CN104834796B - A kind of carburetor parameter design optimization method - Google Patents
A kind of carburetor parameter design optimization method Download PDFInfo
- Publication number
- CN104834796B CN104834796B CN201510279009.5A CN201510279009A CN104834796B CN 104834796 B CN104834796 B CN 104834796B CN 201510279009 A CN201510279009 A CN 201510279009A CN 104834796 B CN104834796 B CN 104834796B
- Authority
- CN
- China
- Prior art keywords
- carburetor
- matrix
- parameter
- pipe diameter
- design
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
Abstract
The present invention relates to a kind of carburetor parameter design optimization methods, based on variance analysis theory and matrix picture theory, decomposition operator in matrix picture theory is mixed into measurement applied to what factorial effect in mixed-level carburetor parameter designing was asked, while giving corresponding calculation method.The optimal selection algorithm can be suitable for carburetor parameter designing mixed-level situation, it meets carburetor test field and selects the requirement of optimal design and pass through the above method, have effectively achieved the optimal selection problem of mixed-level situation orthogonal design, through practice test, it especially can correctly identify carburetor parameter designing table OA (18,3127, 2) and the optimal design of situation, and oneself is successfully applied in the analysis of specific fuel consumption data, substantially increases the validity and reliability of experimental data.
Description
Technical field
The present invention relates to carburetors to test field, joins more particularly, to a kind of carburetor based on mixed-level orthogonal design
Number design optimization method.
Background technique
All kinds of carburetors test field, it is often necessary to using test by the way of come explore specific fuel consumption significantly affect because
Element comprising big throat pipe diameter, middle throat pipe diameter, annular small larynx pipe diameter, air jet diameter, air pressure etc., routine test is past
The experiment number very big toward needs.It is a kind of excellent there is an urgent need to use to the arrangement of experiment due to the limitation of human and material resources and financial resources
The mode of change carries out.Wherein, mixed-level orthogonal design OA (18,3 used by carburetor parameter designing127, 2) be it is a kind of
The design scheme being used widely tends to obtain more reliable data information using less experiment number.Japan
This table is often recommended to carry out parameter designing in quality control analyst field mouthful profound one.
For the regular design part in carburetor parameter designing, can be set using international resolution topology degree
The optimal selection of meter, however, for informal design, especially mixed-level orthogonal design, it is domestic at present on this level
Application study be more deficient.
Summary of the invention
The technical problem to be solved in the present invention is to provide a kind of carburetor test fields that meets to select wanting for optimal design
It seeks and substantially increases the validity and reliability of experimental result while reducing the carburetor parameter design optimization side of specific fuel consumption
Method.
Realize that one of the technical solution of the object of the invention is to provide a kind of optimal selection method of carburetor parameter designing, packet
Include (1) " generating algorithm of each controlling elements matrix picture ", (2) " optimizing word length type algorithm ", (3) " reality of the generation of matrix picture
It is existing " and (4) " realization for optimizing word length type " Four processes;
Wherein (1) " generating algorithm of each controlling elements matrix picture " uses incidence matrix and carrys out calculating matrix picture, can be with
Suitable for carburetor parameter designing;
(2) it " optimizes word length type algorithm " to calculate the measurement that mixes in any t dimension projection, and giving
The broad sense word length type of oily device parameter designing;
(3) " realization of the generation of matrix picture " calculates the matrix picture of each higher-order effect based on each column matrix picture;
(4) realization of word length type " optimize " uses random search techniques, and it is less to test number (TN) when can exhaustion transport
It calculates, obtains absolute optimal design.
Realize a kind of the two of the technical solution of the object of the invention selection method for being to provide carburetor parameter designing, this method
By parameter model, the parameterized model of carburetor is established, utilizes mixed-level carburetor parameter designing OA (18,3127, 2) and
Variance analysis is theoretical, analyzes corresponding data, finally determines that optimal level combines according to Significance factors.
