CN115901281A - Fault diagnosis method and system for gas injection valve of LNG engine - Google Patents
Fault diagnosis method and system for gas injection valve of LNG engine Download PDFInfo
- Publication number
- CN115901281A CN115901281A CN202211130142.0A CN202211130142A CN115901281A CN 115901281 A CN115901281 A CN 115901281A CN 202211130142 A CN202211130142 A CN 202211130142A CN 115901281 A CN115901281 A CN 115901281A
- Authority
- CN
- China
- Prior art keywords
- signal
- fault
- fault diagnosis
- injection valve
- gas injection
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
- 238000003745 diagnosis Methods 0.000 title claims abstract description 77
- 238000002347 injection Methods 0.000 title claims abstract description 44
- 239000007924 injection Substances 0.000 title claims abstract description 44
- 238000000034 method Methods 0.000 title claims abstract description 41
- 238000004422 calculation algorithm Methods 0.000 claims abstract description 30
- 238000012549 training Methods 0.000 claims abstract description 24
- 239000013598 vector Substances 0.000 claims abstract description 21
- 238000001914 filtration Methods 0.000 claims abstract description 16
- 238000012360 testing method Methods 0.000 claims abstract description 14
- 238000012545 processing Methods 0.000 claims abstract description 13
- 238000004458 analytical method Methods 0.000 claims description 13
- 238000005457 optimization Methods 0.000 claims description 9
- 241000272168 Laridae Species 0.000 claims description 8
- 238000000354 decomposition reaction Methods 0.000 claims description 8
- 238000012843 least square support vector machine Methods 0.000 claims description 8
- 238000003909 pattern recognition Methods 0.000 claims description 6
- 238000004364 calculation method Methods 0.000 claims description 4
- 238000004590 computer program Methods 0.000 claims description 4
- 239000011159 matrix material Substances 0.000 claims description 3
- 238000012935 Averaging Methods 0.000 claims description 2
- 238000001228 spectrum Methods 0.000 claims description 2
- 230000000737 periodic effect Effects 0.000 abstract description 9
- 239000000446 fuel Substances 0.000 abstract description 5
- 239000007789 gas Substances 0.000 description 32
- 230000000694 effects Effects 0.000 description 8
- 208000036075 Autosomal dominant tubulointerstitial kidney disease Diseases 0.000 description 7
- 101001133056 Homo sapiens Mucin-1 Proteins 0.000 description 7
- 102100034256 Mucin-1 Human genes 0.000 description 7
- 239000003795 chemical substances by application Substances 0.000 description 5
- 230000007547 defect Effects 0.000 description 5
- 230000006399 behavior Effects 0.000 description 4
- 238000000605 extraction Methods 0.000 description 4
- 230000008569 process Effects 0.000 description 4
- 241000272161 Charadriiformes Species 0.000 description 3
- 230000008901 benefit Effects 0.000 description 3
- 239000002737 fuel gas Substances 0.000 description 3
- 230000005012 migration Effects 0.000 description 3
- 238000013508 migration Methods 0.000 description 3
- 230000009467 reduction Effects 0.000 description 3
- 239000000243 solution Substances 0.000 description 3
- 230000001174 ascending effect Effects 0.000 description 2
- 238000000889 atomisation Methods 0.000 description 2
- 238000002485 combustion reaction Methods 0.000 description 2
- 238000010586 diagram Methods 0.000 description 2
- 238000012804 iterative process Methods 0.000 description 2
- 238000012986 modification Methods 0.000 description 2
- 230000004048 modification Effects 0.000 description 2
- 230000002159 abnormal effect Effects 0.000 description 1
- 230000009471 action Effects 0.000 description 1
- 230000003044 adaptive effect Effects 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 230000008859 change Effects 0.000 description 1
- 230000007423 decrease Effects 0.000 description 1
- 230000006866 deterioration Effects 0.000 description 1
- 230000002708 enhancing effect Effects 0.000 description 1
- 230000006870 function Effects 0.000 description 1
- 230000004927 fusion Effects 0.000 description 1
- 230000006872 improvement Effects 0.000 description 1
- 238000005259 measurement Methods 0.000 description 1
- 230000000877 morphologic effect Effects 0.000 description 1
- 230000003287 optical effect Effects 0.000 description 1
- 230000011218 segmentation Effects 0.000 description 1
- 230000035945 sensitivity Effects 0.000 description 1
- 230000035939 shock Effects 0.000 description 1
- 238000004088 simulation Methods 0.000 description 1
- 239000000779 smoke Substances 0.000 description 1
- 230000009466 transformation Effects 0.000 description 1
- 230000001052 transient effect Effects 0.000 description 1
- 230000007704 transition Effects 0.000 description 1
Images
Classifications
-
- 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
- Electrical Control Of Air Or Fuel Supplied To Internal-Combustion Engine (AREA)
Abstract
The invention provides a method and a system for diagnosing faults of a gas injection valve of an LNG engine. The method comprises the following steps: acquiring a pressure fluctuation signal of an air supply pipe, and dividing the acquired pressure fluctuation signal into a training signal and a test signal; filtering the pressure fluctuation signal by using an SOA-CYCBD algorithm to obtain a pressure signal with enhanced impact components; processing the filtered pressure signal to obtain fault characteristics; and after the fault characteristic vector is trained, fault diagnosis and mode recognition are carried out, and a diagnosis result is output. The technical scheme of the invention is suitable for fault diagnosis of the LNG engine fuel injection valve in the field industrial environment with strong noise interference, can reduce the noise interference, enhance the periodic impact component and improve the fault diagnosis precision of the LNG engine fuel injection valve.
