阅读说明：本技术 在线柴油机活塞环状态检测方法 (On-line diesel engine piston ring state detection method ) 是由 蒋大伟 罗辉 丁乐中 于 2019-10-23 设计创作，主要内容包括：本发明公开了一种在线柴油机活塞环状态检测方法,包括：通过传感器对柴油机的瞬时转速、缸盖振动、油耗、排温、冷却水温监测；将步骤1采集的信号转化成标准电信号；由数据采集卡对标准电信号进行采集及模数转换；通过对柴油机瞬时转速及气缸振动的耦合,将振动在时域内转化到与曲轴转角对应的角度域内进行阶比分析；测取正常情况下柴油机的阶比数据矩阵<Image he="87" wi="426" file="DDA0002243974100000011.GIF" imgContent="drawing" imgFormat="GIF" orientation="portrait" inline="no"></Image>带入公式将油耗x g/kWh、排温y K及冷却水温度z K变成裕度系数η<Sub>1</Sub>＝((x/y)×z)/H<Sub>1</Sub>,η<Sub>2</Sub>＝((x/y)×z)/H<Sub>2</Sub>；测取任意状态下阶比数据矩阵B,与ηA对比,如<Image he="69" wi="380" file="DDA0002243974100000012.GIF" imgContent="drawing" imgFormat="GIF" orientation="portrait" inline="no"></Image>即判断柴油机活塞环存在异常。本发明实现对活塞环漏气故障进行预判。(The invention discloses an online detection method for a piston ring state of a diesel engine, which comprises the following steps: monitoring instantaneous rotating speed, cylinder cover vibration, oil consumption, exhaust temperature and cooling water temperature of the diesel engine through a sensor; converting the signals collected in the step 1 into standard electric signals; the data acquisition card is used for acquiring and carrying out analog-to-digital conversion on the standard electric signal; converting the vibration into an angle domain corresponding to the crank angle in a time domain for order ratio analysis by coupling the instantaneous rotating speed of the diesel engine and the vibration of the air cylinder; measuring the order data matrix of the diesel engine under normal condition The fuel consumption x g/kWh, the exhaust temperature y K and the cooling water temperature z K are changed into a margin coefficient eta by substituting into a formula 1 ＝((x/y)×z)/H 1 ，η 2 ＝((x/y)×z)/H 2 (ii) a Measuring the ratio data matrix B at any state, comparing with eta A, e.g. Namely, the piston ring of the diesel engine is judged to be abnormal. The invention realizes the prejudgment of the air leakage fault of the piston ring.)

1. An online detection method for the piston ring state of a diesel engine is characterized by comprising the following steps:

step 1, monitoring instantaneous rotating speed, cylinder cover vibration, oil consumption, exhaust temperature and cooling water temperature of a diesel engine through a sensor;

step 2, converting the signals collected in the step 1 into standard electric signals;

step 3, a data acquisition card is used for acquiring and carrying out analog-to-digital conversion on the standard electric signal;

step 4, converting the vibration of the diesel engine into an angle domain corresponding to the crank angle in a time domain for order ratio analysis by coupling the instantaneous rotating speed of the diesel engine and the vibration of the air cylinder;

step 5, measuring the order ratio data matrix of the diesel engine under normal condition

Step 6, substituting the formula to change the oil consumption x g/kWh, the exhaust temperature y K and the cooling water temperature z K into margin coefficients

η₁＝((x/y)×z)/H₁，

η₂＝((x/y)×z)/H₂，

H is a constant and is determined according to each model;

step 7, measuring the order ratio data matrix B in any state, comparing with eta A, for example

2. The method for detecting the state of a piston ring of an online diesel engine as claimed in claim 1, wherein the step 4 is performed by the steps of:

1) instantaneous speed outlier rejection

During the measurement of the instantaneous rotating speed of the diesel engine, the pulse frequency converted by the signal conditioning board is measured, however, in the actual measurement, because the vibration of the sensor or the electromagnetic interference on the conditioning board, the phenomenon of pulse loss or pulse increase can occur, which is shown on the data, namely, the instantaneous rotating speed value or the period value between two top dead center signals is not the number of teeth of the speed measuring gear; however, the increased pulse period value is independent of the diesel engine speed and usually represents a particularly large abnormal value, and these reasons exist, so that the direct use of the measured data for order ratio tracking has a great influence on the calculation accuracy. After removing obvious abnormal value, when searching other abnormal values, using difference statistics, firstly, making instantaneous rotating speed backward difference

X(i)＝x(i)-x(i+1) (1)

X (i) is the ith instantaneous rotating speed value, and X (i) is the rotating speed variation between the ith tooth and the (i-1) th tooth, because the variation of the rotating speed cannot be infinitely large, the rotating speed is in accordance with normal distribution under normal conditions, the problem that the removed abnormal value belongs to the single normal total variance confidence interval is solved, and the total X-N (mu, sigma) is set²) Where μ, σ²Unknown, X₁,X₂,…,X_nIs taken from a sample of the total X.the variance σ is calculated²Has a confidence interval of 1-alpha. sigma²Is estimated as S²Is provided with

For a given confidence level 1- α, from

The variance σ is then²Has a 1-alpha confidence interval of

And 1-alpha confidence interval of variance sigma

When the abnormal point is found, the time occupied by the abnormal point is merged to the next normal measurement value, all the values are moved forward by one, let x (i) be the abnormal value, and x (i) be the reduced value

2) Computed order ratio based on resampling

Converting the vibration signal from a time domain into an angle domain is used for accurately positioning and acquiring a signal in a fault section, and is the first step of feature extraction; the method is a process of obtaining equal angle interval data with angle identification by performing digital resampling on the data sampled at equal time intervals, namely calculating order ratio tracking; by using resampling calculation, the algorithm combines interpolation and extraction uniquely, and the process is as follows:

(1) intercepting instantaneous rotating speed data N and vibration signals X (t) within 720 degrees by using the up-to-point signals;

(2) from the instantaneous speed N_iDetermining the time interval T of each pulse_iWherein i ═ 1,2,3 … 269, 269;

wherein z-number of teeth of gear plate

(3) Determining the instantaneous period value gamma in each tooth angle segment_iI.e. the time required for the diesel engine to rotate by a unit of degrees in each tooth segment

(4) Determining the number of intermediate crank angles R for each tooth_i；

(5) Determining a sequence of interpolation points by using the set angular resolution value delta

cubic spline interpolation is carried out on the abscissa of R and the ordinate of gamma to obtain a new coordinate array

(6) determining the sampling interval time according to the sampling frequency of the vibration signal X (t)

t＝1/f (10)

Wherein f-vibration sampling frequency

(7) Computing

k＝T₀/t,m＝T₀％t (12)

Technical Field

The invention relates to an online detection method for a piston ring state of a diesel engine, and belongs to the technical field of internal combustion engines and marine equipment.

Background

The technical core of diesel engine piston ring detection is actually feature extraction and feature analysis, and the method is the main direction of intelligent diesel engine development.

The modern large diesel engine is a complex nonlinear system which integrates multiple subjects such as materials science, mechanics, electronic control and the like, and comprises multiple subsystems, all the subsystems interact with each other, during the operation of the diesel engine, a mechanical structure and an electric control unit coordinate to work according to a certain time sequence, however, along with the abrasion of the mechanical structure or the inaccuracy or instability of the electric control unit, all the systems can not work effectively in time, the systems are mutually coupled, and the abnormality of one system can affect other systems and show other fault phenomena. The causes of the faults and the fault expressions can not correspond to each other, and the phenomena of one cause and more results and one cause and more reasons occur, so that some parts of the diesel engine can be found or positioned after being completely damaged in many cases, and great loss is caused at the moment. The management of diesel engines is shifted from "after repair" to "prior maintenance".

Disclosure of Invention

The invention aims to provide an online detection method for the piston ring state of a diesel engine, which can be used for prejudging the air leakage fault of the piston ring, so that a user is reminded to overhaul and replace in advance, and the maintenance cost is reduced.