Realize that the two of the technical solution of the object of the invention is to provide a kind of carburetor parameter design optimization method, including matrix
The generation of elephant and optimization word length type two parts, wherein the generation of matrix picture is described below:
3) generation of incidence matrix;
Firstly the need of the incidence matrix for generating each controlling elements in carburetor parameter designing, including big throat pipe diameter, middle larynx
Pipe diameter, annular small larynx pipe diameter, air jet diameter, the incidence matrix of five controlling elements of air pressure;Any s in note design table
Horizontal row a=(a1, a2..., an), wherein a1, a2..., anIt is the element on galois field GF (s), n designs the examination of table for this
Number is tested, then the incidence matrix X of the column is the matrix that an order is n × s, as the horizontal value j-1 of i-th test, then
(i, j) element of incidence matrix is 1, and otherwise value is 0;
4) the matrix picture of carburetor parameter is found out;
By variance analysis theory it is found that the sum of square of deviations of each factor can be expressed as one two in carburetor parameter designing
Secondary type;That is SS2=YTAY, wherein Y=(y1, y2..., yn) it is data vector, the square as corresponding to any one quadratic form
Battle array A is an orthogonal intersection cast shadow matrix, and A can use the incidence matrix calculated in previous step and obtain, its calculation formula is:
A=XT(XTX)-1X-Pn,Wherein 1n1 n-dimensional vector is all for element;
Mix measurement in t dimension projection due to needing to calculate, we need to calculate the matrix picture of each higher-order factor analysis effect, and note is appointed
The matrix picture of meaning column is Aj, M=(j1,..., jm) it is an index set, the then specific table of the corresponding matrix picture of m rank interaction
It is shown as
Wherein, | M | indicate the element number in index set M,The Kronecker product of representing matrix;
3) mix index between calculating carburetor parameter effects;
By matrix picture theory it is recognised that the orthogonality between carburetor test parameters is equivalent to the orthogonality between matrix picture,
Mix measurement using between following indexs calculating effect;
According to projection matrixProperty, can construct and mix Measure Indexes α (u1, u2..., um), it should
Index describes any column u1, u2..., umBetween mix metric, as α (u1, u2..., um) be 0 when, illustrate this
u1, u2..., umBetween completely it is orthogonal, mix without any information, as α (u1, u2..., um) be greater than 0 when, illustrate u1,
u2..., umBetween there are information to mix;
Optimize being described below for word length type: using the method for t dimension projection, specific decoding step are as follows: first according to each rank
Mixing between carburetor parameter effects mixes measurement in Measure Indexes calculating t dimension projection;Secondly by the hybrid UV curing in t dimension projection
Amount calculates broad sense word length type;Finally by the sequence to broad sense word length type, carburetor parameter optimal design is obtained.
Further, the detailed process for optimizing word length type using the method for t dimension projection is as follows:
1) mix measurement in any t dimension projection;
Because of the design for n row p column, according to low order effect principle, each factor is in low-dimensional in carburetor parameter designing
Mixing on hybrid ratio higher dimensional space spatially is more important, so the value that mixes in each t dimension projection is calculated, t=2,
3 ... p. and that t column are chosen in p column is sharedKind situation, so total mixes metric are as follows:
u1Represent big throat pipe diameter, u2Throat pipe diameter in representative, u3Represent annular small larynx pipe diameter, u4Represent air jet
Diameter, u5Air pressure is represented, then α (u1, u2, u3) indicate big throat pipe diameter, middle throat pipe diameter and annular small larynx pipe diameter three's effect
Between mix value;
2) mix the sequence of index in the projection of t dimension;
Mix the situation that mixes measured and described between all main effects in 2 dimension projections, mixes measurement in 3 dimension projections
Describe the situation that mixes of all main effects and second order interaction, mixing measurement while describing all masters in 4 dimension projections
The situation that mixes of effect and three rank interactions mixed between situation and all second order interactions, higher order, which mixes, to be contained
Justice and so on, write as expression formula are as follows:
...,
SAp=α (u1, u2..., up),
SA2Mix situation describing following 10 betweens of parameters, they are:
Throat pipe diameter in big throat pipe diameter vs representative,
Big throat pipe diameter vs annular small larynx pipe diameter,
Big throat pipe diameter vs air jet diameter,
Big throat pipe diameter vs air pressure,
Middle throat pipe diameter vs annular small larynx pipe diameter,
Middle throat pipe diameter vs air jet diameter,
Middle throat pipe diameter vs air pressure,
Annular small larynx pipe diameter vs air jet diameter,
Annular small larynx pipe diameter vs air pressure,
Air jet diameter vs air pressure,
And make the broad sense word length type of carburetor parameter designing,
SA=(SA2, SA3..., SAp);
3) selection of carburetor parameter optimal design;
The broad sense word length type of carburetor parameter designing is compared one by one according to succession, if a design
D1Broad sense word length type, meet claimed below, be then a carburetor parameter optimal design;
D is designed for other same parameters2, there are a real number t0, so thatAnd SAt(D1)
=SAt(D2), t≤t0。
Realize that the three of the technical solution of the object of the invention is to provide a kind of carburetor Parameters design, including the life of matrix picture
At realization and optimal design selection realization two parts:
One, the realization that matrix picture generates;
Corresponding mixed-level carburetor parameter designing is read first, and each carburetor parameter is then calculated according to respective formula
Matrix picture, the matrix picture of each high-order control factorial effect and the hybrid UV curing of carburetor parameter index set M intrinsic factor effect
Magnitude;
Two, the realization of word length type is optimized;
1) word length type: word length type SA=(SA is sought2, SA3..., SAp) it is the key that find carburetor parameter optimal design,
It, which needs to calculate, mixes metric SA in each t dimension projectiont;
2) it seeks the optimal word length type of carburetor parameter designing: for given design parameter n, p, it is all to calculate itKind feelings
Condition, and more corresponding word length type one by one, finally take word length type optimal.
