CN114810448A - Wavelet transform-based online analysis method for fuel gas injection process time of high-pressure natural gas direct injection engine - Google Patents
Wavelet transform-based online analysis method for fuel gas injection process time of high-pressure natural gas direct injection engine Download PDFInfo
- Publication number
- CN114810448A CN114810448A CN202210380960.XA CN202210380960A CN114810448A CN 114810448 A CN114810448 A CN 114810448A CN 202210380960 A CN202210380960 A CN 202210380960A CN 114810448 A CN114810448 A CN 114810448A
- Authority
- CN
- China
- Prior art keywords
- pressure
- time
- natural gas
- injection process
- wavelet
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Granted
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
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02T—CLIMATE CHANGE MITIGATION TECHNOLOGIES RELATED TO TRANSPORTATION
- Y02T10/00—Road transport of goods or passengers
- Y02T10/10—Internal combustion engine [ICE] based vehicles
- Y02T10/30—Use of alternative fuels, e.g. biofuels
Landscapes
- Engineering & Computer Science (AREA)
- Chemical & Material Sciences (AREA)
- Combustion & Propulsion (AREA)
- Mechanical Engineering (AREA)
- General Engineering & Computer Science (AREA)
- Combined Controls Of Internal Combustion Engines (AREA)
Abstract
The invention discloses a wavelet transform-based online analysis method for time of a fuel gas injection process of a high-pressure natural gas direct injection engine. Step 1: installing and debugging equipment, wherein an inlet pressure sensor is installed at a gas inlet of a natural gas HPDI (high pressure differential) ejector, and meanwhile, a pressure sensor is installed at an air hole of the ejector; step 2: performing wavelet transform-based processing on the pressure signals acquired by the pressure sensor installed in the step 1; and step 3: performing a time characteristic identification algorithm based on the pressure signal processed based on the wavelet transform in the step 2; and 4, step 4: and (4) comparing and analyzing the time characteristics obtained by time characteristic identification in the step (3) with the air injection rule measured by an off-line experiment, so as to realize the on-line analysis of the time of the fuel gas injection process of the direct injection engine in the high-pressure natural gas cylinder. The method is used for solving the problem that the time characteristics of the existing natural gas and fuel gas injection process are not observable; and the online analysis of the time characteristics of the gas injection process is realized.
Description
Technical Field
The invention belongs to the field of power energy; in particular to a wavelet transform-based online analysis method for the fuel gas injection process time of a direct injection engine in a high-pressure natural gas cylinder.
Background
In recent years, with the emergence of national emission regulations, the internal combustion engine industry faces various difficulties due to the shortage of traditional fuel resources. Natural gas fuel is considered as a very promising alternative energy source, and has the advantages of small exhaust pollution, high octane number, high antiknock property, abundant reserves and reproducibility. From statistical data, it has been shown that natural gas fuel has increased in industrial use in recent years year by year, compared to conventional diesel engines. Natural gas engines typically have lower carbon emissions. However, the traditional premixed combustion natural gas engine has the advantages that the charge coefficient of the engine is reduced and the efficiency is reduced due to factors such as the knock limit and the pre-ignition, and the natural gas is directly sprayed into the cylinder by the high-pressure natural gas in-cylinder direct injection (HPDI) technology, so that the good air-fuel ratio is ensured, the charging efficiency is improved, the combustion is promoted, and the premixed combustion natural gas engine has a good application prospect.
The injector is the most important part of the natural gas direct injection technology as an actuator of a fuel supply system. Due to the complicated pneumatic hydraulic-electric structure in the injector, the change rule of the injection characteristic of the injector is quite complicated, and the HPDI injector has the difficulty in the development of the HPDI technology when the injection characteristic can not be directly observed in the actual working process.
Because the injection characteristic of the injector can not be directly observed in the actual operation process, open-loop control is adopted for controlling the HPDI electronic control injector, and the injection strategies under different working conditions of the engine are difficult to realize by realizing a calibrated MAP. Therefore, it is necessary to provide an online testing method which can realize online feedback of the injector, further realize closed-loop control of the fuel injection quantity, and simultaneously provide real-time monitoring, fault early warning and residual life prediction of the injector according to the running state of the injector.
