature,Kthe temperature of the bottom of the tank and the temperature of the top of the tank are averaged to obtain the temperature;
Figure 40281DEST_PATH_IMAGE005
the distance between the inner surface temperature monitoring point and the inner wall of the tank is shown,m。
as can be seen from the above, it is shown that,λ e 、A tank and
Figure 878924DEST_PATH_IMAGE005
all are constants, so the above formula can be expressed as follows:
Figure 97416DEST_PATH_IMAGE006
wherein the content of the first and second substances,
Figure 363312DEST_PATH_IMAGE007
is composed ofλ e 、A tank And
Figure 237727DEST_PATH_IMAGE005
the product of (a).
Combing out the first characteristic value based on the preprocessing operationX 1 The expression (2), i.e. the outer wall temperature of the tank and the tankLNGDifference in temperatureT W -T L Guarantee the first characteristic valueX 1 Can correspond to field data measurement points. Wherein the content of the first and second substances,T W obtaining the temperature monitoring point of the outer wall of the storage tank;T L in a storage tankLNGThe temperature is obtained by averaging a temperature monitoring point at the bottom of the storage tank and a temperature monitoring point at the top of the storage tank.T W AndT L the difference value of the two reflects the amount of heat led into the storage tank to a certain extent, and if the difference value of the two is larger, the more heat is led into the storage tank; otherwise, the less.
LNGThe data of the cold/heat input part of the storage tank feeding initiation is preprocessed through the flow of the feeding hole and the feeding based on the following calculation equationLNGTemperature and interior of original storage tankLNGProduct measurement of temperature differenceLNGThe cold/heat is initiated by the feeding of the storage tank, and a second characteristic value is extractedX 2 ：Q I （T I -T L ）。
Figure 739116DEST_PATH_IMAGE008
Wherein the content of the first and second substances,q I indicating the heat input initiated by the feed,W；Q I which indicates the feed inlet flow rate,m 3 /s；
Figure 218639DEST_PATH_IMAGE009
to representLNGThe density of the mixture is higher than the density of the mixture,kg/m 3 ；T I indicating the feedLNGThe temperature of the mixture is controlled by the temperature,K；T L indicating in the original tankLNGThe temperature of the mixture is controlled by the temperature,K；cto representLNGThe specific heat capacity of the heat exchanger is improved,J/( kgK)。
it can be seen that the specific heat capacityc、LNGDensity of
Figure 452174DEST_PATH_IMAGE009
Approximated as a constant value, neglectingcAnd
Figure 377667DEST_PATH_IMAGE009
a change in (c). The above formula can be expressed as follows:
Figure 558112DEST_PATH_IMAGE010
wherein the content of the first and second substances,
Figure 688879DEST_PATH_IMAGE011
is composed ofcAnd
Figure 155633DEST_PATH_IMAGE012
the product of (a).
Combing out a second characteristic value based on the preprocessing operationX 2 The expression (2), i.e. feed inlet flow and feedLNGTemperature and in the storage tankLNGProduct of temperature differenceQ I (T I -T L ) Guarantee the characteristic valueX 2 Can correspond to field data measurement points. Wherein the content of the first and second substances,T I is composed ofLNGStorage tank feed temperature from the fieldLNGAcquiring a storage tank feeding temperature monitoring point;Q I the flow of the feed inlet is obtained by a field feeding flow monitoring point;T L in a storage tankLNGThe temperature is obtained by averaging a temperature monitoring point at the bottom of the storage tank and a temperature monitoring point at the top of the storage tank.
Second characteristic valueX 2 ：Q I （T I -T L ) I.e. byLNGThe difference value of the feeding temperature monitoring point of the storage tank and the average value of the temperature monitoring point at the bottom of the storage tank and the temperature monitoring point at the top of the storage tank is multiplied by the feeding flow monitoring point.
The operation heat production amount of the tank bottom pump is partially preprocessed, based on the following calculation equation, the operation heat production amount of the tank bottom pump is measured through the discharge capacity of the tank bottom pump, and a third characteristic value is extractedX 3 ：Q P 。
Figure 207902DEST_PATH_IMAGE013
Wherein the content of the first and second substances,
Figure 723197DEST_PATH_IMAGE014
the pump head at the bottom of the tank is shown,m；Q P the pump displacement at the bottom of the tank is shown,m 3 /s，
Figure 36367DEST_PATH_IMAGE015
indicating tank bottom pump efficiency, related to flow.