It realizes that the four of the technical solution of the object of the invention is to provide a kind of carburetor Parameters design, including walks as follows
It is rapid:
1. establishing the parameterized model of carburetor design: by the analysis to carburetor influence factor, in order to be subsequently generated
The incidence matrix of each controlling elements in carburetor parameter designing, confirmation selection are small including big throat pipe diameter, middle throat pipe diameter, annular
Five throat pipe diameter, air jet diameter and air pressure controlling elements;
2. establishing the incidence matrix that carburetor designs each parameter: arranging a=(a in note design table1, a2..., an), wherein a1,
a2..., anIt is the element on galois field GF (s), n designs the test number (TN) of table for this, then the incidence matrix X of the column is one
Order is the matrix of n × s, and as the horizontal value j-1 of i-th test, then (i, j) element of incidence matrix is 1, is otherwise taken
Value is 0;
3. find out the matrix of carburetor parameter as: it is theoretical by variance analysis it is found that each factor in carburetor parameter designing
Sum of square of deviations can be expressed as a quadratic form;That is SS2=YTAY, wherein Y=(y1, y2..., yn) it is data vector, by
For matrix theory it is found that matrix A corresponding to quadratic form is an orthogonal intersection cast shadow matrix, A can use the pass calculated in previous step
Join matrix to obtain, its calculation formula is:
A=XT(XTX)-1 X-Pn,
Wherein A representing matrix picture, X be each incidence matrix obtained in the previous step, 1n1 n-dimensional vector is all for element;
Then the matrix picture for calculating each higher-order factor analysis effect, its calculation formula is
AjFor matrix picture, M=(j1,..., jm) it is an index set, | M | indicate the element number in index set M,
The Kronecker product of representing matrix;
By matrix picture theory it is recognised that the orthogonality between carburetor test parameters is equivalent to the orthogonality between matrix picture,
Mix measurement using between following indexs calculating effect;
According to projection matrixProperty, as α (u1, u2..., um) be 0 when, illustrate the u1,
u2..., umBetween completely it is orthogonal, mix without any information, as α (u1, u2..., um) be greater than 0 when, illustrate u1, u2..., um
Between there are information to mix;According to low order effect principle, each factor is mixed on lower dimensional space in carburetor parameter designing
It is miscellaneous more important than mixing on higher dimensional space, mix value in each t dimension projection so calculating, so always mix metric
Are as follows:
u1Represent big throat pipe diameter, u2Throat pipe diameter in representative, u3Represent annular small larynx pipe diameter, u4Represent air jet
Diameter, u5Air pressure is represented, such as α (u1, u2, u3) indicate big throat pipe diameter, middle throat pipe diameter and annular small larynx pipe diameter three's effect
Between mix value;
4. the selection of carburetor parameter designing:
Mix the situation that mixes measured and described between all main effects in 2 dimension projections, mixes measurement in 3 dimension projections
Describe the situation that mixes of all main effects and second order interaction, higher order mixes meaning, is write as expression formula are as follows:
…….
SAp=α (u1, u2..., up),
For example, SA2Mix situation describing following 10 betweens of parameters, they are:
Throat pipe diameter in big throat pipe diameter vs representative,
Big throat pipe diameter vs annular small larynx pipe diameter,
Big throat pipe diameter vs air jet diameter,
Big throat pipe diameter vs air pressure,
Middle throat pipe diameter vs annular small larynx pipe diameter,
Middle throat pipe diameter vs air jet diameter,
Middle throat pipe diameter vs air pressure,
Annular small larynx pipe diameter vs air jet diameter,
Annular small larynx pipe diameter vs air pressure,
Air jet diameter vs air pressure,
And make the broad sense word length type of carburetor parameter designing,
SA=(SA2, SA3..., SAp);
5. obtaining the parameter combination of carburetor design by observing selection by the result of step 4..
The present invention has the effect of positive: (1) optimal selection that the present invention has effectively achieved carburetor experimental design is asked
Topic, through practice test, can correctly identify the optimal design of common situation, and be successfully applied to carburetor parameter designing
Data analysis in, substantially increase the validity and reliability of specific fuel consumption data.The ratio oil of carburetor test is reduced simultaneously
Consumption.
(2) parameter model of specific fuel consumption of the invention is simple and convenient, using five main controlling elements: big throat pipe diameter,
Middle throat pipe diameter, annular small larynx pipe diameter, air jet diameter, air pressure.Parameter model use main effects model, avoid it is complete because
Deficiency of the sub- interaction models in data analysis.
(3) the mixed-level carburetor parameter designing table OA that the present invention uses carburetor design using matrix picture theory
(18,3127, 2) and optimal selection has been carried out, the design watch test number is less, and can obtain and test almost the same knot completely
By can satisfy the actual needs of engineering.
(4) present invention is less using obtained optimal control parameter number, and chooses to the level of other non-significant factors
Range is relatively broad, i.e., the Parameter Conditions that the present invention obtains are more steady.Also it is easy to be verified with other design tables
Test.
Detailed description of the invention
Fig. 1 is that the matrix picture of carburetor parameter designing generates implementation flow chart;
Fig. 2 is the optimal word length type implementation flow chart of carburetor parameter designing;
Fig. 3 is carburetor experimental design table OA (18,3127, 2);
Fig. 4 is carburetor experimental design table OA (18,3127, 2) optimal selection result.