Disclosure of Invention
The invention provides a wavelet transform-based online analysis method for time of a gas injection process of a direct injection engine in a high-pressure natural gas cylinder, which is used for solving the problem of unobservability of time characteristics of the existing gas injection process; by analyzing the time-frequency characteristics of the inlet pressure signal, the identification of the injector time characteristics is realized on the inlet pressure signal wavelet change and the signals after the dimensionality reduction treatment, and the online analysis of the gas injection process time characteristics is realized.
The invention is realized by the following technical scheme:
a wavelet transform-based online analysis method for the fuel gas injection process time of a high-pressure natural gas direct injection engine in a cylinder comprises the following steps:
step 1: installing and debugging equipment, wherein an inlet pressure sensor is installed at a gas inlet of a natural gas HPDI ejector 1, and meanwhile, a pressure sensor is installed at an air hole of the ejector;
step 2: performing wavelet transform-based processing on the pressure signals acquired by the pressure sensor installed in the step 1;
and step 3: performing a time characteristic identification algorithm based on the pressure signal processed based on the wavelet transform in the step 2;
and 4, step 4: and (4) comparing and analyzing the time characteristics obtained by the time characteristic identification in the step (3) with the air injection rule measured by an off-line experiment, so as to realize the on-line analysis of the time of the fuel injection process of the direct injection engine in the high-pressure natural gas cylinder.
The step 1 of installing equipment specifically comprises a natural gas HPDI ejector 1, a PXI2, an upper computer 3, a pressure control device 4, a pressure sensor 5, an air source 6, an air rail 7, an air compressor 8 and an amplifier module 9.
The natural gas HPDI ejector 1 is respectively connected with PXI2 and a pressure sensor 5, the pressure sensor 5 is connected with an upper computer 3 through an amplifier module 9, the PXI2 is connected with the upper computer 3, the pressure sensor 5 is connected with a gas compressor 8 through a gas rail 7, the gas compressor 8 is connected with a gas source 6, and the gas rail 7 is connected with a pressure control device 4.
The step 1 of installing equipment specifically comprises a natural gas HPDI ejector 1, a PXI2, an upper computer 3, a pressure control device 4, a pressure sensor 5, an air source 6, an air rail 7, an air compressor 8 and an amplifier module 9.
The gas source 6 is pressurized by the compressor 8 and then ejected out of the natural gas HPDI ejector 1 through the gas rail 7 and the pressure sensor 5;
the signal of the pressure sensor 5 is amplified by a charge amplifier 9; then, acquiring an inlet pressure signal through a data acquisition module in the upper computer; the collected inlet pressure signal is stored in an upper computer;
the inlet pressure sensor 5 of the natural gas HPDI ejector 1 transmits signals to PXI2, and the PXI2 transmits processed signals to the upper computer 3;
the air pressure of the air rail 7 is monitored by the pressure control device 4;
the pressure control device 4 transmits a signal to the upper computer 3.
A wavelet transform-based method for analyzing the fuel gas injection process time of a direct injection engine in a high-pressure natural gas cylinder on line is characterized in that a force sensor is installed at an air hole of a natural gas HPDI (hot plug direct injection) injector 1, the time characteristic of the injection process is defined, and the injection starting time t is 0 Time t of full opening of needle valve 1 Time t at which needle valve starts to seat 2 End of injection time t 3 。
A process based on wavelet transform for analyzing time of gas injection process of direct injection engine in high pressure natural gas cylinder based on wavelet transform is carried out on pressure signals in step 2, specifically, derivation is carried out on inlet pressure signals, correlation between change rate of the derived inlet pressure signals and air injection rule is analyzed, the change rate of the inlet pressure signals is obtained to be a time-varying signal, the time-varying signal is divided into three stages of rising, stabilizing and falling in time domain, the stabilizing stage is extracted, and the wavelet transform of the signals is defined as:
wherein a is a scaling factor; τ is a translation factor; p (t) is the inlet pressure signal; psi is a window function; t is time;
extracting time domain characteristics and wavelet domain characteristics of the change rate of the segmented stationary section inlet pressure signals; in order to adapt to the computing power of an electronic control unit of the engine, the feature dimension reduction processing is carried out, and the time domain feature is saved for facilitating the subsequent time feature identification.
