CN110397495B - A data-driven performance estimation method for engine aftertreatment device - Google Patents

A data-driven performance estimation method for engine aftertreatment device Download PDF

Info

Publication number
CN110397495B
CN110397495B CN201910590191.4A CN201910590191A CN110397495B CN 110397495 B CN110397495 B CN 110397495B CN 201910590191 A CN201910590191 A CN 201910590191A CN 110397495 B CN110397495 B CN 110397495B
Authority
CN
China
Prior art keywords
performance
engine
post
processing device
temperature
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
Application number
CN201910590191.4A
Other languages
Chinese (zh)
Other versions
CN110397495A (en
Inventor
胡杰
廖健雄
颜伏伍
蔡之洲
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Wuhan University of Technology WUT
Original Assignee
Wuhan University of Technology WUT
Priority date (The priority date 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 date listed.)
Filing date
Publication date
Application filed by Wuhan University of Technology WUT filed Critical Wuhan University of Technology WUT
Priority to CN201910590191.4A priority Critical patent/CN110397495B/en
Publication of CN110397495A publication Critical patent/CN110397495A/en
Application granted granted Critical
Publication of CN110397495B publication Critical patent/CN110397495B/en
Active legal-status Critical Current
Anticipated expiration legal-status Critical

Links

Images

Classifications

    • F—MECHANICAL ENGINEERING; LIGHTING; HEATING; WEAPONS; BLASTING
    • F01—MACHINES OR ENGINES IN GENERAL; ENGINE PLANTS IN GENERAL; STEAM ENGINES
    • F01N—GAS-FLOW SILENCERS OR EXHAUST APPARATUS FOR MACHINES OR ENGINES IN GENERAL; GAS-FLOW SILENCERS OR EXHAUST APPARATUS FOR INTERNAL COMBUSTION ENGINES
    • F01N11/00—Monitoring or diagnostic devices for exhaust-gas treatment apparatus, e.g. for catalytic activity
    • F—MECHANICAL ENGINEERING; LIGHTING; HEATING; WEAPONS; BLASTING
    • F01—MACHINES OR ENGINES IN GENERAL; ENGINE PLANTS IN GENERAL; STEAM ENGINES
    • F01N—GAS-FLOW SILENCERS OR EXHAUST APPARATUS FOR MACHINES OR ENGINES IN GENERAL; GAS-FLOW SILENCERS OR EXHAUST APPARATUS FOR INTERNAL COMBUSTION ENGINES
    • F01N11/00—Monitoring or diagnostic devices for exhaust-gas treatment apparatus, e.g. for catalytic activity
    • F01N11/002—Monitoring or diagnostic devices for exhaust-gas treatment apparatus, e.g. for catalytic activity the diagnostic devices measuring or estimating temperature or pressure in, or downstream of the exhaust apparatus
    • F—MECHANICAL ENGINEERING; LIGHTING; HEATING; WEAPONS; BLASTING
    • F01—MACHINES OR ENGINES IN GENERAL; ENGINE PLANTS IN GENERAL; STEAM ENGINES
    • F01N—GAS-FLOW SILENCERS OR EXHAUST APPARATUS FOR MACHINES OR ENGINES IN GENERAL; GAS-FLOW SILENCERS OR EXHAUST APPARATUS FOR INTERNAL COMBUSTION ENGINES
    • F01N11/00—Monitoring or diagnostic devices for exhaust-gas treatment apparatus, e.g. for catalytic activity
    • F01N11/002—Monitoring or diagnostic devices for exhaust-gas treatment apparatus, e.g. for catalytic activity the diagnostic devices measuring or estimating temperature or pressure in, or downstream of the exhaust apparatus
    • F01N11/005—Monitoring or diagnostic devices for exhaust-gas treatment apparatus, e.g. for catalytic activity the diagnostic devices measuring or estimating temperature or pressure in, or downstream of the exhaust apparatus the temperature or pressure being estimated, e.g. by means of a theoretical model
    • 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

  • Engineering & Computer Science (AREA)
  • Chemical & Material Sciences (AREA)
  • Chemical Kinetics & Catalysis (AREA)
  • Combustion & Propulsion (AREA)
  • Mechanical Engineering (AREA)
  • General Engineering & Computer Science (AREA)
  • Combined Controls Of Internal Combustion Engines (AREA)
  • Exhaust Gas After Treatment (AREA)