Specific embodiment
(embodiment 1)
A kind of carburetor parameter design optimization method of the present embodiment establishes the parametrization mould of carburetor by parameter model
Type is shown in Fig. 3, utilizes mixed-level carburetor parameter designing OA (18,3127, 2) and variance analysis theory, corresponding data is carried out
Analysis finally determines that optimal level combines according to Significance factors.It is compared with existing optimum condition, to determine whether reality
Now reduce the purpose of specific fuel consumption, existing optimum is 218.1.Specifically comprise the following steps:
1. establishing the parameterized model of carburetor design.By the analysis to carburetor influence factor, the shadow that has found that it is likely that
It rings because usually including big throat pipe diameter, middle throat pipe diameter, annular small larynx pipe diameter, air jet diameter and air pressure five, in order to
It is subsequently generated the incidence matrix of each controlling elements in carburetor parameter designing, confirmation selection is straight including big throat pipe diameter, middle trunnion
Diameter, five annular small larynx pipe diameter, air jet diameter and air pressure controlling elements, it includes four 3 horizontal factors and 2 water
The flat factor, design parameter see the table below 1 in the present embodiment:
Table 1
Parameter | Level 1 | Level 2 | Level 3 |
Big throat pipe diameter (Φ) | 32 | 34 | 36 |
Middle throat pipe diameter (Φ) | 22 | 21 | 20 |
Annular small larynx pipe diameter (Φ) | 10 | 9 | 8 |
Air jet diameter (Φ) | 1.2 | 1.0 | 0.8 |
Air pressure | It is high | It is low |
2. establishing the incidence matrix that carburetor designs each parameter: arranging a=(a in note design table1, a2..., an), wherein a1,
a2..., anIt is the element on galois field GF (s), n designs the test number (TN) of table for this, then the incidence matrix X of the column is one
Order is the matrix of n × s, and as the horizontal value j-1 of i-th test, then (i, j) element of incidence matrix is 1, is otherwise taken
Value is 0;Using each test in the present embodiment:
A. the design behavior of big throat pipe diameter:
1 | 2 | 3 | 1 | 2 | 3 | 1 | 2 | 3 | 1 | 2 | 3 | 1 | 2 | 3 | 1 | 2 | 3 |
According to the generation of carburetor design relationship matrix, its incidence matrix can be obtained are as follows:
1 | 0 | 0 | 1 | 0 | 0 | 1 | 0 | 0 | 1 | 0 | 0 | 1 | 0 | 0 | 1 | 0 | 0 |
0 | 1 | 0 | 0 | 1 | 0 | 0 | 1 | 0 | 0 | 1 | 0 | 0 | 1 | 0 | 0 | 1 | 0 |
0 | 0 | 1 | 0 | 0 | 1 | 0 | 0 | 1 | 0 | 0 | 1 | 0 | 0 | 1 | 0 | 0 | 1 |
B. the design behavior of middle throat pipe diameter:
1 | 2 | 3 | 1 | 2 | 3 | 2 | 3 | 1 | 3 | 1 | 2 | 2 | 3 | 1 | 3 | 1 | 2 |
According to the generation of carburetor design relationship matrix, its incidence matrix can be obtained are as follows:
1 | 0 | 0 | 1 | 0 | 0 | 0 | 0 | 1 | 0 | 1 | 0 | 0 | 0 | 1 | 0 | 1 | 0 |
0 | 1 | 0 | 0 | 1 | 0 | 1 | 0 | 0 | 0 | 0 | 1 | 1 | 0 | 0 | 0 | 0 | 1 |
0 | 0 | 1 | 0 | 0 | 1 | 0 | 1 | 0 | 1 | 0 | 0 | 0 | 1 | 0 | 1 | 0 | 0 |
C. the design behavior of small larynx pipe diameter:
1 | 2 | 3 | 2 | 3 | 1 | 1 | 2 | 3 | 3 | 1 | 2 | 3 | 1 | 2 | 2 | 3 | 1 |
According to the generation of carburetor design relationship matrix, its incidence matrix can be obtained are as follows:
1 | 0 | 0 | 0 | 0 | 1 | 1 | 0 | 0 | 0 | 1 | 0 | 0 | 1 | 0 | 0 | 0 | 1 |
0 | 1 | 0 | 1 | 0 | 0 | 0 | 1 | 0 | 0 | 0 | 1 | 0 | 0 | 1 | 1 | 0 | 0 |
0 | 0 | 1 | 0 | 1 | 0 | 0 | 0 | 1 | 1 | 0 | 0 | 1 | 0 | 0 | 0 | 1 | 0 |
D. the design behavior of air jet diameter:
1 | 2 | 3 | 2 | 3 | 1 | 3 | 1 | 2 | 2 | 3 | 1 | 1 | 2 | 3 | 3 | 1 | 2 |
According to the generation of carburetor design relationship matrix, its incidence matrix can be obtained are as follows:
1 | 0 | 0 | 0 | 0 | 1 | 0 | 1 | 0 | 0 | 0 | 1 | 1 | 0 | 0 | 0 | 1 | 0 |
0 | 1 | 0 | 1 | 0 | 0 | 0 | 0 | 1 | 1 | 0 | 0 | 0 | 1 | 0 | 0 | 0 | 1 |
0 | 0 | 1 | 0 | 1 | 0 | 1 | 0 | 0 | 0 | 1 | 0 | 0 | 0 | 1 | 1 | 0 | 0 |
E. the design behavior of air pressure:
1 | 1 | 1 | 1 | 1 | 1 | 1 | 1 | 1 | 2 | 2 | 2 | 2 | 2 | 2 | 2 | 2 | 2 |
According to the generation of carburetor design relationship matrix, its incidence matrix can be obtained are as follows:
1 | 1 | 1 | 1 | 1 | 1 | 1 | 1 | 1 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 |
0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 1 | 1 | 1 | 1 | 1 | 1 | 1 | 1 | 1 |
3. find out the matrix of carburetor parameter as: it is theoretical by variance analysis it is found that each factor in carburetor parameter designing
Sum of square of deviations can be expressed as a quadratic form.That is SS2=YTAY, wherein Y=(y1, y2..., yn) it is data vector, by
For matrix theory it is found that matrix A corresponding to quadratic form is an orthogonal intersection cast shadow matrix, A can use the pass calculated in previous step
Join matrix to obtain, its calculation formula is:
A=XT(XTX)-1X-Pn,
Wherein A representing matrix picture, X be each incidence matrix obtained in the previous step, 1n1 n-dimensional vector is all for element.
