CN114970363A - Low-calorific-value gas turbine characteristic control method based on knowledge graph and machine learning - Google Patents

Low-calorific-value gas turbine characteristic control method based on knowledge graph and machine learning Download PDF

Info

Publication number
CN114970363A
CN114970363A CN202210644972.9A CN202210644972A CN114970363A CN 114970363 A CN114970363 A CN 114970363A CN 202210644972 A CN202210644972 A CN 202210644972A CN 114970363 A CN114970363 A CN 114970363A
Authority
CN
China
Prior art keywords
low
characteristic
gas turbine
bleed
compressor
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
CN202210644972.9A
Other languages
Chinese (zh)
Other versions
CN114970363B (en
Inventor
辛小鹏
刘振宇
谭建荣
撒国栋
张栋豪
刘惠
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.)
Zhejiang University ZJU
Original Assignee
Zhejiang University ZJU
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 Zhejiang University ZJU filed Critical Zhejiang University ZJU
Priority to CN202210644972.9A priority Critical patent/CN114970363B/en
Priority claimed from CN202210644972.9A external-priority patent/CN114970363B/en
Publication of CN114970363A publication Critical patent/CN114970363A/en
Application granted granted Critical
Publication of CN114970363B publication Critical patent/CN114970363B/en
Active legal-status Critical Current
Anticipated expiration legal-status Critical

Links

Images

Classifications

    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F30/00—Computer-aided design [CAD]
    • G06F30/20—Design optimisation, verification or simulation
    • G06F30/27—Design optimisation, verification or simulation using machine learning, e.g. artificial intelligence, neural networks, support vector machines [SVM] or training a model
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
    • G06F16/30—Information retrieval; Database structures therefor; File system structures therefor of unstructured textual data
    • G06F16/36—Creation of semantic tools, e.g. ontology or thesauri
    • G06F16/367—Ontology
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F2119/00—Details relating to the type or aim of the analysis or the optimisation
    • G06F2119/02—Reliability analysis or reliability optimisation; Failure analysis, e.g. worst case scenario performance, failure mode and effects analysis [FMEA]
    • Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
    • Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
    • Y02T—CLIMATE CHANGE MITIGATION TECHNOLOGIES RELATED TO TRANSPORTATION
    • Y02T10/00—Road transport of goods or passengers
    • Y02T10/10—Internal combustion engine [ICE] based vehicles
    • Y02T10/12—Improving ICE efficiencies

Landscapes

  • Engineering & Computer Science (AREA)
  • Theoretical Computer Science (AREA)
  • Physics & Mathematics (AREA)
  • Evolutionary Computation (AREA)
  • General Engineering & Computer Science (AREA)
  • General Physics & Mathematics (AREA)
  • Artificial Intelligence (AREA)
  • Computational Linguistics (AREA)
  • Databases & Information Systems (AREA)
  • Animal Behavior & Ethology (AREA)
  • Data Mining & Analysis (AREA)
  • Computer Vision & Pattern Recognition (AREA)
  • Life Sciences & Earth Sciences (AREA)
  • Medical Informatics (AREA)
  • Software Systems (AREA)
  • Computer Hardware Design (AREA)
  • Geometry (AREA)
  • Control Of Positive-Displacement Air Blowers (AREA)

Abstract

The invention discloses a low-calorific-value gas turbine characteristic control method based on knowledge graph and machine learning. Acquiring operation data of the low-heat-value gas turbine in real time, extracting characteristic variables of four major components, and establishing a knowledge graph of the characteristic variables of the components; analyzing real-time operation data of the characteristic variables to establish a characteristic map of part of working conditions; optimizing the characteristic map of the part of the working condition part by combining the incidence relation of the characteristic variables in the part knowledge map to form a full-working condition part characteristic map; and establishing a variable working condition process control function, carrying out overall variable working condition calculation processing on the real-time operation data by using the variable working condition process control function, and predicting control parameters in the variable working condition process so as to carry out adjustment control. The invention reduces the high test cost of the components, avoids the coal gas leakage risk existing in the test of the single components of the coal gas compressor, avoids the problems of long simulation calculation time and low calculation accuracy of three-dimensional modeling of the characteristics of the components, and improves the reliability, the economy and the safety.