Description
Technical Field
The invention relates to the technical field of mechanical fault diagnosis, in particular to a fault diagnosis method and system for a fuel gas injection valve of an LNG engine.
Background
The quality of the fuel atomization is determined by the working quality of the gas injection valve of the LNG engine, so that the whole combustion process is influenced. If the gas injection valve breaks down, poor fuel atomization can cause combustion deterioration, and the problems of engine power reduction, oil consumption increase, black smoke emission, excessive emission, difficult starting, even abnormal operation and the like are caused. It is therefore necessary to study and analyze the early failure phenomenon of the gas injection valve.
In order to extract early failure characteristics of mechanical systems, researchers have proposed many effective methods such as wavelet transformation, empirical mode decomposition, fuzzy theory, morphological filtering, and Blind Deconvolution (BD). The process in which the source signal excited by the fault of the gas injection valve is transmitted to the sensor through the signal channel can be regarded as a convolution process of the source signal and the signal channel. The principle of blind deconvolution theory is to extract the fault pulse by solving a deconvolution filter that maximizes or minimizes the convolution target. Therefore, blind deconvolution has unique advantages for dealing with gas injection valve failure signals. Since the Minimum Entropy De-convolution (MED) is proposed by Ralph Wiggins, the fault diagnosis method based on deconvolution draws the attention of many experts and scholars, and quickly promotes the application of the deconvolution method in the field of fault diagnosis. However, the MED often has only one or a few pulses, which is prone to other shock losses. To avoid this problem, mcdonald proposes an MCKD (Maximum corrected Kurtosis deconstruction, MCKD) algorithm. MCKD highlights continuous impact pulses submerged by noise through deconvolution operation, improves correlation kurtosis value of original signals, and is suitable for extracting continuous transient impact of weak fault signals. However, although the MCKD can extract periodic pulses, it can extract only a limited number of pulses, and the number of shifts greatly limits the filtering effect of the MCKD. Mcdonald therefore proposes a Minimum Entropy Deconvolution of multipoint optimum adjustment (MOMEDA). The process of solving the inverse filter by MOMEDA is a non-iterative process, so that the operation time of the algorithm is reduced. However, MOMEDA can reduce the amplitude of pulses in the signal greatly while reducing noise.
Based on this, marco Buzzoni proposed a new Deconvolution method, maximum cyclostationary Deconvolution (CYCBD). CYCBD overcomes the defects that MED recovers a single dominant pulse and MCKD can only extract a limited number of pulses, and can well extract continuous periodic pulses. Compared with MOMEDA, the extracted periodic impact can enhance impact at the same time, and the noise reduction performance is good. But similar to MED, MCKD, MOMEDA, which recovers the source of the fault by deconvoluting through solving a finite filter, the length of the filter has a large impact on its results. The SOA (SOA) algorithm is a novel group intelligent optimization algorithm proposed by Gaurav Dhiman in 2019, and the algorithm mainly simulates gull migration in nature and attack behaviors in the migration process. Compared with other optimization algorithms, the SOA algorithm can solve the problem of large-scale constraint and has high competitiveness. Moreover, due to the simple algorithm, the global optimal solution can be obtained, the searching precision and the searching efficiency are high, and the method has a good application prospect in the aspect of parameter self-optimization. In view of this, the patent adaptively seeks CYCBD optimum filter lengths using SOA algorithms. The optimal CYCBD filter is used for filtering fault signals of the oil injector, noise interference is reduced, and continuous impact components submerged in noise are highlighted.
