KR102473942B1 - Diesel Particulate Filter device and operating method therefor - Google Patents

Abstract

According to one embodiment of the present invention, a diesel burner configured to increase the temperature of the exhaust gas introduced by the engine of the vehicle; A diesel oxidation catalyst oxidizing the exhaust gas passing through the diesel burner; a filter for collecting particulate matter in the exhaust gas that has passed through the diesel oxidation catalyst; Located between the engine and the diesel burner, a first exhaust temperature (Te) of the exhaust gas, a second exhaust temperature (Tb) between the diesel burner and the diesel oxidation catalyst, and a temperature between the diesel oxidation catalyst and the filter a temperature measuring unit measuring at least one of a third exhaust temperature (Tf) and a fourth exhaust temperature (T4) behind the filter; and a processor configured to predict an operating state of the engine based on the measured exhaust temperature, wherein the processor is configured to stop the operation of the diesel burner when the operating state of the engine is predicted to be abnormal.

Description

Soot reduction device and operation method therefor {Diesel Particulate Filter device and operating method therefor}

Embodiments of the present invention relate to a particulate matter reduction device for predicting engine failure and controlling a post-processing device, and an operating method therefor.

Devices installed to reduce air pollutants emitted from diesel engines include SCR (Selective Catalytic Reduction), EGR (Exhaust Gas Recirculation), and DPF (Diesel Particulate Filter). SCR is a catalyst, EGR is circulating, and DPF is a device that reduces exhaust gas through a filter.

Specifically, the DPF is a kind of exhaust post-processing device that physically captures and burns PM (particulate matter) in the exhaust gas of a diesel engine to remove them. The DPF's PM burning operation does not always run while the car is running, and it works when the filter is filled with dust to some extent.

When this function is activated, the temperature of the filter is raised to 600 degrees Celsius to burn the collected PM. This burning of PM is called regeneration.

However, if an engine malfunction occurs, the amount of PM emitted from the engine can increase more than twice as much as in normal operation, and the amount of PM collected by the DPF filter may exceed the allowable collection limit. When the function is activated, the excessively collected PM burns instantaneously and generates high heat of 1000 degrees Celsius or more, which can damage the filter and have additional adverse effects on the engine.

Therefore, it is necessary to detect abnormal symptoms of the engine early and to warn the user (driver) in order to protect the engine and DPF filter when an abnormality occurs in the engine, and to control DPF regeneration in this case.

Reference may be made to Korean Patent Publication No. 10-2015-0173025 filed on 2015.12.07 as a prior proposal.

Embodiments of the present invention are aimed at providing a smoke reduction device and an operating method therefor that accurately predict the operating state of the engine using big data, warn when an abnormality occurs, and accordingly control the operation of the post-processing device. .

It relates to a method for predicting engine failure of a vehicle equipped with a diesel particular filter (DPF). In normal vehicles, exhaust temperatures are almost similar under the same vehicle speed and back pressure conditions, and engine abnormalities such as engine fuel injection system, cooling system, and lubrication system exhaust temperature increases. Using this, engine (cooling system, lubrication system, injection system) failure can be predicted. Specifically,

1) measuring the temperature by installing a temperature sensor in the exhaust pipe and filter unit to measure the exhaust gas temperature;

2) Among the temperatures measured in the measurement step, using a deep learning AI model that combines supervised learning and unsupervised learning under the same vehicle speed and back pressure conditions, whether the exhaust temperature is not common (abnormal high temperature cognition) determining;

3) When an unusual temperature condition occurs, informing OBD (On-Board Diagnosis) of an engine abnormality and blocking the operation of a DPF regeneration device (diesel burner).

According to one embodiment of the present invention, a particulate matter reduction device attached to a vehicle includes a diesel burner configured to increase a temperature of exhaust gas introduced by an engine of the vehicle; A diesel oxidation catalyst that oxidizes the exhaust gas passing through the diesel burner (Diesel Oxidation Catalyst); a filter for collecting particulate matter in the exhaust gas that has passed through the diesel oxidation catalyst; Located between the engine and the diesel burner, the first exhaust temperature (Te) of the exhaust gas, the second exhaust temperature (Tb) between the diesel burner and the diesel oxidation catalyst, and the third exhaust temperature (Tf) between the diesel oxidation catalyst and the filter and a temperature measuring unit measuring at least one of the fourth exhaust temperature T4 behind the filter. and a processor unit predicting an operating state of the engine based on the measured exhaust temperature, wherein the processor is configured to stop the operation of the diesel burner when the operating state of the engine is predicted to be abnormal.

