CN115270895B - Fault detection method for diesel engine - Google Patents
Fault detection method for diesel engine Download PDFInfo
- Publication number
- CN115270895B CN115270895B CN202211177738.6A CN202211177738A CN115270895B CN 115270895 B CN115270895 B CN 115270895B CN 202211177738 A CN202211177738 A CN 202211177738A CN 115270895 B CN115270895 B CN 115270895B
- Authority
- CN
- China
- Prior art keywords
- vibration
- signal
- vibration signal
- time
- diesel engine
- 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
- 238000001514 detection method Methods 0.000 title claims abstract description 13
- 238000000034 method Methods 0.000 claims abstract description 44
- 230000008569 process Effects 0.000 claims abstract description 17
- 230000000737 periodic effect Effects 0.000 claims abstract description 14
- 238000007405 data analysis Methods 0.000 claims abstract description 12
- 239000000284 extract Substances 0.000 claims abstract description 5
- 238000006243 chemical reaction Methods 0.000 claims description 3
- 238000010606 normalization Methods 0.000 claims description 3
- 238000007906 compression Methods 0.000 abstract description 21
- 230000006835 compression Effects 0.000 abstract description 20
- 229920005994 diacetyl cellulose Polymers 0.000 abstract description 9
- 230000008859 change Effects 0.000 description 8
- 238000013528 artificial neural network Methods 0.000 description 4
- 238000004364 calculation method Methods 0.000 description 4
- 230000003044 adaptive effect Effects 0.000 description 3
- 238000004458 analytical method Methods 0.000 description 2
- 230000000694 effects Effects 0.000 description 2
- 230000002238 attenuated effect Effects 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 230000007423 decrease Effects 0.000 description 1
- 238000004519 manufacturing process Methods 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 238000005457 optimization Methods 0.000 description 1
- 230000000750 progressive effect Effects 0.000 description 1
- 230000009466 transformation Effects 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01H—MEASUREMENT OF MECHANICAL VIBRATIONS OR ULTRASONIC, SONIC OR INFRASONIC WAVES
- G01H17/00—Measuring mechanical vibrations or ultrasonic, sonic or infrasonic waves, not provided for in the preceding groups
-
- 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
- Y02T10/40—Engine management systems
Landscapes
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Measurement Of Mechanical Vibrations Or Ultrasonic Waves (AREA)
Abstract
The invention relates to the technical field of digital data processing, in particular to a fault detection method for a diesel engine, which collects vibration signals in the working process of the diesel engine; dividing a time region of the vibration signal according to the periodic characteristics of the vibration signal; acquiring a vibration peak of the vibration signal in each time region, and judging the importance degree of the vibration signal in different time regions by comparing the trend degree difference between the vibration peaks at the same position in different time regions; acquiring the size of a bit block of each time region based on the importance degree, and compressing the vibration signal by using the bit blocks of all the time regions; and transmitting the compressed vibration signal to a data analysis system so that the data analysis system extracts the vibration signal through decoding, and further carrying out fault identification. The invention can carry out the compression coding of the self-adaptive DACs algorithm on different vibration signals, thereby achieving the balance between the reading speed and the compression ratio of different vibration signals.
Description
Technical Field
The invention relates to the technical field of digital data processing, in particular to a fault detection method for a diesel engine.
Background
The diesel engine is used as a power source of most mechanical equipment, and if a fault occurs, the diesel engine has great influence on the safety and the efficiency in production. In the diesel engine fault detection process, the vibration sensor is arranged to collect the change of the vibration signal in the diesel engine working process to detect the diesel engine fault in real time, and the compression method of the vibration signal greatly influences the real-time property of detecting the diesel engine fault.
The DACs algorithm is a compression algorithm combining the advantages of fixed-length coding and variable-length coding, and can perform direct decoding while ensuring a larger compression rate. The size of the bit block W in the DACs algorithm determines the compression efficiency and the reading efficiency of the algorithm. If the bit block W is set to be too large, the data quantity needing bit compensation in the algorithm is large, data redundancy is caused, the compression rate is low, and the data reading speed is high due to the small number of the bit blocks when the data is read; if the bit block W is set too small, the amount of data requiring bit padding in the algorithm is small, and the compression rate is high, but the data reading speed is slow because the number of bit blocks is large when reading data.