An on-line analysis method for the time of the gas injection process of a high-pressure natural gas direct injection engine based on wavelet transformation adopts a principal component analysis method for dimension reduction treatment of characteristics, specifically,
data were normalized:
the new coordinate system is { w 1 ，w 2 ，…,w n In which w i Is an orthonormal base, m is a dimensionality reduction dimension, x i As sample data, mean (x) i ) Is an array mean, std (x) i ) Is the standard deviation;
sample data x i The projection in the low-dimensional space coordinate system satisfies the following conditions:
wherein z is ij Is x i Coordinates of the jth dimension in a dimension reduction coordinate system;
according to the nearest reconstruction, the distance between original sample points based on the reconstructed sample points satisfies:
in the formula, W T XX T W is the sample variance, I is the identity matrix, W is { W 1 ，w 2 ，…,w n }，tr(W T XX T W) is a matrix trace;
using the lagrange multiplier for equation (4):
XX T W＝γW (5)
for covariance matrixArray XX T All the eigenvalues are obtained, and the obtained eigenvalue sequence is expressed as gamma 1 ≥γ 2 ≥···≥γ m If the feature vectors corresponding to the first m feature values are the solution solved by the principal component analysis, the formula (6) is shown as follows:
W * ＝(w 1 ,w 2 ,···w m ) (6)
wherein, W * The solution is the solution of principal component analysis.
A wavelet transform-based online analysis method for the time of the gas injection process of a direct injection engine in a high-pressure natural gas cylinder is characterized in that in the injection process of a natural gas injector, the change rate of an inlet pressure signal is divided into three stages, namely an ascending stage, a stable stage and a descending stage; three of which are defined as falling (t) 0 -t 1 ) Rises (t) 2 -t 3 ) And the method is stable, and the time domain characteristics of the inlet pressure signal change rate are still kept after dimension reduction processing, so that the inlet pressure signal change rate can be used in the time characteristic identification process.
The time characteristic identification algorithm of the step 3 is specifically that the first discontinuity point of the change rate of the inlet pressure signal after wavelet transformation and dimensionality reduction is t 1 Forward search for stationary phase point t 0 (ii) a The second discontinuity is t 2 Retrieving the stationary phase t backward 3 If there is only one discontinuity, say t 1 And t 2 Overlap in time domain, so it is determined that the needle begins to seat without opening during injection, and the needle is in motion throughout the injection.
A wavelet transform-based online analysis method for the fuel gas injection process time of a high-pressure natural gas direct injection engine in a cylinder comprises the following steps:
step S1: collecting an inlet pressure signal of a natural gas HPDI ejector 1;
step S2: deriving the pressure signal of step S1;
step S3: performing wavelet transformation on the derivative pressure signal of step S2;
step S4: performing dimension reduction processing on the pressure signal subjected to wavelet change in the step S3;
step S5: the inlet pressure signal change rate of the pressure signal of the dimensionality reduction processing of step S4 is divided into a first discontinuity point t 1 And a second discontinuity t 2 ；
Step S6: first discontinuity t based on step S5 1 Forward search for stationary phase point t 0 ；
Step S7: second discontinuity t based on step S5 2 Retrieving the stationary phase t backward 3 ；
Step S8: judging the first discontinuity t of step S5 1 And a second discontinuity t 2 Whether they are equal; if equal, say t 1 And t 2 The time domain coincides, and the needle valve begins to seat without being opened in the injection process; if not equal, t 1 And t 2 Not coinciding in time domain, the needle valve is already open during injection.
The invention has the beneficial effects that:
the invention ensures the structural integrity of the natural gas engine.
The invention has real-time performance.
The invention does not need to change the structure of the device, has good economical efficiency and long service life.
The invention has rapid signal processing speed and high precision.
Drawings
FIG. 1 is a schematic diagram of an experimental setup of the present invention.
FIG. 2 is a schematic diagram of a time profile of the present invention.
FIG. 3 is a flow chart of the method of the present invention.
Detailed Description
The technical solutions in the embodiments of the present invention will be described clearly and completely with reference to the accompanying drawings in the embodiments of the present invention, and it is obvious that the described embodiments are only a part of the embodiments of the present invention, and not all of the embodiments. All other embodiments, which can be obtained by a person skilled in the art without making any creative effort based on the embodiments in the present invention, belong to the protection scope of the present invention.
