CN110080882A - A kind of the starting method and starter of gas turbine - Google Patents
A kind of the starting method and starter of gas turbine Download PDFInfo
- Publication number
- CN110080882A CN110080882A CN201910303047.8A CN201910303047A CN110080882A CN 110080882 A CN110080882 A CN 110080882A CN 201910303047 A CN201910303047 A CN 201910303047A CN 110080882 A CN110080882 A CN 110080882A
- Authority
- CN
- China
- Prior art keywords
- parameter value
- output
- model
- parameter
- starter
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
Classifications
-
- F—MECHANICAL ENGINEERING; LIGHTING; HEATING; WEAPONS; BLASTING
- F02—COMBUSTION ENGINES; HOT-GAS OR COMBUSTION-PRODUCT ENGINE PLANTS
- F02C—GAS-TURBINE PLANTS; AIR INTAKES FOR JET-PROPULSION PLANTS; CONTROLLING FUEL SUPPLY IN AIR-BREATHING JET-PROPULSION PLANTS
- F02C7/00—Features, components parts, details or accessories, not provided for in, or of interest apart form groups F02C1/00 - F02C6/00; Air intakes for jet-propulsion plants
- F02C7/26—Starting; Ignition
- F02C7/268—Starting drives for the rotor, acting directly on the rotor of the gas turbine to be started
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F17/00—Digital computing or data processing equipment or methods, specially adapted for specific functions
- G06F17/10—Complex mathematical operations
- G06F17/15—Correlation function computation including computation of convolution operations
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Mathematical Optimization (AREA)
- Combustion & Propulsion (AREA)
- Mathematical Analysis (AREA)
- General Engineering & Computer Science (AREA)
- Pure & Applied Mathematics (AREA)
- Computational Mathematics (AREA)
- Data Mining & Analysis (AREA)
- Mathematical Physics (AREA)
- Chemical & Material Sciences (AREA)
- Theoretical Computer Science (AREA)
- Algebra (AREA)
- Software Systems (AREA)
- Databases & Information Systems (AREA)
- Computing Systems (AREA)
- Mechanical Engineering (AREA)
- Control Of Turbines (AREA)
Abstract
The invention discloses the starting methods and starter of a kind of gas turbine, this method comprises: obtaining sample data；Include training data in sample data, includes the first input parameter value and the first output parameter value in training data；First input parameter value in training data is obtained into N number of output result as the input parameter value of N number of kernel function；N is the integer more than or equal to 2；If the first output result matches with the first output parameter value in training data in N number of output result, the corresponding first function of result will be exported as Boot Model with first；Starter is based on Boot Model control gas turbine starting.In this way, starter can choose a more appropriate Boot Model, so as to reduce influence of the Boot Model to the starting performance of gas turbine, and then the starting performance of gas turbine can be improved.
Description
Technical field
The present invention relates to gas turbines to model field, the starting method and starter of espespecially a kind of gas turbine.
Background technique
Gas turbine is the gas impeller high speed rotation continuously to flow, and is useful work by the energy conversion of fuel
Internal combustion type dynamic power machine is a kind of rotary vane type Thermal Motor.Gas turbine start-up course refers to gas turbine and its institute
The equipment of driving, from starting state is prepared to the process for preparing stress state.
In general, gas turbine is started based on Boot Model.Boot Model is the input parameter of gas turbine and defeated
Functional relation between parameter out, i.e. gas turbine are the output valves based on Boot Model to start, and output valve is by starting
Model and input value determine.Therefore, whether Boot Model and input value are suitable, the correlation whether normally started with gas turbine
It is larger.For example, when Boot Model is improper, the starting performance that will lead to gas turbine is bad, affects opportunity adversely etc., it is likely to result in
The problems such as damage of gas turbine.
Summary of the invention
The embodiment of the present invention provides the starting method and starter of a kind of gas turbine, to reduce Boot Model to combustion
The influence of the starting performance of gas-turbine improves the starting performance of gas turbine.
In a first aspect, the embodiment of the present invention provides a kind of starting method of gas turbine, it is applied to a starter, it is described
Starter is for starting gas turbine, which comprises
Obtain sample data；Include training data in the sample data, includes the first input ginseng in the training data
Number value and the first output parameter value；
It is taken using the first input parameter value in the training data as the input parameter of each kernel function in N number of kernel function
Value, calculates each kernel function, obtains N number of output result；N is the integer more than or equal to 2；
If the first output result in N number of output result and the first output parameter value phase in the training data
Match, then it will be with the corresponding first function of the first output result as Boot Model；
The gas turbine starting is controlled based on the Boot Model.
Optionally, further include verify data in the sample data, include that the second input parameter takes in the verify data
Value and the second output parameter value, before the starter is based on the Boot Model control gas turbine starting,
The method also includes:
Second input parameter value in the verify data is calculated into institute as the input parameter value of the Boot Model
Boot Model is stated, obtains the second output as a result, the second output result is taken with the second output parameter in the verify data
Value compares；
If the mean square error between the second output result and the second output parameter value is in the second preset range
It is interior, it is determined that the Boot Model is eligible.
Optionally, the first output result and the first output parameter value in the training data in N number of output result
Match, comprising: the mean square error in the first output result and the training data between the first output parameter value exists
Within first preset range.
It optionally, include model parameter in the first function；Will corresponding with the first output parameter value
Before one function is as Boot Model, the method also includes:
A model parameter value is selected from multiple model parameter values of the first function；
Parameter value is inputted as the input parameter value of N number of kernel function using in the training data first, comprising:
First input parameter value in the training data and the model parameter value selected are brought into respectively described N number of
Kernel function calculates each kernel function, obtains N number of output result.
Optionally, the model parameter includes punishment parameter and kernel functional parameter, and the punishment parameter is the kernel function
Constraint condition, the kernel functional parameter is the coefficient of the kernel function.
Optionally, the input parameter of the Boot Model includes at least one of following parameter:
Atmospheric temperature, atmospheric pressure, amount of natural gas, starting current of electric.
Optionally, the output parameter of the Boot Model includes at least one of following parameter:
Engine speed, compressor delivery temperature, compressor delivery pressure, combustor exit pressure, turbine-exit temperature,
Turbine outlet pressure.