The purpose of the invention is realized by the following technical scheme:

an online detection method for a piston ring state of a diesel engine comprises the following steps:

1. monitoring instantaneous rotating speed, cylinder cover vibration, oil consumption, exhaust temperature and cooling water temperature of the diesel engine through a sensor;

2. converting the signals collected in the step 1 into standard electric signals;

3. the data acquisition card is used for acquiring and carrying out analog-to-digital conversion on the standard electric signal;

4. the method comprises the steps that through the coupling of the instantaneous rotating speed of a diesel engine and the vibration of an air cylinder, the vibration in the diesel engine is converted into an angle domain corresponding to a crank angle in a time domain to carry out order ratio analysis;

5. measuring the order data matrix of the diesel engine under normal conditiona(j))；

6. The oil consumption x g/kWh, the exhaust temperature y K and the cooling water temperature z K are changed into margin coefficients by the substitution formula

η₁＝((x/y)×z)/H₁，

η₂＝((x/y)×z)/H₂，

H is a constant and is determined according to each model;

7. measuring the ratio data matrix B at any state, comparing with eta A, e.g.

Namely, the piston ring of the diesel engine is judged to be abnormal.

The object of the invention can be further achieved by the following technical measures:

the online detection method for the piston ring state of the diesel engine is completed by the following steps in step 4:

1) instantaneous speed outlier rejection

During the measurement of the instantaneous rotating speed of the diesel engine, the pulse frequency converted by the signal conditioning board is measured, however, in the actual measurement, because the vibration of the sensor or the electromagnetic interference on the conditioning board, the phenomenon of pulse loss or pulse increase can occur, which is shown on the data, namely, the instantaneous rotating speed value or the period value between two top dead center signals is not the number of teeth of the speed measuring gear; however, the increased pulse period value is independent of the diesel engine speed and usually represents a particularly large abnormal value, and these reasons exist, so that the direct use of the measured data for order ratio tracking has a great influence on the calculation accuracy. After removing obvious abnormal value, when searching other abnormal values, using difference statistics, firstly, making instantaneous rotating speed backward difference

X(i)＝x(i)-x(i+1) (4-1)

x (i) is the i-th instantaneous speed value, and X (i) is the speed variation between the i-th and the (i-1) th teeth, because the speed variation cannot be infinitely greatThe normal distribution is conformed under normal conditions, the problem that the removed abnormal value belongs to the single normal overall variance confidence interval is solved, and the overall X-N (mu, sigma) is set²) Where μ, σ²Unknown, X₁,X₂,…,X_nIs taken from a sample of the total X.the variance σ is calculated²Has a confidence interval of 1-alpha. sigma²Is estimated as S²Is provided with

For a given confidence level 1- α, from

The variance σ is then²Has a 1-alpha confidence interval of

And 1-alpha confidence interval of variance sigma

When the abnormal point is found, the time occupied by the abnormal point is merged to the next normal measurement value, all the values are moved forward by one, let x (i) be the abnormal value, and x (i) be the reduced value

2) Computed order ratio based on resampling

Converting the vibration signal from a time domain into an angle domain is used for accurately positioning and acquiring a signal in a fault section, and is the first step of feature extraction; the method is a process of obtaining equal angle interval data with angle identification by performing digital resampling on the data sampled at equal time intervals, namely calculating order ratio tracking; by using resampling calculation, the algorithm combines interpolation and extraction uniquely, and the process is as follows:

(1) intercepting instantaneous rotating speed data N and vibration signals X (t) within 720 degrees by using the up-to-point signals;

(2) from the instantaneous speed N_iDetermining the time interval T of each pulse_iWherein i ═ 1,2,3 … 269, 269;

wherein z-number of teeth of gear plate

(3) Determining the instantaneous period value gamma in each tooth angle segment_iI.e. the time required for the diesel engine to rotate by a unit of degrees in each tooth segment

(4) Determining the number of intermediate crank angles R for each tooth_i；

(5) Determining a sequence of interpolation points by using the set angular resolution value delta

Calculating the time required for the diesel engine to rotate per degree in each delta section

