A fault diagnosis method for diesel engine emissions exceeding the standard based on multi-source information

Abstract

The present invention discloses a method for diagnosing diesel engine emission exceeding standard faults based on multi-source information. The method extracts multi-source information from diesel vehicles that have excessive exhaust pollutants or substandard power during loading and deceleration detection, implement National IV or National V emission standards, and use high-pressure common rail and selective catalytic reduction (SCR) technology routes for emission control. The multi-source information includes exhaust pollutant test result information of the inspection station, OBD data stream information of the diesel vehicle itself, and measurement data of maintenance personnel at the maintenance station. The multi-source information is input into a fault diagnosis model established based on fuzzy logic. After fuzzy reasoning calculation, the fault type and fault cause are determined according to the diagnosis output rules. The method can accurately diagnose the fault of diesel engines that have no obvious fault phenomenon in daily use, but have excessive exhaust pollutants or substandard power during regular emission inspection and loading and deceleration detection, and output the fault cause for reference for maintenance personnel.

Description

Technical Field

[0001] The present invention relates to the technical field of vehicle fault diagnosis, and in particular to a fault diagnosis method for diesel engines used in National IV and National V heavy trucks that fail a pollutant emission test (exceed emission standards) using a loading and deceleration method based on multi-source information. Background Art

[0002] Diesel engines are widely used in the field of heavy trucks due to their high thermal efficiency and high power density. However, diesel trucks have large displacement, high usage intensity, and poor driving conditions, making them the focus of automobile emission pollution control and strengthening management. The implementation of the emission inspection and maintenance (I / M) system is one of the important measures to strengthen the emission control of in-use vehicles. The implementation of the standard GB3847-2018 "Emission Limits and Measurement Methods of Pollutants for Diesel Vehicles (Free Acceleration Method and Loaded Deceleration Method)" has improved the ability to identify high-emission vehicles during the regular emission inspection of in-use diesel trucks in my country. However, since the fault phenomenon of diesel vehicles with excessive emission is not obvious in daily use, the exhaust pollutants exceed the standard or the power does not meet the standard during the regular emission inspection and loading deceleration test, and the fault indicator light usually does not alarm and there is no current fault code. Therefore, it is difficult to diagnose the fault of diesel vehicles with excessive emission, and blind maintenance is common in maintenance management, with low maintenance efficiency and high cost. At present, the number of heavy-duty diesel trucks in use in my country is mainly composed of National IV and National V models that adopt the high-pressure common rail fuel injection system + selective catalytic reduction system (SCR) technology route for emission control. The problem of backward fault diagnosis technology for such diesel vehicles with excessive emissions is particularly prominent. Chinese patent CN108825343A proposes a maintenance method for automobile exhaust emission system, which analyzes the components of gasoline engine exhaust gas through automobile exhaust gas analyzer and determines the fault in combination with automobile OBD diagnostic instrument. However, the above method is for gasoline engines and does not specify the specific working conditions of the engine during exhaust gas testing. The fault diagnosis information source is single. When there is an obvious fault in gasoline engine emissions, it has a certain reference value, but it is difficult to meet the needs of regular emission inspection. Fault diagnosis of diesel engines that fail to meet the requirements.

[0003] Therefore, it is necessary to provide a method for accurately diagnosing diesel engine faults of vehicles with excessive emissions based on multi-source information in order to solve the problem that it is difficult to accurately determine the fault location and cause of diesel trucks that fail the loading and deceleration pollutant emission tests of National IV and National V standards using a high-pressure common rail fuel injection system + SCR technology. Summary of the invention

[0004] For diesel vehicles that meet the National IV and National V standards and use a high-pressure common rail fuel injection system + SCR technology, when the diesel vehicle fails to meet the exhaust pollutant standards or the power is not up to standard during the loading and deceleration method, the fault phenomenon is not obvious in daily use, and there is no current OBD fault code, which leads to the difficulty of fault diagnosis when the maintenance station (M station) repairs vehicles with excessive emissions, the phenomenon of blind maintenance is common, and the maintenance efficiency is low and the cost is high. The present invention provides a method for diagnosing diesel engine emission excessive faults based on multi-source information. The method makes full use of the pollutant emission detection result information under the working conditions specified by the loading and deceleration method of the inspection station (I station), combines the data stream information read from the diesel engine OBD by the maintenance personnel of the maintenance station (M station) and the manual measurement information of the maintenance personnel, and applies fuzzy analysis and reasoning to improve the accuracy and speed of diesel engine fault diagnosis of vehicles with excessive emissions, and improve the effectiveness of diesel truck emission control. In order to achieve the above-mentioned invention purpose, the technical solution adopted by the present invention is:

