DE102005016132A1 - Exhaust system for internal combustion (IC) engine, makes particulate sensors on upstream side of particulate filter independently operable with respect to their reference conditions - Google Patents

Abstract

Particulate sensors (5-7) for detecting amount of particulates in exhaust gas lie upstream and downstream of a particulate filter (4) in an exhaust system (A). The sensors (5,6) on the upstream side are independently operable with respect to their reference conditions, and have sensing regions located on a common substrate. An independent claim is also included for a method for collecting particulates in exhaust gas of an IC engine.

Description

The The invention relates to an internal combustion engine with an exhaust system comprising a particulate filter and an exhaust gas of the internal combustion engine exposed first and second particle sensor for detecting a Exhaust particulate loading and a method for detecting a Particle load of an exhaust gas of an internal combustion engine, in which detects a deposited on a particle sensor amount of particles becomes.

From the EP 0 525 566 A1 a sensor is known which is exposed to the exhaust gas of an internal combustion engine and can be loaded during operation of the internal combustion engine with the particles contained in the exhaust gas. The level of particle loading can be measured by measuring the electrical resistance between two spaced-apart electrodes of the sensor. When the sensor is loaded with a certain amount of particulates, the sensor is regenerated by activating a heating element and burning off the particulate matter that is predominantly soot so that the sensor is returned to an initial state. If a predetermined number of regenerations of the sensor has taken place, regeneration of the particle filter is initiated. In this way, a particle load of the exhaust gas or a particle loading of a particle filter arranged downstream of the sensor can be determined. During the regeneration phases of the sensor, however, this is not able to detect the particulate load of the exhaust gas. The determination of the load of the particulate filter is therefore subject to inaccuracies and the need for a particulate filter regeneration can not be determined with a desired reliability. The particulate filter regeneration may therefore be delayed, resulting in degraded engine operation.

task The invention is therefore an internal combustion engine and a method indicate which one is more reliable Allow detection of exhaust particulate loading.

These Task is by an internal combustion engine with the characteristics of Claim 1 and by a method having the features of the claim 7 solved.

The inventive internal combustion engine records according to the claim 1 characterized in that a first particle sensor and a second particle sensor are provided, wherein the first and the second particle sensor independently can be brought into a reference state. As a result of this independent operation can the particle sensors delayed in their respective reference state be brought, thereby ensuring is that at least one of the sensors is always ready for use and can detect the particulate load of the exhaust gas. It therefore arises no measuring breaks, so that the loading condition of the particulate filter especially reliable can be determined. It also provides redundancy in monitoring the exhaust quality achieved, so that, for example, in a total failure of one of both sensors detecting the exhaust particulate loading by the other can continue to take place. This way is after a total failure one of the sensors low-emission operation of the internal combustion engine at least for a certain amount of time.

Under an exhaust particulate load is in this context a preferred amount of particulate exhaust gas constituents based on the time or amount of exhaust gas understood which the exhaust system at the installation of the respective sensor happens. A proportion correlating with this amount is stored on the particle sensors, with the sensors being designed that they can deliver a measuring signal which coincides with the deposited Particle quantity correlates. Due to the time-varying Quantity of deposited particles therefore generate the particle sensors each a corresponding, time-varying signal. It is therefore provided that the sensors from time to time in a reference state can be brought. This reference state is to be understood as the state of the particle sensor, in which the sensor signal clearly to a certain amount deposited particles can be assigned. Preferably this is a condition in which the sensor essentially free of loaded particles, in particular free of loaded soot particles is. Accordingly, the sensor signal obtained in this state a zero point signal. In the reference state is thus preferably a zero point calibration of the particle sensor.

to Obtaining the reference state is preferably a conditioning provided such that particles deposited on the particle sensor at least partially removed. It is particularly preferred, the reference state by burning off particles deposited on the respective sensor bring about. For this purpose, the sensor for a predetermined time to an increased Temperature heated so that the predominantly made of soot Burn off particles. In this way, a reset takes place the sensor signal.