Where j is 1,2,3 … 720/Δ,

k＝0,1,2,…,n-1，

cubic spline interpolation is carried out on the abscissa of R and the ordinate of gamma to obtain a new coordinate array

And

(6) determining the sampling interval time according to the sampling frequency of the vibration signal X (t)

t＝1/f (4-10)

Wherein f-vibration sampling frequency

(7) Computing

Vibration semaphore at angle:

k＝T₀/t,m＝T₀％t (4-12)

and a (j) the horizontal and vertical coordinates after resampling,

is the sampling angle a (j) is the vibration amplitude.

Compared with the prior art, the invention has the beneficial effects that: the method for detecting the state of the piston ring of the on-line diesel engine converts the vibration signal from the time domain into the angle domain so as to accurately position and obtain the signal in the fault section and realize the prejudgment of the air leakage fault of the piston ring, thereby reminding a user to repair and replace in advance so as to reduce the maintenance cost.

Drawings

FIG. 1 is a schematic diagram of data acquisition and delivery according to the present invention;

FIG. 2 is a graph of instantaneous diesel engine speed measurements over a cycle of operation;

FIG. 3 is a flow chart of the resampling calculation according to the invention;

FIG. 4 is a flow chart of the system data processing of the present invention.

Detailed Description

The invention is further described with reference to the following figures and specific examples.

The data acquisition and transmission of the online diesel engine piston ring state detection method are shown in figure 1, and the method comprises the following steps:

1. the instantaneous rotating speed, the vibration of a cylinder cover, the oil consumption, the exhaust temperature and the cooling water temperature of the diesel engine are monitored through sensors.

2. The instantaneous rotating speed is monitored, the flywheel gear ring is detected through an MP-981 magneto sensor, sine wave signals with the frequency related to the rotating speed are output, converted into pulse signals through a Smith trigger, and then input into a data acquisition card. Gear shaft vibration monitoring, install the GST piezoelectric type acceleration sensor model at the gear shaft tip through the monitoring: CA-YD-107CM, output charge signal, and directly access to data acquisition card. The fuel consumption monitoring adopts an HZB2000 mass fuel consumption meter connected in series in a fuel pipeline, data can be highlighted in real time or input into a data acquisition card through a serial port, the exhaust temperature and the cooling water temperature are respectively detected by a K-type thermocouple and a PT100 temperature sensor, and output signals are input into the data acquisition card after being conditioned.

3. The data acquisition card is used for acquiring and carrying out analog-to-digital conversion on the standard electric signal; PXI series 8106 of NI company which is a mature product in the market is adopted, a host machine is PXI series 1050, an analog input data acquisition card adopts 6251 for acquiring temperature and pressure, a pulse timer adopts 6602 for acquiring instantaneous rotating speed, and 4472 for acquiring vibration acceleration.

4. The method comprises the steps that through the coupling of the instantaneous rotating speed of a diesel engine and the vibration of an air cylinder, the vibration in the diesel engine is converted into an angle domain corresponding to a crank angle in a time domain to carry out order ratio analysis; it is completed by the following steps:

4.1 removal of abnormal value of instantaneous speed

During the measurement of the instantaneous rotating speed of the diesel engine, the pulse frequency converted by the signal conditioning board is measured, however, in the actual measurement, because the vibration of the sensor or the electromagnetic interference on the conditioning board, the phenomenon of pulse loss or pulse increase can occur, which is shown on the data, namely, the instantaneous rotating speed value or the period value between two top dead center signals is not the number of teeth of the speed measuring gear. However, the increased pulse period value is independent of the diesel engine speed and usually represents a particularly large abnormal value, and these reasons exist, so that the direct use of the measured data for order ratio tracking has a great influence on the calculation accuracy. As shown in FIG. 2, the instantaneous rotation speed measurement value in one working cycle of the diesel engine is shown, 296 values should be obtained by measuring 148 teeth of the gear disc in the experiment, 301 measurement values are obtained in the actual measurement, 5 abnormal values are added, the rotation speed of the diesel engine does not change suddenly, so that three abnormal values can be obviously seen, two abnormal values are removed besides the three abnormal values, and differential statistics is used when other abnormal values are searched. Firstly, the instantaneous rotating speed is differenced backwards