Description

Low-calorific-value gas turbine characteristic control method based on knowledge graph and machine learning
Technical Field
The invention relates to a gas turbine control method in the field of low-calorific-value gas turbine characteristic modeling, in particular to a low-calorific-value gas turbine characteristic control method based on knowledge graph and machine learning.
Background
The low-heat value gas turbine has low fuel heat value and frequent variable working conditions, and the full-working-condition, high-precision and dynamically updated low-heat value gas turbine characteristic map is particularly important for the design and safe operation of control system parameters. The characteristic maps of the low-calorific-value gas turbine mainly comprise an air compressor characteristic map, a gas compressor characteristic map, a combustion chamber characteristic map and a turbine characteristic map.
The existing full three-dimensional calculation simulation technology is limited by a turbulence model, a boundary layer, boundary conditions and calculation grid numbers, and a full-working-condition, high-precision and dynamically-updated low-calorific-value gas turbine component characteristic map cannot be obtained through calculation. The existing air compressor part test, gas compressor part test, combustor part test and turbine part test are limited by large load (more than 50MW), gas toxicity, high combustion temperature (the highest temperature is more than 1300 ℃) and the like, the test cost is high, and in the single part test, due to different boundary conditions, the characteristic diagram of the part on the whole machine cannot be completely represented. The direct whole-process parameter test of the whole machine has high cost, and the four parts are matched together, so that the characteristic map of the whole working condition cannot be directly obtained.
The characteristic map of the low-heat-value gas turbine with full working condition, high precision and dynamic update is difficult to obtain by the prior art.
Disclosure of Invention
In order to solve the problems in the background art, the invention provides a low-calorific-value gas turbine characteristic control method based on knowledge graph and machine learning, the accuracy of the full-working-condition characteristic graph is improved, the control precision of a dynamic process is increased, the dynamic update of the low-calorific-value gas turbine characteristic graph is realized, the high component test cost is reduced, the gas leakage risk in the single component test of a gas compressor is avoided, and the problems of long time for three-dimensional modeling simulation calculation of the component characteristics and low calculation accuracy are avoided by continuously iterating the knowledge graph of the low-calorific-value gas turbine component and the machine learning model of the running data in real time.
In order to achieve the purpose, the technical scheme adopted by the invention for solving the technical problems is as follows:
the method comprises the following steps: acquiring operation data of the low-heat-value gas turbine in real time, extracting characteristic variables of four parts, namely an air compressor, a gas compressor, a combustion chamber and a turbine in the low-heat-value gas turbine, analyzing the correlation of the characteristic variables, extracting the physical mechanism relation of the characteristic variables, and establishing a knowledge graph of the characteristic variables of the parts in the low-heat-value gas turbine;
step two: analyzing real-time operation data of the characteristic variables of the low-heat-value gas turbine according to the knowledge graph, and establishing a characteristic graph of part of working condition components;
step three: optimizing the characteristic maps of the parts under the partial working conditions by adopting a random forest integrated machine learning algorithm in combination with the incidence relation of characteristic variables in the knowledge maps of the parts of the low-calorific-value gas turbine, and further predicting the characteristic maps of the parts under the full working conditions to form the characteristic maps of the parts under the full working conditions;
step four: the characteristic prediction function in the full working condition characteristic map formed by four parts is adopted to establish a variable working condition process control function, the variable working condition process control function is used for carrying out overall variable working condition calculation processing on the real-time operation data of the low-heat-value gas turbine, and the control parameters in the variable working condition prediction process are further adjusted and controlled.
And dynamically updating the characteristic map of the all-condition component of the low-calorific-value gas turbine under frequent variable working conditions.
The low heat value refers to the range of the heat value of the fuel being 2.5-4.5 MJ/Nm 3 In the meantime.
In the first step, the extracted characteristic variables comprise thermal characteristic variables, output characteristic variables, motion characteristic variables and geometric characteristic variables, and a knowledge graph among the four types of characteristic variables is established.
The thermal characteristic variables are pressure, temperature and flow of inlets and outlets of the four components respectively, the output characteristic variables are loads of the four components respectively, the motion characteristic variables are rotating speeds of the four components respectively, the four components are different in geometric characteristic variables, the geometric characteristic variables of the air compressor are IGV opening of an inlet adjustable guide vane, high-pressure anti-surge valve opening and low-pressure anti-surge valve opening of the air compressor, the geometric characteristic variables of the gas compressor are VGV opening of a rotatable guide vane, high-pressure anti-surge valve opening, medium-pressure anti-surge valve opening and low-pressure anti-surge valve opening of the gas compressor, the geometric characteristic variables of a combustion chamber are combustion bypass valve opening, and the turbine is a fixed geometric component and is not used for controlling and adjusting the geometric characteristic variables.
In the second step, establishing the characteristic map of the part of the working condition components comprises,
establishing a working condition characteristic map of the pressure ratio and the efficiency of the air compressor and the outlet folded flow of the air compressor respectively;
establishing a working condition characteristic map of the pressure ratio and the efficiency of the gas compressor and the outlet reduced flow of the gas compressor respectively;
establishing a working condition characteristic map of the pressure loss coefficient and efficiency of the combustion chamber and the rotating speed and load of the low-heat-value gas turbine respectively;
