FR3123089A1 - METHOD FOR IMPROVING A MEASUREMENT SIGNAL OF AN OXYGEN SENSOR - Google Patents

Abstract

The invention relates to a method for improving the accuracy of a measurement signal from an oxygen sensor (22, 23) associated with a device (21) for depolluting the exhaust gases of a heat engine ( 10) gasoline comprising: - a step of storing a map of measurement signal transfer functions of the oxygen sensor (22, 23) as a function of a lambda coefficient representative of a richness of air/fuel mixture for different temperatures of the oxygen sensor (22, 23), - a step for determining the temperature of the oxygen sensor (22, 23), - a step for selecting, in the map of transfer functions, the function transfer function closest to the temperature of the oxygen sensor (22, 23) previously determined, and - a step of using the transfer function selected to obtain a measurement signal from the oxygen sensor (22, 23) . Figure 3

Description

METHOD FOR IMPROVING A MEASUREMENT SIGNAL OF AN OXYGEN SENSOR

The present invention relates, in general, to the post-treatment of exhaust gases from a combustion engine, in particular in a motor vehicle.

The invention relates more particularly to a method for improving a measurement signal of an oxygen sensor.

A motor vehicle with a heat engine emits gases, called exhaust gases, following the combustion of fuel and air in the heat engine. The emission of these exhaust gases is regulated by standards aimed at limiting the content of certain pollutants, by post-treatment of the exhaust gases from an engine.

Also, it is known to treat the exhaust gases in the gas exhaust line, in other words after their emission. Such an exhaust line generally comprises a catalyst and a particulate filter, also designated GPF, for "Gasoline Particulate Filters" in English, in the case of a gasoline engine.

In order to be able to effectively measure the oxygen content of the exhaust gases as well as the performance of the oxidation catalyst, oxygen sensors, also called "lambda sensor" are placed respectively upstream and downstream of the catalyst.

As illustrated by the showing the evolution of the voltage signals Us from the upstream probe (curve C1) and from the downstream probe (curve C2), the operation of the catalyst is diagnosed via the creation of a rich phase Ph_r upstream of the catalyst to purge it then of a poor phase Ph_p upstream of the catalyst. The purge is maintained during the MP period. We then wait to observe the signal from the downstream probe switch to a lean regime.

We then measure the duration between the instant t1 of switching of the upstream probe and the instant t2 of switching of the downstream probe. The oxygen flow rate is integrated over the duration Pint between t1 and t2 to determine the oxygen storage capacity called “OSC” for “Oxygen Storage Capacity” in English of the catalyst. Catalyst OSC is the ability of the catalyst to store oxygen. The OSC is representative of the state of old age of the catalyst because the older the catalyst, the more the OSC drops over time. When the OSC falls below a threshold value, it is considered that the catalyst is faulty and therefore needs to be changed.

Oxygen sensors are based on the Zirconia principle and platinum electrodes. Following a rise in temperature, Zirconia has the ability to let oxygen ions through. The current associated with this passage of ions is representative of the presence of oxygen.

The characteristic operating curve of an oxygen sensor, called an "S" curve due to its shape, is shown on the . This curve indicates the voltage level returned by the probe according to a "Lambda" ratio. This lambda ratio is the ratio between an ideal fuel dosage and an actual fuel dosage. The Lambda ratio therefore represents the inverse of wealth.

The probe sees its voltage switch according to the richness of the medium. The voltage is close to 1V in rich mode and close to 0V in lean mode. The Preg range corresponds to an adjustment range between 0.97 and 1.04.

The problem is that this characteristic operating curve of the oxygen sensor is dependent on the temperature of the sensor, which can induce significant dispersions of the measurement signal and therefore an erroneous diagnosis of the operating state of the catalyst.

The invention aims to effectively remedy this drawback by proposing a method for improving the accuracy of a measurement signal from an oxygen sensor associated with a device for depolluting the exhaust gases of a gasoline combustion engine. , said method comprising:

- a step of storing a map of measurement signal transfer functions of the oxygen sensor as a function of a lambda coefficient representative of a richness of air/fuel mixture for different temperatures of the oxygen sensor,

- a step for determining the temperature of the oxygen sensor,

- a selection step, in the map of transfer functions, of the transfer function closest to the temperature of the previously determined oxygen sensor, and

- a step of using the selected transfer function to obtain a measurement signal from the oxygen sensor.