Abstract

本发明提出一种基于数据驱动的发动机后处理装置的性能估计方法，基于发动机排放控制后处理装置的远程监测系统获取发动机后处理装置使用过程中入口和出口处的实时温度值；基于获得的实时温度大数据通过数据驱动方法获取后处理装置温差累计值、标准化系数和温度信号延迟时间等性能特征；通过不同性能特征值与性能的对应关系可以较为准确地识别出后处理装置性能和进行故障预测。本发明提出的基于数据驱动的发动机后处理装置性能估计方法，能充分利用实时状态数据和历史数据快速准确预测发动机后处理装置性能，诊断故障形式及程度，且该方法受工况的影响较小，能够较好地应用于实际后处理诊断和性能估计中。

Figure 201910590191

The invention proposes a performance estimation method of an engine after-treatment device based on data driving. Based on the remote monitoring system of the engine emission control after-treatment device, the real-time temperature values at the inlet and the outlet during the use of the engine after-treatment device are obtained; based on the obtained real-time temperature values The temperature big data obtains the performance characteristics of the post-processing device such as the accumulated value of the temperature difference, the normalization coefficient and the temperature signal delay time through the data-driven method; the performance of the post-processing device can be more accurately identified and fault prediction can be performed through the corresponding relationship between different performance characteristic values and performance. . The method for estimating the performance of the engine post-processing device based on the data drive proposed by the present invention can make full use of real-time state data and historical data to quickly and accurately predict the performance of the engine post-processing device, diagnose the form and degree of failure, and the method is less affected by working conditions , which can be better applied to practical post-processing diagnosis and performance estimation.