Disclosure of Invention
In order to solve the problem that the compression method influences the real-time performance of diesel engine fault detection, the invention provides a fault detection method for a diesel engine, which adopts the following technical scheme:
one embodiment of the present invention provides a fault detection method for a diesel engine, including the steps of:
collecting vibration signals in the working process of the diesel engine; dividing a time region of the vibration signal according to the periodic characteristics of the vibration signal;
acquiring a vibration peak of the vibration signal in each time region, and judging the importance degree of the vibration signal in different time regions by comparing the trend degree difference between the vibration peaks at the same position in different time regions; acquiring the size of a bit block of each time region based on the importance degree, and compressing the vibration signal by using the bit blocks of all the time regions;
and transmitting the compressed vibration signal to a data analysis system so that the data analysis system extracts the vibration signal through decoding and then carries out fault identification.
Preferably, the method for acquiring the periodic characteristics comprises the following steps:
acquiring the peak value and the valley value of the vibration signal, and carrying out absolute value conversion on the valley value smaller than zero to obtain a vibration curve; selecting different time periods to divide the vibration curve to obtain a plurality of sections of sub-curves corresponding to each division, and obtaining the optimal degree of the time period corresponding to each division based on the similarity degree of each two sections of adjacent sub-curves; selecting a time period with the maximum preferred degree as a cycle length, and dividing the time region of the vibration signal; the period length is the period characteristic.
Preferably, the method for acquiring the similarity between each two adjacent sub-curves comprises:
and for each two adjacent sub-curves, acquiring the Euclidean distance between every two sequence points at the same position, and taking the Euclidean distance as a negative index of a preset value to obtain the similarity degree of the sequence points between the corresponding two sequence points, wherein the average value of the similarity degrees of all the sequence points is the similarity degree of the corresponding two adjacent sub-curves.
Preferably, the method for acquiring the vibration peak comprises the following steps:
and selecting a region with fluctuation from the vibration signals corresponding to each time region as the vibration peak.
Preferably, the method for acquiring the trend degree comprises the following steps:
and for each time region, acquiring the fluctuation range of each vibration peak, acquiring a straight line formed by every two adjacent vibration signal data points of each vibration peak, and calculating the slope average value of all the formed straight lines to be used as the trend degree of the corresponding vibration peak in the current time region.
Preferably, the acquiring the fluctuation range of each vibration peak includes:
and for each vibration peak, taking a vibration signal larger than zero as a signal area of the vibration peak, calculating an average signal amplitude of the signal area, and when the amplitude difference of two adjacent signal amplitudes in the signal area is larger than 2 times of the average signal amplitude, taking the former signal amplitude of the two adjacent signal amplitudes as a range point, wherein all the range points form the fluctuation range of the vibration peak.
Preferably, the method for acquiring the importance degree comprises the following steps:
calculating the trend mean value of each vibration peak in all time regions, calculating the difference value of the trend of each vibration peak in each time region and the corresponding trend mean value, and taking the normalization result of the sum of all the difference values in each time region as the corresponding importance degree.
Preferably, the method for obtaining the bit block size includes:
setting an initial bit block size and an adjustment coefficient to add the product of the adjustment coefficient and the importance level to the sum of the initial bit block size as a bit block size of a corresponding time zone.
The embodiment of the invention at least has the following beneficial effects:
according to the collected diesel engine vibration signal data, time regions are divided according to the periodic characteristics of the diesel engine vibration signal data, if a certain component in the diesel engine breaks down, wrong characteristics can appear in a certain section in a corresponding periodic time period, and the time regions are divided based on the periodic characteristics, so that a comparison basis can be provided for the follow-up determination of fault positions; the method comprises the steps of analyzing vibration signal characteristics of different time areas respectively, calculating importance degrees in different time areas through the overall variation trend of vibration peaks of vibration signals in different time areas, further adapting the sizes of bit blocks for different time areas, and easily receiving the influence of noise if the importance degrees are judged only through the variation of the amplitude of the vibration signals, so that the influence of the noise can be greatly reduced by obtaining the importance degrees through the variation of the overall trend, then carrying out adaptive obtaining on the sizes of the bit blocks based on the importance degrees, ensuring that the reading speed of important data is high, ensuring that the compression ratio of unimportant data is high, and carrying out compression coding of an adaptive DACs algorithm according to the requirements of different data on the reading speed and the compression ratio to achieve balance between the reading speed and the compression ratio of different vibration signals.