[0005] A method for diagnosing diesel engine exhaust emission failures based on multi-source information, characterized in that it includes: for diesel vehicles that exceed exhaust pollutants or do not meet power standards during load deceleration detection, implement National IV or National V emission standards, and use high-pressure common rail and selective catalytic reduction (SCR) technology routes for emission control, based on the exhaust pollutant test result information of the diesel vehicle at Station I, the exhaust pollutant test result information of the diesel vehicle at Station I includes: the smoke value at 100% maximum power speed point with the throttle fully open, the smoke value and the measured maximum wheel-side power at 80% maximum power speed point, and the rated power of the engine; the diesel engine's own OBD data stream information, the diesel engine's own OBD data stream information includes, when the diesel engine is in a hot engine, no-load, and full-throttle working condition, using an OBD diagnostic instrument to read the corresponding intake flow value or intake The air pressure value, the common rail pressure value, and the intake flow rate or intake pressure and the lower limit of the common rail pressure of a diesel engine of the same model and good technical condition as the tested diesel engine under the same working conditions are obtained; the maintenance personnel of the M station measure the diesel vehicle data, and the maintenance personnel of the M station measure the diesel vehicle data, including the exhaust back pressure value (relative pressure of the engine exhaust) measured by a barometer on the exhaust pipe upstream of the SCR carrier under high-speed (full throttle speed) working conditions when the diesel engine is hot and idling; the exhaust pollutant test result information of the I station, the diesel vehicle's own OBD data stream information, and the measurement data of the maintenance personnel of the M station are input into the fault diagnosis model established based on fuzzy logic, and after fuzzy reasoning calculation, the fault type and fault cause are determined according to the diagnosis output rules; the fault diagnosis model established based on fuzzy logic includes:

[0006] A fuzzy mathematical model for diesel engine fault diagnosis is established. The relationship between the fault type and diagnostic parameters of diesel vehicles is described based on fuzzy mathematics, thereby realizing fault diagnosis:

[0007] Assume that the set U represents different fault types that cause diesel engine exhaust pollutants to exceed the standard or power to fail to meet the standard, including: Y1 is poor injection atomization, Y2 is too low injection pressure, Y3 is too large injection volume, Y4 is low SCR conversion efficiency, Y5 is SCR carrier blockage, Y6 is intake system failure, Y7 is combustion chamber carbon deposition, and the set U is expressed as follows:

[0008] U={Y1,...,Y j ,...,Y7}

[0009] Among them, Y j is the jth fault type;

[0010] Assume that set V is the fault symptom set. Since the fault symptom is determined by the diagnostic parameter value, the fault symptom is represented by the diagnostic parameter value, including: X1: the nitrogen oxide concentration at 80% maximum power speed point (ppm), X2: the larger smoke value between the 100% maximum power speed point and the 80% maximum power speed point (m -1 ), X3: the ratio of the measured maximum wheel power at 100% maximum power speed point to the rated power of the engine, X4: the ratio of the intake flow rate or intake pressure value to the lower limit of the intake flow rate or intake pressure of a diesel engine of the same model and good technical condition under the same working conditions of the tested diesel engine, X5: the exhaust back pressure value (bar), X6: the ratio of the actual common rail pressure value to the lower limit of the common rail pressure of a diesel engine of the same model and good technical condition under the same working conditions of the tested diesel engine, and the set V is expressed as follows:

[0011] U={X1,...,X i ,...,X6}

[0012] Among them, X i is the i-th diagnostic parameter;

[0013] Let x i For X i The specific diagnostic parameter values ​​for (x1,...x i ,...,x6), establish the membership function To characterize x i Belongs to X i The degree of membership of the diagnostic parameter can be expressed as:

[0014]

[0015] Wherein, the membership function is a piecewise function, and the specific expression is:

[0016]

[0017]

[0018]

[0019]