After the fault signal is filtered, the feature extraction of the fault signal is a key step of fault diagnosis. In recent years, numerous methods for measuring the nonlinear time series complexity of mechanical dynamics systems have been proposed in succession and applied to the field of fault diagnosis, such as approximate entropy, sample entropy, fuzzy entropy, permutation entropy, and the like. Entropy (PE) quantifies dynamic changes based on ordered patterns of time series structures, and PE has found widespread application in time series complexity analysis due to its theoretical simplicity and fast computational power. However, the PE algorithm only uses the ordinal number structure of the time series and ignores the amplitude information, so Bilal et al propose Weighted Entropy (WPE) based on PE. However, WPE only considers complexity of time series on a single scale and ignores useful information on other scales, so YIN et al combines WPE with multi-scale Entropy to propose multi-scale Weighted Permutation Entropy (MWPE), but the multi-scale Weighted Permutation Entropy only considers low frequency components of the time series and ignores high frequency parts of the time series. Based on the above, the patent proposes a Hierarchical Weighted Permutation Entropy (HWPE) based on Hierarchical analysis and Weighted Permutation Entropy fusion, which can not only consider the high-frequency and low-frequency components of the original sequence, but also improve the anti-interference performance and the sensitivity of signal bandwidth change. The effectiveness and superiority of the proposed method are verified through the analysis of the simulation signals and the test data.
Through the above analysis, the problems and defects of the prior art are as follows:
1. although the conventional weak fault diagnosis method can effectively enhance the periodic pulse, the preset parameters of the algorithm have decisive influence on the enhancement effect, the inaccurate selection of the preset parameters leads to unsatisfactory enhancement effect, and the weak fault diagnosis precision is low.
2. The existing fault feature extraction method only considers the complexity of time sequences on a low scale, ignores useful information on other scales, easily ignores effective fault information, and is difficult to improve the fault diagnosis precision.
The difficulty in solving the above problems and defects is:
1. aiming at different weak fault signals, how to realize the diagnosis and identification of the weak fault by using a weak fault diagnosis method in a self-adaptive manner is an urgent problem to be solved.
2. The existing multi-scale analysis method can only consider low-frequency components, but cannot consider high-frequency components, so that a new method needs to be developed to consider both the low-frequency components and the high-frequency components.
The significance for solving the problems and the defects is as follows:
the algorithm provided by the patent is used for adaptively enhancing weak impact, reducing noise interference, highlighting continuous impact components submerged in noise, analyzing enhanced fault signals from the angles of high frequency and low frequency, extracting effective fault characteristics of the LNG engine gas injection valve, improving fault diagnosis and identification precision, and providing the LNG engine gas injection valve fault diagnosis method suitable for the field industrial environment with strong noise interference.
Disclosure of Invention
Therefore, the technical problem to be solved by the invention is to overcome the defects that although the commonly used weak fault diagnosis method in the prior art can effectively enhance the periodic pulse, the preset parameters of the algorithm have decisive influence on the enhancement effect, the inaccurate selection of the preset parameters causes the unsatisfactory enhancement effect, and the weak fault diagnosis precision is low; the existing fault feature extraction method only considers the complexity of a time sequence on a low scale, ignores useful information on other scales, easily ignores effective fault information, and is difficult to improve the fault diagnosis precision, so that the LNG engine gas injection valve fault diagnosis method and the system which are suitable for the LNG engine gas injection valve fault diagnosis under the field industrial environment with strong noise interference are provided, the noise interference can be reduced, the periodic impact component can be enhanced, and the LNG engine gas injection valve fault diagnosis precision can be improved.
In order to solve the above problems, the present invention provides a method for diagnosing a fault of a gas injection valve of an LNG engine, comprising:
s1, acquiring a pressure fluctuation signal of a gas supply pipe, and dividing the acquired pressure fluctuation signal into a training signal and a test signal;
s2, filtering the pressure fluctuation signal by using an SOA-CYCBD algorithm to obtain a pressure signal with enhanced impact components;
s3, processing the filtered pressure signal to obtain fault characteristics;
and S4, after the fault characteristic vector is trained, fault diagnosis and mode recognition are carried out, and a diagnosis result is output.
Optionally, the processing the filtered pressure signal in step S3 specifically includes the following steps:
performing hierarchical analysis on the filtered pressure signal;
and calculating the weighted arrangement entropy of each layer to form a fault feature subset.