and establishing a working condition characteristic map of the turbine outlet reduced flow and efficiency and the turbine expansion ratio respectively.
The partial working conditions refer to the variation ranges of the geometric characteristic variable, the motion characteristic variable and the output characteristic variable in the actual operation process of the low-heat-value gas turbine within a certain period of time.
And extracting the physical mechanism relation of the characteristic variables from the knowledge graph, and establishing the knowledge graph of the incidence relation of the characteristic variables of the low-heat-value gas turbine components through flow balance, load balance and pressure balance of cross section nodes of inlets and outlets of the four components.
In the third step, the incidence relation of the characteristic variables in the knowledge graph of the low-heat-value gas turbine component comprises the following steps:
A) air compressor component characteristic prediction function:
π c ＝f 1 (m c ,IGV,Bleed cH ,Bleed cL ,n)，η c ＝f 2 (m c ,IGV,Bleed cH ,Bleed cL ,n)
wherein f is 1 、f 2 For a prediction function, pi, of the characteristics of the first and second air compressor components c Is the pressure ratio of the air compressor eta c For air compressor efficiency, m c The flow rate is converted for the outlet of the air compressor, and the IGV is the opening degree of an adjustable guide vane at the inlet of the air compressor, Bleed cH 、Bleed cL Respectively the high-pressure anti-surge valve opening and the low-pressure anti-surge valve opening of the air compressor, wherein n is the rotating speed of the low-heat-value gas turbine;
B) gas compressor component characteristic prediction function:
π gc ＝f 3 (m gc ,VGV,Bleed gcH ,Bleed gcM ,Bleed gcL ,n)
η gc ＝f 4 (m gc ,VGV,Bleed gcH ,Bleed gcM ,Bleed gcL ,n)
wherein f is 3 、f 4 For a prediction function of the characteristics of the first and second gas compressor parts, pi gc Is the pressure ratio, eta, of the gas compressor gc For gas compressor efficiency, m gc For the outlet of the gas compressor to reduce the flow, and the VGV is the opening degree of the adjustable guide vanes at the inlet of the gas compressor, blade gcH 、Bleed gcM 、Bleed gcL Respectively the high-pressure anti-surge valve opening, the medium-pressure anti-surge valve opening and the low-pressure anti-surge valve opening of the gas compressor;
C) combustor component characteristic prediction function:
dp＝f 5 (Bypass,n,P)，η cb ＝f 6 (Bypass,n,P)
wherein f is 5 、f 6 Dividing the combustion chamber into two parts of an increasing speed process and an increasing load process for the characteristic prediction function of the first combustion chamber component and the second combustion chamber component, d p Is pressure loss of combustion chamber eta cb For combustor efficiency, Bypass is the combustor Bypass valve opening, P is the low calorific value gas turbineThe load of (2);
D) turbine characteristic prediction function:
m T ＝f 7 (π T ,n)，η T ＝f 8 (π T ,n)
wherein f is 7 、f 8 For the first and second turbine component characteristic prediction functions, m T For turbine outlet reduced flow, pi T Is the turbo expansion ratio, η T Is the turbine efficiency.
The third step is specifically as follows:
characteristic prediction function f of each part of low-heat-value gas turbine by adopting random forest integrated machine learning algorithm 1 、f 2 、f 3 、f 4 、f 5 、f 6 、f 7 、f 8 Respectively optimizing and determining parameters in the characteristic prediction function, and using the obtained characteristic prediction function to perform all-condition modeling prediction:
f i ＝Γ(D,D bs )
wherein f is i Is a characteristic prediction function of the low-heat value gas turbine, gamma is a random forest machine learning algorithm, D is a gas turbine characteristic sample, D bs Are random self-sampling gas turbine characteristic samples.
The fourth step is specifically as follows:
establishing the following variable working condition process control functions of the low-heat-value gas turbine according to characteristic prediction functions of all parts of the low-heat-value gas turbine:
(IGV,VGV,Bypass)＝f 9 (n,P)
IGV＝f 10 (m c ,π c ,η c ,Bleed cH ,Bleed cL ,n)
VGV＝f 11 (m gc ,π gc ,η gc ,Bleed gcH ,Bleed gcM ,Bleed gcL ,n)
Bypass＝f 12 (dp,η cb ,n,P)
π T ＝π c ×(1-dp)，m T ＝m c +m gc
wherein f is 9 As a function of the process control of the varying conditions of the gas turbine with low calorific value, f 10 As a calculated function of the air compressor IGV, f 11 As a function of the calculation of the VGV of the gas compressor, f 12 Calculating functions of the combustion chamber Bypass, including variable load and variable rotating speed process control functions, wherein n is the rotating speed of the low-heat-value gas turbine, and P is the load of the low-heat-value gas turbine; IGV denotes the opening degree of an adjustable guide vane at the inlet of the air compressor, Bleed cH Indicating the degree of opening of the high-pressure surge-proof valve of the air compressor, Bleed cL Showing the opening of the low-pressure surge-proof valve of the air compressor, VGV showing the opening of the adjustable guide vane of the inlet of the gas compressor, Bleed gcH Indicating the degree of opening of the high-pressure surge-proof valve of the gas compressor, Bleed gcM Indicating the degree of opening of the gas compressor medium pressure surge-proof valve, Bleed gcL The opening of the low-pressure anti-surge valve of the gas compressor is represented, and Bypass represents the opening of a Bypass valve of the combustion chamber;
and processing the variable working condition process control function of the low-heat value gas turbine by using real-time operation data to obtain the control parameters.
The control parameters comprise an opening IGV of an adjustable guide vane at the inlet of the air compressor and an opening Bleed of a high-pressure anti-surge valve of the air compressor cH Low-pressure anti-surge valve opening degree Bleed cL Guide vane opening degree VGV adjustable at inlet of gas compressor and high-pressure anti-surge valve opening degree Bleed of gas compressor gcH Middle pressure surge-proof valve opening degree Bleed gcM Low-pressure anti-surge valve opening degree Bleed gcL And the opening degree Bypass of the combustion chamber Bypass valve.
Surging of the air compressor and the gas compressor mainly occurs in the process of increasing the rotating speed, a high-pressure anti-surge valve and a low-pressure anti-surge valve of the air compressor are arranged in the process of increasing the rotating speed, and are firstly and completely opened, and are completely closed when the rotating speed reaches 90 percent of the rated rotating speed; the high-pressure anti-surge valve, the medium-pressure anti-surge valve and the low-pressure anti-surge valve of the gas compressor are arranged to be opened completely at first, and the high-pressure anti-surge valve, the medium-pressure anti-surge valve and the low-pressure anti-surge valve are closed linearly when the rotating speed reaches 60% of the rated rotating speed and are closed completely when the rotating speed reaches 90% of the rated rotating speed. In the process of load increase, the high-pressure anti-surge valve and the low-pressure anti-surge valve of the air compressor, and the high-pressure anti-surge valve, the medium-pressure anti-surge valve and the low-pressure anti-surge valve of the gas compressor are all closed. The low-calorific-value gas turbine acquires more operation data in real time under frequent variable working conditions, and the accuracy of the full-working-condition characteristic spectrogram is improved and the control precision of a dynamic process is increased through continuous real-time iteration of the knowledge spectrogram and machine learning.