Second aspect, the embodiment of the present invention provide a kind of starter, and the starter is for starting gas turbine, institute
Stating starter includes:
Acquiring unit, for obtaining sample data；Include training data in the sample data, is wrapped in the training data
Include the first input parameter value and the first output parameter value；
Processing unit, for using the first input parameter value in the training data as core letter each in N number of kernel function
Several input parameter values, calculates each kernel function, obtains N number of output result；N is the integer more than or equal to 2；
If the first output result in N number of output result and the first output parameter value phase in the training data
Match, the processing unit is also used to using first function corresponding with the first output result as Boot Model；
The processing unit is also used to control the gas turbine starting based on the Boot Model.
Optionally, further include verify data in the sample data, include that the second input parameter takes in the verify data
Value and the second output parameter value, the processing unit are starting it for controlling the gas turbine based on the Boot Model
Before, it is also used to:
Second input parameter value in the verify data is calculated into institute as the input parameter value of the Boot Model
Boot Model is stated, obtains the second output as a result, the second output result is taken with the second output parameter in the verify data
Value compares；
If the mean square error between the second output result and the second output parameter value is in the second preset range
It is interior, it is determined that the Boot Model is eligible.
Optionally, the first output result and the first output parameter value in the training data in N number of output result
Match, comprising: the mean square error in the first output result and the training data between the first output parameter value exists
Within first preset range.
It optionally, include model parameter in the first function；The processing unit with described first for that will export
Before the corresponding first function of parameter value is as Boot Model, it is also used to:
A model parameter value is selected from multiple model parameter values of the first function；
The processing unit is for inputting parameter value as N number of kernel function in the training data first
When inputting parameter value, it is specifically used for:
First input parameter value in the training data and the model parameter value selected are brought into respectively described N number of
Kernel function calculates each kernel function, obtains N number of output result.
Optionally, the model parameter includes punishment parameter and kernel functional parameter, and the punishment parameter is the kernel function
Constraint condition, the kernel functional parameter is the coefficient of the kernel function.
Optionally, the input parameter of the Boot Model includes at least one of following parameter:
Atmospheric temperature, atmospheric pressure, amount of natural gas, starting current of electric.
Optionally, the output parameter of the Boot Model includes at least one of following parameter:
Engine speed, compressor delivery temperature, compressor delivery pressure, combustor exit pressure, turbine-exit temperature,
Turbine outlet pressure.
The third aspect, the embodiment of the present invention provide a kind of starter, and the starter is for starting gas turbine, institute
Stating starter includes:
Getter, for obtaining sample data；Include training data in the sample data, includes in the training data
First input parameter value and the first output parameter value；
Processor, for using the first input parameter value in the training data as each kernel function in N number of kernel function
Input parameter value, calculate each kernel function, obtain N number of output result；N is the integer more than or equal to 2；
If the first output result in N number of output result and the first output parameter value phase in the training data
Match, the processor is also used to using first function corresponding with the first output result as Boot Model；
The processor is also used to control the gas turbine starting based on the Boot Model.
Optionally, further include verify data in the sample data, include that the second input parameter takes in the verify data
Value and the second output parameter value, the processor are starting it for controlling the gas turbine based on the Boot Model
Before, it is also used to:
Second input parameter value in the verify data is calculated into institute as the input parameter value of the Boot Model
Boot Model is stated, obtains the second output as a result, the second output result is taken with the second output parameter in the verify data
Value compares；
If the mean square error between the second output result and the second output parameter value is in the second preset range
It is interior, it is determined that the Boot Model is eligible.
Optionally, the first output result and the first output parameter value in the training data in N number of output result
Match, comprising: the mean square error in the first output result and the training data between the first output parameter value exists
Within first preset range.
It optionally, include model parameter in the first function；The processor with first output for that will join
Before the corresponding first function of number value is as Boot Model, it is also used to:
A model parameter value is selected from multiple model parameter values of the first function；
The processor is for inputting parameter value as the defeated of N number of kernel function in the training data first
When entering parameter value, it is specifically used for:
First input parameter value in the training data and the model parameter value selected are brought into respectively described N number of
Kernel function calculates each kernel function, obtains N number of output result.
Optionally, the model parameter includes punishment parameter and kernel functional parameter, and the punishment parameter is the kernel function
Constraint condition, the kernel functional parameter is the coefficient of the kernel function.
Optionally, the input parameter of the Boot Model includes at least one of following parameter:
Atmospheric temperature, atmospheric pressure, amount of natural gas, starting current of electric.
Optionally, the output parameter of the Boot Model includes at least one of following parameter:
Engine speed, compressor delivery temperature, compressor delivery pressure, combustor exit pressure, turbine-exit temperature,
Turbine outlet pressure.
Fourth aspect, the embodiment of the present invention provide a kind of starter, including processor and memory；Wherein, described to deposit
Reservoir is for storing one or more computer programs；When one or more computer programs of memory storage are described
When processor executes, so that the starter realizes any one possible design of first aspect or above-mentioned first aspect
Method.
5th aspect, the embodiment of the present invention provide a kind of computer readable storage medium, the computer-readable storage medium
Matter is stored with computer program, and the computer program includes program instruction, and described program instructs when executed by a computer, makes
The method that the computer executes any one possible design of first aspect or above-mentioned first aspect.
6th aspect, the embodiment of the present invention provide a kind of computer program product, and the computer program product is stored with
Computer program, the computer program include program instruction, and described program instructs when executed by a computer, make the calculating
The method that machine executes any one possible design of first aspect or above-mentioned first aspect.
The present invention has the beneficial effect that:
In the present invention in the technical solution of embodiment, the starting method of gas turbine includes: acquisition sample data；Sample number
Include training data in, includes the first input parameter value and the first output parameter value in training data；By training data
In first input parameter value respectively as the input parameter value of N number of kernel function, obtain N number of output result；N be more than or equal to
2 integer；It, will be with if the first output result matches with the first output parameter value in training data in N number of output result
The corresponding first function of first output result is as Boot Model；Starter is based on Boot Model control gas turbine starting.
In this way, starter can choose a more appropriate Boot Model, so as to reduce Boot Model to combustion
The influence of the starting performance of gas-turbine, and then the starting performance of gas turbine can be improved.