Optionally, in step S4, training is performed by using the processed pressure signal as a fault feature vector, fault diagnosis and pattern recognition are performed, and a diagnosis result is output, which specifically includes:
inputting all the hierarchical weighted permutation entropies of the training pressure signals serving as training samples into a multi-classifier of a least square support vector machine for training;
and carrying out fault diagnosis and pattern recognition on the hierarchical weighted arrangement entropy of the test signal sample by adopting the trained least square support vector machine multi-classifier, and outputting a diagnosis result.
Optionally, the SOA-CYCBD in step S2 is to use a gull optimization algorithm to find the optimal values of the CYCBD parameter filter length L and the fault period T with the harmonic significance index HSI as an objective function, where the HSI is calculated as follows:
where ω is frequency, F (ω) represents the amplitude at frequency ω in the envelope spectrum, Q is the largest harmonic used for calculation, and usually Q =5, n (ω) is the noise amount around frequency ω, which can be calculated by a moving average filter and the harmonic amplitude needs to be removed in advance; q is a constant with a value range of [1, Q ]. p (ω) = F (ω)/N (ω) is an amplitude ratio for representing the degree of significance of a specific frequency component.
Optionally, the hierarchical analysis is performed on the filtered pressure signal, specifically, the hierarchical analysis is performed on the pressure signal with the signal length of N:
based on vectors
The node components defining the decomposition of each layer of the time series u (i) are as follows:
Q γn as average operator, gamma n =0 or 1,Q 0 And Q 1 The following:
wherein N =2 n And n is a positive integer. Operator Q 0 Sum operator Q 1 Has a length of 2 n-1 (ii) a k is the number of decomposition layers, u k,e Constructing an n-dimensional vector [ gamma ] for the node components of the k-layer decomposition of the time sequence u, e being a positive integer 1 ,γ 2 ,...,γ n ]E {0,1}, then the integer e can be expressed as:
wherein the vector corresponding to the positive integer e is [ gamma ] 1 ,γ 2 ,...,γ n ]。
Optionally, the weighted permutation entropy of each level is calculated, and the calculation result of the entropy value is as follows:
HWPE＝E(u k,e ,m,d)＝[E 1 ,E 2 ,...,E e ] T (7)
where m is the embedding dimension, d is the time delay, T is the matrix transpose, E 1 ,E 2 ,...,E e Entropy values are arranged for weighting.
The invention also provides a fault diagnosis system for the gas injection valve of the LNG engine, which comprises the following components:
the signal acquisition system is used for acquiring a pressure fluctuation signal of the gas supply pipe and dividing the acquired pressure fluctuation signal into a training signal and a test signal;
the filtering system is used for filtering the pressure fluctuation signal by using an SOA-CYCBD algorithm to obtain a pressure signal with enhanced impact components;
the signal processing system is used for processing the filtered pressure signal to obtain fault characteristics;
and the fault analysis system is used for carrying out fault diagnosis and pattern recognition after training the fault feature vector and outputting a diagnosis result.
The present invention also provides a computer readable storage medium having stored thereon a computer program which, when being executed by a processor, implements the LNG engine gas injection valve fault diagnosis method as described above.
The invention has the following advantages:
firstly, the SOA-CYCBD is effectively utilized to carry out filtering processing on the pressure signal of the fuel gas injection valve of the LNG engine in a self-adaptive manner, so that the periodic impact component is enhanced, and the noise interference is reduced;
and secondly, the fault information of the pressure signal of the gas supply pipe is comprehensively and accurately reflected through the hierarchical weighted arrangement entropy, the fault diagnosis of the gas injection valve of the LNG engine is completed under the strong noise environment, the fault diagnosis rate of the gas injection valve is improved, and the misdiagnosis rate and the missed diagnosis rate are reduced.
Drawings
In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings used in the description of the embodiments or the prior art will be briefly described below, and it is obvious that the drawings in the following description are some embodiments of the present invention, and other drawings can be obtained by those skilled in the art without creative efforts.