Figure 201910590191

Description

Performance estimation method of engine post-processing device based on data driving
Technical Field
The invention belongs to the technical field of engines, and particularly relates to a performance estimation method of an engine post-processing device based on data driving.
Background
The engine is one of high-efficiency power sources, and is widely applied to the fields of transportation, engineering machinery, agricultural machinery, ships and the like. With the high-speed increase of the engine holding capacity, the problem of engine emission pollution is increasingly serious, becomes a main contributor of air pollution emission, becomes one of main factors restricting the development of the engine, and needs to be solved urgently. In order to meet the national VI and stricter emission standards, the addition of an aftertreatment device is a necessary choice for controlling the emission of the engine, and the aftertreatment device can effectively reduce the emission of the engine while ensuring the performance of the engine. The exhaust emission components of the gasoline engine and the diesel engine have difference due to different working principles. Aftertreatment devices for different exemplary engine types are shown in table 1. The working principle of different after-treatment devices is also different, the three-way catalytic converter TWC mainly passes NOXCO and HC react with each other to reduce exhaust emission, the particulate matter trap GPF and the DPF trap particulate matter by adopting mutually staggered closed pore passages, the oxidation type catalyst DOC achieves the aim of removing HC and CO through a catalyst coated on the surface of a carrier, and the selective reduction catalysis technology SCR is adoptedThe NOx emission values are reduced by injecting a urea solution to react with the NOx.
TABLE 1 aftertreatment devices for different engine types
Figure GDA0002186793590000011
Because of the wide breadth of our country, factors such as fuel quality, lubricating oil quality, road conditions, legal supervision and the like, and the fact that the performance of an after-treatment device is reduced or deteriorated due to reasons such as thermal failure, chemical poisoning (DOC), blockage/crystallization/fouling, mechanical Damage (DPF) and the like in the actual use process, the engine emission is rapidly deteriorated, the real-time monitoring management and estimation of the working state and the performance of the after-treatment device are very necessary, and currently, for the performance estimation of the after-treatment device, three types of methods are mainly adopted at home and abroad: (1) direct sensor-based measurements, such as PM sensors, radio frequency sensors, etc.; (2) model-based estimation, such as a pressure drop model, a temperature gradient model, etc.; (3) based on sensor signals, such as differential pressure sensors, temperature sensors, etc. The method is mainly used for estimating the performance based on the working condition of the engine, the post-processing temperature, the pressure drop and the like, so that a real-time state signal (such as temperature and pressure drop) acquisition device of the post-processing device needs to be additionally arranged to monitor the post-processing state and estimate the performance.
At present, a vehicle-mounted diagnosis system has good real-time performance in the aspect of automobile emission control fault diagnosis, can detect corresponding faults immediately and remind a driver through a fault indicator lamp, but when sub-systems or components related to an emission control system are in fault, normal work of an engine is possibly not influenced, so that timely processing and maintenance of the faults are not mandatory, a vehicle is still in a state that emission seriously exceeds the standard, and the expected target of reducing the vehicle emission control cannot be achieved.
With the rapid development of technologies such as 5G communication, cloud computing, big data and artificial intelligence, the car networking technology is widely applied, can conveniently acquire running state data of various systems or subsystems or components in the running process of a car, and how to generate values of the data is a problem to be faced by people. In recent years, data-driven-based methods have been rapidly developed in other fields to fully correlate and apply the large data generated on the engineering machinery without establishing an accurate mathematical model.
Disclosure of Invention
The technical problem to be solved by the present invention is to provide a performance estimation method of an engine post-processing device based on data driving, which analyzes and reuses real-time and historical big data to obtain the performance change of an engine.
The technical scheme adopted by the invention for solving the technical problems is as follows: a method for estimating performance of a data-driven engine aftertreatment device, comprising the steps of:
s1) acquiring real-time temperature values of front and rear measuring points of the engine exhaust control post-processing device based on a monitoring system of the engine exhaust control post-processing device;
s2) obtaining a temperature difference accumulated value fitting line and the slope of the fitting line by using a time series model and a fitting method according to the inlet and outlet temperature values of the current engine post-processing device;
s3) calculating and obtaining a covariance function, a normalized correlation coefficient and delay time of inlet and outlet temperature signals of the engine post-processing device by using a temperature correlation-based method;
s4) analyzing the obtained performance characteristic values, and combining all the performance characteristic values to predict and correct the performance, if the engine post-processing device is in a normal state, returning to the step I, and if the engine post-processing device is out of a normal range, utilizing the performance characteristic values to identify and diagnose faults.
According to the scheme, the calculation formula of the temperature difference accumulated value in the step S2) is as follows:
Tsum(n)＝Tsum(n-1)+(T2(n)-T1(n))
wherein n is a sequence of samples, Tsum(n) is the temperature difference integrated value of the sampling point n, T1And T2Respectively, the aftertreatment device inlet temperature and the outlet temperature.
According to the above scheme, the covariance function in step S3) is defined as follows:
Figure GDA0002186793590000021
wherein, N is the length of the signal,
Figure GDA0002186793590000022
and
Figure GDA0002186793590000023
is the average of x (n) and y (n), x (n) and y (n) are two signals of equal and finite length,
the dimensionless normalized correlation coefficient can be expressed as:
Figure GDA0002186793590000031