It is envisaged that the first particle sensor and the second particle sensor are operated independently of each other and can be brought independently according to the invention in their respective reference state. In this case, the soot burn-off process of a sensor, which is preferably provided for this purpose, is carried out in such a way that the measuring operation of the respective other sensor is not impaired.

In Embodiment of the invention are the first particle sensor and the second particle sensor disposed adjacent to each other in the exhaust system. Both Sensors thus detect the same particulate load of the exhaust gas. The signals or measured values of the two sensors can therefore be at least temporarily be compared with each other, whereby a particularly accurate detection allows the exhaust particulate loading is.

In Further embodiment of the invention, the first particle sensor and the second particle sensor each have a sensory area on, with the sensory areas on a common substrate are arranged. In this way is the space required for the sensors and their connections minimized. Furthermore is guaranteed that the sensors of exhaust gas with almost the same nature flows against so that they capture the same particle load. To this Way you can their signals are compared with each other.

In Further embodiment of the invention are the first particle sensor and the second particulate sensor upstream of the particulate filter in the exhaust system arranged. In this way, the amount of particles flowing into the particulate filter is determined. With known or determined degree of separation of the particulate filter Thus, its loading with particles can be determined continuously and an optimal time for a particle filter regeneration can be determined.

In Another embodiment of the invention is downstream of the particulate filter a third particle sensor arranged in the exhaust system. This allows a determination the filter efficiency or the Partikelabscheidegrads. Thus is allows a diagnosis of the particulate filter. Is exceeded by the third particle sensor determined a predetermined exhaust particulate loading, so is a Filter defect diagnosed.

In Another embodiment of the invention is at least one of the particle sensors designed as impedometric sensor. Preferably, both the first and the second as well as an optionally existing one third particle sensor designed as impedometric sensors and the amount of particulate deposited on a respective sensitive area can over the impedance of an electrode arrangement arranged on a sensor carrier be recorded. The impedance also includes a DC resistance. Preferably, however, the impedometric sensor is provided with an alternating voltage operated. That way you can also low soot loadings be detected, which is not yet to form a DC conductive bridge between two adjacent electrodes and thus an observable change lead the DC resistance.

In Another embodiment of the invention is at least one of particle sensors heated formed. Preferably, both the first and the second as well as an optionally existing third particle sensor can be heated educated. Thus, a burn-off of the respective sensor self-sufficient and independent be carried out by the temperature of the exhaust gas. The sensor can thus at any time in its respective reference state to be brought.

The inventive method is according to claim 7, characterized in that a first particle sensor and a second particle sensor are used and for the first and the second Particle sensor recurring conditioning operations are performed so that the respective particle sensor brought into a reference state becomes. Due to the conditioning processes, the sensors of Deposits free and it becomes a reference state in this way brought about in which the sensors are substantially free of deposited particles are. As typically with increasing particle loading of a sensor whose sensitivity decreases, becomes so recurrent in this way intervals restored the maximum sensitivity of the sensors. Preferably if the conditioning operations are each a soot burn, which by activating a sensor integrated in the heating element with corresponding heating takes place. The time for one each conditioning process is expediently depends on the loading of the sensor. The frequency the conditioning operations is therefore dependent from the particulate load of the exhaust gas to which the sensor is exposed is.

In Embodiment of the method, the conditioning operations for the first and the second particle sensor carried out anticyclically. To this Way is guaranteed that the conditioning operations the two sensors never coincide in time or overlap in time. Consequently, it is constant at least one of the two sensors ready to measure, so that no Measurement breaks arise.

In a further embodiment of the method, the first and the second particle sensor be arranged adjacent to each other in the exhaust system. Thus, the two sensors detect the same exhaust particulate loading and a plausibility check of their signals can be performed. A defect or failure of a sensor can therefore be detected quickly. In this case, however, the exhaust particulate load may still be due to the intact sensor. In this way, the reliability of the measurements is significantly improved.