As can be seen,gis a constant number of times, and is,LNGdensity of
Figure 814967DEST_PATH_IMAGE016
Approximated as a constant value, neglecting
Figure 151271DEST_PATH_IMAGE017
Variation of (2), and tank bottom pump efficiency
Figure 798153DEST_PATH_IMAGE018
Pump head at bottom of tank
Figure 841195DEST_PATH_IMAGE019
Related to the tank bottom pump displacement. The above formula can be expressed as follows:
Figure 587434DEST_PATH_IMAGE020
wherein the content of the first and second substances,
Figure 974815DEST_PATH_IMAGE021
is composed ofg、
Figure 35175DEST_PATH_IMAGE022
、
Figure 322937DEST_PATH_IMAGE019
And
Figure 240078DEST_PATH_IMAGE023
the product of (a).
Combing out a third characteristic value based on the preprocessing operationX 3 Expressed in terms of canister bottom pump displacementQ P Guarantee the third characteristic valueX 3 Can correspond to field data measurement points. Wherein, the first and the second end of the pipe are connected with each other,Q P the low-pressure pump flow is determined by the sum of two low-pressure pump displacement monitoring points on site.
LNGPreprocessing the data of the heat storage part, based on the following calculation equation, neglectingLNGDensity of
Figure 488656DEST_PATH_IMAGE024
Specific heat capacity ofcBy the interior of the tankLNGHeight of liquid level andLNGmeasurement of the product of the rate of temperature changeLNGExtracting a fourth characteristic value according to the heat quantity corresponding to the temperature riseX 4 ：
Figure 477341DEST_PATH_IMAGE025
。
Figure 291713DEST_PATH_IMAGE026
Wherein the content of the first and second substances,q tem to representLNGThe temperature change absorbs or dissipates heat that,W；rwhich represents the inner radius of the tank,m；Lto representLNGThe height of the liquid level is higher than the standard value,m；tthe time is represented by the time of day,s。
as can be seen,ris constant, specific heat capacityc、LNGDensity of
Figure 848596DEST_PATH_IMAGE027
Approximated as a constant value, neglectingcAnd
Figure 443526DEST_PATH_IMAGE027
so the above formula can be expressed as follows:
Figure 111268DEST_PATH_IMAGE028
wherein the content of the first and second substances,
Figure 780146DEST_PATH_IMAGE029
is composed of
Figure 868450DEST_PATH_IMAGE030
、π、r 2 Andcthe product of (a).
Combing out a fourth characteristic value based on the preprocessing operationX 4 By means of expression of
Figure 91621DEST_PATH_IMAGE031
Guarantee the fourth eigenvalueX 4 Can correspond to field data measurement points. Wherein the content of the first and second substances,Lis composed ofLNGThe liquid level is higher than that in the storage tankLNGThe liquid level sensor is used for monitoring the liquid level,T L in a storage tankLNGThe temperature is obtained by averaging a temperature monitoring point at the bottom of the storage tank and a temperature monitoring point at the top of the storage tank.
According to the interior of the storage tankLNGThe variation of temperature and time can be fitted to obtain the variation function of temperature and time, and the first derivative value corresponding to any time can be obtained, and simultaneously combinedLNGThe monitoring data of the liquid level sensor is used as a characteristic value, and the characteristic value can be well reflectedLNGThe temperature rises by the corresponding amount of heat.
Phase-change thermal partial data preprocessing, which is to performLNGDividing all data collected by the phase change heat absorption part in the storage tank into two subdata sets for data preprocessing in sequence, wherein the data preprocessing comprises storage tank gas phase space temperature, pressure and height change part data preprocessing and compressor air extraction amount part data preprocessing, and extracting fifth characteristic values respectivelyX 5 And a sixth characteristic valueX 6 。
Partial data preprocessing of the storage tank gas phase space temperature, pressure and altitude changes by the following calculation equationLNGThe change rate of the product of the absolute pressure of the gas phase space in the storage tank and the inverse number of the height and the temperature of the gas phase space is measuredLNGRate of change of quantity of gaseous substances inside the tank, i.e. per unit of timeLNGExtracting a fifth characteristic value from the variation of the amount of the gas-phase substance in the storage tankX 5 ：
Figure 94212DEST_PATH_IMAGE032
。
Based on the gas phase state equation:
Figure 945494DEST_PATH_IMAGE033
wherein the content of the first and second substances,
Figure 844180DEST_PATH_IMAGE034