Then the matrix picture for calculating each higher-order factor analysis effect, its calculation formula is
AjFor matrix picture, M=(j1,..., jm) it is an index set, | M | indicate the element number in index set M,Table
Show existence and unigueness.
By matrix picture theory it is recognised that the orthogonality that carburetor test parameters is asked is equivalent to the orthogonality that matrix picture is asked,
Mix measurement using what following indexs calculating effect was asked.
According to projection matrixProperty, as α (u1, u2..., um) be 0 when, illustrate the u1,
u2..., umBetween completely it is orthogonal, mix without any information, as α (u1, u2..., um) be greater than 0 when, illustrate u1, u2..., um
Between there are information to mix.According to low order effect principle, each factor is mixed on lower dimensional space in carburetor parameter designing
It is miscellaneous more important than mixing on higher dimensional space, mix value in each t dimension projection so calculating, so always mix metric
Are as follows:
In the present embodiment, u1Represent big throat pipe diameter, u2Throat pipe diameter in representative, u3Represent annular small larynx pipe diameter, u4Generation
Table air jet diameter, u5Air pressure is represented, such as α (u1, u2, u3) indicating big throat pipe diameter, middle throat pipe diameter and annular secondary venturi are straight
Mix value between diameter three's effect.
4. the selection of carburetor parameter designing:
Mix the situation that mixes measured and described between all main effects in 2 dimension projections, mixes measurement in 3 dimension projections
Describe the situation that mixes of all main effects and second order interaction, higher order mixes meaning, is write as expression formula are as follows:
…….
SAp=α (u1, u2..., up),
For example, SA2Mix situation describing following 10 betweens of parameters, they are:
Throat pipe diameter in big throat pipe diameter vs representative,
Big throat pipe diameter vs annular small larynx pipe diameter,
Big throat pipe diameter vs air jet diameter,
Big throat pipe diameter vs air pressure,
Middle throat pipe diameter vs annular small larynx pipe diameter,
Middle throat pipe diameter vs air jet diameter,
Middle throat pipe diameter vs air pressure,
Annular small larynx pipe diameter vs air jet diameter,
Annular small larynx pipe diameter vs air pressure,
Air jet diameter vs air pressure,
And make the broad sense word length type of carburetor parameter designing,
SA=(SA2, SA3..., SAp)。
In the present embodiment, to carburetor parameter designing table OA (18,3127, 2) carry out five parameter factors optimal selection,
Obtain optimal word length type are as follows: (7,45.83,3.75), corresponding design is classified as (1,3,4,5,8), i.e., such as the data group of the following table 2
It closes:
Table 2
Serial number | Big trunnion | Middle trunnion | Annular secondary venturi | Air jet | Air pressure | Specific fuel consumption |
1 | 32 | 22 | 10 | 1.2 | It is high | 241.0 |
2 | 34 | 21 | 9 | 1.0 | It is high | 231.2 |
3 | 36 | 20 | 8 | 0.8 | It is high | 234.6 |
4 | 32 | 22 | 9 | 1.0 | It is high | 215.4 |
5 | 34 | 21 | 8 | 0.8 | It is high | 211.3 |
6 | 36 | 20 | 10 | 1.2 | It is high | 311.3 |
7 | 32 | 21 | 10 | 0.8 | It is high | 256.4 |
8 | 34 | 20 | 9 | 1.2 | It is high | 229.7 |
9 | 36 | 22 | 8 | 1.0 | It is high | 240.1 |
10 | 32 | 20 | 8 | 1.0 | It is low | 203.1 |
11 | 34 | 22 | 10 | 0.8 | It is low | 266.8 |
12 | 36 | 21 | 9 | 1.2 | It is low | 241.2 |
13 | 32 | 21 | 8 | 1.2 | It is low | 205.5 |
14 | 34 | 20 | 10 | 1.0 | It is low | 255.6 |
15 | 36 | 22 | 9 | 0.8 | It is low | 233.1 |
16 | 32 | 20 | 9 | 0.8 | It is low | 211.8 |
17 | 34 | 22 | 8 | 1.2 | It is low | 203.1 |
18 | 36 | 21 | 10 | 1.0 | It is low | 325.8 |
The specific fuel consumption data of table 2 are calculated, the range analysis table of carburetor test is obtained, see the table below 3:
Table 3
WhereinRespectively indicate the average value of the specific fuel consumption test data under each level.From above-mentioned carburetor
Range analysis table is tested as it can be seen that the influence of annular small larynx pipe diameter is maximum, the influence of big throat pipe diameter is taken second place, further the side of passing through
Difference analyses conspicuousness, examines the conspicuousness of annular secondary venturi diameter and big throat pipe diameter, see the table below 4:
Table 4
Carburetor tests analysis of variance table
5. by table 4, that carburetor test analysis of variance table has obtained big throat pipe diameter, middle throat pipe diameter, annular secondary venturi is straight
The significance of five diameter, air jet diameter and air pressure influence factors, inspection show annular small larynx pipe diameter and big trunnion
Influence of the diameter to specific fuel consumption is significant.It, can be straight using big trunnion again from the specific fuel consumption test data mean value in table 3
The 1 of the horizontal and annular small larynx pipe diameter in the 1 of diameter is horizontal to carry out verification test, the parameter group of the following carburetor design of final choice
It closes.It see the table below 5:
Table 5
10 tests of verifying are carried out to the Optimal Parameters combination of carburetor design, see the table below 6:
Table 6
The average value of its flat specific fuel consumption test data is 188.84, achieves satisfied effect.It is more best than existing
As a result (218.1) reduce 13.42%.