In Another embodiment of the method is a for filtering Particles from the exhaust gas of the internal combustion engine suitable particle filter provided and the first and the second particle sensor are upstream the particulate filter disposed in the exhaust system of the internal combustion engine. In this way, a purification of the exhaust gas of particles, wherein continuously the load of the particulate filter can be monitored. It is therefore possible To monitor the particulate loading of the particulate filter and if necessary in time to initiate a particle filter regeneration. It can also a total load can be determined so that between a soot loading and an ash charge of the filter can be distinguished.

In Another embodiment of the method is a third particle sensor used downstream of the Partikelfiters. By means of the third particle sensor, in particular correlated with a measured value of an upstream Particle sensor, the filter efficiency and the degree of separation of the particulate filter constantly monitored become. Specifically, the third sensor may cause a particle breakthrough leading Filter defect can be detected.

In Further embodiment of the method, the output signals the first and / or the second and / or the third particle sensor used to control regeneration processes for the particulate filter. It can be both the frequency or the timing of the regeneration processes and their duration controlled or influenced.

In Further embodiment of the method, the output signals the first and / or the second and / or the third particle sensor used to determine a condition of the particulate filter. In addition to the determination of the loading state can also the degree of separation and thus the functionality of the particulate filter are determined. In particular, a suitable evaluation is possible the signals from upstream and downstream of the particulate filter arranged particle sensors a particularly reliable onboard diagnosis of the particulate filter.

advantageous embodiments The invention is illustrated and illustrated in the drawings described below. Here are the above and to be explained below Features not only in the specified feature combination, but also usable in other combinations or in isolation, without to leave the scope of the present invention.

It demonstrate:

1 a schematic block diagram of an internal combustion engine with associated exhaust system,

2 1 schematically illustrated as an exploded view of a preferred embodiment of the particle sensor used according to the invention,

3a the particle sensor of the 2 in a schematic plan view,

3b the loaded with a particle layer particle sensor of 2 in a schematic cross-sectional representation,

4a a simplistic representation of the in 1 illustrated particle sensor in plan view,

4b a highly simplified representation of a second advantageous embodiment of the particle sensor according to the invention used in plan view,

5 a diagram illustrating a typical sensor waveform and

6 a schematic timing diagram of a temporal sequence of operating conditions for two particle sensors.

In 1 is only schematically an internal combustion engine 1 represented with an exhaust system A. According to 1 receives the internal combustion engine 1 a motor vehicle, not shown, combustion air via an intake air line 2 , The combustion exhaust gases are via an exhaust pipe 3 dissipated. In the exhaust pipe 3 are a particle filter 4 and further, not shown, cleaning active components arranged. As particle filter, each filter-effective component can be used, which is suitable for filtering particles from engine exhaust gas. Particularly preferred is the use of a catalytically coated wall-flowed particulate filter honeycomb construction. The internal combustion engine 1 , hereinafter referred to as engine for short, is preferably designed as a diesel engine.

For detecting the exhaust gas particle load, the input side of the particulate filter 4 a first particle sensor 5 and a second particle sensor 6 closely adjacent in the exhaust pipe 3 arranged. The particle sensors 5 . 6 can be arranged on a common carrier or spatially separated from each other on each of a separate carrier. In both cases, it may be appropriate to provide a common housing for both sensors. On the output side of the particle filter 4 is another, third particle sensor 7 intended. The signals of the sensors 5 . 6 . 7 be via signal lines 8th to an electronic control unit 9 directed. The control unit 9 has in addition to means for driving the sensors 5 . 6 . 7 via a computing unit for processing the received data and a memory unit in which, for example, maps are stored and data can be stored, which is not shown in detail. The electronic control unit 9 is also capable of operating the motor depending on the signals 1 and the exhaust system, in particular the particulate filter 4 to control. Representative of the existing control lines is a motor control line 10 drawn to control and detection of engine operation.