According to one embodiment of the present invention, it is possible to accurately predict the operating state of the engine by utilizing big data, and furthermore, it is possible to prevent malfunction of the post-processing device.

1 is a diagram showing the configuration of a particulate matter reduction device according to an embodiment of the present invention.
2 is a block diagram illustrating and describing a computing environment including a computing device suitable for use in example embodiments.
3 is a sequence-chart illustrating an operating method of a particulate matter reduction device according to an embodiment of the present invention.
4 is an exemplary diagram for explaining a method for predicting an engine state according to an embodiment of the present invention.
5 is a conceptual diagram for explaining a threshold range according to an embodiment of the present invention.
6 and 7 are diagrams showing learning patterns according to machine learning according to an embodiment of the present invention.

Hereinafter, various embodiments of the present invention will be described with reference to the accompanying drawings. It should be understood that the present invention is not limited to specific embodiments, and includes various modifications, equivalents, and/or alternatives of the embodiments of the present invention. In connection with the description of the drawings, like reference numerals may be used for like elements.

In this document, expressions such as "has", "may have", "includes", or "may include" refer to the presence of a corresponding feature (eg, numerical value, function, operation, or component such as a part). , which does not preclude the existence of additional features.

In this document, expressions such as "A or B", "at least one of A and/and B", or "one or more of A or/and B" may include all possible combinations of the items listed together. . For example, "A or B", "at least one of A and B", or "at least one of A or B" includes (1) at least one A, (2) at least one B, Or (3) may refer to all cases including at least one A and at least one B.

Expressions such as "first", "second", "first", or "second" as used herein may modify various elements, in any order and/or importance, and may refer to one element as another. It is used to distinguish from components, but does not limit the components. For example, without departing from the scope of rights described in this document, a first element may be called a second element, and similarly, the second element may also be renamed to the first element.

As used in this document, the expression "configured to" means "suitable for", "having the capacity to", depending on the situation. ", "designed to", "adapted to", "made to", or "capable of" can be used interchangeably. The term "configured (or set) to" does not necessarily mean "specifically designed to."

In this document, for example, “command”, “instruction”, “control information”, “message”, “messages” transmitted and received between the first electronic device(s) and the second electronic device(s). "Information", "data", "packet", "data packet", "intent" and/or "signal" means any human-perceivable idea or specific electrical representation (e.g., digital code/analog physical quantity) or may refer to itself. It will be apparent to those skilled in the art that the above-listed exemplary expressions can be interpreted in various ways depending on the context in which they are used. In this document, “A is greater than B” means not only “A is greater than B”, but also “A is greater than or equal to B”.

Terms used in this document are only used to describe a specific embodiment, and may not be intended to limit the scope of other embodiments. Singular expressions may include plural expressions unless the context clearly dictates otherwise. Terms used herein, including technical or scientific terms, may have the same meaning as commonly understood by a person of ordinary skill in the technical field described in this document. Among the terms used in this document, the terms defined in general dictionaries may be interpreted as having the same or similar meaning to the meaning in the context of the related art, and unless explicitly defined in this document, in an ideal or excessively formal meaning. not interpreted In some cases, even terms defined in this document cannot be interpreted to exclude the embodiments of this document.

1 is a diagram showing the configuration of a particulate matter reduction device according to an embodiment of the present invention.

As shown in FIG. 1, the particulate matter reduction device according to an embodiment of the present invention includes a diesel burner 102, a pressure measuring unit 103, a temperature measuring unit 104, a vehicle speed measuring unit 105, and a processor unit ( 106). The exhaust gas reduction device according to an embodiment of the present invention may be attached to a vehicle, and the vehicle includes a diesel fuel tank 108, a fuel borne catalyst tank (FBC tank) 110, and a pump module 112 ), an engine 114, an On-Board Diagnosis (ODB) 116, a Diesel Oxidation Catalyst (DOC) 118, and a filter 120. Note that, according to embodiments, a configuration included in a vehicle may be defined as a configuration of a particulate matter reduction device.