By the software compilation method of above procedure, the optimal selection problem of carburetor experimental design is had effectively achieved,
Through practice test, the optimal design of common situation can be correctly identified, concrete outcome is shown in Fig. 4 and table 2. and has successfully applied
In the data analysis of carburetor parameter designing, the validity and reliability of specific fuel consumption data is substantially increased.It reduces simultaneously
The specific fuel consumption of carburetor test.
(embodiment 2)
A kind of optimal selection method of the carburetor parameter designing of the present embodiment, including (1) " each controlling elements matrix as
Generating algorithm ", (2) " optimizing word length type algorithm ", (3) " realization of the generation of matrix picture " and (4) " optimize the reality of word length type
It is existing " Four processes;
Wherein (1) " generating algorithm of each controlling elements matrix picture " uses incidence matrix and carrys out calculating matrix picture, can be with
Suitable for carburetor parameter designing;
(2) it " optimizes word length type algorithm " to calculate the measurement that mixes in any t dimension projection, and gives carburetion
The broad sense word length type of device parameter designing;
(3) " realization of the generation of matrix picture " calculates the matrix picture of each higher-order effect based on each column matrix picture;
(4) realization of word length type " optimize " uses random search techniques, and it is less to test number (TN) when can exhaustion transport
It calculates, obtains absolute optimal design.
(embodiment 3)
A kind of carburetor parameter design optimization method, generation and optimization word length type two parts including Projection matrix, wherein
The generation of matrix picture is described below:
1) generation of incidence matrix;
Firstly the need of the incidence matrix for generating each controlling elements in carburetor parameter designing, including big throat pipe diameter, middle larynx
Pipe diameter, annular small larynx pipe diameter, air jet diameter, the incidence matrix of five controlling elements of air pressure;Any s in note design table
Horizontal row a=(a1, a2..., an), wherein a1, a2..., anIt is the element on galois field GF (s), n designs the examination of table for this
Number is tested, then the incidence matrix X of the column is the matrix that an order is n × s, as the horizontal value j-1 of i-th test, then
(i, j) element of incidence matrix is 1, and otherwise value is 0;
2) the matrix picture of carburetor parameter is found out;
By variance analysis theory it is found that the sum of square of deviations of each factor can be expressed as one two in carburetor parameter designing
Secondary type;That is SS2=YTAY, wherein Y=(y1, y2..., yn) it is data vector, the square as corresponding to any one quadratic form
Battle array A is an orthogonal intersection cast shadow matrix, and A can use the incidence matrix calculated in previous step and obtain, its calculation formula is:
A=XT(XTX)-1X-Pn,Wherein 1n1 n-dimensional vector is all for element.
Mix measurement in t dimension projection due to needing to calculate, we need to calculate the matrix picture of each higher-order factor analysis effect, and note is appointed
The matrix picture of meaning column is Aj, M=(j1,..., jm) it is an index set, the then specific table of the corresponding matrix picture of m rank interaction
It is shown as
Wherein, | M | indicate the element number in index set M,The Kronecker product of representing matrix;
3) mix index between calculating carburetor parameter effects;
By matrix picture theory it is recognised that the orthogonality between carburetor test parameters is equivalent to the orthogonality between matrix picture,
Mix measurement using between following indexs calculating effect.
According to projection matrixProperty, can construct and mix Measure Indexes α (u1, u2..., um), it should
Index describes any column u1, u2..., umBetween mix metric, as α (u1, u2..., um) be 0 when, illustrate this
u1, u2..., umBetween completely it is orthogonal, mix without any information, as α (u1, u2..., um) be greater than 0 when, illustrate u1,
u2..., umBetween there are information to mix.
Optimize being described below for word length type: using the method for t dimension projection, specific decoding step are as follows: first according to each rank
Mixing between carburetor parameter effects mixes measurement in Measure Indexes calculating t dimension projection;Secondly by the hybrid UV curing in t dimension projection
Amount calculates broad sense word length type;Finally by the sequence to broad sense word length type, carburetor parameter optimal design is obtained.