In normal operation of the engine 1 the exhaust gases are cleaned by filtering out the soot particles through the particle filter 4 , In general, a gradually increasing soot loading of the particulate filter occurs 4 so that it clogs to an increasing extent and from time to time a particulate filter regeneration, preferably by thermal Rußabbrand is required. The time interval with which these regenerations are required depends largely on the particulate load of the exhaust gas. Although in the control unit 9 a model can be deposited, with which the particle load of the exhaust gas from the respective engine operating point and thus the loading state of the particulate filter 4 However, this is usually not sufficiently accurate. According to the invention, therefore, the particle load of the exhaust gas from the sensors 5 . 6 . 7 determined. In addition, also not shown pressure sensors in front of and behind the particulate filter 4 be provided, which the pressure drop across the particulate filter 4 as a measure of its particle load capture.

The particle sensors 5 . 6 . 7 whose preferred construction in connection with the 2 to 4b is explained in more detail, consists essentially of a platelet-shaped ceramic substrate on which at least two measuring electrodes are applied. They can be applied, for example, in thick film technology, by brushing or spraying or in the form of an interdigital electrode structure. Particles contained in the exhaust gas settle on the surface of the ceramic substrate and thereby change the electrical impedance between the measuring electrodes. The changing impedance is a measure of the amount of settled soot particles. In this way, the particle load of the exhaust gas can be determined. By means of the particle sensors 5 . 6 . 7 However, not only the quality of the exhaust gas, but also the state of the particulate filter 4 be determined, for example, a breakthrough by the particulate filter 4 ,

In particular, the need for particulate filter regeneration by thermal soot burnup by means of the particulate sensors 5 . 6 . 7 detected. By evaluating the corresponding signals is from the control unit 9 the loading condition of the particulate filter 4 continuously determined. If a critical load is detected, as soon as a suitable and from the control unit 9 is recognized as permissible engine operating state, the regeneration of the particulate filter 4 initiated by thermal Rußabbrand. This is done by the control unit 9 the engine operation switched so that an increased temperature of the in the particulate filter 4 incoming exhaust gas results. Preferably, an exhaust gas temperature of about 650 ° C on the input side of the particulate filter 4 set. For this purpose, known measures such as intake air throttling, late fuel post-injection and the like can be taken.

at Regeneration becomes the accumulated particles that become one huge Part consist of carbon or carbon compounds, burned.

Hereinafter, a preferred embodiment of the used according to the invention, executed in planar technology particle sensor based on 2 explained. The particle sensor is on a first, preferably made of alumina ceramic substrate 12 built up.

On the bottom of the first substrate 12 is a heater structure 11 with two associated connections 13 . 14 applied for connection of a heating voltage. Preferably, the heater structure 11 and the connections 13 . 14 in thick film technology, alternatively manufactured in thin film technology. Here are the heater structure 11 and their respective supply voltage designed so that at least 600 ° C, preferably more than 700 ° C can be achieved on the top of the particle sensor. This makes it possible to burn off deposited on the top soot particles.

On the first substrate 12 is a second substrate also preferably formed of alumina ceramic 17 it is advantageous, between the first substrate 12 and the second substrate 17 a preferably closed separating layer 20 from an electrically conductive Ma provide material. In this case, an unillustrated terminal for applying an operating voltage to the separation layer 20 be provided.

On the second substrate 17 is a, preferably also layered executed temperature sensor 16 with two connections 18 . 19 applied. It is advantageous, this example designed as a planar resistance thermometer temperature sensor 16 over the heater structure 11 to arrange.

An insulating layer 15 covers the temperature sensor 16 and the connections 18 . 19 from. On the insulating layer 15 is, preferably also by means of a layer technology, an electrode structure preferably designed as an interdigital structure with comb-like intermeshing conductor tracks 21 (IDK structure) with a first port 22 and a second port 23 applied. Typically, the width of the interconnects of the IDK structure and their spacing is in the range between 1 .mu.m and 1000 .mu.m. Particularly preferred is a distance of a few hundred microns.