wherein, cxxAnd cyyThe variances of the sequences x (n) and y (n), respectively; rhoxyIs a normalized correlation function of variables with a variation range of [ -1,1 [)]For calculating where signal y (n) has the greatest similarity to signal x (n), y (n) is shifted by k forward or backward in time sequence, and then the cross-correlation function between x (n) and y (n + k) is calculated, where ρxyThe lag time of two sets of temperature signals as a function of time shift k may be determined by a cross-correlation function ρxy(k) At p, atxyThe time shift k when taking the maximum value is the delay time, and the formula is as follows:
Figure GDA0002186793590000032
Figure GDA0002186793590000033
according to the scheme, the fault identification and diagnosis by using the performance characteristic value in the step S4) includes the following steps: the temperature difference accumulated values corresponding to different fault types and fault degrees are obviously different, the difference is more obvious along with the increase of the sampling point n, the fitted temperature difference accumulated value is divided into a plurality of areas according to the slopes of the fitted data lines of different fault types and fault degrees, and the performance of the post-processing device is estimated according to the slope of the fitted line of the temperature difference accumulated values of the post-processing device with unknown faults and the area where the fitted line is located; the covariance of the two temperature signals is calculated by utilizing the inlet temperature sensor and the outlet temperature sensor of the post-processing device to represent the variation trend difference of the two temperature signals, the similarity degree of the two different signals is evaluated only through the value of the covariance and is not accurate enough, and the covariance function is standardized by introducing a standardized correlation coefficient, so that the influence caused by the amplitude of the sequence to be compared is avoided; in the process of transmitting the heat of the exhaust gas of the engine from the inlet to the outlet of the aftertreatment device, the temperature signal is correspondingly delayed, and the delay time is different under different fault conditions, so the performance of the aftertreatment device of the engine is estimated according to the corresponding relation between the delay time and the performance state of the engine and the temperature difference accumulation curve.
The invention has the beneficial effects that: the invention provides a performance estimation method of an engine post-processing device based on data driving, the influence of performance state change such as the fault occurrence of the post-processing device on the inlet and outlet temperatures is low-frequency, slow in occurrence and long-term, and a small number of sample points can not clearly represent the working state of the post-processing device; the method fully utilizes a large amount of real-time temperature data generated in the operation process of the post-processing device, and reuses the historical data of the post-processing device again, so that whether the post-processing device fails or not and the failure form and degree can be known in time, the post-processing device can normally work in the whole period, and the method can be widely applied to performance prediction and fault identification and diagnosis of various engine post-processing devices.
Drawings
FIG. 1 is a schematic view of a monitoring system for an engine aftertreatment device in accordance with one embodiment of the invention.
FIG. 2 is a flow chart of one embodiment of the present invention.
FIG. 3 is a flow chart of performance estimation for a diesel particulate trap (DPF) according to one embodiment of the present invention.
Detailed Description
For a better understanding of the present invention, reference is made to the following description taken in conjunction with the accompanying drawings and examples.
As shown in fig. 1 and 2, a performance estimation method of an engine after-treatment device based on data driving comprises the following steps:
step S10, acquiring real-time values of inlet temperature and outlet temperature in the engine post-processing device through a monitoring system of the automobile emission control post-processing device;
step S20, calculating the temperature difference accumulation value of the inlet and outlet temperature of the post-processing device through the time series model, wherein the calculation formula of the temperature difference accumulation value is as follows:
Tsum(n)＝Tsum(n-1)+(T2(n)-T1(n))
wherein n is a sequence of samples, Tsum(n) is the temperature difference integrated value of the sampling point n, T1And T2Respectively, the aftertreatment device inlet temperature and the outlet temperature. The temperature difference value of the normal post-processing device presents a more stable trend, the fitted temperature difference cumulative value is divided into a plurality of areas according to the obtained fitted data line slopes of different fault types and fault degrees, and the performance of the post-processing device is estimated according to the slope of the fitted line of the temperature difference cumulative value of the post-processing device with unknown fault and the area where the fitted line is located;
in step S30, the covariance of the two temperature signals is calculated by the post-processing device inlet and outlet temperature sensors to represent the difference in the trend of the two temperature signals, and the covariance function is defined as follows:
Figure GDA0002186793590000041
where N is the length of the signal,
Figure GDA0002186793590000042
and
Figure GDA0002186793590000043
is the average of x (n) and y (n), x (n) and y (n) being two signals of equal and finite length.
The normalized covariance function is then used to eliminate the effect of the sequence amplitudes to be compared. The dimensionless normalized correlation coefficient can be expressed as:
Figure GDA0002186793590000044
wherein, cxxAnd cyyThe variances of the sequences x (n) and y (n), respectively; rhoxyIs a normalized correlation function of variables with a variation range of [ -1,1 [)]。
Finally, shifting y (n) forward or backward in time sequence by k, and calculating the cross-correlation function of x (n) and y (n + k) when rhoxyThe lag time of two sets of temperature signals as a function of time shift k may be determined by a cross-correlation function ρxy(k) At p, atxyThe time shift k when taking the maximum value is the delay time. The formula is as follows:
Figure GDA0002186793590000051
Figure GDA0002186793590000052
the correlation coefficient and the delay time of the inlet temperature and the outlet temperature of the normal post-processing device keep constant values along with the increase of sampling points, however, the delay time of the post-processing device in different fault states has larger difference, so the performance of the post-processing device can be monitored according to different fault forms and the corresponding relation between the fault degree and the delay time;
and step S40, finally judging the real-time performance, failure fault form and degree of the diesel engine post-processing device by combining the performance characteristic values obtained by utilizing and mining the real-time data and the historical data.