[0020]

[0021]

[0022] The membership of the diagnostic parameters is normalized as follows:

[0023]

[0024] The membership of the normalized diagnostic parameter can be expressed as X' = (x'1, x'2, ..., x'6);

[0025] According to the principle of fuzzy reasoning synthesis, the fuzzy relationship equation between fault type and diagnostic parameters can be obtained as follows:

[0026]

[0027] Among them, Y' is the fuzzy vector of fault type, Y'=(y′1,...,y' j ,...,y'7),y′ j is the membership degree of fault type Y, is a fuzzy logic operator, and the operation rules can be expressed as: Among them, the fuzzy diagnosis matrix R is obtained based on the analysis and collation of actual diagnosis and maintenance data combined with the experience review of maintenance industry experts. The expression of the fuzzy diagnosis matrix R is:

[0028]

[0029] Among them, r ij Represents the membership relationship of the i-th diagnostic parameter to the j-th fault type.

[0030] Therefore, the specific diagnostic parameter values ​​are obtained from the multi-source information consisting of the exhaust pollutant test result information of station I, the diesel vehicle's own OBD data stream information, and the maintenance personnel measurement data of station M. The fuzzy vector of the fault type is calculated by the above method, and the fault type and fault cause are determined according to the diagnostic output rule. The diagnostic output rule is:

[0031] The membership of the fault types in the fuzzy vector after fuzzy operation is sorted from large to small, and the first two are set as A and B respectively, and the output rule of the diagnosis result is determined as follows:

[0032] Rule 1: The diagnostic system generally outputs two fault types;

[0033] Rule 2: If A≥0.2, then directly output the largest fault type corresponding to A; if A≤0.2, then the system outputs "unable to diagnose";

[0034] Rule 3: If A ≥ 0.2, calculate the threshold β of the second fault type output as follows:

[0035]

[0036] If B≥β, the fault type corresponding to B is output; if B<β, the fault type corresponding to B is not output.

[0037] The above method comprehensively uses the test results of the periodic inspection of diesel vehicles at station I and the loading and deceleration method specified in the "Emission Limits and Measurement Methods for Pollutants from Diesel Vehicles (Free Acceleration Method and Loading and Deceleration Method)", OBD diagnostic parameters, and manual measurement data at station M, and can accurately diagnose the fault type and cause of diesel vehicles with excessive exhaust pollutants or substandard power of National IV and National V models with high-pressure common rail fuel injection system + selective catalytic reduction system (SCR) as the technical route. This method can accurately diagnose the fault of diesel engines that are not obvious in daily use, but have excessive exhaust pollutants or substandard power during regular emission inspection and loading and deceleration testing, and output the cause of the fault for reference by maintenance personnel. BRIEF DESCRIPTION OF THE DRAWINGS

[0038] Figure 1 The present invention is a framework diagram of a method for diagnosing excessive diesel engine emissions faults based on multi-source information in an embodiment of the present invention. DETAILED DESCRIPTION

[0039] In order to facilitate the understanding of the present invention, the present invention will be described more fully below in conjunction with the relevant drawings. The following description is essentially only exemplary and is not intended to limit the application or use of the present invention. On the contrary, the purpose of providing embodiments is to make the disclosure of the present invention more thorough and comprehensive.

[0040] like Figure 1As shown, a method for diagnosing diesel engine emission exceeding standard faults based on multi-source information is characterized in that it includes: for diesel vehicles whose exhaust pollutants exceed standards or whose power does not meet standards during loading and deceleration detection, which implement National IV or National V emission standards, and whose emission control adopts high-pressure common rail and selective catalytic reduction (SCR) technology routes, based on the exhaust pollutant test result information of diesel vehicles at Station I, the exhaust pollutant test result information of diesel vehicles at Station I includes: the smoke value at 100% maximum power speed point and the measured maximum wheel-side power, the smoke value at 80% maximum power speed point and the nitrogen oxide concentration, and the rated power of the engine; the diesel engine's own OBD data stream information, the diesel engine's own OBD data stream information includes, when the diesel engine is in the hot engine, no-load, and full-throttle working conditions, using the OBD diagnostic instrument to read the corresponding intake flow value or intake pressure value, common rail pressure value, and obtain the intake flow rate or intake pressure, common rail pressure lower limit value of a diesel engine of the same model and good technical condition as the tested diesel engine under the same working conditions; the M station maintenance personnel measure the diesel vehicle data, the M station maintenance personnel measure the diesel vehicle data, including the exhaust back pressure value (relative pressure of engine exhaust) measured by a barometer on the exhaust pipe upstream of the SCR carrier under high speed (full throttle speed) working conditions when the diesel engine is hot and idling; the exhaust pollutant test result information of the I station, the diesel vehicle's own OBD data stream information, and the M station maintenance personnel's measurement data are input into the fault diagnosis model established based on fuzzy logic, and after fuzzy reasoning calculation, the fault type and fault cause are determined according to the diagnosis output rules; the fault diagnosis model established based on fuzzy logic includes:

[0041] A fuzzy mathematical model for diesel engine fault diagnosis is established. The relationship between the fault type and diagnostic parameters of diesel vehicles is described based on fuzzy mathematics, thereby realizing fault diagnosis:

[0042] Assume that the set U represents different fault types that cause diesel engine exhaust pollutants to exceed the standard or power to fail to meet the standard, including: Y1 is poor injection atomization, Y2 is too low injection pressure, Y3 is too large injection volume, Y4 is low SCR conversion efficiency, Y5 is SCR carrier blockage, Y6 is intake system failure, Y7 is combustion chamber carbon deposition, and the set U is expressed as follows:

[0043] U={Y1,...,Y j ,...,Y7}

[0044] Among them, Y j is the jth fault type;

[0045] Assume that set V is the diagnostic parameter set, which specifically includes: X1: NOx concentration at 80% maximum power speed point (ppm), X2: smoke density value (m -1), X3: the ratio of the measured maximum wheel power at 100% maximum power speed point to the rated power of the engine, X4: the ratio of the intake flow rate or intake pressure value to the lower limit of the intake flow rate or intake pressure of a diesel engine of the same model and good technical condition under the same working conditions of the tested diesel engine, X5: the exhaust back pressure value (bar), X6: the ratio of the actual common rail pressure value to the lower limit of the common rail pressure of a diesel engine of the same model and good technical condition under the same working conditions of the tested diesel engine, and the set V is expressed as follows:

[0046] V={X1,...,X i ,...,X6}

[0047] Among them, X i is the i-th diagnostic parameter;

[0048] Let x i For X i The specific diagnostic parameter values ​​for (x1,...x i ,...,x6), establish the membership function To characterize x i Belongs to X i The degree of membership of the diagnostic parameter can be expressed as:

[0049]

[0050] Wherein, the membership function is a piecewise function, and the specific expression is:

[0051]

[0052]

[0053]

[0054]

[0055]

[0056]

[0057] The membership of the diagnostic parameters is normalized as follows:

[0058]

[0059] The membership of the normalized diagnostic parameter can be expressed as X' = (x'1, x'2, ..., x'6);

[0060] According to the principle of fuzzy reasoning synthesis, the fuzzy relationship equation between fault type and diagnostic parameters can be obtained as follows:

[0061]

[0062] Among them, Y' is the fuzzy vector of fault type, Y'=(y′1,...,y' j ,...,y'7),y′ j Fault type Y j The membership degree of is a fuzzy logic operator, and the operation rules can be expressed as: Among them, the fuzzy diagnosis matrix R is obtained based on the analysis and collation of actual diagnosis and maintenance data combined with the experience review of maintenance industry experts. The expression of the fuzzy diagnosis matrix R is:

[0063]

[0064] Among them, r ij Represents the membership relationship of the i-th diagnostic parameter to the j-th fault type.

[0065] Therefore, the specific diagnostic parameter values ​​are obtained from the multi-source information consisting of the exhaust pollutant test result information of station I, the diesel vehicle's own OBD data stream information, and the maintenance personnel measurement data of station M. The fuzzy vector of the fault type is calculated by the above method, and the fault type and fault cause are determined according to the diagnostic output rule. The diagnostic output rule is:

[0066] The membership degree y′ of the fault type in the fuzzy vector after fuzzy operation j Sort from largest to smallest, the first two are set as A and B respectively, and the output rule of the diagnosis result is determined as follows:

[0067] If A≤0.2, the system output is "unable to diagnose";

[0068] If A≥0.2, calculate the threshold β of the second fault type output as follows:

[0069]

[0070] If B≥β, then the fault types corresponding to A and B are output respectively; if B<β, then only the fault type corresponding to A is output.