The IDK structure 21 forms the actual, with regard to a particle deposition sensory area 24 of the particle sensor. By the formation of the electrode structure 21 As the IDK structure having a plurality of interdigitated electrode strips, high sensitivity and high accuracy of the particle sensor can be achieved in detection of particles deposited between or on the electrodes. Preferably, the sensory area 24 directly above the temperature sensor 16 arranged. In this way, the temperature of the sensory area 24 recorded very precisely and by energizing the heater structure 11 be regulated, so that a temperature-constant measuring operation and a controlled burn-off of the sensory area 24 deposited soot particles can be achieved.

A first supply line 25 and a second supply line 26 are led to a control and evaluation unit, not shown, which supplies the operating voltages necessary for the operation of the particle sensor to the terminals provided and the evaluation of the between the first port 22 and the second port 23 the electrode structure 24 this electrical parameter takes over. In this case, the control and evaluation unit also by the control unit 9 of the 1 be formed.

The 3a and 3b show those related to the 2 explained preferred embodiment of the particle sensor again schematically, with corresponding components are provided with the same reference numerals. It is in 3a a top view of the, the sensory area 24 carrying part of the particle sensor, while in 3b a schematic cross-sectional view is shown. In 3b is the IDK structure 21 with a particle layer 28 covered, which has formed during the operating time there. Through this particle layer 28 the electric field acting between the electrodes is influenced, which manifests itself in a change in the electrical impedance between the electrodes and by the connected control and evaluation unit 29 is detected. Here are only schematic field lines 27 an electrical field between the electrodes caused here by a DC voltage for clarification of the facts drawn.

4a shows again greatly simplified the particle sensor of 2 in plan view. The sensor is preferably designed to be slim with a width in the centimeter range. Preferably, the width is in the range between 0.2 cm and 1.5 cm. According to the invention, as in 1 shown, two independently of each other operating sensors upstream of the particulate filter 4 used. It is advantageous in this context, the two particle sensors on a single substrate 12 to arrange how much simplified in 4b shown. The substrate 12 thus carries two sensory areas 24 . 24 ' and thus two separately operable particle sensors are formed, which the in 1 shown particle sensors 5 . 6 correspond. However, it is also possible to construct the two particle sensors on different substrates, wherein the sensors are preferably arranged closely adjacent in a common housing. In any case, it is provided to operate the particle sensors separately.

The following explains the operation principle and the normal operation of a particle sensor. It is assumed that the sensor is the exhaust of the engine 1 is exposed.

To operate the particle sensor this is first heated. For this purpose, from the control and evaluation unit a heating voltage to the terminals 13 . 14 placed. In this case, a regulation takes place to a predefinable operating temperature of about 300 ° C to 500 ° C by means of the temperature sensor also connected to the control and evaluation unit 16 , From the control and evaluation then an operating voltage to the terminals 22 . 23 the IDK structure 21 created. This operating voltage may be a DC voltage or an AC voltage having a predetermined frequency, which is typically in the range between 10 ² Hz to 10 ⁶ Hz. The measured variable is preferably the complex impedance of the sensory area 24 issued by the control and evaluation unit 29 is evaluated according to real part and / or imaginary part. In the present case, the electrode structure represents 21 a capacitive resistance. By deposition of particles on the IDK structure 21 occurs a change in the imaginary part and / or the real part of the complex impedance, which depends mainly on the amount of deposited particles. By evaluating this measurand can from the control and evaluation 29 therefore, the amount of deposited particles can be determined.

This in 5 Diagram illustrates this situation.

In the in 5 The diagram shown is an example of the resistance or the real part R of the complex impedance of a particle sensor according to 2 to 4b applied as a function of the operating time t. In this case, the particle sensor was exposed to the exhaust gas of a diesel engine with temporally approximately constant particle load. The particles which accumulate on the sensory area over time and consist predominantly of soot or soot-like substances lead to a steady decrease in the resistance R between the electrodes of the IDK structure considered here by way of example. The resulting course is indicated by the in the diagram of 5 represented represented curve. Since the amount of particles depositing per unit of time correlates with the particle load of the exhaust gas, the particulate load of the exhaust gas can be detected by the particle sensor.