Drawings
In order to more clearly illustrate the embodiments of the present invention or the technical solutions and advantages of the prior art, the drawings used in the description of the embodiments or the prior art will be briefly described below, it is obvious that the drawings in the following description are only 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 flowchart illustrating steps of a fault detection method for a diesel engine according to an embodiment of the present invention.
Detailed Description
To further illustrate the technical means and effects of the present invention adopted to achieve the predetermined objects, the following detailed description of a method for detecting a fault of a diesel engine according to the present invention, its specific implementation, structure, features and effects will be provided in conjunction with the accompanying drawings and the preferred embodiments. In the following description, different "one embodiment" or "another embodiment" refers to not necessarily the same embodiment. Furthermore, the particular features, structures, or characteristics may be combined in any suitable manner in one or more embodiments.
Unless defined otherwise, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this invention belongs.
The following describes a specific scheme of the fault detection method for the diesel engine provided by the invention in detail with reference to the accompanying drawings.
Referring to fig. 1, a flow chart illustrating steps of a fault detection method for a diesel engine according to an embodiment of the present invention is shown, the method including the following steps:
s001, collecting a vibration signal in the working process of the diesel engine; and dividing the time region of the vibration signal according to the periodic characteristics of the vibration signal.
The method comprises the following specific steps:
1. and collecting vibration signals in the working process of the diesel engine.
The method comprises the steps that a diesel engine cylinder cover surface vibration signal data acquisition platform is arranged and used for detecting whether a diesel engine breaks down, and vibration signals are acquired in the working process of the diesel engine by installing a vibration sensor on the surface of the diesel engine cylinder cover. The type of the vibration sensor can be determined according to the specific implementation situation of an implementer.
The time range of the collected vibration signals in a certain time period is recorded as
. Because the vibration signal is influenced by the surrounding environment and equipment in the acquisition process, a large amount of noise is contained in the acquired vibration signal, and in order to reduce the influence of the noise as much as possible, the vibration signal is subjected to wavelet denoising pretreatment.
2. And dividing the time region of the vibration signal according to the periodic characteristics of the vibration signal.
In the diesel engine vibration signal of collection, because the work of each part all follows certain theory of operation in the diesel engine, the vibration signal that its appears also follows certain law, promptly along with the vibration of different subassemblies in the diesel engine, the vibration signal of collection presents regular change. If a component in the diesel engine fails, a certain section of the diesel engine can have wrong characteristics in a corresponding period time. Therefore, the collected diesel engine vibration signals in a certain time period are divided into periodic segments.
Acquiring the peak value and the valley value of the vibration signal, and carrying out absolute value conversion on the valley value smaller than zero to obtain a vibration curve; selecting different time periods to divide the vibration curve to obtain a plurality of sections of sub-curves corresponding to each division, and acquiring the optimal degree of the time period corresponding to each division based on the similarity degree of each two sections of adjacent sub-curves; selecting a time period with the maximum optimization degree as a cycle length, and dividing a time region of the vibration signal; the period length is the period characteristic.
And for each two adjacent sub-curves, acquiring the Euclidean distance between every two sequence points at the same position, and taking the Euclidean distance as a negative index of a preset value to obtain the similarity degree of the sequence points between the corresponding two sequence points, wherein the average value of the similarity degrees of all the sequence points is the similarity degree of the corresponding two adjacent sub-curves.