FIG. 1 is a flow chart of an LNG engine gas injection valve fault diagnosis algorithm based on SOA-CYCBD and hierarchical weighted arrangement entropy of the invention;
FIG. 2 is a flow chart of the SOA-CYCBD optimization algorithm of the present invention;
FIG. 3 is a fault diagnosis result diagram of an LNG engine fuel gas injection valve based on SOA-CYCBD and hierarchical weighted arrangement entropy;
FIG. 4 is a pictorial view of the complete machine test stand of the present invention;
FIG. 5 is a pictorial view of the sensor mounting location of the present invention;
FIG. 6 is a graph showing a typical failed high pressure supply conduit vibration signal of the present invention;
FIG. 7 is one of the flow charts of the steps of the LNG engine gas injection valve fault diagnosis method of the present invention;
FIG. 8 is a second flowchart illustrating the steps of the LNG engine fuel injection valve fault diagnosis method of the present invention;
fig. 9 is a connection schematic diagram of the LNG engine gas injection valve fault diagnosis system of the present invention.
Detailed Description
The technical solutions of the present invention will be described clearly and completely with reference to the accompanying drawings, and it is to be understood 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 derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present invention.
Summary of the application
Although the commonly used weak fault diagnosis method can effectively enhance the periodic pulse, the preset parameters of the algorithm have decisive influence on the enhancement effect, the inaccurate selection of the preset parameters leads to the unsatisfactory enhancement effect, and the weak fault diagnosis precision is low. The existing fault feature extraction method only considers the complexity of time sequences on a low scale, ignores useful information on other scales, easily ignores effective fault information, and is difficult to improve fault diagnosis precision.
Fault diagnosis method for gas injection valve of exemplary LNG engine
Fig. 7 is a flowchart of a method for diagnosing a fault of a gas injection valve of an LNG engine, which includes:
s1, acquiring a pressure fluctuation signal of a gas supply pipe, and dividing the acquired pressure fluctuation signal into a training signal and a test signal;
s2, filtering the pressure fluctuation signal by using an SOA-CYCBD algorithm to obtain a pressure signal with enhanced impact components;
s3, processing the filtered pressure signal to obtain fault characteristics;
and S4, training the fault feature vector, performing fault diagnosis and pattern recognition, and outputting a diagnosis result.
The first stage of the fault diagnosis method for the gas injection valve of the LNG engine comprises the following steps: acquiring a pressure fluctuation signal by using a pressure sensor arranged on an air supply pipe, and initially establishing a sample set of a training signal and a test signal; the second stage is as follows: filtering the pressure fluctuation signal by using an SOA-CYCBD algorithm to obtain a pressure signal with enhanced impact components; the third stage is as follows: calculating the hierarchical weighted arrangement entropy of the filtered pressure signals, and taking the hierarchical weighted arrangement entropy as the fault characteristics of the pressure signals; the fourth stage is as follows: inputting the hierarchical weighted arrangement entropies of all training samples as feature vectors into a multi-classifier of a least square support vector machine for training; the fifth stage is: and performing fault diagnosis and mode recognition on the hierarchical weighted arrangement entropy of the test sample by adopting the trained multi-classifier of the least square support vector machine, and outputting a diagnosis result.
The pressure sensor is arranged on the air supply pipe to measure the pressure state parameters of the air supply pipe, so that accurate measurement of data can be realized, and the accuracy of fault diagnosis is improved.
Thus, the acquired pressure fluctuation signal is used to form a training signal sample set and a test signal sample set.
Specifically, step S2, using SOA-CYCBD algorithm to carry out adaptive filtering processing on the pressure signal, and obtaining a filtering signal with enhanced impact components. SOA-CYCBD optimization algorithm flow As shown in FIG. 2, the migration behavior of gull is modeled as:
D＝|C+M| (8)
where D represents the distance between the search agent and the best-fit agent, and C is defined as:
C＝A×P(t) (9)
where P (t) is the position of the gull in the t-th generation, and parameter A represents the movement behavior of the search agent in a given search space, which is used to avoid collision between gulls. Also, A is a parameter that decreases linearly as the iteration progresses
A＝a-(t×(a/MAX iteration )) (10)
Where a is a constant that is used to control the frequency of the variable a, and t represents the algebraic number of the current iteration. M is defined as
M＝B×(Pbs(t)-P(t)) (11)
Wherein, M represents the direction of the best position, pbs (t) is the best gull of the tth generation, B is a random parameter responsible for balancing the global search and the local search, and is calculated as follows:
B＝2×A 2 ×rd (12)
wherein rd is a random parameter between [0,1 ]. Upon arrival at a new location, the gulls attack their prey with a spiral action. The attacking behavior of gulls is modeled as
P(t)＝(D×x×y×z)+Pbs(t) (13)
Wherein,
x＝r×cos(θ) (14)
y＝r×sinθ( (15)
z＝r×θ (16)
r＝u×e θv (17)
where u and v are constants, e is the base of the natural logarithm, and θ is a random number between [0,2 π ]. The gull optimization algorithm starts with randomly generated populations and in an iterative process, the search agents can update their locations according to the best search agent. To smoothly transition between global and local searches, the variable B is employed. Therefore, the gull optimization algorithm has better global search and local search capabilities.