X(i)＝x(i)-x(i+1) (4-1)

X (i) is the ith instantaneous rotating speed value, and X (i) is the rotating speed variation between the ith tooth and the (i-1) th tooth, because the variation of the rotating speed cannot be infinitely large, the rotating speed is in accordance with normal distribution under normal conditions, the problem that the removed abnormal value belongs to the single normal total variance confidence interval is solved, and the total X-N (mu, sigma) is set²) Where μ, σ²Unknown, X₁,X₂,…,X_nIs taken from a sample of the total X.the variance σ is calculated²Has a confidence interval of 1-alpha. sigma²Is estimated as S²Is provided with

For a given confidence level 1- α, from

The variance σ is then²Has a 1-alpha confidence interval of

And 1-alpha confidence interval of variance sigma

The confidence interval 1-alpha can be selected according to the number of abnormal values, and when two abnormal values are removed, 1-alpha can be selected to be 0.99, the specific determination method is as follows, the actual measured value is 301, 3 can be obviously removed, and 2 abnormal values also need to be removed, and the ratio of the two abnormal values in the sample needing to be removed is: 2/(301-3) ═ 0.006712, when two out of confidence intervals, i.e., 99% probability, are outliers.

When the abnormal point is found, the time occupied by the abnormal point is merged to the next normal measurement value, all the values are moved forward by one, let x (i) be the abnormal value, and x (i) be the reduced value

4.2 calculated order ratio based on resampling

The vibration signal is converted from the time domain to the angle domain so as to accurately locate and acquire the signal in the fault section, and the method is the first step of feature extraction. The method is a process of obtaining equal angle interval data with angle identification by performing digital resampling on the data sampled at equal time intervals, namely calculating order ratio tracking. Resampling calculation is used, which requires a large amount of computation. Fig. 3 is a calculation process.

The algorithm designed according to specific conditions combines interpolation and extraction uniquely, and the process is as follows:

1) and intercepting the instantaneous rotating speed data N and the vibration signal X (t) within 720 degrees by using the up-to-point signal.

2) From the instantaneous speed N_iDetermining the time interval T of each pulse_iWherein i is 1,2,3 … 269.

Wherein z-number of teeth of gear plate

3) Determining the instantaneous period value gamma in each tooth angle segment_iI.e. the diesel engine in each tooth segment is rotated pastTime required per degree

4) Determining the number of intermediate crank angles R for each tooth_i。

5) Determining a sequence of interpolation points by using the set angular resolution value delta

Calculating the time required for the diesel engine to rotate per degree in each delta section

Where j is 1,2,3 … 720/Δ,

k＝0,1,2,…,n-1，

cubic spline interpolation is carried out on the abscissa of R and the ordinate of gamma. Get the new coordinate array

And

6) determining the sampling interval time according to the sampling frequency of the vibration signal X (t)

t＝1/f (4-10)

Wherein f-vibration sampling frequency

7) ComputingVibration semaphore at angle:

k＝T₀/t,m＝T₀％t (4-12)

and a (j) the horizontal and vertical coordinates after resampling,

is the sampling angle a (j) is the vibration amplitude.

5. Measuring the order data matrix of the diesel engine under normal conditiona(j))。

6. The oil consumption x g/kWh, the exhaust temperature y K and the cooling water temperature z K are changed into margin coefficients by the substitution formula

η₁＝((x/y)×z)/H₁，

η₂＝((x/y)×z)/H₂，

H is a constant and depends on each model.

7. Measuring the ratio data matrix B at any state, comparing with eta A, e.g.

Namely, the piston ring of the diesel engine is judged to be abnormal.

The whole system data processing flow of the invention is shown in fig. 4.

In addition to the above embodiments, the present invention may have other embodiments, and any technical solutions formed by equivalent substitutions or equivalent transformations fall within the scope of the claims of the present invention.