which represents the pressure of the gas phase,Pa；V g the volume of the gas phase is expressed,m 3 ；Zrepresents a compression factor;Rthe molar gas constant is expressed as the molar gas constant,
Figure 85805DEST_PATH_IMAGE035
；TgrepresentBOGThe temperature of the mixture is controlled by the temperature,K；H tank which is indicative of the height of the tank,m；ngrepresenting gas phase space in storage tanksBOGThe amount of the substance(s) to be administered,mol。
deducing the time unitLNGCaused by variations in pressure, temperature, height of the tankBOGAmount of substance varied:
Figure 219983DEST_PATH_IMAGE036
wherein the content of the first and second substances,
Figure 66716DEST_PATH_IMAGE037
which represents the pressure of the gas phase,Pa；Zrepresents a compression factor;Rthe molar gas constant is expressed as the molar gas constant,
Figure 667462DEST_PATH_IMAGE038
；T g to representBOGThe temperature of the mixture is controlled by the temperature,K；H tank which is indicative of the height of the tank,m；ngrepresenting gas phase space in storage tanksBOGThe amount of the substance(s) to be administered,mol。
as can be seen,r、Ris a constant number of times, and is,Zthe fluctuation is not large and neglectedZSo the above formula can be expressed as follows:
Figure 989859DEST_PATH_IMAGE039
wherein the content of the first and second substances,
Figure 271936DEST_PATH_IMAGE040
is composed ofπ、r 2 、
Figure 35492DEST_PATH_IMAGE042
And
Figure 636500DEST_PATH_IMAGE044
the product of (a).
Combing out a fifth characteristic value based on the preprocessing operationX 5 Expression ofI.e. by
Figure 55980DEST_PATH_IMAGE045
Guarantee the fifth eigenvalueX 5 Can correspond to field data measurement points. Wherein the content of the first and second substances,
Figure 531961DEST_PATH_IMAGE046
the absolute pressure is obtained from a gas phase pressure monitoring point in the storage tank;Lis composed ofLNGThe liquid level is higher than that in the storage tankLNGMonitoring by a liquid level sensor;T g is composed ofBOGTemperature from the storage tankBOGAnd obtaining a temperature monitoring point.
Monitoring point and method according to gas phase pressure in storage tankLNGA liquid level sensor,BOGThe change and time of the reciprocal product of the temperature monitoring point data can be fitted to obtain a change function of the temperature monitoring point data with respect to time, a first derivative value corresponding to any time can be obtained, and the first derivative value is used as a characteristic value to better reflect the change and time of the reciprocal product of the temperature monitoring point data in unit timeLNGThe amount of change in the amount of gas phase species inside the tank.
The data of the suction air quantity part of the compressor is preprocessed, and the suction and the discharge of the compressor from a storage tank are measured by the product of the outlet pressure of the compressor and the outlet flow and the reciprocal of the outlet temperature of the compressor based on the following calculation equationBOGRate of change of mass, i.e. compressor suction discharge from reservoir per unit timeBOGThe amount of change of the amount of the substance of (1), and a sixth characteristic value is extractedX 6 ：
Figure 150024DEST_PATH_IMAGE047
。
Figure 295835DEST_PATH_IMAGE048
Wherein the content of the first and second substances,n dis indicating compressor suction from reservoir to dischargeBOGThe amount of the substance(s) to be administered,mol；
Figure 327245DEST_PATH_IMAGE049
which is indicative of the compressor outlet pressure,MPa。
as can be seen,Ris a constant number of times, and is,Zthe fluctuation is not large and neglectedZSo the above formula can be expressed as follows:
Figure 13441DEST_PATH_IMAGE050
wherein the content of the first and second substances,
Figure 954852DEST_PATH_IMAGE051
is composed of
Figure 130618DEST_PATH_IMAGE053
And
Figure 524691DEST_PATH_IMAGE054
the product of (a).
Based on the preprocessing operation, the characteristic value is combedX 6 The expression of (1), i.e. the product of the compressor outlet pressure and its outlet flow, the reciprocal of the outlet temperature
Figure 14578DEST_PATH_IMAGE055
Guarantee the characteristic value X 6 Can correspond to field data measurement points. Wherein the content of the first and second substances,p dis is monitored and obtained by an on-site compressor outlet pressure sensor,Q dis is monitored and obtained by an on-site compressor outlet flow sensor,T dis and monitoring and acquiring by an on-site compressor outlet temperature sensor.
In order to verify the effect of the method in this embodiment, the experimental result of this embodiment is described, specifically:
LNGraw eigenvalues in tank operational dataX 1 ~X 7 Mainly comprisesT W 、T L 、Q P 、L、p dis 、Q dis AndT dis in the present embodiment, the characteristic value is usedQ dis I.e. compressor outlet displacement as a labelYThe remaining seven groups of characteristic values are takenAs input variablesX 1 ~X 7 Constructing a data model; in other embodiments, the data model may be constructed by using one or more other characteristic values as tags and using the other characteristic values as input variables.
The physical model of thermal equilibrium is:
Figure 995234DEST_PATH_IMAGE056
Figure 748426DEST_PATH_IMAGE057
wherein the content of the first and second substances,HPTis the heat of phase change,J mol。
based on the data preprocessing method of the patent, characteristic values are extracted, namelyT W -T L 、Q I （T I -T L ）、Q P 、
Figure 426532DEST_PATH_IMAGE058
、
Figure 516848DEST_PATH_IMAGE059
And
Figure 167272DEST_PATH_IMAGE060
in the present embodiment, the
Figure 888103DEST_PATH_IMAGE061
I.e. the compressor draws from the reservoir and discharges outwardsBOGRate of change of amount of substance as labelYThe remaining five sets of eigenvalues are used as input variablesX 1 ~X 5 Constructing a data model; understandably, the labelYSelecting a target to be predicted for the result of machine learning output
Figure 381402DEST_PATH_IMAGE062
Physical significance of labeling: predictionLNGOutside the tankBOGRate of change of amount of substance; in other embodiments, the data model may be constructed by using one or more other characteristic values as tags and using the other characteristic values as input variables. Wherein gradient lifting regression is adopted (Gradient boosting regression，GBR) Fitting by the method, as shown in fig. 1, which is an original feature fitting effect graph after fitting by a gradient lifting regression method; as shown in fig. 2, the result is a graph of the post-fitting effect of the recombination eigenvalues after fitting by the gradient lifting regression method; determining coefficients using error indicatorsR 2 Maximum relative errorMax Relative ErrorAverage relative errorMean Relative ErrorThe predicted performance is judged, and the fitting effect pair after the original characteristic value and the recombined characteristic value is shown in the table 1:
TABLE 1 comparison of the fitting of the original characteristics with the post-recombination characteristic values
Maximum relative error Error index determination coefficient R2 Average relative error
Fitting of raw features 442.424% 0.843 4.742%
Recombination eigenvalue post-fitting 37.801% 0.908 4.045%
Example 2:
the embodiment provides a data processing system for liquefied natural gas storage tank operation data mining, including:
a feature value construction module configured to: obtaining a first characteristic value according to the difference value between the temperature inside the storage tank and the temperature outside the storage tank; obtaining a second characteristic value according to the difference between the temperature of the fed liquefied natural gas and the temperature of the liquefied natural gas in the storage tank; constructing a third characteristic value based on the equipment operation parameters; obtaining a fourth characteristic value according to the derivative of the temperature of the liquefied natural gas in the tank to the time; obtaining a fifth characteristic value according to the derivative of the algebraic operation result of the physical state parameter and the volume related parameter of the gas phase in the tank to the time; obtaining a sixth characteristic value according to algebraic operation of gas-phase physical state parameters and discharge capacity at the outlet of the compressor;
a data fitting module configured to: according to a machine learning algorithm, one or more of the first characteristic value, the second characteristic value, the third characteristic value, the fourth characteristic value, the fifth characteristic value and the sixth characteristic value are used as labels, and other characteristic values are used as input to perform data fitting;
the working method of the system is the same as the data processing method for the liquefied natural gas storage tank operation data mining in the embodiment 1, and the detailed description is omitted here.
The above description is only a preferred embodiment of the present invention, and is not intended to limit the present invention, and those skilled in the art can make various modifications and variations. Any modification, equivalent replacement, or improvement made within the spirit and principle of the present embodiment should be included in the protection scope of the present embodiment.