A wavelet transform-based online analysis method for the fuel gas injection process time of a high-pressure natural gas direct injection engine in a cylinder comprises the following steps:
step 1: installing and debugging equipment, wherein an inlet pressure sensor is installed at a gas inlet of a natural gas HPDI ejector 1, and meanwhile, a pressure sensor is installed at an air hole of the ejector;
step 2: performing wavelet transform-based processing on the pressure signals acquired by the pressure sensor installed in the step 1;
and step 3: performing a time characteristic identification algorithm based on the pressure signal processed based on the wavelet transform in the step 2;
and 4, step 4: and (4) comparing and analyzing the time characteristics obtained by time characteristic identification in the step (3) with the air injection rule measured by an off-line experiment, so as to realize the on-line analysis of the time of the fuel gas injection process of the direct injection engine in the high-pressure natural gas cylinder.
The step 1 of installing equipment specifically comprises a natural gas HPDI ejector 1, a PXI2, an upper computer 3, a pressure control device 4, a pressure sensor 5, an air source 6, an air rail 7, an air compressor 8 and an amplifier module 9.
The natural gas HPDI ejector 1 is respectively connected with PXI2 and a pressure sensor 5, the pressure sensor 5 is connected with an upper computer 3 through an amplifier module 9, the PXI2 is connected with the upper computer 3, the pressure sensor 5 is connected with a gas compressor 8 through a gas rail 7, the gas compressor 8 is connected with a gas source 6, and the gas rail 7 is connected with a pressure control device 4.
A wavelet transform-based online analysis method for the fuel gas injection process time of a high-pressure natural gas direct injection engine in a cylinder is characterized in that in step 1, equipment is installed, wherein the equipment comprises a natural gas HPDI (high pressure direct injection) injector 1, a PXI2, an upper computer 3, a pressure control device 4, a pressure sensor 5, a gas source 6, a gas rail 7 and a gas compressor 8;
the gas source 6 is pressurized by the compressor 8 and then ejected out of the natural gas HPDI ejector 1 through the gas rail 7 and the pressure sensor 5;
the signal of the pressure sensor 5 is amplified by a charge amplifier 9; then, acquiring an inlet pressure signal through a data acquisition module in the upper computer; the collected inlet pressure signal is stored in an upper computer;
the inlet pressure sensor 5 of the natural gas HPDI ejector 1 transmits signals to PXI2, and the PXI2 transmits processed signals to the upper computer 3;
the air pressure of the air rail 7 is monitored by the pressure control device 4;
the pressure control device 4 transmits a signal to the upper computer 3.
A wavelet transform-based online analysis method for the fuel gas injection process time of a direct injection engine in a high-pressure natural gas cylinder is characterized in that a force sensor is installed at an air hole of a natural gas HPDI (hot plug and direct injection) injector 1 and used for measuring momentum at an outlet of an orifice, and then an air injection rule is obtained through a momentum method; as shown in fig. 1; time characteristic definition of the injection process, the starting moment t of the injection 0 Time t of full opening of needle valve 1 Time t at which needle valve starts to seat 2 End of injection time t 3 . As shown in fig. 2.
A process based on wavelet transform for the time online analysis of the fuel gas injection process of a direct injection engine in a high-pressure natural gas cylinder is carried out, wherein the process 2 based on wavelet transform for pressure signals is specifically that the corresponding relation with the time characteristics of an injector cannot be found only according to the inlet pressure signals of the injector, so that the derivation is carried out on the inlet pressure signals, the correlation between the change rate of the derived inlet pressure signals and the air injection rule is analyzed, the change rate of the inlet pressure signals is obtained to be a time-varying signal, the time-varying signal is divided into three stages of rising, stabilizing and falling in the time domain, the stabilizing stage is extracted, and the time domain division of the change rate of the inlet pressure signals is realized;
the wavelet transform of a signal is defined as:
wherein a is a scale factor (scale parameter); τ is a translation factor (position parameter); p (t) is the inlet pressure signal; psi is a window function; t is time;
performing effective feature extraction on the change rate of the inlet pressure signal of the divided stationary section, wherein the effective feature extraction is mainly used for extracting time domain features and wavelet domain features; in order to adapt to the computing power of an electronic control unit of the engine, the feature dimension reduction processing is carried out, and the time domain feature is saved for facilitating the subsequent time feature identification.