Detailed description of the invention
Fig. 1 is a kind of schematic diagram of application scenarios provided in an embodiment of the present invention；
Fig. 2 is a kind of flow diagram of the starting method of gas turbine provided in an embodiment of the present invention；
Fig. 3 is a kind of schematic diagram of the method for determining Boot Model provided in an embodiment of the present invention；
Fig. 4 is a kind of schematic diagram of the engine speed comparing result provided in the embodiment of the present invention；
Fig. 5 is a kind of schematic diagram of the compressor delivery temperature comparing result provided in the embodiment of the present invention；
Fig. 6 is a kind of schematic diagram of the compressor delivery pressure comparing result provided in the embodiment of the present invention；
Fig. 7 is a kind of schematic diagram of the combustor exit pressure comparison result provided in the embodiment of the present invention；
Fig. 8 is a kind of schematic diagram of the turbine-exit temperature comparing result provided in the embodiment of the present invention；
Fig. 9 is a kind of schematic diagram of the turbine outlet pressure comparison result provided in the embodiment of the present invention；
Figure 10 is a kind of structural schematic diagram of starter provided in an embodiment of the present invention；
Figure 11 is a kind of structural schematic diagram of starter provided in an embodiment of the present invention；
Figure 12 is a kind of structural schematic diagram of starter provided in an embodiment of the present invention.
Specific embodiment
To make the objectives, technical solutions, and advantages of the present invention clearer, below in conjunction with attached drawing to the present invention make into
It is described in detail to one step, it is clear that the described embodiment is only a part of the embodiment of the present invention, instead of all the embodiments.
Based on the embodiments of the present invention, obtained by those of ordinary skill in the art without making creative efforts all
Other embodiments shall fall within the protection scope of the present invention.
The shapes and sizes of each component do not reflect actual proportions in attached drawing, and purpose is schematically illustrate the content of present invention.
Referring to FIG. 1, being a kind of schematic diagram of application scenarios provided in an embodiment of the present invention.As shown in Figure 1, the applied field
It include gas turbine 100 and starter 101 in scape, wherein gas turbine 100 and starter 101 connect.Starter
101 can control the starting of gas turbine 100 according to Boot Model.In Fig. 1, with starter 101 and gas turbine 100
It is different for equipment, in fact, starter 101 and gas turbine 100 can also be same equipment, such as starter
101 can integrate in gas turbine 100, and the embodiment of the present invention does not limit.
Optionally, starter 101 can select a more appropriate kernel function from multiple kernel functions, as starting
Model starts gas turbine, to reduce the influence to the starting performance of gas turbine, and then can be improved gas turbine
Starting performance.
In addition, the Boot Model that the start-up course of gas turbine relies on is based on gas turbine due in the prior art
Inherent mechanism (working principle of gas turbine start-up course) is modeled.However, being based on needing when modelling by mechanism to combustion gas
Each stage in turbine start-up course is studied, and because gas turbine start-up course is complicated, so that being based on mechanism
The difficulty of modeling is big, expends time and efforts.For example, technical staff needs when modeling to gas turbine start-up course
Consider the working principle (gas turbine start-up course may include multiple stages) in each stage of gas turbine, it is also necessary to examine
Consider whether each stage can interact between each other.Therefore, technical staff is based on the inherent mechanism of gas turbine to combustion gas
When turbine start-up course is modeled, difficulty is larger, and the time and efforts spent is larger.And it provided in an embodiment of the present invention opens
Movable model is not necessarily to the inherent mechanism without the concern for gas turbine, the work in each stage without considering gas turbine
Make principle, also can choose more appropriate Boot Model, facilitates the starting process modeling method that can reduce gas turbine
Difficulty, the time and efforts of saving technique personnel.
The process that gas turbine 100 shown in FIG. 1 starts is described below.
Referring to FIG. 2, being a kind of flow diagram of the starting method of gas turbine provided in an embodiment of the present invention.Such as figure
Shown in 2, this method comprises:
S201, sample data is obtained；Include training data in sample data, includes that the first input parameter takes in training data
Value and the first output parameter value.
S202, parameter value is inputted as the input of each kernel function in N number of kernel function using in training data first
Parameter value calculates each kernel function, obtains N number of output result；N is the integer more than or equal to 2.
If in S203, N number of output result first output result and training data in the first output parameter value match,
It will be with the first corresponding first function of output result as Boot Model.
S204, starter are based on Boot Model control gas turbine starting.
Optionally, sample data is the test data in gas turbine, i.e., before determining Boot Model, technical staff can
Gas turbine start-up course is tested with choosing different input parameter values, obtains different output parameter values
(i.e. S201).Wherein, with the input parameter of gas turbine start-up course, (the input parameter is also starting mould in the embodiment of the present invention
The input parameter of type) it include atmospheric temperature T1, atmospheric pressure P1, amount of natural gas FCH4 and starting current of electric I these four input ginseng
It include starting with the output parameter (output parameter that the output parameter is also Boot Model) of gas turbine start-up course for number
Machine revolving speed N, compressor delivery temperature T2, compressor delivery pressure P2, combustor exit pressure P3, turbine-exit temperature T3The whirlpool and
Take turns outlet pressure P4For this six kinds of output parameters.Certainly, the input parameter of gas turbine can also include other parameters, defeated
Parameter can also include other parameters out, and the embodiment of the present invention does not limit.
It should be noted that the selection about input and output parameter can follow the two principles: parameter wants energy
It enough include all physical quantities of Study On Start-up Process For Gas Turbines；Correlation between parameter is small as far as possible.For example, being risen in gas turbine
The physical quantity that dynamic process is related to includes atmospheric temperature, atmospheric pressure, amount of natural gas and starting current of electric etc. for input
Physical quantity and engine speed, compressor delivery temperature, compressor delivery pressure, combustor exit pressure, turbine outlet temperature
The physical quantity for output such as degree and turbine outlet pressure.Therefore, Study On Start-up Process For Gas Turbines can be related to by starter
Input physical quantity for input parameter as gas turbine start-up course and Boot Model, physical quantity will be exported and be used for conduct
The output parameter of gas turbine start-up course and Boot Model.For another example, the relevance between atmospheric temperature and atmospheric pressure compared with
Small, i.e., interacting property between the two is smaller, and others input physical quantity and output physical quantity are also similar.
Optionally, sample data can be divided into training data and verify data by technical staff, wherein training data and be tested
It may include the value of known input parameter and the value of known output parameter in card data.
Optionally, input parameter value known in training data (the first input parameter value) is distinguished in starter
Before the input parameter value of N number of kernel function, technical staff can be by the value of input parameter known in training data
It is pre-processed.For example, preprocessing process may include: to kick known input parameter value abnormal in training data
It removes, and will kick except the known input parameter value after abnormal known input parameter value is normalized.Certainly,
Known output parameter value (the first output parameter value) can use above-mentioned identical mode or similar in training data
Mode is pre-processed.Training data by normalized is for the convergence rate of start quickly model.