The invention thus makes it possible, by taking into account the temperature of the oxygen sensor, to improve the precision of the measurement signal and therefore the quality of the operating diagnosis carried out from the measurements returned by the oxygen sensors.

According to one implementation of the invention, the step of determining the temperature of the oxygen sensor comprises a sub-step of generating a current peak in the oxygen sensor.

According to one implementation of the invention, the step of determining the temperature of the oxygen sensor comprises a sub-step of measuring the value of the Nernst resistance of the oxygen sensor during said current peak.

According to an implementation of the invention, the Nernst resistance is calculated from the following formula:

- U(t0) being the voltage measured at time t0 corresponding to a time at which the voltage peak begins,

- U(t1) being a voltage measured at a time t1 corresponding to a time when the current peak has reached its maximum value, and

- lpulse being the intensity of a current peak.

According to one implementation of the invention, each step of determining the temperature of the oxygen sensor comprises a sub-step of determining the temperature from said value of the measured Nernst resistance and from a predetermined curve representing the Nernst resistance as a function of temperature.

According to one implementation of the invention, the pollution control device comprises a catalyst and a particulate filter.

The invention also relates to a computer comprising a memory storing software instructions for implementing the method for improving the precision of a measurement signal from an oxygen sensor as defined above.

The invention also relates to a motor vehicle comprising a gasoline combustion engine and a device for depolluting the exhaust gases of said gasoline combustion engine, said exhaust gases being adapted to pass through said pollution control device from upstream to downstream , said motor vehicle comprising an upstream oxygen sensor placed upstream of the pollution control device and a downstream oxygen sensor placed downstream of the pollution control device, said motor vehicle comprising a computer as defined above.

According to one embodiment of the invention, the pollution control device comprises a catalyst and a particulate filter.

According to one embodiment of the invention, the oxygen sensors comprise Zirconia.

The invention will be better understood on reading the following description and on examining the accompanying figures. These figures are given only by way of illustration but in no way limit the invention.

The , already described, shows the evolution as a function of time of voltage signals from an upstream oxygen sensor and from a downstream oxygen sensor observable during a diagnosis of the operation of the catalyst;

The , already described, is an operating characteristic curve of an oxygen sensor;

The is a schematic representation of an exhaust line of the combustion engine of a motor vehicle implementing the method according to the invention for improving a measurement signal from an oxygen sensor;

The is a cross-sectional view of an oxygen sensor from the and the electrical connection circuit of said probe;

The shows curves representing measurements made by oxygen probes of the ;

The shows a curve of a voltage measured by an oxygen probe of the ;

The shows a plot of oxygen sensor resistance as a function of temperature;

The shows a map of measurement signal transfer functions of the oxygen sensor as a function of a coefficient representative of a richness of air/fuel mixture for different temperatures of the oxygen sensor.

As illustrated in the , a motor vehicle 1 comprises a heat engine 10 and an exhaust line 20 for the exhaust gases emitted by said heat engine 10.

The heat engine 10 comprises a combustion chamber (not shown) in which fuel and oxidant are mixed. The fuel can, for example, be gasoline, and be injected into the combustion chamber by an injector (not shown), in particular by a direct injection system. The oxidant can, for example, be air.

The combustion of the mixture in the combustion chamber drives a piston (not shown) in translation, which makes it possible to transform the thermal energy of combustion into mechanical energy in order to drive the shaft of the motor 10 in rotation.

When burning the mixture, exhaust gases are formed. Such exhaust gases include in particular gases, such as, for example, nitrogen oxide, the content of which is regulated by standards.

When the quantity of air injected into the combustion chamber is greater than the quantity of air necessary for the combustion of the fuel to be complete, it is called a lean mixture. When the amount of air injected is less than the amount of air needed, it is called a rich mixture. In the latter case, all the fuel is not burned, we speak of incomplete combustion and the unburnt fuel can be evacuated through the exhaust line 20, which increases the quantity of polluting gas rejected by the engine 10.

After combustion, the exhaust gases are evacuated through the exhaust line 20 also making it possible to post-treat these exhaust gases at least partially.