Specifically, as shown in fig. 8, the LNG engine gas injection valve fault diagnosis system method further includes:
the step S3 of processing the filtered pressure signal specifically includes the following steps:
step S301: performing hierarchical analysis on the filtered pressure signal;
given aA time series of length N { u (i), i =1, 2., N }, defining an averaging operator Q 0 And Q 1 The following were used:
wherein N =2 n And n is a positive integer. Operator Q 0 Sum operator Q 1 Has a length of 2 n-1 . According to the average operator Q 0 And Q 1 The original sequence can be reconstructed into
u＝{(Q 0 (u) j +Q 1 (u) j ),(Q 0 (u) j -Q 1 (u) j )},j＝0,1,2,...,2 n-1
When j =0 or j =1, a matrix Q is defined j The operators are as follows
Step S302: calculating the weighted arrangement entropy of each layer to form a fault feature subset:
the second step: construct an n-dimensional vector [ gamma ] 1 ,γ 2 ,...,γ n ]E {0,1}, then the integer e can be expressed as
Wherein the vector corresponding to the positive integer e is [ gamma ] 1 ,γ 2 ,...,γ n ]。
The third step: based on the vector [ gamma ] 1 ,γ 2 ,...,γ n ]The node components of each layer decomposition defining the time series u (i) are as follows
Where k denotes the k layers in the hierarchical segmentation, and the original time series u (i) is divided into u for the low-frequency and high-frequency parts of the k +1 layer k,0 And u k,1 And (4) showing.
The fourth step: for time series u k,e Performing phase space reconstruction to obtain a series of subsequences
Where m is the embedding dimension, τ is the time delay, K is the number of reconstructed components, and K = N- (m-1) τ, N is the time series length. The elements in each reconstruction subsequence are arranged in ascending order according to the magnitude of the numerical value, and a group of symbol sequences pi can be obtained after each reconstruction subsequence is arranged in ascending order i ＝[k 1 ,k 2 ,...,k m ]。
Calculating the weight value omega of each subsequence i ；
Arbitrary subsequence
The characteristic information of (A) is represented by a weight value omega i And arrangement pattern pi i And (4) performing representation. There are K permutation patterns in total for the time series U, each permutation pattern pi i The weighted probability value of (a);
calculating a weighted arrangement entropy WPE value of the time series U according to the definition of the information entropy;
the hierarchical discrete entropy may be expressed as:
HWPE＝E(u k,e ,m,d)＝[E 1 ,E 2 ,...,E e ] T 。
and S4, inputting the feature vector into a multi-classifier of the least square support vector machine for training by taking the hierarchical weighted arrangement entropies of all the training samples as the feature vector.
And S5, carrying out fault diagnosis and mode recognition on the hierarchical weighted arrangement entropy of the test sample by adopting the trained least square support vector machine multi-classifier, and outputting a diagnosis result, wherein the classification result is shown in figure 3.
As shown in fig. 9, the fault diagnosis system for the gas injection valve of the LNG engine according to the present invention sequentially obtains a voltage fluctuation signal, a voltage filtering, a fault feature analysis and training, a fault diagnosis, and an output result, thereby achieving reduction of noise interference, enhancement of a cycle impact component, and improvement of the fault diagnosis accuracy for the gas injection valve of the LNG engine.
Exemplary computer readable storage Medium
The computer readable storage medium has a computer program stored thereon, and the computer program, when executed by a processor, implements the LNG engine gas injection valve fault diagnosis method described above.
The computer readable storage medium can implement the steps of the method for diagnosing the fault of the gas injection valve of the LNG engine in any of the embodiments, and can achieve the same technical effects, so that all the beneficial effects of any of the embodiments are achieved, and the detailed description is omitted here.
The computer-readable storage medium in this embodiment is, for example, a Read-Only Memory (ROM), a Random Access Memory (RAM), a magnetic disk or an optical disk.
According to the above description, the present application has the following advantages:
1. the algorithm provided by the patent is utilized to adaptively enhance weak impact, reduce noise interference, highlight continuous impact components submerged in noise,
2. and analyzing the enhanced fault signal from the high-frequency and low-frequency angles, extracting effective fault characteristics of the gas injection valve of the LNG engine, and improving the fault diagnosis and identification precision.