The method can reduce the high component test cost, avoids the coal gas leakage risk existing in the individual component test of the coal gas compressor, avoids the problems of long time and low calculation accuracy of three-dimensional modeling simulation calculation of component characteristics, and improves the reliability, economy and safety of the low-calorific-value gas turbine under the conditions of low fuel calorific value and frequent variable working conditions.
The method reduces the high component test cost, avoids the coal gas leakage risk existing in the individual component test of the coal gas compressor, and avoids the problems of long time and low calculation accuracy of three-dimensional modeling simulation calculation of component characteristics. The reliability, the economy and the safety of the low-heat-value gas turbine under the conditions of low fuel heat value and frequent variable working conditions are improved.
Compared with the prior art, the invention has the beneficial effects that:
the low-calorific-value gas turbine component characteristic map modeling method based on the knowledge map and machine learning can widen the working condition range of the low-calorific-value gas turbine component characteristic map, improve the accuracy and improve the dynamic updating performance.
The method has the advantages of low cost and high safety for acquiring the characteristic map of the low-calorific-value gas turbine component, and particularly avoids the gas leakage risk and high component test cost in the test of the single component of the gas compressor. The problems of long time and low calculation accuracy of three-dimensional modeling simulation calculation of the part characteristics are solved.
The invention can update the characteristic map of the low-heat value gas turbine component on line and realize the dynamic adjustment of the control system parameter. Therefore, the reliability, the economy and the safety of the low-heat-value gas turbine under the conditions of low fuel heat value and frequent variable working conditions are improved.
The invention realizes the modeling and dynamic control of the characteristics of the variable geometric components of the air compressor IGV, the gas compressor VGV and the combustion chamber Bypass of the low-heat-value gas turbine, and improves the overall matching performance of the low-heat-value gas turbine.
Drawings
FIG. 1 is a flow chart of a method of controlling low heating value gas turbine characteristics based on knowledge-mapping and machine learning.
FIG. 2 is a schematic illustration of a low heating value gas turbine component knowledge-map topology.
Fig. 3 is a schematic diagram of a characteristic map of partial working conditions and full working conditions of the air compressor.
Fig. 4 is a schematic diagram of characteristic maps of partial working conditions and full working conditions of the gas compressor.
FIG. 5 is a schematic representation of a characteristic map of partial and full conditions of a combustion chamber.
FIG. 6 is a schematic representation of a characteristic map of part-operating versus full operating conditions of the turbine.
FIG. 7 is a general block diagram of a low heating value gas turbine thermodynamic architecture.
FIG. 8 is a typical variable operating load and speed trend graph for a lower heating value gas turbine.
FIG. 9 is a typical variable operating condition control law map for a low heating value gas turbine.
Detailed Description
The present invention will be described in detail below with reference to the attached drawings, and it should be noted that the described embodiments are only intended to facilitate understanding of the present invention, and do not have any limiting effect thereon.
As shown in fig. 7, the embodied gas turbine includes a radial air inlet chamber 1, an air compressor 2, a combustion chamber 3, a turbine 4, a generator 5, and a blast furnace gas compressor 6; the outlet of the radial air inlet chamber 1 is connected with the inlet of an air compressor 2, the outlet of the air compressor 2 is respectively connected with the inlet of a combustion chamber 3 and the air inlet of a turbine 4, the rotating shaft of the air compressor 2, the turbine 4, the rotating shaft of a generator 5 and the rotating shaft of a blast furnace gas compressor 6 are synchronously connected, the outlet of the blast furnace gas compressor 6 is connected with the inlet of the combustion chamber 3, and the outlet of the combustion chamber 3 is connected with the air inlet of the turbine 4.
The low-calorific-value blast furnace gas is input from an inlet of a blast furnace gas compressor 6, the low-calorific-value gas is input into a combustion chamber 3 after being compressed by the blast furnace gas compressor 6, the medium-calorific-value coke oven gas 7 is input into the combustion chamber 3 when the combustion chamber 3 is ignited, air is input from an inlet of a radial air inlet chamber 1, the air enters an air compressor 2 through the radial air inlet chamber 1, the air is compressed by the air compressor 2 and outputs high-pressure air, the high-pressure air is divided into two paths, one path of the high-pressure air is used as main combustion air and directly input into the combustion chamber 3, the combustion chamber 3 outputs high-temperature gas, the other path of the high-pressure air is used as secondary air 8 and is input into a turbine 4 together with the high-temperature gas output by the combustion chamber 3, the turbine 4 works through the high-temperature gas and is cooled through the secondary air 8, and the outlet of the turbine 4 exhausts the gas.
The gas turbine comprises four large rotating parts of an air compressor 2, a turbine 4, a generator 5 and a blast furnace gas compressor 6. In fig. 2, the dashed lines represent the thermodynamic flow of air and fuel, and the solid lines represent the rotor shaft connections.
As shown in fig. 1, the embodiment of the present invention and the implementation process thereof are specifically as follows:
the method comprises the following steps: acquiring operation data of the low-heat-value gas turbine in real time, extracting characteristic variables of four parts, namely an air compressor, a gas compressor, a combustion chamber and a turbine in the low-heat-value gas turbine, analyzing the correlation of the characteristic variables, extracting the physical mechanism relation of the characteristic variables, and establishing a knowledge graph of the characteristic variables of the parts in the low-heat-value gas turbine;