The detailed process that word length type is optimized using the method for t dimension projection is as follows:
1) mix measurement in any t dimension projection;
Because of the design for n row p column, according to low order effect principle, each factor is in low-dimensional in carburetor parameter designing
Mixing on hybrid ratio higher dimensional space spatially is more important, so the value that mixes in each t dimension projection is calculated, t=2,
3 ... p. and that t column are chosen in p column is sharedKind situation, so total mixes metric are as follows:
u1Represent big throat pipe diameter, u2Throat pipe diameter in representative, u3Represent annular small larynx pipe diameter, u4Represent air jet
Diameter, u5Air pressure is represented, then α (u1, u2, u3) indicate big throat pipe diameter, middle throat pipe diameter and annular small larynx pipe diameter three effect
Mix value between answering;
2) mix the sequence of index in the projection of t dimension;
Mix the situation that mixes measured and described between all main effects in 2 dimension projections, mixes measurement in 3 dimension projections
Describe the situation that mixes of all main effects and second order interaction, mixing measurement while describing all masters in 4 dimension projections
The situation that mixes of effect and three rank interactions mixed between situation and all second order interactions, higher order, which mixes, to be contained
Justice and so on, write as expression formula are as follows:
...,
SAp=α (u1, u2..., up),
SA2Mix situation describing following 10 betweens of parameters, they are:
Throat pipe diameter in big throat pipe diameter vs representative,
Big throat pipe diameter vs annular small larynx pipe diameter,
Big throat pipe diameter vs air jet diameter,
Big throat pipe diameter vs air pressure,
Middle throat pipe diameter vs annular small larynx pipe diameter,
Middle throat pipe diameter vs air jet diameter,
Middle throat pipe diameter vs air pressure,
Annular small larynx pipe diameter vs air jet diameter,
Annular small larynx pipe diameter vs air pressure,
Air jet diameter vs air pressure,
And make the broad sense word length type of carburetor parameter designing,
SA=(SA2, SA3..., SAp);
3) selection of carburetor parameter optimal design;
The broad sense word length type of carburetor parameter designing is compared one by one according to succession, if a design
D1Broad sense word length type, meet claimed below, be then a carburetor parameter optimal design;
D is designed for other same parameters2, there are a real number t0, so thatAnd SAt(D1)
=SAt(D2), t≤t0。
(embodiment 4)
Realization two of a kind of carburetor Parameters design, the realization generated including matrix picture and optimal design selection
Point:
One, the realization that matrix picture generates;
Corresponding mixed-level carburetor parameter designing is read first, and each carburetor parameter is then calculated according to respective formula
Matrix picture, the matrix picture of each high-order control factorial effect and the hybrid UV curing of carburetor parameter index set M intrinsic factor effect
Magnitude;See Fig. 1.
Two, the realization of word length type is optimized;
1) word length type: word length type SA=(SA is sought2, SA3..., SAp) it is the key that find carburetor parameter optimal design,
It, which needs to calculate, mixes metric SA in each t dimension projectiont;
2) it seeks the optimal word length type of carburetor parameter designing: for given design parameter n, p, it is all to calculate itKind feelings
Condition, and more corresponding word length type one by one, finally take word length type optimal.See Fig. 2.
Obviously, the above embodiment is merely an example for clearly illustrating the present invention, and is not to of the invention
The restriction of embodiment.For those of ordinary skill in the art, it can also be made on the basis of the above description
Its various forms of variation or variation.There is no necessity and possibility to exhaust all the enbodiments.And these belong to this hair
The obvious changes or variations that bright spirit is extended out are still in the protection scope of this invention.
Claims (1)
1. a kind of carburetor parameter design optimization method, it is characterised in that generation and optimization two, word length type including Projection matrix
Point, wherein the generation of matrix picture is described below:
1) generation of incidence matrix;
It is straight firstly the need of the incidence matrix for generating each controlling elements in carburetor parameter designing, including big throat pipe diameter, middle trunnion
Diameter, annular small larynx pipe diameter, air jet diameter, the incidence matrix of five controlling elements of air pressure;Any s is horizontal in note design table
Arrange a=(a1,a2,...,an), wherein a1,a2,...,anIt is the element on galois field GF (s), n designs the test time of table for this
It counts, then horizontal row a=(a1,a2,...,an) incidence matrix X be an order be n × s matrix, when i-th test water
When flat value j-1, then (i, j) element of incidence matrix is 1, and otherwise value is 0;
2) the matrix picture of carburetor parameter is found out;
It is theoretical by variance analysis it is found that in carburetor parameter designing the sum of square of deviations of each factor can be expressed as one it is secondary
Type;That is SS2=YTAY, wherein Y=(y1,y2,...,yn) it is data vector, the matrix A as corresponding to any one quadratic form
It is an orthogonal intersection cast shadow matrix, A can use the incidence matrix calculated in previous step and obtain, its calculation formula is:
A=XT(XTX)-1X-Pn,Wherein 1n1 n-dimensional vector is all for element;
It calculates and mixes measurement in t dimension projection, calculate the matrix picture of each higher-order factor analysis effect, remember that the matrix picture of any column is Aj, M