Example one
As shown in fig. 3, applying the performance estimation method of the data-driven engine aftertreatment device to the performance estimation of a Diesel Particulate Filter (DPF) includes:
(1) acquiring real-time values of DPF inlet temperature and outlet temperature in an automobile emission control post-processing device monitoring system;
(2) calculating the temperature difference accumulation value of the DPF inlet and outlet temperatures in the diesel engine post-treatment device through a time series model, wherein the calculation formula of the temperature difference accumulation value is as follows:
Tsum(n)＝Tsum(n-1)+(T2(n)-T1(n))
wherein n is a sequence of samples, Tsum(n) is the temperature difference integrated value of the sampling point n, T1And T2DPF inlet and outlet temperatures, respectively. The fitting curves of the temperature difference accumulated value corresponding to different fault types and fault degrees are obviously different, and the difference is more obvious as the sampling point n is increased. DPF plugging failure can result in a drop in DPF outlet temperature and therefore exhibit a decreasing trend (increasing absolute value) in its difference from the DPF inlet temperature. Failure to break can cause the DPF outlet temperature to rise and thus its difference from the DPF inlet temperature exhibits an increasing trend (decreasing absolute value). And dividing the fitted temperature difference accumulated value into a plurality of regions according to the slopes of the fitted data lines of different fault types and fault degrees, and estimating the performance of the post-processing device according to the slope of the fitted line of the temperature difference accumulated value of the post-processing device with unknown fault and the region where the fitted line is located.
(3) In the process of transmitting engine exhaust from the DPF inlet to the DPF outlet, corresponding delay occurs in the temperature signals at the DPF inlet and the DPF outlet, the delay time is different under different fault conditions, and the covariance of the two temperature signals is calculated by using the temperature sensors at the DPF inlet and the outlet to represent the variation trend difference of the two temperature signals, and the covariance function is defined as follows:
Figure GDA0002186793590000053
where N is the length of the signal,
Figure GDA0002186793590000054
and
Figure GDA0002186793590000055
is the average of x (n) and y (n), x (n) and y (n) being two signals of equal and finite length.
Evaluating the similarity of two different signals by the value of the covariance alone is not accurate enough. The normalized covariance function can be used to avoid the effect of the sequence amplitudes to be compared. The dimensionless normalized correlation coefficient can be expressed as:
Figure GDA0002186793590000061
wherein, cxxAnd cyyThe variances of the sequences x (n) and y (n), respectively; rhoxyIs a normalized correlation function of variables with a variation range of [ -1,1 [)]。
For calculating where signal y (n) has the greatest degree of similarity to signal x (n), y (n) is shifted by k forward or backward in time sequence, and then the cross-correlation function between y (n) and y (n + k) is calculated, where pxyAs a function of the time shift k. The lag time of two sets of temperature signals can be determined by the cross-correlation function ρxy(k) At p, atxyThe time shift k when taking the maximum value is the delay time. The formula is as follows:
Figure GDA0002186793590000062
Figure GDA0002186793590000063
when the filter body is broken and has a fault, the DPF inlet temperature and the DPF outlet temperature can obtain a larger cross correlation coefficient on a smaller translation phase k, and when the filter body is blocked and has a fault, the DPF inlet temperature and the DPF outlet temperature can obtain the largest cross correlation coefficient on a larger translation phase. The correlation coefficient of the DPF inlet temperature and outlet temperature increases along with the increase of the damage degree, and the delay time is reduced; the correlation coefficient of the DPF inlet temperature and outlet temperature decreases as the degree of clogging increases, and the delay time increases. It is therefore possible to set limits according to the delay times for different fault forms and fault degrees, in order to monitor the performance of the aftertreatment device.
(4) And finally judging the real-time performance, the failure fault form and the failure fault degree of the diesel engine post-processing device by combining the performance characteristic values obtained by utilizing and mining the real-time data and the historical data of the DPF.
As a fault diagnosis and performance estimation technology, the method has high fault identification accuracy and does not need to establish an accurate mathematical model, and the data-driven method is rapidly developed in other fields. The data driving mainly adopts various data mining technologies to obtain useful information implied in online data and offline data, the useful information represents normal and fault states of a current system, and finally fault detection and diagnosis are realized. Therefore, the invention utilizes the engine emission control post-processing monitoring system to acquire the signals of the temperature sensor, and utilizes the data driving method to acquire the characteristic quantity representing the real-time performance of the engine post-processing device. In the post-processing system, the inlet and outlet temperature differences corresponding to a normal post-processing device and post-processing devices with different fault types and fault degrees are obviously different, meanwhile, in the process of transmitting the heat of engine exhaust from the inlet to the outlet of the post-processing device, corresponding delay occurs to a temperature signal, and the delay time is different under different fault conditions, so that a time sequence method is used for obtaining a temperature difference accumulated value, a temperature correlation coefficient and the delay time as characteristic quantities reflecting the performance of the post-processing device, a temperature difference accumulated value fitting curve is obtained by fitting, different performance characteristic quantities and the temperature difference accumulated value fitting curve can reflect different real-time performances (such as normality, damage types, damage degrees and the like) of the engine, and the performance of the post-processing device is estimated according to the corresponding relation between the characteristic quantities and the performance of the post-processing device of the engine.
The above description is only for the preferred embodiment of the present invention and is not intended to limit the scope of the present invention, and all equivalent structural changes made by using the contents of the present specification and the drawings, or any other related technical fields, are intended to be covered by the scope of the present invention.