step two: analyzing real-time operation data of the characteristic variables of the low-heat-value gas turbine according to the knowledge graph, and establishing a characteristic graph of part of working condition components;
step three: optimizing the characteristic maps of the parts under the partial working conditions by adopting a random forest integrated machine learning algorithm in combination with the incidence relation of characteristic variables in the knowledge maps of the parts of the low-calorific-value gas turbine, and further predicting the characteristic maps of the parts under the full working conditions to form the characteristic maps of the parts under the full working conditions;
step four: the characteristic prediction function in the full working condition characteristic map formed by four parts is adopted to establish a variable working condition process control function, the variable working condition process control function is used for carrying out overall variable working condition calculation processing on the real-time operation data of the low-heat-value gas turbine, and the control parameters in the variable working condition prediction process are further adjusted and controlled.
And dynamically updating the characteristic map of the all-condition component of the low-calorific-value gas turbine under frequent variable working conditions.
Further: in the first step, the first step is carried out,
the extracted characteristic variables comprise thermal characteristic variables, output characteristic variables, motion characteristic variables and geometric characteristic variables, and a knowledge graph among the four types of characteristic variables is established. The method comprises the steps that heat characteristic variables in operation data of the low-calorific-value gas turbine are pressure, temperature and flow of inlets and outlets of four components, output characteristic variables are load, motion characteristic variables are rotating speed, geometric characteristic variables of the air compressor are IGV opening of inlet adjustable guide vanes, high-pressure anti-surge valve opening and low-pressure anti-surge valve opening, geometric characteristic variables of the gas compressor are VGV opening of rotatable guide vanes, high-pressure anti-surge valve opening, medium-pressure anti-surge valve opening and low-pressure anti-surge valve opening, and geometric characteristic variables of a combustion chamber are combustion bypass valve opening.
Further: the establishment of the knowledge graph among the four types of characteristic variables is characterized in that,
extracting the physical mechanism relation of the characteristic variables, and establishing a knowledge graph of the correlation relation of the characteristic variables of the low-calorific-value gas turbine components through flow balance, load balance and pressure balance of cross section nodes of inlets and outlets of the four components, wherein the topological structure schematic diagram of the knowledge graph of the low-calorific-value gas turbine components is shown in FIG. 2.
Further: in the second step, establishing a part working condition component characteristic map comprises,
establishing a working condition characteristic map of the pressure ratio and the efficiency of the air compressor and the outlet folded flow of the air compressor respectively;
establishing a working condition characteristic map of the pressure ratio and the efficiency of the gas compressor and the outlet reduced flow of the gas compressor respectively;
establishing a working condition characteristic map of the pressure loss coefficient and efficiency of the combustion chamber and the rotating speed and load of the low-heat-value gas turbine respectively;
and establishing a working condition characteristic map of the turbine outlet reduced flow and efficiency and the turbine expansion ratio respectively.
The partial working conditions refer to the variation ranges of the geometric characteristic variable, the motion characteristic variable and the output characteristic variable in the actual operation process of the low-heat-value gas turbine within a certain period of time.
The partial working condition characteristic map of the low-calorific-value gas turbine component comprises, as shown by a shaded part in figure 3, establishing partial working condition characteristic maps of the pressure ratio and the efficiency of an air compressor and the outlet reduced flow respectively, as shown by a shaded part in figure 4, establishing partial working condition characteristic maps of the pressure ratio and the efficiency of a gas compressor and the outlet reduced flow respectively, as shown by a shaded part in figure 5, establishing partial working condition characteristic maps of the pressure loss coefficient and the efficiency of a combustion chamber and the rotating speed and the load respectively, as shown by a shaded part in figure 6, and establishing partial working condition characteristic maps of the outlet reduced flow and the efficiency of a turbine and the turbine expansion ratio respectively. The partial working condition refers to the variation range of the geometric characteristic variable, the motion characteristic variable and the output characteristic variable in the actual operation process of the low-heat-value gas turbine within a certain period of time.
Further: in the third step, the first step is that,
the incidence relation of the characteristic variables in the knowledge graph of the low-heat-value gas turbine component comprises the following steps:
A) air compressor component characteristic prediction function:
π c ＝f 1 (m c ,IGV,Bleed cH ,Bleed cL ,n)，η c ＝f 2 (m c ,IGV,Bleed cH ,Bleed cL ,n)
wherein f is 1 、f 2 For a prediction function, pi, of the characteristics of the first and second air compressor components c Is the pressure ratio of the air compressor eta c For air compressor efficiency, m c The flow rate is converted for the outlet of the air compressor, and the IGV is the opening degree of an adjustable guide vane at the inlet of the air compressor, Bleed cH 、Bleed cL Respectively the high-pressure anti-surge valve opening and the low-pressure anti-surge valve opening of the air compressor, wherein n is the rotating speed of the low-heat-value gas turbine;
B) gas compressor component characteristic prediction function:
π gc ＝f 3 (m gc ,VGV,Bleed gcH ,Bleed gcM ,Bleed gcL ,n)
η gc ＝f 4 (m gc ,VGV,Bleed gcH ,Bleed gcM ,Bleed gcL ,n)
wherein f is 3 、f 4 For a prediction function of the characteristics of the first and second gas compressor parts, pi gc Is the pressure ratio, eta, of the gas compressor gc For gas compressor efficiency, m gc For the outlet of the gas compressor to reduce the flow, and the VGV is the opening degree of the adjustable guide vanes at the inlet of the gas compressor, blade gcH 、Bleed gcM 、Bleed gcL Respectively the high-pressure anti-surge valve opening, the medium-pressure anti-surge valve opening and the low-pressure anti-surge valve opening of the gas compressor;
