CN110555470B - Oil sprayer grouping method - Google Patents
Oil sprayer grouping method Download PDFInfo
- Publication number
- CN110555470B CN110555470B CN201910765326.6A CN201910765326A CN110555470B CN 110555470 B CN110555470 B CN 110555470B CN 201910765326 A CN201910765326 A CN 201910765326A CN 110555470 B CN110555470 B CN 110555470B
- Authority
- CN
- China
- Prior art keywords
- oil
- fuel
- injectors
- fuel injector
- injection quantity
- 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
Images
Classifications
-
- F—MECHANICAL ENGINEERING; LIGHTING; HEATING; WEAPONS; BLASTING
- F02—COMBUSTION ENGINES; HOT-GAS OR COMBUSTION-PRODUCT ENGINE PLANTS
- F02M—SUPPLYING COMBUSTION ENGINES IN GENERAL WITH COMBUSTIBLE MIXTURES OR CONSTITUENTS THEREOF
- F02M65/00—Testing fuel-injection apparatus, e.g. testing injection timing ; Cleaning of fuel-injection apparatus
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/21—Design or setup of recognition systems or techniques; Extraction of features in feature space; Blind source separation
- G06F18/213—Feature extraction, e.g. by transforming the feature space; Summarisation; Mappings, e.g. subspace methods
- G06F18/2135—Feature extraction, e.g. by transforming the feature space; Summarisation; Mappings, e.g. subspace methods based on approximation criteria, e.g. principal component analysis
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/23—Clustering techniques
- G06F18/232—Non-hierarchical techniques
- G06F18/2321—Non-hierarchical techniques using statistics or function optimisation, e.g. modelling of probability density functions
- G06F18/23213—Non-hierarchical techniques using statistics or function optimisation, e.g. modelling of probability density functions with fixed number of clusters, e.g. K-means clustering
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/24—Classification techniques
Landscapes
- Engineering & Computer Science (AREA)
- Data Mining & Analysis (AREA)
- Theoretical Computer Science (AREA)
- Computer Vision & Pattern Recognition (AREA)
- General Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Evolutionary Biology (AREA)
- Evolutionary Computation (AREA)
- Bioinformatics & Computational Biology (AREA)
- Artificial Intelligence (AREA)
- General Physics & Mathematics (AREA)
- Life Sciences & Earth Sciences (AREA)
- Probability & Statistics with Applications (AREA)
- Chemical & Material Sciences (AREA)
- Combustion & Propulsion (AREA)
- Mechanical Engineering (AREA)
- Electrical Control Of Air Or Fuel Supplied To Internal-Combustion Engine (AREA)
Abstract
The invention relates to the technical field of engine part tests and discloses a fuel injector grouping method which comprises the following steps: the method comprises the steps of utilizing a fuel injector performance debugging rack to carry out fuel injection quantity test, and sequentially determining actual fuel injection quantity data Q corresponding to N fuel injectors produced in batch under a plurality of different debugging working conditionsij(ii) a Q for all debugging operating pointsijPerforming standardization processing to form oil quantity standardization data Sij(ii) a Standardizing the QijAs input variables, S is converted by orthonormal conversion according to principal component analysisi1、Si2…SijThe variables with correlation are converted into M linearly uncorrelated principal components; performing cluster analysis on the obtained dimensionalities of the M main components, and dividing the N oil injectors into k groups; and obtaining an optimal k value to finish oil quantity grouping of the oil injectors. The method can group the oil mass of the fuel injector assemblies produced in batch, and can ensure the consistency of the fuel injectors in the same group to the maximum extent on the basis of not changing the technical conditions of the existing production process.
Description
Technical Field
The invention relates to the technical field of engine part tests, in particular to a fuel injector grouping method.
Background
In an electric control high-pressure common rail fuel system, the consistency of fuel injectors has great influence on the emission level of a diesel engine and the fuel consumption and economy. In the mass production process, because the mechanical manufacturing errors of all parts of the fuel injector are inevitable and are limited by the processes and implementation conditions of all stages of production, processing and assembly, the fuel injection quantity difference exists among fuel injector assemblies, the output power of an engine is also different among different cylinders, and accordingly the vehicle emission power performance is poor and the fuel consumption is high.