The diesel burner 102 is a component of a particulate reduction device, and may be a point where exhaust gas from the engine 114 of the vehicle is first introduced. For convenience, one end into which exhaust gas is introduced from the particulate abatement device may be defined as a front end, and the other end through which exhaust gas is discharged through the particulate abatement device may be defined as a rear end.

The diesel burner 102 may be configured to increase the temperature of the incoming exhaust gas. That is, the diesel burner 102 may forcibly increase the temperature of the exhaust gas for filter regeneration.

The pressure measurement unit 103 may measure engine back pressure or differential pressure of a filter.

The temperature measuring unit 104 may measure the temperature of the exhaust gas at at least one point in the path through which the exhaust gas passes in the particulate reduction device. According to one embodiment, the temperature measurement unit 104 may include a plurality of sensors for measuring the temperature of the exhaust gas in the exhaust gas reduction device. The plurality of sensors may be installed on a path through which exhaust gas in the exhaust gas reduction device passes.

The temperature measurement unit 104 may be located between the engine 114 and the diesel burner 102 to measure the exhaust temperature of the exhaust gas. Referring to FIG. 1 , the exhaust temperature of the exhaust gas between the engine 114 and the diesel burner 102 may be referred to as Te.

According to an embodiment, the temperature measuring unit 104 measures the exhaust gas temperature (Tb) between the diesel burner 102 and the diesel oxidation catalyst, and the temperature between the diesel oxidation catalyst 118 and the filter 120. At least one of the exhaust gas temperature Tf and the exhaust gas temperature T4 behind the filter 120 may be further measured.

The speed measurement unit 105 may measure the vehicle speed or the number of revolutions of the engine.

The processor unit 106 may predict an operating state of the engine based on the measured exhaust temperature Te. Specifically, as shown in the graph of the vehicle in which the actual engine abnormality occurred in FIG. 4, a specific range of exhaust gas temperature and pressure exists at a specific speed. ), it is possible to predict the operating state of the engine based on.

For example, the processor unit 106 may be implemented to predict the operating state of the engine through a comparison between a preset threshold and the exhaust temperature Te, or to predict the operating state of the engine through a procedure of setting and updating a threshold range. can

In addition, the processor unit 106 may predict the operating state of the engine by applying at least one of deep learning and machine learning and continuous data of various sensors such as vehicle speed and exhaust temperature.

A method of predicting operating states of various engines performed by the processor unit 106 will be described in detail with reference to drawings to be described later.

The processor unit 106 may stop the operation of the diesel burner 102 when the engine state is predicted to be abnormal. Also, when the engine state is predicted to be abnormal, the processor unit may generate a warning message. A warning message may be transmitted to the self-diagnosis device 116 to indicate an engine abnormality. That is, damage to the engine 114 and the filter 120 that may occur due to excessive combustion of soot generated when the engine 114 operates abnormally can be prevented in advance.

Also, when the engine state is predicted to be abnormal, the processor unit 106 may control the operation of the pump module 112 to stop or avoid regeneration.

The processor unit 106 determines the control states of the diesel burner 102, the oxidation catalyst 118, and the filter 120 based on the exhaust gas temperatures Tb, Tf, and T4 for each location measured by the temperature processing unit. can be checked. Tf and T4 may be used to determine whether or not the regeneration temperature is reached at the rear end of the DOC and whether or not the regeneration temperature is reached at the rear end of the DPF, respectively.

According to one embodiment, the processor unit 106 may be an electronic control unit (ECU) mounted on a vehicle.

The diesel fuel tank 108 may store fuel. According to one embodiment, the fuel may be diesel.

The fuel-added catalyst tank 110 may store a fuel-added catalyst (FBC, Fuel Borne Catalyst) supplied to diesel fuel. The fuel-added catalyst tank 110 may supply fuel supplied from the diesel fuel tank 108 to the engine 114 under the control of the processor unit 106 .

The pump module 112 may control the diesel burner 102 . Specifically, the filter is naturally regenerated when the exhaust temperature exceeds a certain temperature condition (about 350 degrees or more), but when the constant temperature is not reached at low speed, etc., the exhaust temperature is forcibly raised using a diesel burner. Regeneration temperatures can be reached.