Claims (8)

1. A data processing method for liquefied natural gas storage tank operation data mining is characterized by comprising the following steps:
obtaining a first characteristic value according to the difference value between the temperature inside the storage tank and the temperature outside the storage tank; obtaining a second characteristic value according to the difference between the temperature of the fed liquefied natural gas and the temperature of the liquefied natural gas in the storage tank; constructing a third characteristic value based on the equipment operation parameters; obtaining a fourth characteristic value according to the derivative of the temperature of the liquefied natural gas in the tank to the time; obtaining a fifth characteristic value according to the derivative of the algebraic operation result of the physical state parameter and the volume related parameter of the gas phase in the tank to the time; obtaining a sixth characteristic value according to algebraic operation of gas-phase physical state parameters and discharge capacity at the outlet of the compressor;
and according to a machine learning algorithm, performing data fitting by taking one or more of the first characteristic value, the second characteristic value, the third characteristic value, the fourth characteristic value, the fifth characteristic value and the sixth characteristic value as a label and taking other characteristic values as input.
2. The data processing method for liquefied natural gas storage tank operation data mining of claim 1, wherein the first characteristic value is equal to a difference between a temperature of an outer wall of the storage tank and a temperature of liquefied natural gas in the storage tank.
3. The data processing method for liquefied natural gas storage tank operation data mining of claim 1, wherein the second characteristic value is equal to a difference between a temperature of the fed liquefied natural gas and a temperature of the liquefied natural gas in the storage tank multiplied by a flow rate of the feed port.
4. The data processing method for liquefied natural gas storage tank operation data mining of claim 1, wherein the third characteristic value is equal to a tank bottom pump displacement.
5. The data processing method for liquefied natural gas storage tank operation data mining of claim 1, wherein the fourth characteristic value is equal to a derivative of a liquefied natural gas temperature in the tank with respect to time multiplied by a liquid level height of the liquefied natural gas.
6. The data processing method for liquefied natural gas storage tank operation data mining according to claim 1, wherein the fifth characteristic value is a value obtained by multiplying a difference between a height of the storage tank and a height of a liquefied natural gas liquid level by a gas phase pressure and then deriving a time by a ratio of the obtained product to a boil-off gas temperature.
7. The data processing method for liquefied natural gas storage tank operation data mining of claim 1, wherein the sixth characteristic value is equal to a ratio of a product of a compressor outlet pressure and a compressor outlet flow rate to a compressor outlet temperature.
8. A data processing system for liquefied natural gas storage tank operational data mining, comprising:
a feature value construction module configured to: obtaining a first characteristic value according to the difference value between the temperature inside the storage tank and the temperature outside the storage tank; obtaining a second characteristic value according to the difference between the temperature of the fed liquefied natural gas and the temperature of the liquefied natural gas in the storage tank; constructing a third characteristic value based on the equipment operation parameters; obtaining a fourth characteristic value according to the derivative of the temperature of the liquefied natural gas in the tank to the time; obtaining a fifth characteristic value according to the derivative of the algebraic operation result of the physical state parameter and the volume related parameter of the gas phase in the tank to the time; obtaining a sixth characteristic value according to algebraic operation of gas-phase physical state parameters and discharge capacity at the outlet of the compressor;
a data fitting module configured to: and according to a machine learning algorithm, performing data fitting by taking one or more of the first characteristic value, the second characteristic value, the third characteristic value, the fourth characteristic value, the fifth characteristic value and the sixth characteristic value as a label and taking other characteristic values as input.
CN202210744755.7A 2022-06-29 2022-06-29 Data processing method and system for liquefied natural gas storage tank operation data mining Active CN114819193B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN202210744755.7A CN114819193B (en) 2022-06-29 2022-06-29 Data processing method and system for liquefied natural gas storage tank operation data mining