In the description of the present invention, it should be noted that the terms "center", "upper", "lower", "left", "right", "vertical", "horizontal", "inner", "outer", etc. indicate orientations or positional relationships based on the orientations or positional relationships shown in the drawings, and are only for convenience of description and simplification of description, but do not indicate or imply that the device or element referred to must have a specific orientation, be constructed and operated in a specific orientation, and thus, should not be construed as limiting the present invention. Furthermore, the terms "first," "second," and "third" are used for descriptive purposes only and are not to be construed as indicating or implying relative importance.
In the description of the present invention, it should be noted that, unless otherwise explicitly specified or limited, the terms "mounted," "connected," and "connected" are to be construed broadly and may be, for example, fixedly connected, detachably connected, or integrally connected; can be mechanically or electrically connected; they may be connected directly or indirectly through intervening media, or they may be interconnected between two elements. The specific meanings of the above terms in the present invention can be understood by those skilled in the art according to specific situations.
In addition, the technical features involved in the different embodiments of the present invention described below may be combined with each other as long as they do not conflict with each other.
It should be understood that the above examples are only for clarity of illustration and are not intended to limit the embodiments. Other variations and modifications will be apparent to persons skilled in the art in light of the above description. And are neither required nor exhaustive of all embodiments. And obvious variations or modifications derived therefrom are intended to be within the scope of the invention.
Claims (8)
1. A fault diagnosis method for a gas injection valve of an LNG engine is characterized by comprising the following steps:
acquiring a pressure fluctuation signal of a gas supply pipe, and dividing the acquired pressure fluctuation signal into a training signal and a test signal;
filtering the pressure fluctuation signal by using an SOA-CYCBD algorithm to obtain a pressure signal with enhanced impact components;
processing the filtered pressure signal to obtain fault characteristics;
and after the fault characteristic vector is trained, fault diagnosis and mode recognition are carried out, and a diagnosis result is output.
2. The LNG engine gas injection valve fault diagnosis method according to claim 1, wherein the step S3 of processing the filtered pressure signal specifically comprises the steps of:
performing hierarchical analysis on the filtered pressure signal;
and calculating the weighted arrangement entropy of each layer to form a fault feature subset.
3. The LNG engine gas injection valve fault diagnosis method according to claim 2, wherein after the training of the fault feature vector in step S4, fault diagnosis and pattern recognition are performed, and a diagnosis result is output, and the method specifically includes:
inputting the hierarchical weighted arrangement entropies of all training pressure signals serving as training samples into a multi-classifier of a least square support vector machine for training by taking the hierarchical weighted arrangement entropies as feature vectors;
and performing fault diagnosis and mode recognition on the hierarchical weighted arrangement entropy of the test signal sample by adopting the trained multi-classifier of the least square support vector machine, and outputting a diagnosis result.
4. The LNG engine gas injection valve fault diagnosis method according to claim 2, wherein the SOA-CYCBD in step S2 is implemented by using a gull optimization algorithm to find optimal values of the CYCBD parameter filter length L and the fault period T using a harmonic significance indicator HSI as an objective function, and the HSI is calculated as follows:
where ω is frequency, F (ω) represents the amplitude at frequency ω in the envelope spectrum, Q is the largest harmonic used for calculation, and usually Q =5, n (ω) is the noise amount around frequency ω, which can be calculated by a moving average filter and the harmonic amplitude needs to be removed in advance; q is a constant with a value range of [1, Q ]; p (ω) = F (ω)/N (ω) is an amplitude ratio for representing the degree of significance of a specific frequency component.
5. The LNG engine gas injection valve fault diagnosis method of claim 3, characterized in that the filtered pressure signal is analyzed hierarchically, in particular, a pressure signal with a signal length N,
based on vectors
The node components of each layer decomposition defining the time series u (i) are as follows:
wherein N =2 n And n is a positive integer. Operator Q 0 Sum operator Q 1 Has a length of 2 n-1 . k is the number of decomposition layers, u k,e Constructing an n-dimensional vector for the node component of the k-layer decomposition of the time sequence u, e being a positive integer
The integer e can be expressed as: />
Wherein the vector corresponding to the positive integer e is
6. The LNG engine gas injection valve fault diagnosis method as claimed in claim 4, wherein weighted permutation entropies of the respective levels are calculated, and the entropy value calculation results are as follows:
HWPE＝E(u k,e ,m,d)＝[E 1 ,E 2 ,...,E e ] T (7)
where m is the embedding dimension, d is the time delay, T is the matrix transpose, E 1 ,E 2 ,...,E e Entropy values are arranged for weighting.