The engine 114 is a device that provides power to the vehicle, and can generate power by receiving fuel and then discharge exhaust gas. Engine 114 according to an embodiment may be a diesel engine.

The self-diagnosis device (OBD unit: 116) indicates whether the product is operating or has an abnormality and can store operation/abnormality data like a black box. Specifically, the measured exhaust temperature and the predicted operating state of the engine may be stored in the self-diagnosis device 116 . The operating state of each of the oxidation catalyst and the filter may be stored in the self-diagnosis device 116 .

The oxidation catalyst (DOC: 118) may perform a function of oxidizing total hydrocarbons and carbon monoxide in exhaust gas using a catalyst and oxidizing nitrogen monoxide to nitrogen dioxide.

The filter 120 means a Diesel Particulate Filter and can reduce the collected PM under a certain temperature condition after collecting particulate matter (PM).

2 is a block diagram illustrating and illustrating a computing environment 10 including a computing device suitable for use in example embodiments. In the illustrated embodiment, each component may have different functions and capabilities other than those described below, and may include additional components other than those described below.

The illustrated computing environment 10 includes a computing device 12 . In one embodiment, computing device 12 may be processor unit 106 .

Computing device 12 includes at least one processor 14 , a computer readable storage medium 16 and a communication bus 18 . Processor 14 may cause computing device 12 to operate according to the above-mentioned example embodiments. For example, processor 14 may execute one or more programs stored on computer readable storage medium 16 . The one or more programs may include one or more computer-executable instructions, which when executed by processor 14 are configured to cause computing device 12 to perform operations in accordance with an illustrative embodiment. It can be.

Computer-readable storage medium 16 is configured to store computer-executable instructions or program code, program data, and/or other suitable form of information. Program 20 stored on computer readable storage medium 16 includes a set of instructions executable by processor 14 . In one embodiment, computer readable storage medium 16 includes memory (volatile memory such as random access memory, non-volatile memory, or a suitable combination thereof), one or more magnetic disk storage devices, optical disk storage devices, flash memory devices, other forms of storage media that can be accessed by computing device 12 and store desired information, or any suitable combination thereof.

Communications bus 18 interconnects various other components of computing device 12, including processor 14 and computer-readable storage medium 16.

Computing device 12 may also include one or more input/output interfaces 22 and one or more network communication interfaces 26 that provide interfaces for one or more input/output devices 24 . An input/output interface 22 and a network communication interface 26 are connected to the communication bus 18 . Input/output device 24 may be coupled to other components of computing device 12 via input/output interface 22 . Exemplary input/output devices 24 include a pointing device (such as a mouse or trackpad), a keyboard, a touch input device (such as a touchpad or touchscreen), a voice or sound input device, various types of sensor devices, and/or a photographing device. input devices, and/or output devices such as display devices, printers, speakers, and/or network cards. The exemplary input/output device 24 may be included inside the computing device 12 as a component constituting the computing device 12, or may be connected to the processor unit 106 as a separate device distinct from the computing device 12. may be

3 is a sequence diagram illustrating an operating method of a particulate matter reduction device according to an embodiment of the present invention.

Referring to FIGS. 1 to 3 , the processor unit 106 operates at a predetermined operation interval (eg, 60 ms) from a plurality of sensors included in the pressure measurement unit 103, the temperature measurement unit 104, and the vehicle speed measurement unit 105. ), it may be implemented to predict the operating state of the engine by processing the measured sensing data.

For reference, the sensing data measured by the plurality of sensors included in the pressure measuring unit 103, the temperature measuring unit 104, and the vehicle speed measuring unit 105 is stored in an On-board Diagnosis unit (OBD unit). can be stored and updated.

In step S302, the temperature measurement unit 104 is located between the engine 114 of the vehicle and the diesel burner 102 configured to increase the temperature of the exhaust gas introduced by the engine 114, and the exhaust gas of the exhaust temperature (Te) can be measured.

In step S304, the processing unit 106 can predict whether the operating state of the engine 114 is in an abnormal state based on the measured exhaust temperature.

A method of predicting the operating state of the engine 114 using the exhaust temperature Te will be described in more detail with reference to FIGS. 4 to 7 .