C) combustor component characteristic prediction function:
dp＝f 5 (Bypass,n,P)，η cb ＝f 6 (Bypass,n,P)
wherein f is 5 、f 6 Dividing the combustion chamber into two parts of an increasing speed process and an increasing load process for the characteristic prediction function of the first combustion chamber component and the second combustion chamber component, d p Is pressure loss of combustion chamber eta cb For the efficiency of the combustion chamber, Bypass is the opening of a Bypass valve of the combustion chamber, and P is the load of the low-heat-value gas turbine;
D) turbine characteristic prediction function:
m T ＝f 7 (π T ,n)，η T ＝f 8 (π T ,n)
wherein f is 7 、f 8 For the first and second turbine component characteristic prediction functions, m T For turbine outlet reduced flow, pi T Is the turbo expansion ratio, η T Is the turbine efficiency.
Pair f by random forest integrated machine learning algorithm 1 、f 2 、f 3 、f 4 、f 5 、f 6 、f 7 、f 8 And (4) carrying out full-working-condition modeling prediction to obtain a complete characteristic map as shown in the figures 3, 4, 5 and 6, wherein the characteristic map comprises a partial-working-condition operation area and an expansion area. f. of i ＝Γ(D,D bs ) Wherein f is i Is a gas turbine component characteristic prediction function, gamma is a random forest machine learning algorithm, D is a gas turbine characteristic sample, D bs For random self-helpA sample of gas turbine characteristics.
Further: in the fourth step of the method, the first step of the method,
establishing the following variable working condition process control functions of the low-heat-value gas turbine according to characteristic prediction functions of all parts of the low-heat-value gas turbine:
(IGV,VGV,Bypass)＝f 9 (n,P)
IGV＝f 10 (m c ,π c ,η c ,Bleed cH ,Bleed cL ,n)
VGV＝f 11 (m gc ,π gc ,η gc ,Bleed gcH ,Bleed gcM ,Bleed gcL ,n)
Bypass＝f 12 (dp,η cb ,n,P)
π T ＝π c ×(1-dp)，m T ＝m c +m gc
wherein, f 9 As a function of the process control of the varying conditions of the gas turbine with low calorific value, f 10 As a calculated function of the air compressor IGV, f 11 As a function of the calculation of the VGV of the gas compressor, f 12 Calculating functions of the combustion chamber Bypass, including variable load and variable rotating speed process control functions, wherein n is the rotating speed of the low-heat-value gas turbine, and P is the load of the low-heat-value gas turbine; IGV denotes the opening degree of an adjustable guide vane at the inlet of the air compressor, Bleed cH Indicating the degree of opening of the high-pressure surge-proof valve of the air compressor, Bleed cL Indicating the opening of the low-pressure surge-proof valve of the air compressor, and VGV indicating the opening of the adjustable guide vane of the inlet of the gas compressor, Bleed gcH Indicating the degree of opening of the high-pressure surge-proof valve of the gas compressor, Bleed gcM Indicating the degree of opening of the medium-pressure surge-preventing valve of the gas compressor, Bleed gcL The opening degree of the low-pressure anti-surge valve of the gas compressor is shown, and Bypass shows the opening degree of a Bypass valve of the combustion chamber.
And processing the variable working condition process control function of the low-heat value gas turbine by using real-time operation data to obtain the control parameters.
The control parameters comprise an opening IGV of an adjustable guide vane at the inlet of the air compressor and an opening Bleed of a high-pressure anti-surge valve of the air compressor cH Low-pressure anti-surge valve opening degree Bleed cL Guide vane opening degree VGV adjustable at inlet of gas compressor and high-pressure anti-surge valve opening degree Bleed of gas compressor gcH Middle pressure surge-proof valve opening degree Bleed gcM Low-pressure anti-surge valve opening degree Bleed gcL And the opening degree Bypass of the combustion chamber Bypass valve.
Surging of the air compressor and the gas compressor mainly occurs in the process of increasing the rotating speed, a high-pressure anti-surge valve and a low-pressure anti-surge valve of the air compressor are arranged in the process of increasing the rotating speed, and are firstly and completely opened, and are completely closed when the rotating speed reaches 90 percent of the rated rotating speed; the high-pressure anti-surge valve, the medium-pressure anti-surge valve and the low-pressure anti-surge valve of the gas compressor are arranged to be opened completely at first, and the high-pressure anti-surge valve, the medium-pressure anti-surge valve and the low-pressure anti-surge valve are closed linearly when the rotating speed reaches 60% of the rated rotating speed and are closed completely when the rotating speed reaches 90% of the rated rotating speed. In the process of load increase, the high-pressure anti-surge valve and the low-pressure anti-surge valve of the air compressor, and the high-pressure anti-surge valve, the medium-pressure anti-surge valve and the low-pressure anti-surge valve of the gas compressor are all closed.
FIG. 8 is a typical variable operating load and speed trend graph of a low heating value gas turbine over a period of time. FIG. 9 is a typical behavior control diagram of a low heating value gas turbine over a period of time. During the speed raising process, the output load of the gas turbine is zero, and when the speed is raised to the full speed of 100% (3000 rpm), the load raising process is started. The opening degree of the adjustable guide vane at the inlet of the air compressor, the adjustable guide vane at the inlet of the gas compressor and the bypass valve of the combustion chamber realize dynamic control and adjustment according to the change of the rotating speed and the load.
Further:
the low-calorific-value gas turbine acquires more operation data in real time under frequent variable working conditions, and the accuracy of the full-working-condition characteristic spectrogram is improved and the control precision of a dynamic process is increased through continuous real-time iteration of the knowledge spectrogram and machine learning.
The low-calorific-value gas turbine component characteristic map modeling method based on the knowledge map and machine learning can widen the working condition range of the low-calorific-value gas turbine component characteristic map, improve the accuracy and improve the dynamic updating performance.
The low-calorific-value gas turbine component characteristic map is low in acquisition cost and high in safety, and particularly avoids gas leakage risk and high component test cost in a gas compressor single component test. The problems of long time and low calculation accuracy of three-dimensional modeling simulation calculation of the part characteristics are solved.
The characteristic map of the low-calorific-value gas turbine component can be updated on line, and the dynamic adjustment of the parameters of the control system is realized. Therefore, the reliability, the economy and the safety of the low-heat-value gas turbine under the conditions of low fuel heat value and frequent variable working conditions are improved.