An on-line analysis method for the time of the gas injection process of a high-pressure natural gas direct injection engine based on wavelet transformation adopts a principal component analysis method for dimension reduction treatment of characteristics, specifically,
data were normalized:
the new coordinate system is { w 1 ，w 2 ，…,w n In which w i Is an orthonormal base, m is a dimensionality reduction dimension, x i As sample data, mean (x) i ) Is an array mean, std (x) i ) Is the standard deviation;
sample data x i The projection in the low-dimensional space coordinate system satisfies the following conditions:
wherein z is ij Is x i Coordinates of the jth dimension in a dimension reduction coordinate system;
according to the nearest reconstruction, the distance between original sample points based on the reconstructed sample points satisfies:
in the formula, W T XX T W is the sample variance, I is the identity matrix, W is { W 1 ，w 2 ，…,w n }，tr(W T XX T W) is a matrix trace;
using the lagrange multiplier for equation (4):
XX T W＝γW (5)
for covariance matrix XX T All the eigenvalues are obtained, and the obtained eigenvalue sequence is expressed as gamma 1 ≥γ 2 ≥···≥γ m If the feature vectors corresponding to the first m feature values are the solution solved by the principal component analysis, the formula (6) is shown as follows:
W * ＝(w 1 ,w 2 ,···w m ) (6)
wherein, W * The solution is the solution of principal component analysis.
After the dimension reduction method for performing principal analysis on the signals is adopted, the main information of the data is enhanced, the characteristics are better embodied, and the subsequent time characteristic identification process is easier to realize.
A wavelet transform-based online analysis method for the time of the gas injection process of a direct injection engine in a high-pressure natural gas cylinder is characterized in that in the injection process of a natural gas injector, the change rate of an inlet pressure signal is divided into three stages, namely an ascending stage, a stable stage and a descending stage. Three of which are defined as falling (t) 0 -t 1 ) Rises (t) 2 -t 3 ) And the method is stable, and the time domain characteristics of the inlet pressure signal change rate are still kept after dimension reduction processing, so that the inlet pressure signal change rate can be used in the time characteristic identification process.
The time characteristic identification algorithm of the step 3 is specifically that the first discontinuity point of the change rate of the inlet pressure signal after wavelet transformation and dimensionality reduction is t 1 Forward search for stationary phase point t 0 (ii) a The second discontinuity is t 2 Retrieving the stationary phase t backward 3 If there is only one discontinuity, say t 1 And t 2 Coincide in the time domain, thus determining that the needle valve is not open during injectionThe needle valve is always in a moving state in the whole injection process after the needle valve is seated. The algorithm flow is shown in FIG. 3;
a wavelet transform-based online analysis method for the fuel gas injection process time of a high-pressure natural gas direct injection engine in a cylinder comprises the following steps:
step S1: collecting an inlet pressure signal of a natural gas HPDI ejector 1;
step S2: deriving the pressure signal of step S1;
step S3: performing wavelet transformation on the derivative pressure signal of step S2;
step S4: performing dimension reduction processing on the pressure signal subjected to wavelet change in the step S3;
step S5: the inlet pressure signal rate of change of the pressure signal for the dimension reduction processing at step S4 is divided into the first discontinuity t 1 And a second discontinuity t 2 ；
Step S6: first discontinuity t based on step S5 1 Forward search for stationary phase point t 0 ；
Step S7: second discontinuity t based on step S5 2 Retrieving the stationary phase t backward 3 ；
Step S8: judging the first discontinuity t of step S5 1 And a second discontinuity t 2 Whether they are equal; if equal, say t 1 And t 2 The time domain coincides, and the needle valve begins to seat without opening in the injection process; if not equal, t 1 And t 2 The time domain is not coincident, and the needle valve is opened during the injection process.
The inlet pressure sensor of fig. 1 is installed at the inlet of the injector in a clamping manner, and signals are acquired through a data acquisition module.
The inlet pressure derivative signal is subjected to wavelet transformation and dimension reduction processing, and the time characteristic is identified by the method of figure 3.
And the online analysis of the fuel gas injection process time of the direct injection engine in the high-pressure natural gas cylinder based on wavelet transformation is realized through the comparative analysis of the injection law diagram measured under the identified time characteristic collinear.