Claims (4)

1.一种基于数据驱动的发动机后处理装置的性能估计方法，其特征在于，包括如下步骤：1. a performance estimation method based on a data-driven engine aftertreatment device, is characterized in that, comprises the steps: S1)基于发动机排放控制后处理装置的监测系统获取发动机后处理装置前后测点的实时温度值；S1) obtain the real-time temperature value of the measuring points before and after the engine after-treatment device based on the monitoring system of the engine emission control after-treatment device; S2)根据当前发动机后处理装置入口和出口温度值利用时间序列模型和拟合方法获取温差累计值拟合线及其拟合线斜率；S2) using a time series model and a fitting method to obtain a fitting line of the accumulated temperature difference and the slope of the fitting line according to the current inlet and outlet temperature values of the engine post-processing device; S3)利用基于温度相关性的方法计算得到发动机后处理装置的入口及出口温度信号的协方差函数、标准化相关系数以及延迟时间；S3) using the method based on temperature correlation to obtain the covariance function, standardized correlation coefficient and delay time of the inlet and outlet temperature signals of the engine aftertreatment device; S4)通过对上述步骤得到温差累计值、温度相关系数和延迟时间的性能特征值并进行分析，利用拟合得到的温差累计值拟合曲线，不同性能特征量和温差累计值拟合曲线能够反映发动机不同的实时性能，通过性能特征量与发动机后处理装置性能的对应关系对后处理装置进行性能估计，若发动机后处理装置处于正常范围，则返回步骤S1)，若超出正常范围，则利用性能特征值进行故障识别和诊断。S4) By obtaining and analyzing the performance characteristic values of the accumulated temperature difference value, temperature correlation coefficient and delay time in the above steps, and using the fitting curve of the accumulated temperature difference value obtained by fitting, the fitting curve of different performance characteristic quantities and accumulated temperature difference value can reflect the Different real-time performance of the engine, the performance of the post-processing device is estimated by the corresponding relationship between the performance feature quantity and the performance of the engine post-processing device, if the engine post-processing device is in the normal range, then return to step S1), if it exceeds the normal range, then use the performance Eigenvalues for fault identification and diagnosis. 2.根据权利要求1所述的一种基于数据驱动的发动机后处理装置的性能估计方法，其特征在于，步骤S2)中所述温差累计值的计算公式为：2. The performance estimation method of a data-driven engine post-processing device according to claim 1, wherein the calculation formula of the temperature difference accumulated value in step S2) is: Tsum(n)＝Tsum(n-1)+(T2(n)-T1(n))T sum (n)=T sum (n-1)+(T 2 (n)-T 1 (n)) 其中，n为样点序列，Tsum(n)为采样点n的温差累计值，T1和T2分别为后处理装置入口温度和出口温度。Among them, n is the sampling point sequence, T sum (n) is the accumulated value of the temperature difference of the sampling point n, and T 1 and T 2 are the inlet temperature and outlet temperature of the post-processing device, respectively. 3.根据权利要求1或2所述的一种基于数据驱动的发动机后处理装置的性能估计方法，其特征在于，步骤S3)中所述协方差函数定义如下：3. a kind of performance estimation method based on data-driven engine post-processing device according to claim 1 and 2, is characterized in that, the covariance function described in step S3) is defined as follows:
Figure FDA0002933764390000011
Figure FDA0002933764390000011
 其中，N为信号的长度，
Figure FDA0002933764390000012
和
Figure FDA0002933764390000013
为x(n)和y(n)的平均值，x(n)和y(n)是两个长度相同且有限的信号，where N is the length of the signal,
Figure FDA0002933764390000012
and
Figure FDA0002933764390000013
is the average value of x(n) and y(n), x(n) and y(n) are two signals of the same length and finite, 无量纲的标准化相关系数可以表示为：The dimensionless standardized correlation coefficient can be expressed as:
Figure FDA0002933764390000014
Figure FDA0002933764390000014
 其中，cxx和cyy分别为序列x(n)和y(n)的方差；ρxy为变量的标准化相关函数，变化范围为[-1,1]，Among them, c xx and c yy are the variances of the sequences x(n) and y(n), respectively; ρ xy is the standardized correlation function of the variable, and the variation range is [-1, 1], 对于计算信号y(n)在何处与信号x(n)有最大的相似程度，需将y(n)在时序上向前或向后平移k，再计算x(n)和y(n+k)的互相关函数，此时ρxy为时移k的函数，两组温度信号的滞后时间可以通过互相关函数ρxy(k)的峰值获取，在ρxy取最大值时的时移k即为延迟时间，其公式如下：To calculate where the signal y(n) is most similar to the signal x(n), it is necessary to shift y(n) forward or backward by k in time series, and then calculate x(n) and y(n+ The cross-correlation function of k), at this time ρ xy is a function of time shift k, the lag time of the two groups of temperature signals can be obtained by the peak value of the cross-correlation function ρ xy (k), the time shift k when ρ xy takes the maximum value is the delay time, and its formula is as follows:
Figure FDA0002933764390000021
Figure FDA0002933764390000021
Figure FDA0002933764390000022
Figure FDA0002933764390000022
 4.根据权利要求3所述的一种基于数据驱动的发动机后处理装置的性能估计方法，其特征在于，步骤S4)中所述利用性能特征值进行故障识别和诊断包括如下内容：不同的故障类型和故障程度所分别对应的温差累计值明显不同，且随着采样点n增加，区别越明显，根据不同故障类型和故障程度的拟合数据线斜率将拟合的温差累计值划分为多个区域，针对未知故障的后处理装置温差累计值拟合线的斜率及所在的区域对后处理装置性能进行估计；利用后处理装置入口和出口温度传感器计算出两个温度信号的协方差以表示两个温度信号的变化趋势差异，仅通过协方差的值来评价两个不同信号的相似程度并不够精确，通过引入标准化相关系数使协方差函数标准化，避免待比较的序列幅值带来的影响；在发动机排气热量从后处理装置入口传递至出口的过程中，温度信号会出现相应的延迟，且不同故障条件下延迟时间也有所不同，因此通过延迟时间与发动机性能状态的对应关系并结合温差累计曲线对发动机后处理装置性能进行估计。4. The performance estimation method based on a data-driven engine after-processing device according to claim 3, wherein the use of performance characteristic values to identify and diagnose faults described in step S4) includes the following: different faults The accumulated temperature difference values corresponding to the type and fault degree are obviously different, and as the sampling point n increases, the difference becomes more obvious. According to the slope of the fitting line of the accumulated temperature difference value of the post-processing device and the region where the unknown fault is located, the performance of the post-processing device is estimated; the covariance of the two temperature signals is calculated by using the inlet and outlet temperature sensors of the post-processing device to represent the two It is not accurate enough to evaluate the similarity of two different signals only by the value of the covariance. The covariance function is standardized by introducing a standardized correlation coefficient to avoid the influence of the amplitude of the sequence to be compared; In the process of engine exhaust heat transferring from the inlet to the outlet of the aftertreatment device, the temperature signal will be delayed accordingly, and the delay time will be different under different fault conditions. Therefore, the corresponding relationship between the delay time and the engine performance state is combined with the temperature difference. The cumulative curve provides an estimate of engine aftertreatment device performance.
CN201910590191.4A 2019-07-02 2019-07-02 A data-driven performance estimation method for engine aftertreatment device Active CN110397495B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN201910590191.4A CN110397495B (en) 2019-07-02 2019-07-02 A data-driven performance estimation method for engine aftertreatment device