When the period division of the vibration signal is performed, the difference of the vibration signal at the corresponding time points in different periods is small. According to the characteristic, the scheme carries out time zone section on the periodic signal collected in a certain time zone
Iteration of size, according to the time zone of each iteration
And dividing the acquired vibration signals, calculating the characteristics of the divided vibration signals, determining the optimal time zone section, and further performing the next analysis as the cycle size.
By first iterating in order to reduce the number of computations
The value is limited in range, and is too small because the period of the vibration signal shows a certain time region
And is too large
All belonging to invalid calculations, e.g. too small
The calculation is less necessary, namely the difference between sporadic data points is larger when the similarity is calculated; is too large
A large error occurs when calculating the comparison similarity, i.e., the difference between the excessive data points is large when calculating the similarity. Therefore, the scheme is set
In an iterative range of
Wherein
Is the time range of the time period of acquisition.
According to the scheme, each iteration is calculated through a DTW dynamic programming algorithm
And (4) the data similarity of each divided time section under the value. But due to the characteristics of the diesel engine vibration signal: the peak value and the valley value appear in sequence, the frequency of the appearance is high, the size change of the corresponding peak value and the valley value is large, the peak value and the valley value are distributed on two sides of the amplitude value of 0, and when the similarity is calculated, the peak value and the valley value are extremely easy to be influenced by noise to cause large errors due to overlarge amplitude value change. Therefore, when the DTW dynamic programming algorithm is performed, the vibration signal is processed: that is, the valley value less than zero is absolute-valued, and when calculating the DTW algorithm, the similarity of the curve formed by each peak point (after the valley value change) after each time zone is analyzed after the divided time zones is used as the similarity of the curve of the current iteration
Selecting the degree of preference under the value corresponding to the highest degree of preference
The value is the determined cycle size.
First, the acquired vibration signal data is subjected to valley point change, i.e., changeAfter that
Data of vibration signal
The calculation expression of (a) is:
wherein,
for the first of the acquired vibration signals
Signal amplitude at each time point.
The changed vibration signal is only a value which is more than or equal to 0, and according to the characteristics of the peak point and the amplitude difference between each time point and the adjacent left and right time points, if the amplitude of the current time point is more than the amplitude of the adjacent left and right time points, the current time point is the peak point. And connecting each peak point to obtain a peak point curve.
To a first order
Sub-iteration
For example, the time is taken
Is divided into
The peak point curve of each time region is recorded as
. According to the DTW algorithm, calculating the adjacent timeThe curve similarity in the inter-region is calculated, and the mean value of the curve similarity is used as the second
At a sub-iteration
Preferred degree value of
. The DTW algorithm is a known technique, and is not described in detail in this document.
Wherein the first step
At a sub-iteration
Preferred degree value of
The computational expression of (a) is:
wherein,
represent according to
The number of divided time regions;
to indicate the first after division
The first time zone acquired by DTW algorithm
A time zone and
the number of matched sequence pairs in each time region;
to indicate the divided first
A time zone and
in a time zone
And matching Euclidean distances between the sequence point pairs after matching. In each iteration process, the Euclidean distance mean value between sequence point pairs which are matched through a DTW algorithm and between the divided adjacent time regions is calculated to represent the similarity between the adjacent time regions. The smaller the Euclidean distance mean value between the matched sequence point pairs is, the smaller the difference between the peak point (changed) curves between two adjacent time regions is, and the greater the similarity between the corresponding two adjacent time regions is. Obtaining the division time period under the current iteration through the similarity between every two adjacent time regions
The greater the degree of similarity, the greater the indication
The greater the degree of preference.
By calculating at each iteration
Selecting the value corresponding to the maximum value of the preference degree
As the length of the period of the acquired vibration signal, and based on
And dividing the acquired vibration signal into time regions.
And according to the acquired diesel engine vibration signal data and the periodic characteristics, dividing a time region. And respectively analyzing the vibration signal characteristics of different time regions, and further self-adapting the bit block sizes for the different time regions.
Step S002, obtaining the vibration peak of the vibration signal in each time region, and judging the importance degree of the vibration signal in different time regions by comparing the trend degree difference between the vibration peaks at the same position in different time regions; and acquiring the bit block size of each time region based on the importance degree, and compressing the vibration signal by using the bit blocks of all the time regions.