At present, the method of grouping the oil quantities of the oil injectors is commonly adopted in the industry to ensure the consistency of the electric control oil injectors in the same group, and the specific method is as follows: firstly, determining the oil mass range of all the injectors to be tested at a certain debugging working point, then dividing the oil mass range into a plurality of continuous smaller oil mass grouping ranges, and finally dividing the injectors to be grouped into the groups according to the oil mass of the injectors to be grouped at the debugging working point. Fuel system suppliers provide host plants with fuel injectors in sets selected from the same set. Therefore, all the oil injectors arranged on the same engine are ensured to be selected as the oil injectors in the same group, so that the consistency of the oil injectors on all the cylinders of the engine is better than that of the oil injectors without grouping.
The existing oil quantity grouping method also has obvious problems, for example, grouping is only carried out according to the oil quantity of a certain working condition point, so that the consistency of the oil injectors in the same group is better when the engine operates at the working condition point, but the consistency of the oil injectors cannot be ensured when the engine operates at other working condition points. Therefore, a more scientific and reasonable fuel injector grouping method is needed to ensure the consistency of the fuel injectors in the same group under more operation conditions.
Disclosure of Invention
The invention aims to provide an oil sprayer grouping method which can group the oil quantity of mass-produced oil sprayer assemblies and can furthest ensure the consistency of the oil sprayers in the same group on the basis of not changing the technical conditions of the existing production process.
In order to realize the purpose, the following technical scheme is provided:
a fuel injector grouping method comprising the steps of:
step a, carrying out an oil injection quantity test by using an oil injector performance debugging rack, and sequentially determining actual oil injection quantities of N oil injectors produced in batch under a plurality of different debugging working conditions, wherein the actual oil injection quantity of the ith oil injector at the jth debugging working condition point is Qij；
Step b, the actual fuel injection quantity data Q of all the debugging working condition pointsijPerforming standardization processing to form oil quantity standardization data Sij；
Step c, standardizing the actual fuel injection quantity data Q after the step bijAs input variables, S is converted by orthonormal conversion according to principal component analysisi1、Si2…SijThe variables with correlation are converted into M linearly uncorrelated principal components;
d, performing cluster analysis on the dimensionalities of the M main components obtained in the step c, and dividing the N oil injectors into k groups;
and e, obtaining the optimal k value by adopting a method for determining the k value, and finishing the grouping of the oil quantity of the oil injectors.
Further, in step a, the commissioning condition point includes:
and determining an idle operation point, a rated power point, a partial load point, a low-speed torque point and/or a small oil quantity point according to the requirement of the engine matched with the oil injector.
Further, the actual fuel injection quantity data QijThe average of the individual injection quantity data of the N injections.
Further, in step b, the normalization process is a z-score method.
Further, in step b, the normalized data
Wherein the content of the first and second substances,
the average fuel injection quantity, sigma, of 500 fuel injectors at the jth debugging pointjIs the standard deviation.
Further, in step c, the principal component analysis method is a linear principal component analysis method.
Further, in step c, the principal component analysis method includes:
step c1, calculating the actual fuel injection quantity S of each debugging working condition pointi1、Si2…SijA matrix of correlation coefficients therebetween;
step c2, calculating characteristic values and the contribution rate and the accumulated contribution rate of each principal component according to the correlation coefficient matrix;
and c3, obtaining M main components according to the accumulated contribution rate being larger than a set threshold value.
Further, in step d, the cluster analysis comprises a K-means clustering method, or a K-means + + clustering method, or a K-center point algorithm.
Further, the K-means clustering method comprises the following steps:
d1, selecting k initial clustering centers from the N fuel injectors, and dividing the N fuel injectors into k groups;
d2, calculating the mean value of all the objects in the k group classes;
and d3, repeating the step d2 until the value using the mean square error as the standard measure function converges.