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN201910590191.4A CN110397495B (en) 2019-07-02 2019-07-02 A data-driven performance estimation method for engine aftertreatment device

Publications (2)

Publication Number Publication Date
CN110397495A CN110397495A (en) 2019-11-01
CN110397495B true CN110397495B (en) 2021-05-11

Family

ID=68323682

Family Applications (1)

Application Number Title Priority Date Filing Date
CN201910590191.4A Active CN110397495B (en) 2019-07-02 2019-07-02 A data-driven performance estimation method for engine aftertreatment device

Country Status (1)

Country Link
CN (1) CN110397495B (en)

Families Citing this family (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN113339115B (en) * 2021-05-25 2022-08-02 潍柴动力股份有限公司 Diesel engine, and diesel engine post-processing fault detection system and method
CN114294084B (en) * 2021-12-29 2023-01-06 潍柴动力股份有限公司 Regeneration control method and device and vehicle
CN115059532B (en) * 2022-07-14 2023-05-12 东风商用车有限公司 Method for diagnosing damage of pore canal of particle catcher carrier
CN116066221A (en) * 2022-11-30 2023-05-05 东风商用车有限公司 A DPF operation state monitoring method, device, electronic equipment and storage medium
CN116304939A (en) * 2023-02-22 2023-06-23 中广核智能科技(深圳)有限责任公司 Multi-monitoring data evolution rule identification method and system based on waveform alignment