Claims (7)

1. A low-calorific-value gas turbine characteristic control method based on knowledge graph and machine learning is characterized by comprising the following steps: the method comprises the following steps:
the method comprises the following steps: acquiring operation data of a low-calorific-value gas turbine in real time, extracting characteristic variables of four parts, namely an air compressor, a gas compressor, a combustion chamber and a turbine in the low-calorific-value gas turbine, and establishing a knowledge graph of the characteristic variables of the parts in the low-calorific-value gas turbine according to the characteristic variables;
step two: analyzing real-time operation data of the characteristic variables of the low-heat-value gas turbine according to the knowledge graph, and establishing a characteristic graph of part of working condition components;
step three: optimizing the characteristic maps of the parts under the partial working conditions by adopting a random forest integrated machine learning algorithm in combination with the incidence relation of characteristic variables in the knowledge maps of the parts of the low-calorific-value gas turbine, and further predicting the characteristic maps of the parts under the full working conditions to form the characteristic maps of the parts under the full working conditions;
step four: and establishing a variable working condition process control function by adopting the full working condition characteristic map, carrying out overall variable working condition calculation processing on the real-time operation data of the low-heat-value gas turbine by using the variable working condition process control function, and predicting control parameters in the variable working condition process so as to carry out adjustment control.
2. The method of controlling low heating value gas turbine characteristics based on knowledge-graph and machine learning of claim 1, wherein: in the first step, the extracted characteristic variables comprise thermal characteristic variables, output characteristic variables, motion characteristic variables and geometric characteristic variables, and a knowledge graph among the four types of characteristic variables is established.
3. The method of controlling low heating value gas turbine characteristics based on knowledge-graph and machine learning of claim 2, wherein: the thermal characteristic variables are pressure, temperature and flow of inlets and outlets of the four components respectively, the output characteristic variables are loads of the four components respectively, the motion characteristic variables are rotating speeds of the four components respectively, the geometric characteristic variables of the air compressor are IGV opening of inlet adjustable guide vanes of the air compressor, high-pressure anti-surge valve opening and low-pressure anti-surge valve opening of the air compressor, the geometric characteristic variables of the gas compressor are VGV opening of rotatable guide vanes of the gas compressor, high-pressure anti-surge valve opening, medium-pressure anti-surge valve opening and low-pressure anti-surge valve opening of the gas compressor, and the geometric characteristic variables of the combustion chamber are combustion bypass valve opening.
4. The method of controlling low heating value gas turbine characteristics based on knowledge-graph and machine learning of claim 1, wherein: in the second step, establishing a part working condition component characteristic map comprises,
establishing a working condition characteristic map of the pressure ratio and the efficiency of the air compressor and the outlet folded flow of the air compressor respectively;
establishing a working condition characteristic map of the pressure ratio and the efficiency of the gas compressor and the outlet reduced flow of the gas compressor respectively;
establishing a working condition characteristic map of the pressure loss coefficient and efficiency of the combustion chamber and the rotating speed and load of the low-heat-value gas turbine respectively;
and establishing a working condition characteristic map of the turbine outlet reduced flow and efficiency and the turbine expansion ratio respectively.
5. The method of claim 1, wherein the knowledge-graph between four types of feature variables is established by: in the third step, the incidence relation of the characteristic variables in the knowledge graph of the low-heating-value gas turbine component comprises the following steps:
A) air compressor component characteristic prediction function:
π c ＝f 1 (m c ,IGV,Bleed cH ,Bleed cL ,n)，η c ＝f 2 (m c ,IGV,Bleed cH ,Bleed cL ,n)
wherein f is 1 、f 2 For a prediction function, pi, of the characteristics of the first and second air compressor components c Is the pressure ratio of the air compressor eta c For air compressor efficiency, m c The flow rate is converted for the outlet of the air compressor, and the IGV is the opening degree of an adjustable guide vane at the inlet of the air compressor, Bleed cH 、Bleed cL Respectively the high-pressure anti-surge valve opening and the low-pressure anti-surge valve opening of the air compressor, wherein n is the rotating speed of the low-heat-value gas turbine;
B) gas compressor component characteristic prediction function:
π gc ＝f 3 (m gc ,VGV,Bleed gcH ,Bleed gcM ,Bleed gcL ,n)
η gc ＝f 4 (m gc ,VGV,Bleed gcH ,Bleed gcM ,Bleed gcL ,n)
wherein f is 3 、f 4 For a prediction function of the characteristics of the first and second gas compressor parts, pi gc Is the pressure ratio, eta, of the gas compressor gc For gas compressor efficiency, m gc For the outlet of the gas compressor to reduce the flow, and the VGV is the opening degree of the adjustable guide vanes at the inlet of the gas compressor, blade gcH 、Bleed gcM 、Bleed gcL Respectively the high-pressure anti-surge valve opening, the medium-pressure anti-surge valve opening and the low-pressure anti-surge valve opening of the gas compressor;
C) combustor component characteristic prediction function:
dp＝f 5 (Bypass,n,P)，η cb ＝f 6 (Bypass,n,P)
wherein f is 5 、f 6 The different characteristics of the combustion chamber are divided into an increasing speed process and an increasing load process as a function of the characteristics of the first and second combustion chamber parts, d p Is pressure loss of combustion chamber eta cb For the efficiency of the combustion chamber, Bypass is the opening of a Bypass valve of the combustion chamber, and P is the load of the low-heat-value gas turbine;
D) turbine characteristic prediction function:
m T ＝f 7 (π T ,n)，η T ＝f 8 (π T ,n)
wherein f is 7 、f 8 For the first and second turbine component characteristic prediction functions, m T For turbine outlet reduced flow, pi T Is the turbo expansion ratio, η T Is the turbine efficiency.
6. The method of controlling low heating value gas turbine characteristics based on knowledge-graph and machine learning of claim 1, wherein: the third step is specifically as follows: characteristic prediction function f of each part of low-heat-value gas turbine by adopting random forest integration machine learning algorithm 1 、f 2 、f 3 、f 4 、f 5 、f 6 、f 7 、f 8 Respectively optimizing and determining parameters in the characteristic prediction function, and using the obtained characteristic prediction function to perform all-condition modeling prediction:
f i ＝Γ(D,D bs )
wherein, f i Is a characteristic prediction function of the low-heat value gas turbine, gamma is a random forest machine learning algorithm, D is a gas turbine characteristic sample, D bs Are random self-sampling gas turbine characteristic samples.
7. The method of controlling low heating value gas turbine characteristics based on knowledge-graph and machine learning of claim 1, wherein: the fourth step is specifically as follows: establishing the following variable working condition process control functions of the low-heat-value gas turbine according to characteristic prediction functions of all parts of the low-heat-value gas turbine:
(IGV,VGV,Bypass)＝f 9 (n,P)
IGV＝f 10 (m c ,π c ,η c ,Bleed cH ,Bleed cL ,n)
VGV＝f 11 (m gc ,π gc ,η gc ,Bleed gcH ,Bleed gcM ,Bleed gcL ,n)
Bypass＝f 12 (dp,η cb ,n,P)
π T ＝π c ×(1-dp)，m T ＝m c +m gc
wherein f is 9 As a function of the process control of the varying conditions of the gas turbine with low calorific value, f 10 As a calculated function of the air compressor IGV, f 11 As a function of the calculation of the VGV of the gas compressor, f 12 The calculation function of the combustion chamber Bypass is adopted, n is the rotating speed of the low-heat-value gas turbine, and P is the load of the low-heat-value gas turbine; IGV denotes the opening degree of an adjustable guide vane at the inlet of an air compressor, Bleed cH Indicating the degree of opening of the high-pressure surge-proof valve of the air compressor, Bleed cL Showing the opening of the low-pressure surge-proof valve of the air compressor, VGV showing the opening of the adjustable guide vane of the inlet of the gas compressor, Bleed gcH Indicating the degree of opening of the high-pressure surge-proof valve of the gas compressor, Bleed gcM Indicating the degree of opening of the medium-pressure surge-preventing valve of the gas compressor, Bleed gcL The opening of the low-pressure anti-surge valve of the gas compressor is represented, and Bypass represents the opening of a Bypass valve of the combustion chamber;
and processing the variable working condition process control function of the low-heat value gas turbine by using real-time operation data to obtain the control parameters.
Surging of the air compressor and the gas compressor mainly occurs in the process of increasing the rotating speed, a high-pressure anti-surge valve and a low-pressure anti-surge valve of the air compressor are arranged in the process of increasing the rotating speed, and are firstly and completely opened, and are completely closed when the rotating speed reaches 90 percent of the rated rotating speed; the high-pressure anti-surge valve, the medium-pressure anti-surge valve and the low-pressure anti-surge valve of the gas compressor are arranged to be opened completely at first, and the high-pressure anti-surge valve, the medium-pressure anti-surge valve and the low-pressure anti-surge valve are closed linearly when the rotating speed reaches 60% of the rated rotating speed and are closed completely when the rotating speed reaches 90% of the rated rotating speed. In the process of load increase, the high-pressure anti-surge valve and the low-pressure anti-surge valve of the air compressor, and the high-pressure anti-surge valve, the medium-pressure anti-surge valve and the low-pressure anti-surge valve of the gas compressor are all closed.
CN202210644972.9A 2022-06-08 Low-heating-value gas turbine characteristic control method based on knowledge graph and machine learning Active CN114970363B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN202210644972.9A CN114970363B (en) 2022-06-08 Low-heating-value gas turbine characteristic control method based on knowledge graph and machine learning