Optionally, N number of kernel function includes multiple model parameters, for example, at least may include penalty parameter c and core letter
Number parameter g, wherein punishment parameter is the constraint condition of kernel function, and kernel functional parameter is the coefficient of kernel function.
Starter is using input parameter value known in training data as the input parameter value of N number of kernel function
Before, a model parameter value can be selected from the value of multiple model parameters, for example, taking from penalty parameter c
It is worth in range and selects a penalty parameter c value, a kernel functional parameter g is selected out of kernel functional parameter g value range
Value.
Optionally, starter can take input parameter value known in training data and the model parameter selected
Value brings N number of kernel function into respectively, obtains N number of output result (i.e. S202).Wherein, N is the integer more than or equal to 2.When N number of output
As a result (the i.e. first output result and training when the first output result matches with known output parameter value in training data in
Within the first preset range, the first preset range can be set mean square error in data between known output parameter value
For [0,0.1], certain first preset range may be arranged as other mean square error ranges, and the embodiment of the present invention does not limit),
Starter can will be with the first corresponding first function of output result as Boot Model.
For example, illustratively, referring to FIG. 3, being a kind of side of determining Boot Model provided in an embodiment of the present invention
The schematic diagram of method.As shown in figure 3, technical staff can be embedded in Boot Model as a black box 300 in starter.Its
In, black box 300 may include N number of kernel function 301, penalty parameter c (with 302 marks in Fig. 3) and kernel functional parameter g
(with 303 marks in Fig. 3).When known input parameter is more in training data, starter can will input parameter value
(T11..., T1m), (P11..., P1n), (FCH41..., FCH4j), (I1..., Ik) as (such as the f of N number of kernel function 3011(T1,
P1, FCH4, I), f2(T1, P1, FCH4, I), f3(T1, P1, FCH4, I) etc.) input parameter value.For example, will
(T11..., T1m), (P11..., P1n), (FCH41..., FCH4j), (I1..., Ik) it is used as f1The input of (T1, P1, FCH4, I) is joined
Number values, then by (T11..., T1m), (P11..., P1n), (FCH41..., FCH4j), (I1..., Ik) it is used as f2(T1, P1,
FCH4, I) input parameter value, and so on, by (T11..., T1m), (P11..., P1n), (FCH41..., FCH4j),
(I1..., Ik) it is used as fNThe input parameter value of (T1, P1, FCH4, I) (being not shown in Fig. 3).Wherein, m, n, j and k are big
, can be equal or unequal between m, n, j and k in the integer for being equal to 1, the embodiment of the present invention does not limit.
Optionally, each kernel function can also be trained by following mode.
Optionally, starter an optional value (can compare from the value range [- 5,5] of penalty (or parameter) c
Such as c1) value as penalty parameter c, can from the value range [- 5,5] of kernel functional parameter g an optional value (such as
G1) the value as kernel functional parameter g.Starter is according to input parameter value (T11..., T1m), (P11..., P1n),
(FCH41..., FCH4j), (I1..., Ik), value and each core letter of the value of penalty parameter c1 and kernel functional parameter g1
Number is trained (such as using c1 as constraint condition, using g1 as the parameter coefficient of kernel function), obtains N number of output result
(T21..., T2m)、(T31..., T3m)、(P21..., P2n)、(P31..., P3n)、(P41..., P4n)、(N1..., Nr).Wherein, r is
Integer more than or equal to 1, r and m, n, j, k can be equal or unequal, and the embodiment of the present invention does not limit.
Optionally, each kernel function can carry out each penalty (or parameter) c and each kernel functional parameter g
Training process, i.e., for each penalty (or parameter) c and each kernel functional parameter g can obtain it is different as a result,
After obtaining multiple results, suitable Boot Model (i.e. suitable kernel function) can be selected based on the result.
Optionally, if the first output result and output parameter value phase known in training data in N number of output result
With (matching process is introduced below), starter can will be with the first corresponding model parameter value (i.e. kernel function of output result
Model parameter of the parameter g) as Boot Model, can will be with the first corresponding first function of output result as Boot Model
(i.e. S203).
In the above manner, can determine the preferable kernel function of an output result as starting mould from multiple functions
Type.
Optionally, it can also include verify data in sample data, include known input parameter value in verify data
(the second input parameter value) and known output parameter value (the second output parameter value).It is opened when determining that first function is used as
When movable model, which verify whether eligible.
For example, starter can be using input parameter value known in verify data as the input of Boot Model
Parameter value obtains the second output as a result, by output parameter value progress known in the second output result and verify data
With (process and second that known output parameter value matches in the first output result and training data export result and test
It is similar or identical to demonstrate,prove the known matched process of output parameter value in data), if in the second output result and verify data
(the second preset range can be with the first preset range phase in the second preset range for the mean square error between output parameter value known
Together, can not also be identical, the embodiment of the present invention does not limit.With the second preset range and the first preset range in inventive embodiments
For identical) in, it is determined that Boot Model is eligible (certainly, if known output in the second output result and verify data
Not in the second preset range, starter can also repeat the above steps until selecting symbol mean square error between parameter value
The Boot Model of conjunction condition).For example, starter can be by the known output parameter in the second output result and verify data
Value is compared, and calculates the mean square error between the known output parameter value in the second output result and verify data
Difference.Illustratively, Fig. 4-Fig. 9 is please referred to, Fig. 4-Fig. 9 is that a kind of engine speed provided in an embodiment of the present invention, compressor go out
Mouthful temperature, compressor delivery pressure, combustor exit pressure, turbine-exit temperature and turbine outlet pressure respectively with it is known defeated
(test data in Fig. 4-Fig. 9 is known output ginseng in verify data to the schematic diagram of the comparing result between parameter value out
Number value, identification result are the second output result).
Optionally, starter can calculate the output parameter value (hair of Boot Model according to the comparing result of Fig. 4-Fig. 9
Motivation revolving speed, compressor delivery temperature, compressor delivery pressure, combustor exit pressure, turbine-exit temperature and turbine outlet
Pressure) and the value of known output parameter between mean square error, i.e. starter can be according to the comparing result of Fig. 4-Fig. 9
Calculate the mean square error between the second output result and the value of known output parameter.Illustratively, table 1 is please referred to, for this
Comparing result analysis between the output parameter and known output parameter value of a kind of Boot Model that inventive embodiments provide.