Further, in step e, the method for determining the optimal k value includes a contour coefficient method, an elbow rule or a Calinski-Harabasz rule.
Compared with the prior art, the invention has the beneficial effects that:
according to the fuel injector grouping method provided by the invention, the magnitude difference of different fuel quantities is eliminated by utilizing data standardization processing based on the fuel quantity characteristics of the electric control common rail fuel injectors under different debugging working conditions; the principal component analysis method can play a role in reducing dimensionality, multiple indexes are combined into a few mutually-independent comprehensive indexes, and each principal component can reflect most of information of an original variable; grouping of the fuel injectors is realized based on a clustering analysis algorithm, and the method is wide in application range, safe and reliable. The fuel injector grouping method provided by the invention can be used for grouping the fuel amount of fuel injector assemblies produced in batches, and can be used for ensuring the consistency of fuel injectors in the same group to the maximum extent on the basis of not changing the technical conditions of the existing production process.
Drawings
In order to more clearly illustrate the technical solutions in the embodiments of the present invention, the drawings used in the description of the embodiments of the present invention will be briefly described below, and it is obvious that the drawings in the following description are only some embodiments of the present invention, and for those skilled in the art, other drawings can be obtained according to the contents of the embodiments of the present invention and the drawings without creative efforts.
FIG. 1 is a schematic flow chart of a fuel injector grouping method according to an embodiment of the present invention;
FIG. 2 is a schematic flow chart of a K-means clustering method according to a second embodiment of the present invention;
fig. 3 is a schematic flow chart of determining an optimal k value by using a contour coefficient method according to a second embodiment of the present invention;
FIG. 4 is a diagram illustrating a contour coefficient method for various sets of k values according to a second embodiment of the present invention;
fig. 5 shows a grouping of 500 injectors when k is 4 according to the second embodiment of the present invention.
Detailed Description
In order to make the technical solutions of the present invention better understood by those skilled in the art, the technical solutions of the present invention are further described below by way of specific embodiments with reference to the accompanying drawings.
Example one
As shown in fig. 1, the present embodiment provides a fuel injector grouping method, including the steps of:
step a, carrying out an oil injection quantity test by using an oil injector performance debugging rack, and sequentially determining actual oil injection quantities corresponding to batch-produced N oil injectors under debugging working condition points determined by different rail pressures and control pulse widths;
the actual fuel injection quantity of the ith fuel injector at the jth debugging working condition point is Qij；
Step b, actual fuel injection quantity data Q of all debugging working condition pointsijPerforming standardization processing to form oil quantity standardization data Sij；
Step c, standardizing the actual fuel injection quantity data Q after the step bijAs input variables, S is converted by orthonormal conversion according to principal component analysisi1、Si2…SijIn which the dependent variables are converted into M linearly independent variablesA main component;
d, performing cluster analysis on the dimensionalities of the M main components obtained in the step c, and dividing the N oil injectors into k groups;
and e, obtaining the optimal k value by using a method for determining the k value so as to obtain a corresponding classification result and finish oil quantity grouping of the oil injectors.
According to the fuel injector grouping method provided by the embodiment, the magnitude difference of different fuel quantities is eliminated by utilizing data standardization processing based on the fuel quantity characteristics of the electric control common rail fuel injector under different debugging working conditions; the principal component analysis method can play a role in reducing dimensionality, multiple indexes are combined into a few mutually-independent comprehensive indexes, and each principal component can reflect most of information of an original variable; grouping of the fuel injectors is realized based on a clustering analysis algorithm, and the method is wide in application range, safe and reliable.
The fuel injector grouping method provided by the embodiment can be used for grouping fuel quantities of fuel injector assemblies produced in batches, and can ensure the consistency of fuel injectors in the same group to the maximum extent on the basis of not changing the technical conditions of the existing production process. The fuel injector grouping method is simple in steps and convenient to operate, is realized based on a data analysis algorithm, does not need to add extra hardware equipment, is low in cost and is suitable for popularization.