Claims (9)
1. A wavelet transform-based online analysis method for the fuel gas injection process time of a direct injection engine in a high-pressure natural gas cylinder is characterized by comprising the following steps:
step 1: installing and debugging equipment, wherein an inlet pressure sensor is installed at a gas inlet of a natural gas HPDI ejector (1), and meanwhile, a pressure sensor is installed at an air hole of the ejector;
step 2: performing wavelet transform-based processing on the pressure signals acquired by the pressure sensor installed in the step 1;
and step 3: performing a time characteristic identification algorithm based on the pressure signal processed based on the wavelet transform in the step 2;
and 4, step 4: and (4) comparing and analyzing the time characteristics obtained by time characteristic identification in the step (3) with the air injection rule measured by an off-line experiment, so as to realize the on-line analysis of the time of the fuel gas injection process of the direct injection engine in the high-pressure natural gas cylinder.
2. The wavelet-transform-based online analysis method for the fuel gas injection process time of the high-pressure natural gas direct injection engine in the cylinder is characterized in that the installation equipment in the step 1 specifically comprises a natural gas HPDI (high pressure direct injection) injector (1), a PXI (PXI) injector (2), an upper computer (3), a pressure control device (4), a pressure sensor (5), an air source (6), an air rail (7), an air compressor (8) and an amplifier module (9);
the natural gas HPDI ejector (1) is connected with PXI (2) and pressure sensor (6) respectively, pressure sensor (5) are connected with host computer (3) through amplifier module (9), PXI (2) are connected with host computer (3), pressure sensor (5) are connected with compressor (8) through gas rail (7), compressor (8) are connected with air supply (6), gas rail (7) are connected with pressure control device (4).
3. The wavelet transform-based online analysis method for the fuel gas injection process time of the high-pressure natural gas direct injection engine in the cylinder is characterized in that in the step 1, the installation equipment specifically comprises a natural gas HPDI (high pressure direct injection) injector (1), a PXI (PXI) injector (2), an upper computer (3), a pressure control device (4), a pressure sensor (5), a gas source (6), a gas rail (7) and a compressor (8);
the gas source (6) is pressurized by the compressor (8) and then ejected by the natural gas HPDI ejector (1) through the gas rail (7) and the pressure sensor (5);
the signal of the pressure sensor (5) is amplified by a charge amplifier (9); then, acquiring an inlet pressure signal through a data acquisition module in the upper computer; the collected inlet pressure signal is stored in an upper computer;
an inlet pressure sensor (5) of the natural gas HPDI ejector (1) transmits a signal to the PXI (2), and the PXI (2) transmits a processed signal to the upper computer (3);
the air pressure of the air rail (7) is monitored by a pressure control device (4);
the pressure control device (4) transmits signals to the upper computer (3).
4. The wavelet-transform-based online analysis method for the fuel gas injection process time of the high-pressure natural gas direct injection engine in the cylinder is characterized in that a force sensor is installed at an air hole of a natural gas HPDI (high pressure direct injection) injector (1), the injection process time characteristic is defined, and the injection starting time t is defined 0 Time t of full opening of needle valve 1 Time t at which needle valve starts to seat 2 End of injection time t 3 。
5. The wavelet-transform-based on-line analysis method for fuel gas injection process time of high-pressure natural gas direct injection engine according to claim 1, wherein the wavelet-transform-based processing of the pressure signal in the step 2 is specifically to derive the inlet pressure signal, analyze the correlation between the change rate of the derived inlet pressure signal and the injection law, obtain that the change rate of the inlet pressure signal is a time-varying signal, divide the time domain into three stages of rising, stationary and falling, extract the stationary stage,
the wavelet transform of a signal is defined as:
wherein a is a scaling factor; τ is a translation factor; p (t) is the inlet pressure signal; psi is a window function; t is time;
extracting time domain characteristics and wavelet domain characteristics of the inlet pressure signal change rate of the divided stationary section; in order to adapt to the computing power of an electronic control unit of the engine, the feature dimension reduction processing is carried out, and the time domain feature is saved for facilitating the subsequent time feature identification.