In step S306, the process unit 106 may stop the operation of the diesel burner 102 when the operating state of the engine 114 is predicted to be abnormal. The processor unit 106 may transmit a warning message to an on-board diagnosis unit (OBD unit) when the operating state of the engine is predicted to be abnormal.

The particulate matter reduction device may include a diesel oxidation catalyst oxidizing the exhaust gas passing through the diesel burner; and a filter for collecting particulate matter from exhaust gas that has passed through the diesel oxidation catalyst, wherein the temperature measurement unit 104 measures the exhaust gas temperature (Tb) between the diesel burner and the diesel oxidation catalyst. , at least one of an exhaust gas temperature (Tf) between the diesel oxidation catalyst and the filter and an exhaust gas temperature (T4) behind the filter may be further measured.

The processor unit 106 determines the diesel burner 102 based on the pressure, temperature, and speed information collected by the exhaust pressure measuring unit 103, the temperature measuring unit 104, and the speed measuring unit 105. , can be configured to predict the operating state of each of the oxidation catalyst 118 and the filter 120.

Operation states of each of the diesel burner 102, the oxidation catalyst 118, and the filter 120 may be stored in an on-board diagnostic unit (OBD unit).

4 is an exemplary diagram for explaining a method for predicting an engine state according to an embodiment of the present invention.

The X-axis of FIG. 4 corresponds to the time t associated with the engine operation, and may be associated with a plurality of sections T1 to T5 associated with the operating cycle of the engine. Meanwhile, the time shown on the X-axis may be associated with a preset operation interval (eg, 60 ms) for a plurality of sensors included in the pressure measurement unit 103, the temperature measurement unit 104, and the vehicle speed measurement unit 105. have.

For example, the preset operation interval may be preferably 60 ms, and in this case, the time interval (4 minutes and 15 seconds) shown on the X-axis of FIG. 4 is 4,250 times (ie, 255 seconds) the preset operation interval (60 ms). /0.06 sec) may mean a time interval enlarged to a time scale.

The left side of the y-axis in FIG. 4 corresponds to the unit (°C) for the exhaust temperature (Te), the unit (kg/h) associated with the intake air flow rate (MAF) of the engine, and the unit (mbar) associated with the pressure (P1) of the exhaust gas. can do. In addition, the right side of the y-axis may correspond to a unit (km/h) associated with the rotational speed (ie, rpm) of the engine.

When it is determined that the operating state of the engine is abnormal, the processor unit 106 may control the operation of at least one of the diesel burner 102, the diesel oxidation catalyst 118, and the filter 120, which are post-processing devices. .

For example, each time period T1 to T5 corresponding to one engine cycle shown in FIG. 4 may be set based on the rotational speed (ie, rpm) of the engine.

As shown in FIG. 4, the first section (T1 to T2) and the second section (T2 to T3) are sections in the normal state of the engine, the third section (T3 to T4), and the fourth section (T4 to T5). ) and the fifth period (T5 to T6) can be understood as a period in which the engine is in an abnormal state.

For example, it can be seen that the intake air flow rate (MAF) and the exhaust gas pressure (P1) of the engine tend to be similar according to the rotation speed of the engine in the first section (T1 to T2) and the second section (T2 to T3). can In addition, the exhaust temperature (Te) in the first section (T1 to T2) and the second section (T2 to T3) shows a tendency to fluctuate around 200 °C to 400 °C.

Unlike this, in the third section (T3 to T4) to the fifth section (T5 to T6), even though the rotational speed of the engine is set lower than the first section (T1 to T2) and the second section (T2 to T3), the exhaust temperature ( Te) tends to fluctuate around 400°C to 600°C, which is higher than before. In addition, it can be seen that similar tendencies of the engine intake air flow rate (MAF) and exhaust gas pressure (P1) drop in the third period (T3 to T4) and the fourth period (T4 to T5).

In particular, around the sixth time point T6 , engine failures may be accumulated and the engine may be stopped. Accordingly, the third period ( T3 to T4 ) to the fifth period ( T5 to T6 ) may be regarded as a time point at which an abnormality warning according to an engine operating state is required.

In this embodiment, predicting that the operating state of the engine is abnormal means that a warning about the engine's operation is required.

In this embodiment, the prediction of the operating state of the engine can be largely implemented in four ways as follows.