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN202210644972.9A CN114970363B (en) 2022-06-08 Low-heating-value gas turbine characteristic control method based on knowledge graph and machine learning

Publications (2)

Publication Number Publication Date
CN114970363A true CN114970363A (en) 2022-08-30
CN114970363B CN114970363B (en) 2024-07-02

Family

ID=

Citations (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JP2001123852A (en) * 1999-10-27 2001-05-08 Hitachi Ltd Gas turbine power generation control device
EP3540542A1 (en) * 2018-03-13 2019-09-18 Siemens Aktiengesellschaft Method and apparatus for providing an instantiated industrial semantic model for an industrial infrastructure
CN111177394A (en) * 2020-01-03 2020-05-19 浙江大学 Knowledge map relation data classification method based on syntactic attention neural network
CN111241295A (en) * 2020-01-03 2020-06-05 浙江大学 Knowledge map relation data extraction method based on semantic syntax interactive network
CN113740042A (en) * 2021-07-30 2021-12-03 西安交通大学 Device and method for testing vibration contact characteristic of gas turbine blade damper system
CN114970035A (en) * 2022-06-08 2022-08-30 浙江大学 Forward design method of low-calorific-value gas turbine based on performance knowledge graph

Patent Citations (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JP2001123852A (en) * 1999-10-27 2001-05-08 Hitachi Ltd Gas turbine power generation control device
EP3540542A1 (en) * 2018-03-13 2019-09-18 Siemens Aktiengesellschaft Method and apparatus for providing an instantiated industrial semantic model for an industrial infrastructure
CN111177394A (en) * 2020-01-03 2020-05-19 浙江大学 Knowledge map relation data classification method based on syntactic attention neural network
CN111241295A (en) * 2020-01-03 2020-06-05 浙江大学 Knowledge map relation data extraction method based on semantic syntax interactive network
CN113740042A (en) * 2021-07-30 2021-12-03 西安交通大学 Device and method for testing vibration contact characteristic of gas turbine blade damper system
CN114970035A (en) * 2022-06-08 2022-08-30 浙江大学 Forward design method of low-calorific-value gas turbine based on performance knowledge graph

Non-Patent Citations (7)

* Cited by examiner, † Cited by third party
Title
A. MEHRPANAHI 等: "Multi-objective optimization of IGV position in a heavy-duty gas turbine on part-load performance", APPLIED THERMAL ENGINEERING 125 (2017), 15 July 2017 (2017-07-15), pages 1478 *
GUANGQIAN LU 等: "Application of Automatic Completion Algorithm of Power Professional Knowledge Graphs in View of Convolutional Neural Network", INTERNATIONAL JOURNAL OF INFORMATION TECHNOLOGIES AND SYSTEMS APPROACH, vol. 16, no. 2, 23 May 2023 (2023-05-23), pages 1 - 14 *
张栋豪 等: "知识图谱在智能制造领域的研究现状及其应用前景综述", 机械工程学报, vol. 57, no. 5, 20 February 2021 (2021-02-20), pages 90 - 113 *
王明达 等: "燃气轮机故障知识图谱构建方法与应用研究", 中国安全生产科学技术, vol. 19, no. 11, 30 November 2024 (2024-11-30), pages 121 - 128 *
王永真;朱轶林;潘利生;李成宇;罗向龙;赵军;: "基于知识图谱的有机朗肯循环研究概览", 太阳能, no. 02, 28 February 2020 (2020-02-28), pages 20 - 34 *
谭建荣;刘振宇;徐敬华;: "新一代人工智能引领下的智能产品与装备", 中国工程科学, no. 04, 15 August 2018 (2018-08-15), pages 43 - 51 *
辛小鹏 等: "燃气轮机正向设计研究进展", 机械工程学报, vol. 58, no. 17, 31 May 2022 (2022-05-31), pages 191 - 205 *

Similar Documents

Publication Publication Date Title
CN111914362B (en) Self-adaptive method for turbofan engine model in research and development stage
CN108223140B (en) A kind of three shaft gas turbine power turbine control of inlet guide vanes rule optimization method
CN108223139B (en) Deflation Rule adjusting optimization method before a kind of split gas turbine power turbine
CN111623988B (en) Safety evaluation method for load shedding test of high-position-arrangement steam turbine set
CN112253515A (en) State adjusting method for performance test of double-duct combined type gas compressor
CN106126902A (en) Gas turbine component characteristic line modification method based on particle swarm optimization algorithm identification
Hao et al. Off-design performance of 9F gas turbine based on gPROMs and BP neural network model
CN113107623A (en) Device and method for improving heat supply steam extraction parameters during low-load operation of double-low-pressure-cylinder steam turbine
Zhu et al. Improvement of part-load performance of gas turbine by adjusting compressor inlet air temperature and IGV opening
Johnson Prediction of gas turbine on-and off-design performance when firing coal-derived syngas
CN114462245A (en) Method and system for determining optimal rotating speed of multi-shaft gas turbine
Kim Generating a virtual physical model through measurement data and reverse engineering: Applying a performance prediction model for an industrial gas turbine during start-up
Long et al. Research on active modulation of gas turbine cooling air flow
CN114970363B (en) Low-heating-value gas turbine characteristic control method based on knowledge graph and machine learning
CN112069616A (en) Intelligent service life prolonging control method for recycling of retired aircraft engine
CN114970363A (en) Low-calorific-value gas turbine characteristic control method based on knowledge graph and machine learning
CN117010099A (en) High-low pressure turbine matching design method for cross-generation small-bypass-ratio turbofan engine
Khormali et al. Identification of an industrial gas turbine based on Rowen's model and using Multi-Objective Optimization method
Macek et al. Calibration and results of a radial turbine 1-d model with distributed parameters
CN115419509A (en) Gas turbine engine dynamic model
CN115758923A (en) Engine overall simulation modeling method and system based on distributed power system
Petrovic et al. New Method for Cycle Performance Prediction Based on Detailed Compressor and Gas Turbine Flow Calculations
Gilani et al. Study the effect of variable vanes on performance of axial compressor for single shaft gas turbine cogeneration plant
Zhang et al. Operational data-based adaptive improvement method of gas turbine component characteristics for performance simulation
Mehrpanahi et al. Dynamic modeling of an industrial gas turbine in loading and unloading conditions using a gray box 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