Preferably, in step a, the debugging operating point includes but is not limited to idle operating point, rated power point, part load point, low speed torque point, small oil quantity point and other characteristic operating points which are determined according to the requirement of the engine matched with the fuel injector.
Optionally, actual fuel injection quantity data QijThe average of the individual injection quantity data of the N injections.
Preferably, in step b, the normalization process is a z-score method. Further, in step b, the data is normalized
Wherein the content of the first and second substances,
average fuel injection quantity, sigma, of N injectors at jth debugging pointjIs the standard deviation. Illustratively, N is 500.
Optionally, in step c, the principal component analysis method is a linear principal component analysis method. Linear principal component analysis is a dimension reduction method that reduces the dimensionality of a data set. It transforms the data into a new coordinate system by linear transformation such that the first large variance of any data projection is on the first coordinate (first principal component), the second large variance is on the second coordinate (second principal component), and so on, to obtain M linearly uncorrelated principal components.
Further, in step c, the principal component analysis method includes:
step c1, calculating the actual fuel injection quantity S of each debugging working condition pointi1、Si2…SijA matrix of correlation coefficients therebetween;
step c2, calculating the eigenvalue and the contribution rate and the accumulated contribution rate of each principal component according to the correlation coefficient matrix;
and c3, obtaining M main components according to the fact that the accumulated contribution rate is larger than the set threshold value.
Optionally, in step d, the cluster analysis comprises a K-means clustering method, or a K-means + + clustering method, or a K-center point algorithm.
Further, the K-means clustering method comprises the following steps:
d1, selecting k initial clustering centers from the N oil injectors, and dividing the N oil injectors into k groups;
d2, calculating the mean value of all the objects in the k group classes;
and d3, repeating the step d2 until the value using the mean square error as the standard measure function converges.
In short, the initial cluster centers of the k group classes are first selected appropriately among the N data objects, and for the remaining other objects, they are respectively assigned to the class represented by the cluster center most similar thereto according to their similarity (distance) to these cluster centers; then, calculating the clustering center of each obtained new data object, namely the mean value of all the objects in the cluster; this process is repeated until convergence of the values begins using the mean square error as a function of the standard measure.
Optionally, in step e, the method of determining the optimal k value includes a contour coefficient method, an elbow rule or the Calinski-Harabasz criterion.
Example two
The embodiment provides a fuel injector grouping method, which comprises the following steps:
step S1, assuming that 500 injectors need to be grouped, first, a bench test is used to determine the actual injection quantities of the injectors under six characteristic conditions determined by preset different rail pressures and pulse widths. The actual fuel injection quantity of the ith fuel injector at the jth debugging working condition point is Qij， i＝1,2,…,500，j＝1,2,…,6。
Step S2, actual fuel injection quantity data Q of all debugging working condition pointsijPerforming standardization processing to form oil quantity standardization data Sij(ii) a The preferred normalization process is the z-score method,
wherein QijThe actual fuel injection quantity of the ith fuel injector at the jth debugging operating point is obtained,
average fuel injection quantity, sigma, of 500 injectors at jth debugging pointjIs the standard deviation. The normalized partial injector fuel quantity data is shown in table 1.