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN202210744755.7A CN114819193B (en) 2022-06-29 2022-06-29 Data processing method and system for liquefied natural gas storage tank operation data mining

Publications (2)

Publication Number Publication Date
CN114819193A true CN114819193A (en) 2022-07-29
CN114819193B CN114819193B (en) 2022-09-27

Family

ID=82523172

Family Applications (1)

Application Number Title Priority Date Filing Date
CN202210744755.7A Active CN114819193B (en) 2022-06-29 2022-06-29 Data processing method and system for liquefied natural gas storage tank operation data mining

Country Status (1)

Country Link
CN (1) CN114819193B (en)

Cited By (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN116705184A (en) * 2023-05-29 2023-09-05 上海海德利森科技有限公司 Liquid hydrogen evaporation loss prediction method, device, equipment and medium

Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN103206614A (en) * 2013-04-01 2013-07-17 中山大学 Early-warning device and early-warning method used for monitoring accidents of boiling liquid expanding vapor explosion (BLEVE) of pressuring-bearing storage tanks
CN107958301A (en) * 2017-10-31 2018-04-24 浙江中控软件技术有限公司 A kind of LNG receiving stations running optimizatin method
CN112990288A (en) * 2021-03-09 2021-06-18 上海应用技术大学 Online monitoring and fault prediction system for large equipment
WO2021170525A1 (en) * 2020-02-25 2021-09-02 Shell Internationale Research Maatschappij B.V. Method and system for production optimization
CN113741362A (en) * 2021-08-17 2021-12-03 中海石油气电集团有限责任公司 Method, system, medium, and computing device for optimizing operation of liquefied natural gas receiving station

Patent Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN103206614A (en) * 2013-04-01 2013-07-17 中山大学 Early-warning device and early-warning method used for monitoring accidents of boiling liquid expanding vapor explosion (BLEVE) of pressuring-bearing storage tanks
CN107958301A (en) * 2017-10-31 2018-04-24 浙江中控软件技术有限公司 A kind of LNG receiving stations running optimizatin method
WO2021170525A1 (en) * 2020-02-25 2021-09-02 Shell Internationale Research Maatschappij B.V. Method and system for production optimization
CN112990288A (en) * 2021-03-09 2021-06-18 上海应用技术大学 Online monitoring and fault prediction system for large equipment
CN113741362A (en) * 2021-08-17 2021-12-03 中海石油气电集团有限责任公司 Method, system, medium, and computing device for optimizing operation of liquefied natural gas receiving station

Cited By (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN116705184A (en) * 2023-05-29 2023-09-05 上海海德利森科技有限公司 Liquid hydrogen evaporation loss prediction method, device, equipment and medium
CN116705184B (en) * 2023-05-29 2024-04-05 上海海德利森科技有限公司 Liquid hydrogen evaporation loss prediction method, device, equipment and medium

Also Published As

Publication number Publication date
CN114819193B (en) 2022-09-27

Similar Documents

Publication Publication Date Title
CN114819193B (en) Data processing method and system for liquefied natural gas storage tank operation data mining
CN102420021B (en) Nuclear power station containment vessel leakage rate monitoring method and system
Barduhn et al. The properties of some new gas hydrates and their use in demineralizing sea water
CN102914108B (en) Apparatus and method for evaluating performance of centrifugal chiller
CN108731190A (en) Detection method, detection device and the detecting system of air-conditioner host parameter
CN104360984B (en) One kind is based on the lower cooling tower filler performance calculation method of Two point boundary value measurement
CN103201581A (en) A method and an apparatus for constructing a heat pipe
CN115127192B (en) Semi-supervised water chilling unit fault diagnosis method and system based on graph neural network
CN113323656A (en) Development index prediction method for closed condensate gas reservoir and computer-readable storage medium
CN104361148A (en) Design method of transverse flow type cooling tower
TWI704319B (en) Intelligent measurement and verification method and system for efficiency of refrigerating and air-conditioning host
CN109342496A (en) A kind of vacuum conveyer tube Cryo Heat Insulation performance measurement test method
CN112945004B (en) Multi-circulation cooling system adopting seawater for cooling
CN111413115B (en) Intelligent measurement verification method and system for host efficiency of refrigeration air conditioner
CN105953386A (en) Control method and control device for two-stage gas replenishing and enthalpy increasing system
CN108595813A (en) A kind of CO based on optimum quality flow velocity2Gas cooler optimum design method
CN114493212A (en) Method for evaluating condensate gas reservoir dynamic reserves
CN209416739U (en) A kind of moderate and high temperature heat property test platform
CN114427449A (en) Abnormal well temperature early warning method
CN205747650U (en) Ammonia refrigeration storage system
CN105043476B (en) A kind of assay method of cold section of double back hot heater reheated steam flow of reheating
CN111351531B (en) Main steam flow online measurement method
CN205402231U (en) Civilian LNG air feeder
CN108680375B (en) Vacuum refrigeration performance test experimental device
CN117391286A (en) Water intrusion type gas storage water inflow and outflow evaluation method based on material balance method

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