[0071] The above method comprehensively utilizes the test results of the periodic inspection of diesel vehicles with loaded deceleration conditions at station I as stipulated in the "Emission Limits and Measurement Methods for Pollutants from Diesel Vehicles (Free Acceleration Method and Loaded Deceleration Method)", OBD diagnostic parameters and manual measurement data at station M. It can accurately diagnose the fault types and causes of diesel vehicles with excessive exhaust pollutants or substandard power of National IV and National V models with a high-pressure common rail fuel injection system + selective catalytic reduction system (SCR) as the technical route.

[0072] In this embodiment, according to the membership function A set of diagnostic parameters X'=(x1,...x i ,...,x6) The fuzzy relationship matrix R is:

[0073]

[0074] In this embodiment, the nitrogen oxide concentration at the 80% maximum power speed point is 367ppm, and the smoke density value at the 100% maximum power speed point and the 80% maximum power speed point, whichever has the larger value, is 1.47m -1 , the ratio of the measured maximum wheel power at 100% maximum power speed point to the engine rated power is 0.43, the ratio of the intake flow rate or intake pressure value to the lower limit of the intake flow rate or intake pressure of the same model and good technical condition diesel engine under the same working conditions is 0.97, the exhaust back pressure value is 6.8bar, the actual common rail pressure value is 0.51 to the lower limit of the common rail pressure of the same model and good technical condition diesel engine under the same working conditions. According to the membership function We get X = [0, 1, 0.91, 0, 0, 0.97], and after normalization X' = [0, 0.3472, 0.3160, 0, 0, 0.3368]. Calculation, that is, the calculation result is Y'=[0.2119, 0.4091, 0.1374, 0, 0.0887, 0.1630, 0]. Taking 0.2119 in Y' as an example, the calculation process is:

[0075] 0.2119=0×0.1+0.3472×0.15+0.316×0.25+0×0+0×0+0.3368×0.24.

[0076] According to the diagnostic output rules, the output fault types are: Y2 injection pressure is too low and Y1 injection atomization is poor. The specific fault locations and causes can be seen from the table below: fuel filter blockage, fuel line dirty blockage or leakage, fuel pump failure, high-pressure pump failure, pressure relief valve failure, excessive fuel return from the injector, fuel metering unit failure or carbon deposits on the injector nozzle, wear or leakage of the injector precision parts, and improper adjustment of the injector with small flow.

[0077]

[0078] In this embodiment, the fault diagnosis model established based on fuzzy logic performs fault diagnosis on 150 diesel vehicles that have excessive exhaust pollutants or power that does not meet the standard during loading and deceleration detection, implement National IV or National V emission standards, and use high-pressure common rail and selective catalytic reduction (SCR) technology routes for emission control. The real fault location is then manually disassembled to check and compare with the diagnosis result of the fault diagnosis model established by fuzzy logic. If the diagnosis output covers the real fault location, it is "diagnosis correct". If it does not meet the diagnosis output rules, it is "unable to diagnose". If the diagnosis output does not cover the real fault location, it is "diagnosis error". The specific diagnosis results are as follows:

[0079] Diagnostic accuracy Undiagnosed rate Diagnostic error rate 90.67％ 8.00％ 1.33％

[0080] The specific implementations described above may have various modifications and changes within the scope of the present invention and the claims. Therefore, the implementations described above do not constitute the protection scope of the claims of the present invention.