TABLE 1 normalized partial injector quantity data
| Injector numbering | |
|
|
|
|
|
| 1 | 2.39 | 1.50 | 0.62 | 1.40 | 1.53 | 1.51 |
| 2 | 0.50 | 1.74 | 0.92 | 1.91 | 1.86 | 1.68 |
| 3 | 1.27 | 1.36 | 0.96 | 1.08 | 1.87 | 1.75 |
| 4 | 0.75 | 1.53 | 1.02 | 1.97 | 1.96 | 1.76 |
| 5 | 1.49 | 1.80 | 0.83 | 1.49 | 1.70 | 1.63 |
| 6 | 1.68 | 1.80 | 0.44 | 1.39 | 1.67 | 1.56 |
| 7 | 1.81 | 1.48 | 0.71 | 1.09 | 1.53 | 1.52 |
| 8 | 1.29 | 0.99 | 0.58 | 1.05 | 1.53 | 1.48 |
| 9 | 0.46 | 0.41 | 0.38 | 0.73 | 1.20 | 1.20 |
| 10 | 0.73 | 1.44 | 0.57 | 1.87 | 1.80 | 1.67 |
Step S3, taking the actual fuel injection quantity data normalized in the step 2 as an input index variable, and converting the actual fuel injection quantity data into S through orthotropic conversion according to a principal component analysis methodi1，Si2，…，Si6These six groups of possibly associated index variables are converted into six linearly uncorrelated principal components. The characteristic value, contribution rate, and cumulative contribution rate of each of the principal components of the partial injector obtained by the calculation are shown in table 2. As can be seen, the cumulative contribution rate of the first two principal components reaches 90.10%, which is greater than the threshold 85%, so that Z1 is selected as the first principal component, Z2 is selected as the second principal component, and the two principal components basically retain the original oil amount Si1，Si2，…，Si6Most of the information of (1). As shown in table 3, the conversion matrix between the six characteristic operating condition data of the injector and the first two principal components reflects the correlation between the selected principal component and each quantity of fuel, and the contribution of each quantity of fuel to the selected principal component. As shown in Table 4, the score values of the first two principal components of the six characteristic working condition data of the partial fuel injector can be used for replacing six original characteristic data, and the score values are used for next clustering analysisThe sample value of (2).
Table 2 shows the characteristic values, contribution rates and cumulative contribution rates of the main components of the partial injectors calculated
| Principal component | Characteristic value | Contribution ratio (%) | Cumulative contribution ratio (%) |
| Z1 | 4.69 | 78.19 | 78.19 |
| Z2 | 0.71 | 11.91 | 90.10 |
| Z3 | 0.35 | 5.89 | 95.99 |
| Z4 | 0.17 | 2.77 | 98.76 |
| Z5 | 0.07 | 1.17 | 99.93 |
| Z6 | 0.00 | 0.07 | 100.00 |
TABLE 3 conversion matrix between six characteristic operating condition data of fuel injector and the first two principal components
| Transformation matrix | Z1 | |
| Working condition | ||
| 1 | 0.38 | -0.21 |
| |
0.44 | -0.11 |
| |
0.27 | 0.95 |
| |
0.42 | -0.17 |
| |
0.45 | -0.03 |
| |
0.45 | -0.10 |
TABLE 4 score values of the first two principal components of six characteristic operating condition data of partial fuel injector
| Injector numbering | Sample | Z1 | Z2 | |
| 1 | X1 | -1.32 | 1.13 | |
| 2 | X2 | -0.19 | 0.71 | |
| 3 | X3 | -1.46 | 0.80 | |
| 4 | X4 | -1.04 | 0.63 | |
| 5 | X5 | -0.77 | 0.13 | |
| 6 | X6 | -0.94 | 0.95 | |
| 7 | X7 | -0.94 | 0.55 | |
| 8 | X8 | -0.80 | 0.88 | |
| 9 | X9 | -1.56 | 1.09 | |
| 10 | X10 | -1.08 | 1.08 |
Step S4, performing cluster classification on the two principal component dimensions obtained in step 3Analysis, 500 injectors are divided into K groups, and a preferred K-means clustering method is taken as an example, as shown in table 4, assuming that X isiThe ith sample data has two principal component dimension variables, i.e., two eigenvalues, obtained as described above. As shown in fig. 2, the K-means clustering method specifically comprises the following steps:
step S41, the initial group k is 2, namely the oil injectors are randomly divided into two groups;
step S42, selecting 1 initial clustering center in each group k, Ci(t), i ═ 1,2, …, k, where i denotes the group class sequence number and t in brackets denotes the number of iterations.