Table 1
| Output parameter | Mean square error |
| Engine speed N | 0.00304527 |
| Compressor delivery temperature T2 | 0.00295662 |
| Compressor delivery pressure P2 | 0.00308422 |
| Combustor exit pressure P3 | 0.0031411 |
| Turbine-exit temperature T3 | 0.00569958 |
| Turbine outlet pressure P4 | 0.0860011 |
As shown in table 1, engine speed N, compressor delivery temperature T2, compressor delivery pressure P2, combustor exit pressure
P3, turbine-exit temperature T3, turbine outlet pressure P4Mean square error between known output parameter value is default second
In range [0,0.1], therefore, the Boot Model is eligible.
Optionally, when determining that Boot Model is eligible, starter can control combustion gas wheel based on the Boot Model
Machine starts (i.e. S204).
It should be noted that joining using input parameter value known in verify data as the input of Boot Model
Before number value, technical staff can be pre-processed input parameter value known in verify data.Wherein, it pre-processes and is
Abnormal known input parameter value in verify data kick removing, and will be kicked except abnormal known input parameter takes
The value of known input parameter after value is normalized.Certainly, known output parameter value can be in verify data
It is pre-processed using above-mentioned identical mode or similar mode.Verify data by normalized is to accelerate to open
The convergence rate of movable model.
In order to determine above-mentioned Boot Model determination process feasibility, it is as an example, real to pass through matrix below
Room (Matlab) is tested to realize that the process of determining Boot Model is introduced.
In Matlab, determining that the process of Boot Model there can be 5 processes, each process respectively corresponds yuchuli.m,
This 5 functions of sfunction_test.m, write4libsvm.m, SVMcg.m and huatu.m.Below to the work of this 5 functions
With being introduced respectively.
The effect of yuchuli function is to read the sample data (including training data and verify data) in Excel table
Enter to Matlab working space (Workspace), be trained using training data in order to sfunction_test function and
It is verified using verify data.
Sfunction_test function (4 input, 6 output is arranged in the function) is used for training data and verify data
Normalized has been carried out, and SVMcg function will have been delivered training data to after normalized, so that SVMcg function uses
Training data, which is trained, selects optimal penalty parameter value c, kernel functional parameter value g.Sfunction_test letter
Number is also used to according to the training number after optimal penalty parameter value c, kernel functional parameter value g, kernel function and normalization
According to predicting the first output as a result, and carrying out renormalization, first normally predicted to the first of prediction the output result
Output is as a result, and pass to huatu function for the normally predict first output result.Sfunction_test function is also used to
By a similar method or identical mode predicts the second output as a result, and carrying out anti-normalizing to the second of prediction the output result
Change, the second output normally predicted is as a result, and pass to huatu function for the normally predict second output result.
Write4libsvm function is before sfunction_test function is trained using training data, for selecting
Select punishment parameter and kernel functional parameter.
The penalty parameter c and kernel functional parameter g that SVMcg function is used to select using training data, write4libsvm function
Be trained, select optimal kernel functional parameter g and penalty parameter c, and by optimal penalty parameter value c and
Kernel functional parameter value g passes to sfunction_test function, so that sfunction_test function is normally predicted
First output result.
The first output result that huatu function is used to by the first output parameter value (i.e. reality output) and normally predict
Curve is presented in same figure that (or the second output result curve by the second output parameter value and normally predicted is presented in
In same figure), allow to more intuitive observation Boot Model precision.Before using huatu function, need to oscillograph
Output is configured, and the first output result and the second output result are saved in the working space of Matlab.
As can be seen from the above description, in the present invention in the technical solution of embodiment, the starting method of gas turbine includes: to obtain
Sample notebook data；Include training data in sample data, includes the first input parameter value and the first output ginseng in training data
Number value；First input parameter value in training data is obtained N number of defeated as the input parameter value of N number of kernel function
Result out；N is the integer more than or equal to 2；If the first output result and the first output parameter in training data in N number of output result
Value matches, then will be with the first corresponding first function of output result as Boot Model；Starter is based on Boot Model
Control gas turbine starting.In this way, starter can choose a more appropriate Boot Model, so as to
Influence of the Boot Model to the starting performance of gas turbine is reduced, and then the starting performance of gas turbine can be improved.
Under based on the same inventive concept, the embodiment of the invention provides a kind of starters.It please refers to shown in Figure 10, for this
A kind of structural schematic diagram for starter that inventive embodiments provide.
As shown in Figure 10, starter 400 includes acquiring unit 401 and processing unit 402.
Optionally, acquiring unit 401, for obtaining sample data；It include training data in sample data, in training data
Including the first input parameter value and the first output parameter value.
Optionally, processing unit 402, for using the first input parameter value in training data as every in N number of kernel function
The input parameter value of a kernel function, calculates each kernel function, obtains N number of output result；N is the integer more than or equal to 2；If N
The first output result in a output result matches with the first output parameter value in training data, and processing unit 402 is also used
In will be with the first corresponding first function of output result as Boot Model；Processing unit 402 is also used to based on Boot Model control
Gas turbine starting processed.
Optionally, further include verify data in sample data, include the second input parameter value and second in verify data
Output parameter value, processing unit 402 are also used to before for based on Boot Model control gas turbine starting:
Second input parameter value in verify data is calculated into Boot Model as the input parameter value of Boot Model,
The second output is obtained as a result, the second output result is compared with the second output parameter value in verify data；
If the mean square error between the second output result and the second output parameter value is in the second preset range, it is determined that
Boot Model is eligible.
Optionally, the first output result matches with the first output parameter value in training data in N number of output result, wraps
Include: the mean square error in the first output result and training data between the first output parameter value is within the first preset range.
It optionally, include model parameter in first function；Processing unit 402 is for will be with the first output parameter value pair
Before the first function answered is as Boot Model, it is also used to:
A model parameter value is selected from multiple model parameter values of first function；
Processing unit 402 is for inputting parameter value as the input of N number of kernel function in training data first
When parameter value, it is specifically used for:
Bring the first input parameter value in training data and the model parameter value selected into N number of kernel function respectively,
Each kernel function is calculated, N number of output result is obtained.
Optionally, model parameter includes punishment parameter and kernel functional parameter, and punishment parameter is the constraint condition of kernel function, core
Function parameter is the coefficient of kernel function.