To do this, the exhaust line 20 comprises, in this example, a pollution control device 21, an upstream oxygen sensor 22 placed upstream of the pollution control device 21 and a downstream oxygen sensor 23 placed downstream of the pollution control device 21.

The depollution device 21 comprises a catalyst and a particulate filter placed downstream of the catalyst. The catalyst, also called oxidation catalyst, makes it possible to oxidize the exhaust gases, in particular the carbon monoxide included in the exhaust gases in order to form carbon dioxide, which is less polluting. In the case of a gasoline engine, the catalyst is a three-way catalyst, also designated TWC for "Three-Way Catalyst" in English. The oxygen storage capacity, also called "OSC", representing the efficiency of the catalyst can vary according to the state of wear of the catalyst.

The particulate filter, also designated GPF for "Gasoline Particulate Filters" in English for a gasoline engine, makes it possible to retain polluting particles from the exhaust gases and burns them during a so-called regeneration phase. Such a particulate filter for gasoline engine 10 is said to be passive because the regeneration phases are carried out passively when the oxygen content of the exhaust gases exceeds a predetermined threshold. The particles retained in the particle filter are then burned by self-combustion.

The upstream oxygen sensor 22 and the downstream oxygen sensor 23 are oxygen sensors suitable for measuring the oxygen level in the exhaust gases upstream and downstream of the pollution control device 21.

As can be seen on the , an oxygen probe 22, 23, also referred to as a "planar probe" in English, comprises a heating element 221, a layer 222, preferably made of ceramic, for protecting the electrical element 221, a first electrode 223 located at the outside of the protective layer 222 in order to be in contact with the exhaust gases to be measured, and a second electrode 224 located inside the protective layer 222. Preferably, the first and the second electrode 223, 224 are made of platinum.

To measure the oxygen level in the exhaust gases, the oxygen sensor 22, 23 measures the voltage Us between the first and the second electrodes 223, 224. Such a voltage Us is representative of a Nernst cell of the oxygen sensor 22, 23.

The second electrode 224 is located in a cavity containing a reference air volume. Preferably the cavity communicates with the atmosphere so that the reference air is the air of the atmosphere. Thus, the measurement of the oxygen level in the exhaust gases is carried out by reference to the air present in the cavity. Such an oxygen sensor 22, 23 with air reference is also referred to as a Zirconia oxygen sensor. Zirconia is a material with the ability to let oxygen ions pass, the electric current associated with this passage of ions is proportional to the oxygen level in the exhaust gases. Thus, the measurement of this electric current allows the oxygen sensor 22, 23 to measure the oxygen level.

Furthermore, the references Vbatt, M, PWM, and ADC correspond respectively to the battery voltage, to the electrical ground, to a pulse width modulator signal (or "Pulse Width Modulation" in English), and to an analog-digital converter ("Analog-Digital Converter" in English). Such an oxygen sensor 22, 23 being known, it will not be described in more detail.

The oxygen sensor 22, 23 is electrically connected to a computer (not shown) of the motor vehicle 1 by an electrical circuit 30. The electrical circuit 30 makes it possible to supply the heating element 221 with electrical energy, in particular by connecting it to the vehicle battery 1.

The electrical circuit 30 also enables the computer to receive the value of the electrical voltage Us measured at the terminals of the electrodes 223, 224 in order to determine the oxygen level in the exhaust gases.

As illustrated by the , the voltage Us1 measured by the upstream oxygen sensor 22 varies according to the richness of the mixture injected into the heat engine 10 while the variation of the voltage Us2 measured by the downstream oxygen sensor 23 is lower due to the depollution of the exhaust gas through the pollution control device 21 between the upstream oxygen sensor 22 and the downstream oxygen sensor 23.

Thus, we speak of a rich mixture when the voltage Us is high, and of a lean mixture when the voltage Us is low.

The resistance R of the Nernst cell being proportional to the temperature, it is possible to use the oxygen probe 22, 23 in order to determine its temperature.

More specifically, to measure the Nernst resistance R, the computer actuates the Port transistor according to a control law in order to send a current peak P (or current pulse) to the level of the Nernst cell of the oxygen sensor 22 , 23.

The pulse P then causes a peak in the voltage Us across the terminals of the electrodes 223, 224, as illustrated in . We note t0 the instant at which the voltage peak begins and t1 the instant at which the peak reaches its maximum value. Preferably, the time interval between instant t0 and instant t1 is of the order of 3 ms.