Step S43, data sample to be clustered (X)1，X2，…,X500) One C of two cluster centers is assigned according to the minimum distance criterioni(t) of (d). Calculating the jth data sample XjWith two cluster centers Ci(t) distance d (X) where i is 1,2j，Ci(t)), i ═ 1,2, …, k, and d (X)j，Ci(t)) is minimum, then XjIs class i, and is recorded as
Step S44, calculating two new cluster centers:
wherein N isiThe number of samples in the i-th class is represented, i is 1,2, …, k.
Step S45, if Ci(t+1)＝CiAnd (t), if i is 1,2, …, k, obtaining k types of sample data, and going to the next step (d6), otherwise, returning to step S43.
Step S46, judging whether the group class k is larger than or equal to the set upper limit 10, if so, going to the next step S5; if no, k is k +1, and the process returns to step S42.
And step S5, obtaining the optimal k value by applying the method for determining the k value. In the embodiment, an optimal k value is preferably determined by using a contour Coefficient method (Silhouette coeffient), and the method is an evaluation method for evaluating the good and bad clustering effect by combining two factors of intra-group similarity and inter-group dissimilarity. And obtaining a classification result corresponding to the k, and finishing oil quantity grouping of the oil injectors. As shown in fig. 3, the contour coefficient method specifically determines the optimal k value as follows:
step S51, where the initial group class k is 2, calculates a certain sample X in the k groups of classification information obtained in step (d)iAverage distance to other samples in the same group a (i).
The smaller the value of the average distance a (i), the more samples are clustered into the group, which represents the intra-group dissimilarity of the samples.
Step S52, calculating a certain sample XiTo other groups CjThe average distance b (ij) of the other samples. The larger the value, the less likely the sample should be clustered into the group, which represents the sample XiGroup CjInter-group dissimilarity of (c). b (i) ═ min { b }i1，bi2，…,bik}。
Step S53, defining the contour coefficient method of the sample i when the group is k according to the intra-group dissimilarity a (i) and the inter-group dissimilarity b (i) of the sample i
For all samples
The mean value of (2) is a contour coefficient method of a clustering result when the group class is k, and is an effective measure for judging whether the clustering is reasonable or not. The closer the value is to 1, the more reasonable the clustering of the sample i is; the closer the value is to-1, the more likely the sample i should be classified into another group;
step S54, as shown in FIG. 4, is a method for calculating contour coefficients for all k groups
Value and take the maximum valueThe corresponding k is the optimal k value, where the maximum profile factor is about 0.9 and the corresponding k value is 4.
As shown in fig. 5, in step S55, the optimal classification information of 500 injectors when k is 4 is obtained, and the final classification result is obtained.
According to the fuel injector grouping method provided by the embodiment, the magnitude difference of different fuel quantities is eliminated by utilizing data standardization processing based on the fuel quantity characteristics of the electric control common rail fuel injector under different debugging working conditions; the principal component analysis method can play a role in reducing dimensionality, multiple indexes are combined into a few mutually-independent comprehensive indexes, and each principal component can reflect most of information of an original variable; grouping of the fuel injectors is realized based on a clustering analysis algorithm, and the method is wide in application range, safe and reliable.
It is to be noted that the foregoing is only illustrative of the preferred embodiments of the present invention and the technical principles employed. It will be understood by those skilled in the art that the present invention is not limited to the particular embodiments described herein, but is capable of various obvious changes, rearrangements and substitutions as will now become apparent to those skilled in the art without departing from the scope of the invention. Therefore, although the present invention has been described in greater detail by the above embodiments, the present invention is not limited to the above embodiments, and may include other equivalent embodiments without departing from the spirit of the present invention, and the scope of the present invention is determined by the scope of the appended claims.