The method comprises the following specific steps:
1. and acquiring a vibration peak of the vibration signal in each time region.
And selecting a region with fluctuation from the vibration signals corresponding to each time region as a vibration peak.
From the a priori knowledge, in the divided time regions, a plurality of fine vibration peaks and larger vibration peaks exist in each region. The vibration peak represents a vibration peak which is in a certain fluctuation range and is generated in a certain range continuously due to the operation of internal components of the diesel engine, such as a cylinder piston and the like, in the working process of the diesel engine.
Since the internal components in different cycles operate identically, the number of vibration peaks generated in different cycles is identical. Moreover, the amplitude in the vibration peak is a trend that is continuously reduced due to the attenuation of the vibration signal. Therefore, the importance degree of the vibration signal in different time regions is determined by comparing the difference between the vibration peaks at the same position in different time regions.
2. And acquiring the trend degree of each vibration peak.
And for each time region, acquiring the fluctuation range of each vibration peak, acquiring a straight line formed by every two adjacent vibration signal data points of each vibration peak, and calculating the slope average value of all the formed straight lines to be used as the trend degree of the corresponding vibration peak in the current time region.
By changing the amplitude of the vibration signal smaller than 0 to a vibration signal equal to or larger than 0, analysis is facilitated. After transformation, the fluctuation range of each vibration peak is not changed, and only the amplitude corresponding to part of the abscissa is changed. Therefore, among the obtained vibration peaks, there appears a vibration peak whose amplitude is continuously attenuated and whose amplitude is equal to or greater than 0.
And for each vibration peak, taking a vibration signal larger than zero as a signal area of the vibration peak, calculating an average signal amplitude of the signal area, and when the amplitude difference of two adjacent signal amplitudes in the signal area is larger than 2 times of the average signal amplitude, taking the former signal amplitude of the two adjacent signal amplitudes as a range point, wherein all the range points form a fluctuation range for the vibration peak.
Considering the attenuation characteristic of the vibration peak, i.e. the process that the initial amplitude of each vibration peak is maximum and gradually decreases, and the influence of noise, a noise threshold value is set first
The vibration signals having amplitudes smaller than the noise threshold in each time region are removed, i.e. the influence of the smaller amplitudes is not taken into account. After the vibration signal with smaller amplitude is removed, the vibration signal larger than 0 is the signal area of the vibration peak, and is recorded as
. For is to
Wherein all signal amplitudes are analyzed and divided, i.e. compared, by attenuation characteristics
If the difference between the amplitudes of adjacent signals is
To middle
Amplitude of signal
And a first step of
Amplitude of signal
Is greater than
In which
Represent
The mean value of the signal amplitudes in (1) indicates the first
The points are range points of the characteristic peak and are recorded as a range point set
. Like this operation, all range points can be obtained. Then it is reacted with
First and last points in and a set of range points
Namely the fluctuation range of all vibration peaks in the current time zone.
To be explainedThe internal components in different time zones work identically, and the number of vibration peaks that may be generated in different time zones is identical. Wherein the noise threshold value
The noise amplitude degree can be determined according to the specific implementation situation of an implementer and historical prior data, and an empirical reference value is given in the embodiment of the invention
。
To a first order
In a time region
Taking a vibration peak as an example, first, the first step is obtained
In a time zone
Forming a straight line by every two adjacent vibration signal data points of each vibration peak, and calculating the slope average value of all the formed straight lines
For characterizing the current time region
Trend of individual vibration peaks. Correspondingly, the first time in different time regions can be acquired
The mean value of the slope of the straight line of each vibration peak is obtained, and then the trend degree used for representing the current vibration peak in different time regions is obtained
Wherein
Indicates the number of divided time regions,
is shown as
In a time region
The mean of the slopes of the straight lines formed by each adjacent vibration signal data point of each vibration peak.
Therefore, the trend degrees of all vibration peaks in different time regions and the corresponding average value of the trend degrees of the vibration peaks in all the time regions are obtained.