Depending on the level of particulate pollution of the exhaust gas, however, particles collect more or less quickly on the sensor, so that the in 5 represented curve may drop more or less steeply. In any case, by appropriate evaluation of the sensor signal, both a currently existing particulate load of the exhaust gas and a particle total emission can be determined over a predefinable time period. Thus, it is possible to detect a received by a downstream particulate filter amount of particulates. As a result, the particle load of the particulate filter can be continuously determined and at given loading values, a particulate filter regeneration can be initiated in order to free the particulate filter from soot burn-up of deposited soot particles and lower its flow resistance, which has increased due to the particulate loading.

As from the graph of the 5 As can be seen, over time, ie with increasing particle loading, the steepness of the curve and thus the sensitivity of the particle sensor decreases. It is therefore intended to condition the particle sensor from time to time or to regenerate in order to restore the output sensitivity and bring the sensor back to a reference state. The time for such regeneration is expediently determined as a function of the size of the sensor signal. For conditioning or regeneration of the heater 11 of the respective sensor activated such that in the associated sensory area 24 a temperature of about 700 ° C is reached, wherein the deposited soot particles burn off. The elevated temperature is maintained until the sensor is free of deposited soot particles. This can be ensured either by specifying a fixed regeneration time or by evaluating the sensor signal during regeneration.

Main task of the particle sensors 5 . 6 . 7 is the rating of the particulate filter 4 in terms of its particle loading and its degree of separation. During a conditioning process, however, the normal measuring operation is interrupted, so that in this time the particulate load of the exhaust gas from the respective particle sensor can not be detected. What the behind the particle filter 4 arranged particle sensor 7 is concerned, so reaches this cleaned of particles of exhaust gas, so that deposit on it particles at a relatively low rate. Accordingly, conditioning operations are for this particle sensor 7 less frequently required. An interruption of readiness to measure means for the monitoring of the particulate filter 4 no drastic loss of functionality. The measurement pause that occurs during a conditioning process is for an onboard diagnosis of the particulate filter 4 thus acceptable.

The particulate load of the exhaust gas is in front of the particulate filter 4 in contrast, much higher. Consequently, for one before the particulate filter 4 arranged particle sensor comparatively often conditioning operations required and the measurement readiness of the sensor is interrupted frequently accordingly. Will only the signal of a single particle sensor in front of the particle filter 4 evaluated, resulting from the measurement breaks an undesirable inaccuracy in determining the condition of the particulate filter 4 , According to the invention, this inaccuracy is due to the use of two independently operable particle sensors 5 . 6 in front of the particle filter 4 avoided. According to the invention are for these particle sensors 5 . 6 provided countercyclical conditioning operations, so that they do not coincide or overlap in time.

The preferred procedure is described below with reference to 6 explained in more detail. In 6 is the time sequence of operating phases of the particle sensors 5 . 6 of the 1 in two corresponding bars 29 . 30 represented over a timeline. areas 31 each identify measuring operating phases in which the respective sensor 5 . 6 is operated normally and the particulate matter of the exhaust gas he summarizes. The phases 31 alternate with contrast comparatively short conditioning phases with Rußabbrand from what through the black marked areas 32 is marked. The particle sensors 5 . 6 are operated according to the invention so that conditioning phases 32 one of the sensors 5 . 6 in measuring operation phases 31 fall from the other sensor.

The procedure according to the invention ensures that the particle load can be detected without interruption, since at least one of the sensors is continuously running 5 . 6 is ready to measure. Since the conditioning phases 32 compared to the measurement operating phases 31 shorter, in addition, for the greater part of the time both sensors are simultaneously available for detecting the particulate load of the exhaust gas. It is therefore envisaged that the signals of the particle sensors 5 . 6 in the phases of joint measurement readiness and linked to a common measured value for the particle load. In this way, the accuracy of determining the exhaust particulate loading and the particulate filter loading can be significantly improved. In addition, it is provided that a plausibility check is performed such that the signal of a sensor 5 . 6 is compared with the signal of the other sensor and checked for plausibility. Dodge those of the sensors 5 . 6 determined values too strong from each other, it can be concluded that a defect of one of the sensors, which is conveniently indicated by a corresponding warning signal. Is a defect in one of the sensors 5 . 6 However, the other is still available to determine the exhaust particulate load and thus to determine the particulate filter state available. Control of particulate filter regeneration is still possible, albeit with a loss of accuracy. In this way it is ensured that in case of a defect of one of the particle sensors 5 . 6 the particulate filter can continue to operate at least approximately properly. In this way, disturbances of the vehicle operation or damage or failure of the particulate filter can be avoided and thus the reliability of the vehicle operation can be improved.