As a first implementation method, the processor unit 106 corresponding to step S304 of FIG. 3 may determine whether the exhaust temperature Te is higher than a preset single threshold value. For example, a single threshold value may be preset to a specific value between 400°C and 600°C.

If it is determined that the exhaust temperature Te is higher than a preset single threshold value (eg, 400° C.), the processor unit 106 performs step S306 and sends a warning message to the self-diagnosis device 116 and performs the post-processing You can control the operation of the device.

As a second method of implementation, the processor unit 106 corresponding to step S304 of FIG. 3 may determine whether the exhaust temperature Te falls within a critical range. For example, the time interval between the time points t−1, t, and t+1 may be understood as the aforementioned preset operation interval (eg, 60 ms).

In this case, the threshold range may be newly defined based on sensing data (eg, Tf and T4) measured at each time point (t-1, t, and t+1).

For example, the middle value M2 of the critical range corresponding to the current time point t may be defined as an average value of Tf and T4 measured at the current time point t. Similarly, the median value (M1) of the critical range corresponding to the previous time point (t-1) is defined as the average value of Tf and T4 measured at the previous time point (t-1), corresponding to the next time point (t+1). The middle value (M3) of the threshold range to be can be defined as the average value of Tf and T4 measured at the next time point (t+1).

That is, the intermediate values M1, M2, and M3 of the threshold range may be newly updated every operation interval (eg, 60 ms).

In other words, the critical range is defined as a value obtained by subtracting and adding the preset boundary value (a) based on the newly defined intermediate value (M1, M2, M3) at each time point (t-1, t, t+1). range can mean. Here, the boundary value (a) may be set to a different value according to an implementation method and an operating environment.

If, as shown in FIG. 5, it is determined that the exhaust temperature Te measured at a specific time point (t+1) is out of the critical range (ie, M3-a to M3+a) corresponding to the specific time point (t+1) , The processor unit 106 may perform step S306, transmit a warning message to the self-diagnosis device 116, and control the operation of the post-processing device.

As a third implementation method, the processor unit 106 corresponding to step S304 of FIG. 3 may predict the operating state of the engine using a pre-learned model as shown in FIGS. 6 and 7 using a machine learning technique.

For example, the pre-learned model uses sensing data (eg, Te, P1) acquired through a plurality of sensors included in the pressure measurement unit 103, the temperature measurement unit 104, and the vehicle speed measurement unit 105. It can be understood that it corresponds to the learning pattern performed by machine learning.

In this case, the learning pattern of the pre-learned model of FIG. 6 may correspond to the normal state of the engine, and the learning pattern of the pre-learned model of FIG. 7 may correspond to the abnormal state of the engine.

Specifically, measuring the distance between the learning pattern of the pre-learned model as shown in FIG. 6 or 7 and the exhaust temperature (Te) corresponding to the same pressure condition measured at each time point (t-1, t, t + 1) The operating state of the engine can be predicted in various ways (Euclidean method, Mahalonobis method).

For example, a Euclidean distance, which is a method of calculating a distance between two points in a multidimensional space, or a Mahalanobis distance, which is a method of calculating a distance on a probability distribution, may be applied.

If the exhaust temperature Te is determined to have an abnormal pattern as shown in FIG. 7, the processor unit 106 performs step S306, sends a warning message to the self-diagnosis unit 116, and controls the operation of the post-processing unit. can

As a fourth implementation method, the processor unit 106 corresponding to step S304 of FIG. 3 may predict the operating state of the engine using the following deep learning technique.

1. Anomaly Detection Technology

- Statistical basis: Method to find outliers that deviate from statistics (mean & variance) - Data must have a normal distribution, and it is difficult to find outliers in high-dimensional data

- Proximity, clustering, tree-based: Determine how far given data is from a lot of sufficiently dense data - Difficult to apply if data density is not high enough, model complexity is high, and initial values are required

- Analysis technology to find the principal component that maximizes data variance - When the data contains outliers, the principal component cannot be found. Labeling of normal data should be performed in a clearly organized state

2. Classification according to abnormal sample definition (Anomaly detection)

- Outlier detection: A method of detecting data that deviate greatly from the normal data distribution as an outlier compared to the existing data distribution. This technique is effective when the deviation of normal data is not large and it is close to normal distribution. It is not appropriate to apply if the abnormal data has characteristics that do not deviate greatly from the distribution of normal interval data, while the daily deviation is large.