6. The wavelet transform-based online analysis method for the fuel gas injection process time of the high-pressure natural gas direct injection engine is characterized in that the characteristic dimension reduction processing is performed by adopting a principal component analysis method, and specifically,
normalizing the data:
the new coordinate system is { w 1 ，w 2 ，…,w n In which w i Is an orthonormal base, m is a dimensionality reduction dimension, x i Is the sample data, mean (x) i ) Is an array mean, std (x) i ) Is the standard deviation;
sample data x i The projection in the low-dimensional space coordinate system satisfies the following conditions:
wherein z is ij Is x i A j-th coordinate in a dimension reduction coordinate system;
according to the nearest reconstruction, the distance between original sample points based on the reconstructed sample points satisfies:
in the formula, W T XX T W is the sample variance, I is the identity matrix, W is { W 1 ，w 2 ，…,w n }，tr(W T XX T W) is a matrix trace;
using the lagrange multiplier for equation (4):
XX T W＝γW (5)
for covariance matrix XX T All the eigenvalues are obtained, and the obtained eigenvalue sequence is expressed as gamma 1 ≥γ 2 ≥···≥γ m If the feature vectors corresponding to the first m feature values are the solution solved by the principal component analysis, the formula (6) is shown as follows:
W * ＝(w 1 ,w 2 ,···w m ) (6)
wherein, W * The solution is the solution of principal component analysis.
7. The wavelet-transform-based online analysis method for the fuel gas injection process time of the high-pressure natural gas direct injection engine in the cylinder is characterized in that in the injection process of the natural gas injector, the change rate of an inlet pressure signal is divided into three stages, namely a rising stage, a stable stage and a falling stage; three of which are defined as falling (t) 0 -t 1 ) Rises (t) 2 -t 3 ) And the inlet pressure signal change rate is stable, and the time domain characteristics of the inlet pressure signal change rate are still kept after dimension reduction processing and are used in the time characteristic identification process.
8. The wavelet-transform-based on-line analysis method for time of gas injection process of direct injection engine in cylinder of high pressure natural gas, according to claim 7, characterized in that the time characteristic identification algorithm in step 3 is specifically that the first discontinuity point of change rate of inlet pressure signal after wavelet transform and dimensionality reduction is t 1 Forward search for stationary phase point t 0 (ii) a Second discontinuityIs t 2 Retrieving the stationary phase t backward 3 If there is only one discontinuity, say t 1 And t 2 Overlap in time domain, so it is determined that the needle begins to seat without opening during injection, and the needle is in motion throughout the injection.
9. The wavelet transform-based online analysis method for the fuel gas injection process time of the high-pressure natural gas direct injection engine is characterized by comprising the following steps of:
step S1: collecting natural gas HPDI ejector (1) inlet pressure signal;
step S2: deriving the pressure signal of step S1;
step S3: performing wavelet transformation on the derivative pressure signal of step S2;
step S4: performing dimension reduction processing on the pressure signal subjected to wavelet change in the step S3;
step S5: the inlet pressure signal change rate of the pressure signal of the dimensionality reduction processing of step S4 is divided into a first discontinuity point t 1 And a second discontinuity t 2 ；
Step S6: first discontinuity t based on step S5 1 Forward search for stationary phase point t 0 ；
Step S7: second discontinuity t based on step S5 2 Retrieving the stationary phase t backward 3 ；
Step S8: judging the first discontinuity t of step S5 1 And a second discontinuity t 2 Whether they are equal; if equal, say t 1 And t 2 The time domain coincides, and the needle valve begins to seat without opening in the injection process; if not, t is indicated 1 And t 2 Not coinciding in time domain, the needle valve is already open during injection.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210380960.XA CN114810448B (en) | 2022-04-12 | 2022-04-12 | Method for online analysis of fuel gas injection process time of direct injection engine in high-pressure natural cylinder based on wavelet transformation |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210380960.XA CN114810448B (en) | 2022-04-12 | 2022-04-12 | Method for online analysis of fuel gas injection process time of direct injection engine in high-pressure natural cylinder based on wavelet transformation |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN114810448A true CN114810448A (en) | 2022-07-29 |
| CN114810448B CN114810448B (en) | 2023-06-13 |
Family
ID=82533994