3. And judging the importance degree of the vibration signals in different time regions by comparing the trend degree difference between the vibration peaks at the same position in different time regions.
Calculating the trend mean value of each vibration peak in all time regions, calculating the difference value of the trend of each vibration peak in each time region and the corresponding trend mean value, and taking the normalization result of the sum of all the difference values in each time region as the corresponding importance degree.
Also in the first place
Taking the time zone as an example, calculating the corresponding importance degree
：
Wherein,
representing the number of acquired vibration peaks;
denotes the first
In a time zone
The trend degree of each vibration peak;
indicating the first in all time regions
The trend mean value of the vibration peak;
a hyperbolic tangent function is represented for normalizing the sum of all differences in the time region.
Comparing the overall variation trend of the vibration peaks in different time areas with the difference of the overall variation trend of the vibration peaks in all corresponding time areas to determine whether the vibration signal in the current time area is important, wherein if the difference is larger, the vibration signal in the current time area is indicated to be more important than the vibration signal in other time areas with smaller difference.
In the process of calculating the importance degrees of the vibration signals in different time areas, if the importance degree is judged only by the amplitude change of the vibration signals, the influence of noise is very easy to occur, so the importance degrees in different time areas are calculated according to the overall change trend of the vibration peaks of the vibration signals in different time areas. The method can greatly reduce the calculation of the importance degree of different time regions of the noise, so that the self-adaptive bit block is more accurate.
4. The bit block size of each time region is acquired based on the degree of importance.
For a time zone with higher importance, it is indicated that the time zone contains more fault information than other time zones, and in the compression process, in order to achieve the purpose of quickly detecting faults, the data reading speed needs to be ensured, so that the size of a bit block in a DACs algorithm needs to be set to be larger; and vice versa.
The initial bit block size and the adjustment coefficient are set so that the sum of the product of the adjustment coefficient and the degree of importance plus the initial bit block size is the bit block size of the corresponding time zone.
Also in the second place
Bit block size, for example per time region
Comprises the following steps:
wherein,
is shown as
Importance in individual time regions; 3 denotes the initial bit block size;
representing a rounding function;
indicating an adjustment coefficient for adjusting the overall size of the bit block.
As an example, the adjustment coefficient in the embodiment of the present invention
In other embodiments, the magnitude of the coefficient is adjustedCan be set according to actual conditions.
Since each bit block in the DACs algorithm contains the original data encoding and identifier, subsequent adjustments are made with a minimum bit block size of 3.
The conventional DACs algorithm uses a fixed-size bit block without considering characteristics of data, and cannot achieve a balance between a reading speed and a compression rate between different data. The reading speed of important data is ensured to be fast, and the compression rate of unimportant data is ensured to be large. According to the method, the size of the self-adaptive bit block is carried out under the condition of fully considering the characteristics of the data, and the compression coding of the self-adaptive DACs algorithm can be carried out according to the requirements of different data on the reading speed and the compression ratio.
And S003, transmitting the compressed vibration signal to a data analysis system so that the data analysis system extracts the vibration signal through decoding and further performs fault identification.
The method comprises the following specific steps:
and transmitting the compressed vibration signal to a data analysis system, decoding the compressed and encoded vibration signal through the data analysis system, extracting characteristics such as a peak value factor and a pulse factor of the vibration signal, and identifying faults.
The specific process of fault identification is as follows: by training the neural network, the peak factor and the pulse factor of the vibration signal are used as the characteristic value of the vibration signal and used as the input data of the neural network, and whether the diesel engine fails or not is used as the output data of the neural network.
The training process of the neural network is trained by taking a peak factor and a pulse factor of historical priori diesel engine vibration signal data and whether a corresponding diesel engine manually marked by a professional fails as a training data set.