=(j1,,...,jm) it is an index set, then m rank interaction corresponding matrix picture is embodied as
Wherein, | M | indicate the element number in index set M,The Kronecker product of representing matrix;
3) mix index between calculating carburetor parameter effects;
By matrix picture theory it is recognised that the orthogonality between carburetor test parameters is equivalent to the orthogonality between matrix picture, use
Mix measurement between following indexs calculating effects;
According to projection matrixProperty, can construct and mix Measure Indexes α (u1,u2,...,um), the index
Describe any column u1,u2,...,umBetween mix metric, as α (u1,u2,...,um) be 0 when, illustrate the u1,
u2,...,umBetween completely it is orthogonal, mix without any information, as α (u1,u2,...,um) be greater than 0 when, illustrate u1,u2,...,um
Between there are information to mix;
Optimize being described below for word length type: using the method for t dimension projection, specific decoding step are as follows: first according to each rank carburetion
Mixing between device parameter effects mixes measurement in Measure Indexes calculating t dimension projection;Secondly by the hybrid UV curing meter in t dimension projection
Calculate broad sense word length type;Finally by the sequence to broad sense word length type, carburetor parameter optimal design is obtained.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201510279009.5A CN104834796B (en) | 2015-05-27 | 2015-05-27 | A kind of carburetor parameter design optimization method |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201510279009.5A CN104834796B (en) | 2015-05-27 | 2015-05-27 | A kind of carburetor parameter design optimization method |
Publications (2)
Publication Number | Publication Date |
---|---|
CN104834796A CN104834796A (en) | 2015-08-12 |
CN104834796B true CN104834796B (en) | 2019-08-23 |
Family
ID=53812680
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201510279009.5A Active CN104834796B (en) | 2015-05-27 | 2015-05-27 | A kind of carburetor parameter design optimization method |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN104834796B (en) |
Citations (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN101676544A (en) * | 2008-09-19 | 2010-03-24 | 株式会社京浜 | Carburetor |
CN102192047A (en) * | 2010-03-08 | 2011-09-21 | 布里格斯斯特拉顿公司 | Carburetor including one-piece fuel metering insert |
-
2015
- 2015-05-27 CN CN201510279009.5A patent/CN104834796B/en active Active
Patent Citations (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN101676544A (en) * | 2008-09-19 | 2010-03-24 | 株式会社京浜 | Carburetor |
CN102192047A (en) * | 2010-03-08 | 2011-09-21 | 布里格斯斯特拉顿公司 | Carburetor including one-piece fuel metering insert |
Non-Patent Citations (1)
Title |
---|
用正交试验法探索化油器最佳生产条件;张明华;《内燃机》;20100415(第2期);第1-5页 * |
Also Published As
Publication number | Publication date |
---|---|
CN104834796A (en) | 2015-08-12 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN103353923B (en) | Adaptive space interpolation method and system thereof based on space characteristics analysis | |
Cleary et al. | A generalised multiple mapping conditioning approach for turbulent combustion | |
CN105574191B (en) | Online community network multi-source point information source tracing system and method | |
Bellio et al. | A pairwise likelihood approach to generalized linear models with crossed random effects | |
CN104699595B (en) | A kind of method for testing software of software-oriented upgrading | |
Zhang et al. | Observability of Boolean control networks: A unified approach based on the theories of finite automata and formal languages | |
Dalbey et al. | Fast generation of space-filling latin hypercube sample designs | |
CN104834796B (en) | A kind of carburetor parameter design optimization method | |
Joyce et al. | Determining the distribution of fitness effects using a generalized Beta-Burr distribution | |
KR100957066B1 (en) | Method of nuclear reactor core analysis using characteristic of porous media | |
Zhou et al. | An new immune genetic algorithm based on uniform design sampling | |
CN103793582A (en) | Optimization method for cylindrical shell large opening connecting pipe structure | |
Wu et al. | Do environmental regulation and foreign direct investment drive regional air pollution in China? | |
CN115687854A (en) | High-precision soil sample parameter measuring method and system thereof | |
Zhang et al. | Constrained ordination analysis in the presence of zero inflation | |
Rouchier et al. | When overconfident agents slow down collective learning | |
CN108268982A (en) | A kind of extensive active power distribution network decomposition strategy evaluation method and device | |
Helluy et al. | OpenCL numerical simulations of two-fluid compressible flows with a 2D random choice method | |
Lachance et al. | Characterization of the water quality in the Saint Lawrence River: determination of homogeneous zones by correspondence analysis | |
Iwai | Family changes and family values in Asian societies: Exploring similarities and differences based on EASS 2006/2016 and CAFS | |
Zheng et al. | Mining the hidden link structure from distribution flows for a spatial social network | |
Feng et al. | Sharing parameter by conjugation for knowledge graph embeddings in complex space | |
Zirwes et al. | Optimizing load balancing of reacting flow solvers in OpenFOAM for high performance computing | |
Nagel et al. | Studying first passage problems using neural networks: A case study in the slit-well microfluidic device | |
Zhao et al. | Spatial distribution and influencing factors analysis of national key rural tourism villages in the Yangtze River Delta region based on geographically weighted regression |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
C06 | Publication | ||
PB01 | Publication | ||
EXSB | Decision made by sipo to initiate substantive examination | ||
SE01 | Entry into force of request for substantive examination | ||
GR01 | Patent grant | ||
GR01 | Patent grant |