Optionally, the input parameter of Boot Model includes at least one of following parameter:
Atmospheric temperature, atmospheric pressure, amount of natural gas, starting current of electric.
Optionally, the output parameter of Boot Model includes at least one of following parameter:
Engine speed, compressor delivery temperature, compressor delivery pressure, combustor exit pressure, turbine-exit temperature,
Turbine outlet pressure.
The starting method of starter 400 and aforementioned gas turbine shown in Fig. 2 in the present embodiment is based on same structure
Invention under thinking, by the detailed description of the aforementioned starting method to gas turbine, those skilled in the art can be clear
The implementation process for solving starter 400 in the present embodiment, so details are not described herein in order to illustrate the succinct of book.
Under based on the same inventive concept, the embodiment of the invention provides a kind of starters.It please refers to shown in Figure 11, for this
A kind of structural schematic diagram for starter that inventive embodiments provide.
As shown in figure 11, starter 500 includes getter 501 and processor 502.
Optionally, getter 501, for obtaining sample data；Include training data in sample data, is wrapped in training data
Include the first input parameter value and the first output parameter value.
Optionally, processor 502, for using the first input parameter value in training data as each in N number of kernel function
The input parameter value of kernel function, calculates each kernel function, obtains N number of output result；N is the integer more than or equal to 2；If N number of
Output result in first output result matches with the first output parameter value in training data, processor 502 be also used to by
First function corresponding with the first output result is as Boot Model；Processor 502 is also used to control combustion gas based on Boot Model
Turbine starting.
Optionally, further include verify data in sample data, include the second input parameter value and second in verify data
Output parameter value, processor 502 are specifically used for before for based on Boot Model control gas turbine starting:
Second input parameter value in verify data is calculated into Boot Model as the input parameter value of Boot Model,
The second output is obtained as a result, the second output result is compared with the second output parameter value in verify data；
If the mean square error between the second output result and the second output parameter value is in the second preset range, it is determined that
Boot Model is eligible.
Optionally, the first output result matches with the first output parameter value in training data in N number of output result, wraps
Include: the mean square error in the first output result and training data between the first output parameter value is within the first preset range.
It optionally, include model parameter in first function；Processor 502 will be for will be corresponding with the first output parameter value
First function as Boot Model before, be specifically used for:
A model parameter value is selected from multiple model parameter values of first function；
Processor 502 using the first input parameter value in training data as the input of N number of kernel function for joining
When number value, it is specifically used for:
Bring the first input parameter value in training data and the model parameter value selected into N number of kernel function respectively,
Each kernel function is calculated, N number of output result is obtained.
Optionally, model parameter includes punishment parameter and kernel functional parameter, and punishment parameter is the constraint condition of kernel function, core
Function parameter is the coefficient of kernel function.
Optionally, the input parameter of Boot Model includes at least one of following parameter:
Atmospheric temperature, atmospheric pressure, amount of natural gas, starting current of electric.
Optionally, the output parameter of Boot Model includes at least one of following parameter:
Engine speed, compressor delivery temperature, compressor delivery pressure, combustor exit pressure, turbine-exit temperature,
Turbine outlet pressure.
The starting method of starter 500 and aforementioned gas turbine shown in Fig. 2 in the present embodiment is based on same structure
Invention under thinking, by the detailed description of the aforementioned starting method to gas turbine, those skilled in the art can be clear
The implementation process for solving starter 500 in the present embodiment, so details are not described herein in order to illustrate the succinct of book.
Under based on the same inventive concept, the embodiment of the invention provides a kind of starters.It please refers to shown in Figure 12, for this
A kind of structural schematic diagram for starter that inventive embodiments provide.As shown in figure 12, starter 600 includes processor 601
With memory 602.Optionally, processor 601 can be general central processing unit (Central Processing Unit,
CPU) or application-specific integrated circuit (Application Specific Integrated Circuit, ASIC), it can be one
It is a or multiple for controlling the integrated circuit of program execution.
Optionally, memory 602 may include high-speed random access memory, can also include nonvolatile storage, example
Such as disk memory, flush memory device or other non-volatile solid state memory parts, the embodiment of the present invention are not construed as limiting.
Optionally, memory 602 is for storing one or more computer programs；When memory 602 store one or
Multiple computer programs by processor 601 execute when so that starter 600 can be realized it is complete in embodiment shown in Fig. 2
Portion or part steps.
The starting method of starter 600 and aforementioned gas turbine shown in Fig. 2 in the present embodiment is based on same structure
Invention under thinking, by the detailed description of the aforementioned starting method to gas turbine, those skilled in the art can be clear
The implementation process for solving starter 600 in the present embodiment, so details are not described herein in order to illustrate the succinct of book.
Under based on the same inventive concept, the embodiment of the invention provides a kind of computer readable storage mediums.Optionally, it counts
Calculation machine readable storage medium storing program for executing has a computer program, and computer program includes program instruction, program instruction when executed by a computer,
The step of making computer execute the starting method of above-mentioned gas turbine.By in this present embodiment computer program and earlier figures 2
Shown in the starting method of gas turbine be that the aforementioned starting method to gas turbine is passed through based on the invention under same design
Detailed description, those skilled in the art can be apparent from the implementation process of computer program in the present embodiment, so in order to
Specification it is succinct, details are not described herein.
Under based on the same inventive concept, the embodiment of the invention provides a kind of computer program product, computer program is produced
Product are stored with computer program, and computer program includes program instruction, program instruction when executed by a computer so that computer
The step of executing the starting method of above-mentioned gas turbine.As in this present embodiment computer program product and earlier figures 2 shown in
The starting method of gas turbine be that the detailed of the aforementioned starting method to gas turbine is passed through based on the invention under same design
Description, those skilled in the art can be apparent from the implementation process of computer program product in the present embodiment, so in order to
Specification it is succinct, details are not described herein.
The present invention be referring to according to the method for the embodiment of the present invention, the process of equipment (system) and computer program product
Figure and/or block diagram describe.It should be understood that every one stream in flowchart and/or the block diagram can be realized by computer program instructions
The combination of process and/or box in journey and/or box and flowchart and/or the block diagram.It can provide these computer programs
Instruct the processor of general purpose computer, special purpose computer, Embedded Processor or other programmable data processing devices to produce
A raw machine, so that being generated by the instruction that computer or the processor of other programmable data processing devices execute for real
The device for the function of being specified in present one or more flows of the flowchart and/or one or more blocks of the block diagram.