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202210380960.XA Active CN114810448B (en) | 2022-04-12 | 2022-04-12 | Method for online analysis of fuel gas injection process time of direct injection engine in high-pressure natural cylinder based on wavelet transformation |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN114810448B (en) |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20140251272A1 (en) * | 2013-03-07 | 2014-09-11 | Cummins Inc. | Hydraulic start-of-injection detecting system and method |
| CN110500217A (en) * | 2019-07-23 | 2019-11-26 | 南京航空航天大学 | Based on can measured data feature common rail for diesel engine system oil pump fault detection method |
| CN111520267A (en) * | 2019-12-30 | 2020-08-11 | 哈尔滨工程大学 | Common rail fuel injector fault diagnosis method based on FOA-VMD and HDE |
| CN112761843A (en) * | 2021-02-18 | 2021-05-07 | 哈尔滨工程大学 | Common rail oil injector fault diagnosis method |
-
2022
- 2022-04-12 CN CN202210380960.XA patent/CN114810448B/en active Active
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20140251272A1 (en) * | 2013-03-07 | 2014-09-11 | Cummins Inc. | Hydraulic start-of-injection detecting system and method |
| CN110500217A (en) * | 2019-07-23 | 2019-11-26 | 南京航空航天大学 | Based on can measured data feature common rail for diesel engine system oil pump fault detection method |
| CN111520267A (en) * | 2019-12-30 | 2020-08-11 | 哈尔滨工程大学 | Common rail fuel injector fault diagnosis method based on FOA-VMD and HDE |
| CN112761843A (en) * | 2021-02-18 | 2021-05-07 | 哈尔滨工程大学 | Common rail oil injector fault diagnosis method |
Also Published As
| Publication number | Publication date |
|---|---|
| CN114810448B (en) | 2023-06-13 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN105003372B (en) | A kind of distributive value method for diagnosing faults based on the observation of rail pressure waveform feature parameter | |
| CN105003373B (en) | A kind of injection timing method for diagnosing faults based on the observation of rail pressure waveform feature parameter | |
| Shen et al. | Gaussian mixture model clustering-based knock threshold learning in automotive engines | |
| CN102095588A (en) | Intelligent fault diagnosis method for diesel engine | |
| Zhao et al. | K-means clustering-driven detection of time-resolved vortex patterns and cyclic variations inside a direct injection engine | |
| CN103900824A (en) | Method for diagnosing faults of diesel engine based on instant rotary speed clustering analysis | |
| Gu et al. | Misfire fault diagnosis method for diesel engine based on MEMD and dispersion entropy | |
| Hellring et al. | Spark advance control using the ion current and neural soft sensors | |
| CN108414230B (en) | A kind of internal combustion engine preignition detecting method | |
| CN110259616B (en) | Diesel engine common rail system fuel injector fault detection method based on measurable data characteristics | |
| Bi et al. | A variety of engine faults detection based on optimized variational mode decomposition-robust independent component analysis and fuzzy C-mean clustering | |
| Ke et al. | Multiscale bidirectional diversity entropy for diesel injector fault-type diagnosis and fault degree diagnosis | |
| CN114810448B (en) | Method for online analysis of fuel gas injection process time of direct injection engine in high-pressure natural cylinder based on wavelet transformation | |
| Rosas et al. | Knock detection method for dual-fuel compression ignition engines based on block vibration analysis | |
| CN113884305B (en) | Diesel engine assembly cold test detection method and system based on SVM | |
| Wang et al. | Effect of port gas injection on the combustion instabilities in a spark-ignition lean-burn natural gas engine | |
| CN105606367A (en) | Misfire detection self-adaptive determination method and device under stable condition of engine | |
| CN113062811B (en) | Method for identifying key time characteristics of oil injection process according to frequency spectrum characteristics of pressure signal at inlet of oil injector | |
| Singh et al. | An Evaluation of Knock Determination Techniques for Diesel-Natural Gas Dual Fuel Engines | |
| Bi et al. | Knock detection based on recursive variational mode decomposition and multilevel semi-supervised local fisher discriminant analysis | |
| CN106884729B (en) | The method of detonation signal detection based on the separation of compression pressure change rate | |
| CN112267998B (en) | Method for testing highest temperature in cylinder and nonuniformity of cylinders of multi-cylinder diesel engine | |
| CN114722334B (en) | STFT-based online identification method for gas injection time characteristics of high-pressure natural gas direct injection engine | |
| Hosseini et al. | Knock probability determination employing convolutional neural network and IGTD algorithm | |
| Aramburu et al. | Knock detection in spark ignited heavy duty engines: An application of machine learning techniques with various knock sensor locations |
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 |