In summary, the embodiment of the invention collects the vibration signal in the working process of the diesel engine; dividing a time region of the vibration signal according to the periodic characteristics of the vibration signal; acquiring a vibration peak of the vibration signal in each time region, and judging the importance degree of the vibration signal in different time regions by comparing the trend degree difference between the vibration peaks at the same position in different time regions; acquiring the size of a bit block of each time region based on the importance degree, and compressing the vibration signal by using the bit blocks of all the time regions; and transmitting the compressed vibration signal to a data analysis system so that the data analysis system extracts the vibration signal through decoding, and further carrying out fault identification. The embodiment of the invention can perform compression coding of the adaptive DACs algorithm according to the requirements of different vibration signals on the reading speed and the compression ratio, thereby achieving the balance between the reading speed and the compression ratio of different vibration signals.
It should be noted that: the sequence of the above embodiments of the present invention is only for description, and does not represent the advantages or disadvantages of the embodiments. And specific embodiments thereof have been described above. In addition, the processes depicted in the accompanying figures do not necessarily require the particular order shown, or sequential order, to achieve desirable results. In some embodiments, multitasking and parallel processing may also be possible or may be advantageous.
The embodiments in the present specification are described in a progressive manner, and the same or similar parts in the embodiments are referred to each other, and each embodiment focuses on differences from other embodiments.
The above-mentioned embodiments are only used for illustrating the technical solutions of the present application, and not for limiting the same; modifications of the technical solutions described in the foregoing embodiments, or equivalents of some technical features may be substituted, and the essential features of the corresponding technical solutions do not depart from the scope of the technical solutions of the embodiments of the present application, and are all included in the scope of the present application.
Claims (8)
1. A fault detection method for a diesel engine, characterized in that it comprises the steps of:
collecting vibration signals in the working process of the diesel engine; dividing a time region of the vibration signal according to the periodic characteristics of the vibration signal;
acquiring a vibration peak of the vibration signal in each time region, and judging the importance degree of the vibration signal in different time regions by comparing the trend degree difference between the vibration peaks at the same position in different time regions; acquiring the size of a bit block of each time region based on the importance degree, and compressing the vibration signal by using the bit blocks of all the time regions;
and transmitting the compressed vibration signal to a data analysis system so that the data analysis system extracts the vibration signal through decoding and then carries out fault identification.
2. The fault detection method for the diesel engine according to claim 1, wherein the periodic characteristics are obtained by:
acquiring the peak value and the valley value of the vibration signal, and carrying out absolute value conversion on the valley value smaller than zero to obtain a vibration curve; selecting different time periods to divide the vibration curve to obtain multiple sections of sub-curves corresponding to each division, and obtaining the optimal degree of the time period corresponding to each division based on the similarity degree of every two adjacent sub-curves; selecting a time period with the maximum optimal degree as a cycle length, and dividing a time region of the vibration signal; the period length is the period characteristic.
3. The method for detecting the fault of the diesel engine according to claim 2, wherein the method for acquiring the similarity degree of each two adjacent sub-curves comprises the following steps:
and for each two adjacent sub-curves, acquiring the Euclidean distance between every two sequence points at the same position, and taking the Euclidean distance as a negative index of a preset value to obtain the similarity degree of the sequence points between the corresponding two sequence points, wherein the average value of the similarity degrees of all the sequence points is the similarity degree of the corresponding two adjacent sub-curves.
4. The method for detecting the fault of the diesel engine according to claim 1, characterized in that the method for obtaining the vibration peak is as follows:
and selecting a region with fluctuation from the vibration signals corresponding to each time region as the vibration peak.
5. The method for detecting the fault of the diesel engine according to claim 1, wherein the trend degree is obtained by:
and for each time region, acquiring the fluctuation range of each vibration peak, acquiring a straight line formed by every two adjacent vibration signal data points of each vibration peak, and calculating the slope average value of all the formed straight lines to be used as the trend degree of the corresponding vibration peak in the current time region.
6. The method of claim 5, wherein the obtaining of the fluctuation range of each vibration peak comprises:
and for each vibration peak, taking a vibration signal larger than zero as a signal area of the vibration peak, calculating an average signal amplitude of the signal area, and when the amplitude difference of two adjacent signal amplitudes in the signal area is larger than 2 times of the average signal amplitude, taking the former signal amplitude of the two adjacent signal amplitudes as a range point, wherein all the range points form the fluctuation range of the vibration peak.