These computer program instructions, which may also be stored in, is able to guide computer or other programmable data processing devices with spy
Determine in the computer-readable memory that mode works, so that it includes referring to that instruction stored in the computer readable memory, which generates,
Enable the manufacture of device, the command device realize in one box of one or more flows of the flowchart and/or block diagram or
The function of being specified in multiple boxes.
These computer program instructions also can be loaded onto a computer or other programmable data processing device, so that counting
Series of operation steps are executed on calculation machine or other programmable devices to generate computer implemented processing, thus in computer or
The instruction executed on other programmable devices is provided for realizing in one or more flows of the flowchart and/or block diagram one
The step of function of being specified in a box or multiple boxes.
Obviously, various changes and modifications can be made to the invention without departing from essence of the invention by those skilled in the art
Mind and range.In this way, if these modifications and changes of the present invention belongs to the range of the claims in the present invention and its equivalent technologies
Within, then the present invention is also intended to include these modifications and variations.
Claims (17)
1. a kind of starting method of gas turbine, which is characterized in that be applied to a starter, the starter is for starting
Gas turbine, which comprises
Obtain sample data；Include training data in the sample data, includes that the first input parameter takes in the training data
Value and the first output parameter value；
Using the first input parameter value in the training data as the input parameter value of each kernel function in N number of kernel function,
Each kernel function is calculated, N number of output result is obtained；N is the integer more than or equal to 2；
If the first output result in N number of output result matches with the first output parameter value in the training data,
It then will be with the corresponding first function of the first output result as Boot Model；
The gas turbine starting is controlled based on the Boot Model.
2. the method as described in claim 1, which is characterized in that it further include verify data in the sample data, the verifying
Include the second input parameter value and the second output parameter value in data, the combustion is controlled based on the Boot Model described
Before gas-turbine starting, the method also includes:
Second input parameter value in the verify data is opened described in calculating as the input parameter value of the Boot Model
Movable model, obtain the second output as a result, by it is described second output result and the verify data in the second output parameter value into
Row comparison；
If it is described second output result and the second output parameter value between mean square error in the second preset range,
Determine that the Boot Model is eligible.
3. the method as described in claim 1, which is characterized in that the first output result and the instruction in N number of output result
Practice the first output parameter value in data to match, comprising: the first output in the first output result and the training data
Mean square error between parameter value is within the first preset range.
4. the method as described in claim 1, which is characterized in that include model parameter in the first function；Will with it is described
Before the corresponding first function of first output parameter value is as Boot Model, the method also includes:
A model parameter value is selected from multiple model parameter values of the first function；
Parameter value is inputted as the input parameter value of N number of kernel function using in the training data first, comprising:
Bring the first input parameter value in the training data and the model parameter value selected into N number of core letter respectively
Number, calculates each kernel function, obtains N number of output result.
5. method as claimed in claim 4, which is characterized in that the model parameter includes punishment parameter and kernel functional parameter,
The punishment parameter is the constraint condition of the kernel function, and the kernel functional parameter is the coefficient of the kernel function.
6. method a method as claimed in any one of claims 1 to 5, which is characterized in that the input parameter of the Boot Model includes following ginseng
At least one of number:
Atmospheric temperature, atmospheric pressure, amount of natural gas, starting current of electric.
7. method a method as claimed in any one of claims 1 to 5, which is characterized in that the output parameter of the Boot Model includes following ginseng
At least one of number:
Engine speed, compressor delivery temperature, compressor delivery pressure, combustor exit pressure, turbine-exit temperature, turbine
Outlet pressure.
8. a kind of starter, which is characterized in that for starting gas turbine, the starter includes: the starter
Acquiring unit, for obtaining sample data；Include training data in the sample data, includes the in the training data
One input parameter value and the first output parameter value；
Processing unit, for using the first input parameter value in the training data as each kernel function in N number of kernel function
Parameter value is inputted, each kernel function is calculated, obtains N number of output result；N is the integer more than or equal to 2；
If the first output result in N number of output result matches with the first output parameter value in the training data,
The processing unit is also used to using first function corresponding with the first output result as Boot Model；
The processing unit is also used to control the gas turbine starting based on the Boot Model.
9. starter as claimed in claim 8, which is characterized in that it further include verify data in the sample data, it is described
It include the second input parameter value and the second output parameter value in verify data, the processing unit based on described for being opened
Before movable model controls the gas turbine starting, it is also used to:
Second input parameter value in the verify data is opened described in calculating as the input parameter value of the Boot Model
Movable model, obtain the second output as a result, by it is described second output result and the verify data in the second output parameter value into
Row comparison；
If it is described second output result and the second output parameter value between mean square error in the second preset range,
Determine that the Boot Model is eligible.
10. starter as claimed in claim 8, which is characterized in that the first output result and institute in N number of output result
It states the first output parameter value in training data to match, comprising: first in the first output result and the training data
Mean square error between output parameter value is within the first preset range.
11. starter as claimed in claim 8, which is characterized in that include model parameter in the first function；The place
Manage unit for will first function corresponding with the first output parameter value as Boot Model before, be also used to:
A model parameter value is selected from multiple model parameter values of the first function；
The processing unit is for inputting parameter value as the input of N number of kernel function in the training data first
When parameter value, it is specifically used for:
Bring the first input parameter value in the training data and the model parameter value selected into N number of core letter respectively
Number, calculates each kernel function, obtains N number of output result.
12. starter as claimed in claim 11, which is characterized in that the model parameter includes punishment parameter and kernel function
Parameter, the punishment parameter are the constraint condition of the kernel function, and the kernel functional parameter is the coefficient of the kernel function.
13. the starter as described in claim 8-12 is any, which is characterized in that the input parameter of the Boot Model includes
At least one of following parameter:
Atmospheric temperature, atmospheric pressure, amount of natural gas, starting current of electric.
14. the starter as described in claim 8-12 is any, which is characterized in that the output parameter of the Boot Model includes
At least one of following parameter:
Engine speed, compressor delivery temperature, compressor delivery pressure, combustor exit pressure, turbine-exit temperature, turbine
Outlet pressure.
15. a kind of starter, which is characterized in that including processor and memory；Wherein,
The memory is for storing one or more computer programs；When one or more computers of memory storage
When program is executed by the processor, so that the starter executes method as claimed in claim 1.