Claims

1. A diesel engine emission excessive fault diagnosis method based on multi-source information, characterized in that: The test results of diesel vehicle exhaust pollutants at the inspection station, the diesel vehicle's own OBD data stream information, and the data measured by the maintenance personnel at the maintenance station are input into the fault diagnosis model established based on fuzzy logic. After fuzzy reasoning calculation, the fault type is determined; The steps include: Assume that the set U represents the different fault types that cause diesel engine exhaust pollutants to exceed the standard or power to fail to meet the standard, including: Y1 is poor injection atomization, Y2 is too low injection pressure, Y3 is too large injection amount, Y4 is low SCR conversion efficiency, Y5 is SCR carrier blockage, Y6 is intake system failure, Y7 is combustion chamber carbon deposition, U={Y1,...,Y j ,...,Y7} Among them, Y j is the jth fault type; j = 1, 2, 3, ... 7; Assume that set V is the diagnostic parameter set, which specifically includes: X1: NOx concentration at 80% maximum power speed point (ppm), X2: smoke density value (m -1 )、X3: Ratio of the measured maximum wheel power at 100% maximum power speed point to the rated power of the engine, X4: Ratio of the intake flow rate or intake pressure value to the lower limit of the intake flow rate or intake pressure of a diesel engine of the same model and good technical condition under the same working conditions of the tested diesel engine, X5: Exhaust back pressure value (bar), X6: Ratio of the actual common rail pressure value to the lower limit of the common rail pressure of a diesel engine of the same model and good technical condition under the same working conditions of the tested diesel engine, V={X1,...,X i ,...,X6} Among them, X i is the i-th fault parameter; i=1, 2, 3, ...6; Let x i For X i The specific diagnostic parameter values ​​for (x1,...x i ,...,x6), establish the membership function To characterize x i Belongs to X i The degree of membership of the diagnostic parameter is expressed as: The membership of the diagnostic parameters is normalized as follows: The membership degree of the normalized fault parameter can be expressed as X' = (x'1, x'2, ..., x'6); The fuzzy relationship equation between fault type and diagnostic parameters is: Among them, Y' is the fuzzy vector of fault type, Y'=(y1',...,y j ′,...,y7′), y j ′ is the fault type Y j The membership degree of is a fuzzy logic operator, and the operation rules can be expressed as: The expression of fuzzy diagnosis matrix R is: Among them, r ij represents the i-th fault parameter X i For the jth fault type Y j The membership relationship of The membership degree y of the fault type in the fuzzy vector after fuzzy operation is j ' Sort from large to small, the first two are set as A and B respectively, and the output rule of the diagnosis result is determined as: If A≤0.2, the system output is "unable to diagnose"; If A≥0.2, calculate the threshold β of the second fault type output as follows: If B≥β, then the fault types corresponding to A and B are output respectively; if B<β, then only the fault type corresponding to A is output.

2. The method for diagnosing diesel engine exhaust gas faults according to claim 1, characterized in that: The membership function is a piecewise function, and the specific expression is:

3. The method for diagnosing diesel engine exhaust gas faults according to claim 1, characterized in that: The diesel engine is a diesel engine that exceeds the exhaust pollutant standard or does not meet the power standard during the loading and deceleration detection, implements the National IV or National V emission standards, and uses high-pressure common rail and selective catalytic reduction (SCR) technology for emission control. The fuzzy diagnosis matrix R is:

4. The method for diagnosing diesel engine exhaust gas faults according to claim 1, characterized in that: When outputting the fault type, the fault location and cause corresponding to the fault type are also output; The fault locations and causes corresponding to different fault types are as follows:

5. The method for diagnosing diesel engine exhaust gas faults according to claim 1, characterized in that: Diagnostic Parameters X i The specific diagnostic parameter value x i Information on diesel vehicle exhaust pollutant test results from inspection stations, diesel vehicle’s own OBD data stream information, and diesel vehicle data measured by maintenance personnel at maintenance stations.

6. The method for diagnosing diesel engine exhaust gas faults according to claim 5, characterized in that: The test results of diesel vehicle exhaust pollutants at the testing station include: smoke density value at full throttle, 100% maximum power speed point and measured maximum wheel-side power, smoke density value and nitrogen oxide concentration at 80% maximum power speed point, and engine rated power; the OBD data stream information of the diesel vehicle itself includes, when the diesel engine is in hot engine, no-load, and full throttle conditions, using the OBD diagnostic instrument to read the corresponding intake flow value or intake pressure value, common rail pressure value, and obtain the lower limit of the intake flow rate or intake pressure and common rail pressure of a diesel engine of the same model and in good technical condition as the tested diesel engine under the same conditions; the maintenance personnel of the maintenance station measure the diesel vehicle data, including the exhaust back pressure value measured by a barometer on the exhaust pipe upstream of the SCR carrier when the diesel engine is in a hot engine and no-load state and under full throttle speed conditions.

Images