7. The method for detecting the fault of the diesel engine according to claim 1, wherein the importance degree obtaining method comprises the following steps:
calculating the trend mean value of each vibration peak in all time regions, calculating the difference value of the trend of each vibration peak in each time region and the corresponding trend mean value, and taking the normalization result of the sum of all the difference values in each time region as the corresponding importance degree.
8. The method for detecting the fault of the diesel engine according to claim 1, wherein the bit block size is obtained by:
setting an initial bit block size and an adjustment coefficient to add the product of the adjustment coefficient and the importance level to the sum of the initial bit block size as a bit block size of a corresponding time zone.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202211177738.6A CN115270895B (en) | 2022-09-27 | 2022-09-27 | Fault detection method for diesel engine |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202211177738.6A CN115270895B (en) | 2022-09-27 | 2022-09-27 | Fault detection method for diesel engine |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN115270895A CN115270895A (en) | 2022-11-01 |
| CN115270895B true CN115270895B (en) | 2022-12-16 |
Family
ID=83756707
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202211177738.6A Active CN115270895B (en) | 2022-09-27 | 2022-09-27 | Fault detection method for diesel engine |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN115270895B (en) |
Families Citing this family (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN118226841B (en) * | 2024-05-23 | 2024-07-26 | 北京七星华创微电子有限责任公司 | Testing method for field programmable gate array FPGA |
Family Cites Families (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US11710560B2 (en) * | 2019-09-11 | 2023-07-25 | Epro Gmbh | Compression of machine health data |
| CN111412978B (en) * | 2020-04-22 | 2021-06-08 | 北京化工大学 | Reciprocating machinery abnormity detection method based on fault-free vibration signal |
| CN115014759B (en) * | 2022-04-29 | 2024-08-23 | 无锡瓦力科技有限公司 | Real-time health state detection method for ball screw pair |
-
2022
- 2022-09-27 CN CN202211177738.6A patent/CN115270895B/en active Active
Also Published As
| Publication number | Publication date |
|---|---|
| CN115270895A (en) | 2022-11-01 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN107316046B (en) | Fault diagnosis method based on incremental compensation dynamic self-adaptive enhancement | |
| CN116828070B (en) | Intelligent power grid data optimization transmission method | |
| CN115270895B (en) | Fault detection method for diesel engine | |
| CN116208172B (en) | Data management system for building engineering project | |
| CN116975008B (en) | Ship meteorological monitoring data optimal storage method | |
| CN115359807A (en) | Noise online monitoring system for urban noise pollution | |
| CN117411947B (en) | Cloud edge cooperation-based water service data rapid transmission method | |
| CN115269526B (en) | Method and system for processing semiconductor production data | |
| CN117376430B (en) | Industrial data rapid transmission method and system based on DCS | |
| CN112070208A (en) | Tool wear prediction method based on encoder-decoder stage attention mechanism | |
| CN116185971B (en) | Intelligent processing system for electronic pressure weighing data | |
| CN116320043A (en) | Method and system for improving transmission efficiency of multi-carrier communication system | |
| CN115184016A (en) | Elevator bearing fault detection method | |
| CN112380992A (en) | Method and device for evaluating and optimizing accuracy of monitoring data in machining process | |
| CN116318172A (en) | Design simulation software data self-adaptive compression method | |
| CN106656201B (en) | Compression method based on amplitude-frequency characteristics of sampled data | |
| JP7026808B2 (en) | Information processing equipment, methods and programs | |
| CN115843366A (en) | Predicting compression ratio of data using compressible decisions | |
| CN117354645B (en) | On-line inspection method for inlet and outlet water quality of water purification plant | |
| CN117459187B (en) | High-speed data transmission method based on optical fiber network | |
| CN117176178B (en) | Data processing method of photoelectric communication system | |
| CN113452379B (en) | Section contour dimension reduction model training method and system and data compression method and system | |
| CN117155402B (en) | Public health intelligent physical examination service system based on RPA technology | |
| CN108668131B (en) | Image compression coding method based on sparse representation | |
| CN118152389B (en) | Intelligent processing method for energy system data |
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 |