Claims (14)

Internal combustion engine ( 1 ) with an exhaust system (A) comprising a particle filter ( 4 ) and the exhaust gas of the internal combustion engine ( 1 ) exposed particle sensor ( 5 ; 6 ; 7 ) for detecting an exhaust gas particle load, characterized in that a first particle sensor ( 5 ) and a second particle sensor ( 6 ), the first particle sensor ( 5 ) and the second particle sensor ( 6 ) can be independently brought into a reference state. Internal combustion engine ( 1 ) according to claim 1, characterized in that the first particle sensor ( 5 ) and the second particle sensor ( 6 ) are arranged adjacent to each other in the exhaust system (A). Internal combustion engine ( 1 ) according to claim 1 or 2, characterized in that the first particle sensor ( 5 ) and the second particle sensor ( 6 ) each have a sensory area ( 24 . 24 ' ), wherein the sensory areas ( 24 . 24 ' ) on a common substrate ( 12 ) are arranged. Internal combustion engine ( 1 ) according to one of claims 1 to 3, characterized in that the first particle sensor ( 5 ) and the second particle sensor ( 6 ) upstream of the particulate filter ( 4 ) are arranged in the exhaust system (A). Internal combustion engine ( 1 ) according to one of claims 1 to 4, characterized in that downstream of the particulate filter ( 4 ) a third particle sensor ( 7 ) is arranged in the exhaust system (A). Internal combustion engine according to one of claims 1 to 5, characterized in that at least one of the particle sensors ( 5 . 6 . 7 ) is designed as impedometric sensor. Internal combustion engine according to one of claims 1 to 6, characterized in that at least one of the particle sensors ( 5 . 6 . 7 ) is formed heated. Method for detecting a particle load of an exhaust gas of an internal combustion engine ( 1 ), in which one on a particle sensor ( 5 . 6 . 7 ) deposited particle amount is detected, characterized in that a first particle sensor ( 5 ) and a second particle sensor ( 6 ) and for the first particle sensor ( 5 ) and the second particle sensor ( 6 ) recurrent conditioning operations are carried out so that the respective particle sensor ( 5 . 6 ) is brought into a reference state. Method according to claim 8, characterized in that the conditioning operations for the first particle sensor ( 5 ) and the second particle sensor ( 6 ) are carried out countercyclically. Method according to claim 8 or 9, characterized in that the first particle sensor ( 5 ) and the second particle sensor ( 6 ) adjacent to each other in an exhaust system (A) of the internal combustion engine ( 1 ) to be ordered. Method according to one of claims 8 to 10, characterized in that a for filtering particles from the exhaust gas of the internal combustion engine ( 1 ) suitable particle filter ( 4 ) is provided and the first particle sensor ( 5 ) and the second particle sensor ( 6 ) upstream of the particulate filter ( 4 ) to be ordered. Method according to claim 11, characterized in that a third particle sensor ( 7 ) downstream of the particulate filter ( 4 ) is used. Method according to claim 12, characterized in that output signals of the first particle sensor ( 5 ) and / or the second particle sensor ( 6 ) and / or the third particle sensor ( 7 ) for controlling regeneration processes for the particulate filter ( 4 ) are used. Method according to claim 12 or 13, characterized in that output signals of the first particle sensor ( 5 ) and / or the second particle sensor ( 6 ) and / or the third particle sensor ( 7 ) for determining a condition of the particulate filter ( 4 ) are used.