3. Classification according to whether or not abnormal samples are used during learning and whether or not there is a label

- Supervised Anomaly Detection: Accuracy is high, but since the frequency of abnormal samples is low, class-imbalance (imbalance) problems are often encountered. To solve this problem, data augmentation, loss function redesign, batch sampling, etc.

- Semi-supervised Anomaly detection (One-class classification): Set a discriminative boundary surrounding normal samples and narrow this boundary as much as possible to consider all samples outside the boundary as abnormal. (Refer to papers such as One-Class SVM and Deep SVDD)

- Unsepervised Anomaly Detection: Assuming that most of the data is a normal sample, learning without acquiring a label. Abnormal data detection using PCA. However, it is sensitive to hyperparameters.

- Isolation forest for anomaly detection: This is a decision tree-based anomaly detection technique that randomly selects dimensions and divides the space based on random criteria to detect outliers.

- LOF (Local Outlier Factor): A learning method that detects abnormal data based on how far away the given data and neighboring data are from the nearby dense data area: Considering all normal values, the computational load is high

- LSTM Autoencoder learning method - It trains normal data and abnormal data with the principle of learning fake data and real data by creating a self-generating model from sequence data. A newer learning method than isolation forest and LOF.

According to an embodiment, the processor unit 106 may perform anomaly detection by performing supervised learning and unsupervised learning in parallel.

Referring to FIG. 4 , after the processor unit 106 of the present invention predicts that the operating state of the engine is abnormal, a phenomenon in which the operation of the engine is actually stopped may occur.

According to an embodiment of the present invention, by utilizing big data, it is possible to accurately predict the operating state of the engine, and furthermore, it is possible to prevent malfunction of the post-processing device.

Although representative embodiments of the present invention have been described in detail above, those skilled in the art will understand that various modifications are possible to the above-described embodiments without departing from the scope of the present invention. . Therefore, the scope of the present invention should not be limited to the described embodiments and should not be defined, and should be defined by not only the claims to be described later, but also those equivalent to these claims.

102: diesel burner
103: pressure measuring unit
104: temperature measuring unit
105: vehicle speed measuring unit
106: processor unit

Claims (6)

In the exhaust gas reduction device attached to the vehicle,
a diesel burner configured to increase the temperature of exhaust gas introduced by the engine of the vehicle;
A diesel oxidation catalyst oxidizing the exhaust gas passing through the diesel burner;
a filter for collecting particulate matter in the exhaust gas that has passed through the diesel oxidation catalyst;
Located between the engine and the diesel burner, a first exhaust temperature (Te) of the exhaust gas, a second exhaust temperature (Tb) between the diesel burner and the diesel oxidation catalyst, and a temperature between the diesel oxidation catalyst and the filter a temperature measuring unit measuring at least one of a third exhaust temperature (Tf) and a fourth exhaust temperature (T4) behind the filter; and
a processor unit predicting an operating state of the engine based on the measured exhaust temperature;
Including,
The processor unit,
configured to stop the operation of the diesel burner when the operating state of the engine is predicted to be abnormal, and to predict the operating state of the engine based on a threshold range associated with the first exhaust temperature Te,
The threshold range is defined based on a boundary value preset in the average values of the third exhaust temperature Tf and the fourth exhaust temperature T4 obtained by the temperature measurement unit, and the threshold range is defined in advance by the temperature measurement unit. updated at each set operation interval,
Smoke abatement device.
According to claim 1,
The processor unit,
When the operating state of the engine is predicted to be abnormal, configured to transmit a warning message to a self-diagnosis device (OBD unit: On-board Diagnosis unit),
Smoke abatement device.
delete delete delete In the operating method of the exhaust gas reduction device according to any one of claims 1 and 2,
Measuring a first exhaust temperature (Te) of the exhaust gas located between the engine of the vehicle and a diesel burner configured to increase the temperature of the exhaust gas introduced by the engine;
determining whether an operating state of the engine is in an abnormal state based on the measured first exhaust temperature (Te); and
stopping the operation of the diesel burner when the operating state of the engine is predicted to be abnormal;
including,
A method of operating a smoke reduction device.

Images