Claims (10)
1. A fuel injector grouping method is characterized by comprising the following steps:
step a, carrying out an oil injection quantity test by using an oil injector performance debugging rack, and sequentially determining the actual oil injection quantity of N oil injectors produced in batch under a plurality of different debugging working condition points, wherein the actual oil injection quantity of the ith oil injector at the jth debugging working condition point is Qij；
Step b, the actual fuel injection quantity data Q of all the debugging working condition pointsijPerforming standardization processing to form oil quantity standardization data Sij；
Step c, marking the mark obtained in the step bStandardized actual fuel injection quantity data QijAs input variables, S is converted by orthonormal conversion according to principal component analysisi1、Si2…SijThe variables with correlation are converted into M linearly uncorrelated principal components;
d, performing cluster analysis on the dimensionalities of the M main components obtained in the step c, and dividing the N oil injectors into k groups;
and e, obtaining the optimal k value by adopting a method for determining the k value, and finishing the grouping of the oil quantity of the oil injectors.
2. The fuel injector grouping method according to claim 1, characterized in that in step a, the commissioning points comprise:
and determining an idle operation point, a rated power point, a partial load point, a low-speed torque point and/or a small oil quantity point according to the requirement of the engine matched with the oil injector.
3. The fuel injector grouping method according to claim 1, characterized in that the actual fuel injection quantity data QijThe average of the individual injection quantity data of the N injections.
4. The fuel injector grouping method according to claim 1, characterized in that in step b, the normalization process is a z-score method.
5. The fuel injector grouping method according to claim 4, characterized in that in step b, the normalized data
Wherein the content of the first and second substances,
the average fuel injection quantity, sigma, of the N fuel injectors at the jth debugging pointjIs the standard deviation.
6. The fuel injector grouping method according to claim 1, characterized in that in step c, the principal component analysis method is a linear principal component analysis method.
7. The fuel injector grouping method according to claim 1, characterized in that in step c, the principal component analysis method includes:
step c1, calculating the actual fuel injection quantity S of each debugging working condition pointi1、Si2…SijA matrix of correlation coefficients therebetween;
step c2, calculating characteristic values and the contribution rate and the accumulated contribution rate of each principal component according to the correlation coefficient matrix;
and c3, obtaining M main components according to the accumulated contribution rate being larger than a set threshold value.
8. The fuel injector grouping method according to claim 1, characterized in that in step d, the cluster analysis comprises a K-means clustering method, or a K-means + + clustering method, or a K-center point algorithm.
9. The fuel injector grouping method according to claim 8, characterized in that the K-means clustering method includes:
d1, selecting k initial clustering centers from the N fuel injectors, and dividing the N fuel injectors into k groups;
d2, calculating the mean value of all the objects in the k group classes;
and d3, repeating the step d2 until the value using the mean square error as the standard measure function converges.
10. The fuel injector grouping method according to claim 1, characterized in that in step e, the method of determining the optimal k-value comprises the contour coefficient method, the elbow rule or the Calinski-Harabasz criterion.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910765326.6A CN110555470B (en) | 2019-08-19 | 2019-08-19 | Oil sprayer grouping method |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910765326.6A CN110555470B (en) | 2019-08-19 | 2019-08-19 | Oil sprayer grouping method |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN110555470A CN110555470A (en) | 2019-12-10 |
| CN110555470B true CN110555470B (en) | 2021-10-01 |
Family
ID=68737733
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201910765326.6A Active CN110555470B (en) | 2019-08-19 | 2019-08-19 | Oil sprayer grouping method |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN110555470B (en) |