Citations (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US5444974A (en) * 1993-07-09 1995-08-29 General Motors Corporation On-board automotive exhaust catalyst monitoring with a calorimetric sensor
KR20050112707A (en) * 2004-05-28 2005-12-01 현대자동차주식회사 System and method for on board diagnosis of diesel particulate filter
WO2006064145A1 (en) * 2004-12-14 2006-06-22 Renault S.A.S Device for controlling regeneration of a particulate filter for an internal combustion engine and corresponding method
CN102828808A (en) * 2011-06-16 2012-12-19 通用汽车环球科技运作有限责任公司 Diagnostic system and method for an oxygen sensor positioned downstream from a catalytic converter
CN103003801A (en) * 2010-05-14 2013-03-27 哈尼施费格尔技术公司 Predictive analysis for remote machine monitoring
KR20160097615A (en) * 2015-02-09 2016-08-18 현대자동차주식회사 Control method for dosing module

Family Cites Families (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20120023911A1 (en) * 2010-07-28 2012-02-02 Gm Global Technology Operations, Inc. Detection of exhaust particulate filter substrate failure

Patent Citations (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US5444974A (en) * 1993-07-09 1995-08-29 General Motors Corporation On-board automotive exhaust catalyst monitoring with a calorimetric sensor
KR20050112707A (en) * 2004-05-28 2005-12-01 현대자동차주식회사 System and method for on board diagnosis of diesel particulate filter
WO2006064145A1 (en) * 2004-12-14 2006-06-22 Renault S.A.S Device for controlling regeneration of a particulate filter for an internal combustion engine and corresponding method
CN103003801A (en) * 2010-05-14 2013-03-27 哈尼施费格尔技术公司 Predictive analysis for remote machine monitoring
CN102828808A (en) * 2011-06-16 2012-12-19 通用汽车环球科技运作有限责任公司 Diagnostic system and method for an oxygen sensor positioned downstream from a catalytic converter
KR20160097615A (en) * 2015-02-09 2016-08-18 현대자동차주식회사 Control method for dosing module

Also Published As

Publication number Publication date
CN110397495A (en) 2019-11-01

Similar Documents

Publication Publication Date Title
CN110397495B (en) A data-driven performance estimation method for engine aftertreatment device
CN109072755B (en) Radio frequency process sensing, control and diagnostic networks and systems
DE10346029B4 (en) Emission control system for an internal combustion engine
US11098630B2 (en) Method and computer program product for diagnosing a particle filter
CN102667090B (en) Apparatus, system, and method for estimating deterioration of a NOX sensor response rate
US9752488B2 (en) Systems and methods of an on-board NOx monitor reset test
CN108952901B (en) SCR catalyst aging correction method based on double NOx sensors
CN113790094B (en) Method, device, vehicle and medium for determining sulfur poisoning of aftertreatment system
CN113669143B (en) Performance detection method, detection device and medium for tail gas aftertreatment system
CN113358363B (en) Method for identifying Urea crystallization critical state of Urea-SCR system
CN109931129B (en) Method and device for monitoring an exhaust gas aftertreatment system of an internal combustion engine
CN106837496B (en) Engine particulate purifying regeneration control system
CN112211705B (en) Method, device and system for monitoring DPF removal
US20230193807A1 (en) Obd sensor response diagnostics and anti tamper device
CN113356988B (en) Online diagnosis method for Urea crystallization risk of Urea-SCR system
CN108825344B (en) NO (nitric oxide)xMethod and device for judging reasonability of sensor measurement value
US20240318585A1 (en) Systems and methods for diagnosing component failure
CN114893280A (en) Estimation method of DPF carbon loading capacity
US12188390B2 (en) Diagnosis device and diagnosis method
CA3125381A1 (en) Methods for diagnostics and operation of an emissions aftertreatment system
CN118309542A (en) Method and device for analyzing overrun of exhaust emission, storage medium and vehicle
CN110325717B (en) Method for monitoring a component in an exhaust aftertreatment system and engine arrangement comprising an exhaust aftertreatment system monitoring arrangement
US12037955B2 (en) Systems and methods for forecasting aftertreatment temperatures over a horizon
CN111927607B (en) Monitoring the status of catalytic converters for reducing nitrogen oxides
CN118997896A (en) Online diagnosis method for urea crystallization state of SDPF (sodium dodecyl sulfate) system

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