7. A LNG engine gas injection valve fault diagnosis system characterized by includes:
the signal acquisition system is used for acquiring a pressure fluctuation signal of the gas supply pipe and dividing the acquired pressure fluctuation signal into a training signal and a test signal;
the filtering system is used for filtering the pressure fluctuation signal by using an SOA-CYCBD algorithm to obtain a pressure signal with enhanced impact components;
the signal processing system is used for processing the filtered pressure signal, and the processed pressure signal is used as a fault characteristic;
and the fault analysis system is used for training the processed pressure signals as fault characteristic vectors, performing fault diagnosis and pattern recognition and outputting diagnosis results.
8. A computer-readable storage medium, characterized in that the computer-readable storage medium has stored thereon a computer program which, when being executed by a processor, implements the LNG engine gas injection valve fault diagnosis method according to any one of claims 1 to 6.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202211130142.0A CN115901281A (en) | 2022-09-16 | 2022-09-16 | Fault diagnosis method and system for gas injection valve of LNG engine |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202211130142.0A CN115901281A (en) | 2022-09-16 | 2022-09-16 | Fault diagnosis method and system for gas injection valve of LNG engine |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN115901281A true CN115901281A (en) | 2023-04-04 |
Family
ID=86484832
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202211130142.0A Pending CN115901281A (en) | 2022-09-16 | 2022-09-16 | Fault diagnosis method and system for gas injection valve of LNG engine |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN115901281A (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN118131605A (en) * | 2024-05-07 | 2024-06-04 | 浙江信合阀门有限公司 | Intelligent ball valve control system |
-
2022
- 2022-09-16 CN CN202211130142.0A patent/CN115901281A/en active Pending
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN118131605A (en) * | 2024-05-07 | 2024-06-04 | 浙江信合阀门有限公司 | Intelligent ball valve control system |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN110361778B (en) | Seismic data reconstruction method based on generation countermeasure network | |
| CN111042917B (en) | Common rail fuel injector weak fault diagnosis method based on GOA-MCKD and hierarchical discrete entropy | |
| CN111929548B (en) | Method for generating discharge and interference signal samples, computer device and storage medium | |
| CN111520267A (en) | Common rail fuel injector fault diagnosis method based on FOA-VMD and HDE | |
| CN114969995A (en) | Rolling bearing early fault intelligent diagnosis method based on improved sparrow search and acoustic emission | |
| CN111914705A (en) | Signal generation method and device for improving health state evaluation accuracy of reactor | |
| CN112013285B (en) | Method and device for detecting pipeline leakage point, storage medium and terminal | |
| CN115901281A (en) | Fault diagnosis method and system for gas injection valve of LNG engine | |
| Yu et al. | A fault diagnosis approach for diesel engine valve train based on improved ITD and SDAG-RVM | |
| CN117009814A (en) | Engine fault diagnosis method and system based on multi-source information fusion | |
| CN114417926A (en) | Power equipment partial discharge pattern recognition method and system based on deep convolution generation countermeasure network | |
| CN103745453A (en) | Town information extraction method based on Google Earth remote sensing image | |
| Yun et al. | A weak fault diagnosis scheme for common rail injector based on MGOA-MOMEDA and improved hierarchical dispersion entropy | |
| CN114818850A (en) | Clustering compression-based network flow space mapping characterization method and device and storage medium | |
| CN114048819A (en) | Power distribution network topology identification method based on attention mechanism and convolutional neural network | |
| CN117633588A (en) | Pipeline leakage positioning method based on spectrum weighting and residual convolution neural network | |
| CN112761843A (en) | Common rail oil injector fault diagnosis method | |
| CN115356599B (en) | Multi-mode urban power grid fault diagnosis method and system | |
| CN116973677A (en) | Distribution network single-phase earth fault line selection method based on cavity convolution and attention mechanism | |
| CN117332268A (en) | GIS partial discharge data enhancement method and system for improving GAN | |
| CN117172375A (en) | Method for predicting remaining service life of mechanical equipment and storage medium | |
| CN112183260A (en) | One-way valve fault diagnosis method based on total variation noise reduction and RQA | |
| CN116610935A (en) | Mechanical fault detection method based on engine vibration signal multi-mode analysis | |
| CN113887452B (en) | Fault diagnosis optimization method based on correlation matrix | |
| US20210270892A1 (en) | Method and system for extracting fault feature of analog circuit based on optimal wavelet basis function |
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 |