16. a kind of computer readable storage medium, which is characterized in that the computer-readable recording medium storage has computer journey
Sequence, the computer program include program instruction, and described program instructs when executed by a computer, execute the computer such as
Any method in claim 1-7.
17. a kind of computer program product, which is characterized in that the computer program product is stored with computer program, described
Computer program includes program instruction, and described program instructs when executed by a computer, executes the computer as right is wanted
Seek any method in 1-7.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910303047.8A CN110080882A (en) | 2019-04-16 | 2019-04-16 | A kind of the starting method and starter of gas turbine |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910303047.8A CN110080882A (en) | 2019-04-16 | 2019-04-16 | A kind of the starting method and starter of gas turbine |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN110080882A true CN110080882A (en) | 2019-08-02 |
Family
ID=67415285
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201910303047.8A Pending CN110080882A (en) | 2019-04-16 | 2019-04-16 | A kind of the starting method and starter of gas turbine |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN110080882A (en) |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111878239A (en) * | 2020-08-06 | 2020-11-03 | 中国人民解放军海军工程大学 | Gas turbine starting control system and method |
| CN112149254A (en) * | 2020-09-24 | 2020-12-29 | 上海电力大学 | Gas turbine starting process prediction method based on Lasso-RF model |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102495939A (en) * | 2011-10-21 | 2012-06-13 | 南京航空航天大学 | SVM solar wing unfolding reliability evaluation method based on kernel optimization |
| US20130097108A1 (en) * | 2011-10-13 | 2013-04-18 | Nec Laboratories America, Inc. | Two-Stage Multiple Kernel Learning Method |
| CN107944173A (en) * | 2017-12-05 | 2018-04-20 | 北京工业大学 | It is a kind of based on selective ensemble least squares support vector machine bioxin hard measurement systems |
| CN108304886A (en) * | 2018-02-27 | 2018-07-20 | 中国民用航空飞行学院 | Aeroengine Performance Monitoring method |
| CN108805217A (en) * | 2018-06-20 | 2018-11-13 | 山东大学 | A kind of health state of lithium ion battery method of estimation and system based on support vector machines |
-
2019
- 2019-04-16 CN CN201910303047.8A patent/CN110080882A/en active Pending
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20130097108A1 (en) * | 2011-10-13 | 2013-04-18 | Nec Laboratories America, Inc. | Two-Stage Multiple Kernel Learning Method |
| CN102495939A (en) * | 2011-10-21 | 2012-06-13 | 南京航空航天大学 | SVM solar wing unfolding reliability evaluation method based on kernel optimization |
| CN107944173A (en) * | 2017-12-05 | 2018-04-20 | 北京工业大学 | It is a kind of based on selective ensemble least squares support vector machine bioxin hard measurement systems |
| CN108304886A (en) * | 2018-02-27 | 2018-07-20 | 中国民用航空飞行学院 | Aeroengine Performance Monitoring method |
| CN108805217A (en) * | 2018-06-20 | 2018-11-13 | 山东大学 | A kind of health state of lithium ion battery method of estimation and system based on support vector machines |
Non-Patent Citations (3)
| Title |
|---|
| 刘建勋等: "某型发动机起动模型的支持向量机辨识及应用", 《推进技术》 * |
| 李辉等: "基于支持向量机的燃气轮机启动过程建模", 《燃气轮机技术》 * |
| 黄冬梅等: "《案例驱动的大数据原理技术及应用》", 30 November 2018, 上海交通大学出版社 * |
Cited By (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111878239A (en) * | 2020-08-06 | 2020-11-03 | 中国人民解放军海军工程大学 | Gas turbine starting control system and method |
| CN111878239B (en) * | 2020-08-06 | 2021-06-22 | 中国人民解放军海军工程大学 | Gas turbine starting control system and method |
| CN112149254A (en) * | 2020-09-24 | 2020-12-29 | 上海电力大学 | Gas turbine starting process prediction method based on Lasso-RF model |
| CN112149254B (en) * | 2020-09-24 | 2023-05-02 | 上海电力大学 | Gas turbine starting process prediction method based on Lasso-RF model |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Camporeale et al. | A modular code for real time dynamic simulation of gas turbines in simulink | |
| EP1882215B1 (en) | Analysis method | |
| EP2177963A1 (en) | Gas turbine model and method for the modeling of a gas turbine | |
| JP2008267382A (en) | Automated engine data diagnostic analysis | |
| MX2012014734A (en) | Turbo-machinery stage families tuning/calibration system and method. | |
| EP2639435A1 (en) | Optimised real-time control of a highly dynamic engine system | |
| CN110080882A (en) | A kind of the starting method and starter of gas turbine | |
| Matveev et al. | The optimization of four-stage low pressure turbine with outlet guide vane | |
| CN110991073B (en) | Simulation method, device and equipment for liquid rocket engine | |
| US20140018973A1 (en) | Systems and methods for liquid fuel modeling | |
| Sankar et al. | Aero-thermodynamic modelling and gas path simulation for a twin spool turbo jet engine | |
| Ito et al. | Development of high level modeling method for rapid modeling process | |
| CN116933693B (en) | Gas turbine performance detection method and device | |
| RU2484441C1 (en) | Gas-turbine engine, test method of gas-turbine engine, production method of batch of gas-turbine engines (versions), and operating method of gas-turbine engine | |
| RU2544638C1 (en) | Gas turbine engine | |
| CN110080886B (en) | Method and device for controlling fuel pressurization system | |
| Rompokos et al. | Application of an advanced adaptation methodology for gas turbine performance monitoring | |
| Kako et al. | Development of High Level Modeling method for rapid modeling process | |
| CN117422017A (en) | Method, device, apparatus and medium for analyzing heat balance of gas turbine | |
| US20220236701A1 (en) | Method for determining a predictive model of a pressure ratio for a dual-flow turbine engine | |
| Biraud et al. | Simulation of the WR-21 Advanced Cycle Engine | |
| Indarti et al. | Integration of Technology Capability for Performance Diagnostics of MS7001EA Using PYTHIA | |
| CN116432336A (en) | Engine performance debugging method and system based on simulation | |
| Hosseini et al. | Model Based Compressor Fault Identification Using Stage Stacking Technique and Nonlinear Diagnostic Systems | |
| Zhang et al. | Steady-state emulation of a diesel engine air charging system through real-time turbocharger modelling |
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 | ||
| RJ01 | Rejection of invention patent application after publication | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20190802 |