The Nernst resistance R can then be calculated from the following formula:

Where U(t1) is the voltage measured at time t1, U(t0) is the voltage measured at time t0 and l _pulse is the intensity of the current peak P.

From the value of the Nernst resistance R thus calculated and from a curve representing the Nernst resistance R as a function of the temperature T, as illustrated in , the computer determines the temperature T of the oxygen sensor 22, 23. Such a graph is predetermined, in particular in the laboratory.

Thus, it is possible to determine the temperature of the oxygen sensor 22, 23 by generating a current pulse P in the oxygen sensor 22, 23.

Furthermore, the computer stores a map Cart of measurement signal transfer functions of the oxygen sensor 22, 23 as a function of a lambda coefficient (λ) representative of a richness of air/fuel mixture for different temperatures of the oxygen sensor 22, 23, as shown in figure .

The computer will be able to select, in the map Cart of transfer functions, the transfer function closest to the temperature of the oxygen sensor 22, 23 determined according to the method described above.

It is then possible to use the selected transfer function to obtain a measurement signal from the oxygen sensor 22, 23.

As can be seen on the , for a difference of 100° C. between the curve obtained at 750° C. and the curve obtained at 850° C., the measurement signal from the oxygen probe 22, 23 has a difference of 40 mV. It is therefore important to be able to correct the measurement signal of the oxygen sensor 22, 23 as a function of its temperature in order to be able to carry out robust diagnostics of its operation.

Claims (10)

Method for improving the accuracy of a measurement signal from an oxygen sensor (22, 23) associated with a device (21) for depolluting the exhaust gases of a gasoline engine (10), characterized in that it includes:
- a step of storing a map (Cart) of measurement signal transfer functions of the oxygen sensor (22, 23) as a function of a lambda coefficient (λ) representative of a richness of air/fuel mixture for different temperatures (T) of the oxygen sensor (22, 23),
- a step for determining the temperature (T) of the oxygen sensor (22, 23),
- a step of selecting, in the cartography (Cart) of transfer functions, the transfer function closest to the temperature (T) of the oxygen sensor (22, 23) previously determined, and
- a step of using the selected transfer function to obtain a measurement signal from the oxygen sensor (22, 23). Method according to claim 1, characterized in that the step of determining the temperature (T) of the oxygen sensor (22, 23) comprises a sub-step of generating a current peak in the oxygen sensor ( 22, 23). Method according to Claim 2, characterized in that the step of determining the temperature (T) of the oxygen sensor (22, 23) comprises a sub-step of measuring the value of the Nernst resistance (R) of the oxygen sensor (22, 23) during said current peak. Method according to Claim 3, characterized in that the Nernst resistance (R) is calculated from the following formula:

- U(t0) being the voltage measured at time t0 corresponding to a time at which the voltage peak begins,
- U(t1) being a voltage measured at a time t1 corresponding to a time when the current peak has reached its maximum value, and
- lpulse being the intensity of a current peak. Method according to Claim 3 or 4, characterized in that each step of determining the temperature (T) of the oxygen sensor (22, 23) comprises a sub-step of determining the temperature (T) from the said value the measured Nernst resistance (R) and a predetermined curve representing the Nernst resistance (R) as a function of temperature (T). Method according to any one of the preceding claims, characterized in that the depollution device (21) comprises a catalyst and a particle filter. Computer comprising a memory storing software instructions for implementing the method for improving the precision of a measurement signal from an oxygen sensor (22, 23) as defined according to any one of the preceding claims. Motor vehicle (1) comprising a petrol engine (10) and a device (21) for depolluting the exhaust gases of said petrol engine (10), said exhaust gases being adapted to pass through said depollution device ( 21) from upstream to downstream, said motor vehicle (1) comprising an upstream oxygen sensor (22) placed upstream of the pollution control device (21) and a downstream oxygen sensor (23) placed downstream of the pollution control device ( 21), characterized in that said motor vehicle (1) includes a computer according to claim 7. Motor vehicle according to Claim 8, characterized in that the depollution device (21) comprises a catalyst and a particle filter. Motor vehicle according to Claim 8 or 9, characterized in that the oxygen sensors (22, 23) comprise Zirconia.