Families Citing this family (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112070142B (en) * | 2020-09-02 | 2024-05-10 | 平安科技（深圳）有限公司 | Grouping method and device for vehicle accessories, electronic equipment and storage medium |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN1283791A (en) * | 1999-07-06 | 2001-02-14 | 中国石油化工集团公司 | Method for measuring contents of components in oil residue |
| CN101285430A (en) * | 2007-04-09 | 2008-10-15 | 山东申普汽车控制技术有限公司 | Method for combined pulse spectrum controlling engine fuel injector |
| CN103195632A (en) * | 2013-04-01 | 2013-07-10 | 中国北方发动机研究所(天津) | Oil injection nozzle matching part flow grouping testing device and testing method |
| CN104481769A (en) * | 2014-12-03 | 2015-04-01 | 中国第一汽车股份有限公司无锡油泵油嘴研究所 | Online diagnosis method for uniformity of common-rail oil injectors |
| CN107368685A (en) * | 2017-07-21 | 2017-11-21 | 重庆工商大学 | Based on intelligent clustering particle filter automotive dampers performance degradation prognosis method |
-
2019
- 2019-08-19 CN CN201910765326.6A patent/CN110555470B/en active Active
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN1283791A (en) * | 1999-07-06 | 2001-02-14 | 中国石油化工集团公司 | Method for measuring contents of components in oil residue |
| CN101285430A (en) * | 2007-04-09 | 2008-10-15 | 山东申普汽车控制技术有限公司 | Method for combined pulse spectrum controlling engine fuel injector |
| CN103195632A (en) * | 2013-04-01 | 2013-07-10 | 中国北方发动机研究所(天津) | Oil injection nozzle matching part flow grouping testing device and testing method |
| CN104481769A (en) * | 2014-12-03 | 2015-04-01 | 中国第一汽车股份有限公司无锡油泵油嘴研究所 | Online diagnosis method for uniformity of common-rail oil injectors |
| CN107368685A (en) * | 2017-07-21 | 2017-11-21 | 重庆工商大学 | Based on intelligent clustering particle filter automotive dampers performance degradation prognosis method |
Non-Patent Citations (2)
| Title |
|---|
| "基于主成分与Adaptive-Lasso的飞机油耗统计分析";谭景宝 等;《长春师范大学学报》;20180831;第37卷(第8期);第8-12页 * |
| "汽车电子喷油器原理及故障诊断实例";邓燕菲 等;《柴油机设计与制造》;20041231(第3期);第54-56页 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN110555470A (en) | 2019-12-10 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN111797567A (en) | Deep learning network-based bearing fault classification method and system | |
| CN112596016A (en) | Transformer fault diagnosis method based on integration of multiple one-dimensional convolutional neural networks | |
| CN110555470B (en) | Oil sprayer grouping method | |
| CN114330042A (en) | Program load spectrum compiling method and system based on SN curve and storage medium | |
| Kim et al. | Deep learning-based data augmentation for hydraulic condition monitoring system | |
| CN114139639A (en) | Fault classification method based on self-walking neighborhood preserving embedding | |
| CN111382792B (en) | Rolling bearing fault diagnosis method based on double-sparse dictionary sparse representation | |
| CN117435992A (en) | Fault prediction method and system for hydraulic propulsion system of shield tunneling machine | |
| CN115828156A (en) | Power grid metadata monitoring-based electricity stealing and leakage identification method and system | |
| CN115879046A (en) | Internet of things abnormal data detection method based on improved feature selection and hierarchical model | |
| CN115169740A (en) | Sequence prediction method and system of pooled echo state network based on compressed sensing | |
| CN101036140A (en) | System and method for generating custom hierarchies in an analytical data structure | |
| CN115358795A (en) | Sales amount prediction method | |
| CN115510740A (en) | Aero-engine residual life prediction method based on deep learning | |
| CN114496068A (en) | Protein secondary structure prediction method, device, equipment and storage medium | |
| CN113537327A (en) | Non-invasive load identification method and system based on Alexnet neural network and color coding | |
| CN113255927A (en) | Logistic regression model training method and device, computer equipment and storage medium | |
| Zhao et al. | Industrial fault diagnosis based on few shot learning | |
| CN113537366B (en) | Transient stability evaluation method for power system | |
| CN117725437B (en) | Machine learning-based data accurate matching analysis method | |
| CN114894480B (en) | Bearing fault diagnosis method and device based on unbalanced data set | |
| CN112650770B (en) | MySQL parameter recommendation method based on query work load analysis | |
| CN116612329A (en) | Point cloud classification neural network optimization method and device based on dimension transformation, electronic equipment and medium | |
| Zhang et al. | Distributed dimensionality reduction of industrial data based on clustering | |
| CN117130942B (en) | Simulation test method for simulating domestic production environment |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |