Method for adjusting richness in an internal combustion engine

Abstract

A method is disclosed for adjusting the richness of a mixture of fresh air and fuel injected into a combustion chamber of an internal combustion engine of the spark-ignition type which comprises an exhaust line equipped with a first catalyst, a second catalyst situated downstream of the first, as well as an upstream richness probe situated upstream of the first catalyst, an intermediate richness probe situated between the first and second catalysts, and a downstream richness probe situated downstream of the second catalyst, the adjustment method comprising: - a step for acquiring the richness of the burnt gases measured by the upstream probe, - a step for calculating a richness set point (CA), and - a step for determining a control parameter (Ti) for the internal combustion engine to regulate the richness of the burnt gases acquired according to said richness set point. The richness set point is modified according to the values of the exhaust gas flow, the richness of the burnt gases measured by the intermediate probe and the richness of the burnt gases measured by the downstream probe.

Description

DESCRIPTIONMethod for adjusting richness in an internal combustion engineTechnical field of the invention

[0001] The present invention generally relates to motor vehicles equipped with an internal combustion engine.

[0002] More specifically, it relates to an internal combustion engine of the spark ignition type which comprises an exhaust line of burnt gases equipped with a first catalyst, a second catalyst situated downstream of the first catalyst, an upstream richness probe (40) situated upstream of the first catalyst, an intermediate richness probe located between the first and second catalysts, and a downstream richness probe located downstream of the second catalyst.

[0003] It relates to a method for adjusting the richness of a mixture of fresh air and fuel injected into the combustion chamber of such an engine. This method comprises:- a step for acquiring the richness of the burnt gases measured by the upstream probe,- a step for calculating a richness set point, and- a step for determining a control parameter for the internal combustion engine to regulate the richness of the burnt gases acquired according to said richness set point.

[0004] More specifically, it relates to an internal combustion engine comprising a computer programmed to implement this method.State of the art

[0005] In an increasingly restrictive legislative framework and with a view to protecting the environment, technical solutions are being sought to improve the performance of internal combustion engines, in particular, to reduce the amount of pollutants released into the atmosphere.

[0006] To reduce polluting emissions, a spark-ignition engine generally comprises a three- way catalyst in the exhaust line in order to oxidise at least part of the unburnt hydrocarbons (HC) and carbon monoxide (CO) and reduce at least part of the nitrogen oxides (NOx) emitted in the combustion gases of the engine.

[0007] There are several known methods and devices for adjusting richness which aim to improve the efficiency of the catalyst.

[0008] By way of example, one known method is to use a control loop aimed at maintaining the richness of the burnt gases at 1 . This is done using a fuel richness probe (commonly known as a "lambda probe") mounted in the exhaust line, upstream of the catalyst. Regulation then consists of taking the probe’s output voltage and subtracting a set point voltage corresponding to a richness value of 1. The error signal is then compared to zero using a binary comparator. Therefore, when the set point voltage is higher than the probe’s output voltage, the mixture air-fuel is enriched by means of a regulator. Conversely, when the set point voltage is lower than the probe's output voltage, the mixture is poorer. The richness of the resulting mixture then oscillates around the stoichiometric value.

[0009] This way, the catalyst operates in its "catalytic window", so it can perform both the aforementioned oxidation and reduction reactions.

[0010] It can be noted that when the catalyst operates outside its catalytic window and is close to oxygen saturation, this favours the oxidisation reactions of carbon monoxide to thedetriment of nitrogen oxides reduction reactions. Conversely, when the catalyst is devoid of oxygen, this favours the reactions of reduction of nitrogen oxides but this situation is unfavourable to reactions of carbon monoxide oxidation.

[0011] The amount of oxygen stored in the catalyst is therefore a very important parameter for ensuring good simultaneous treatment of the aforementioned pollutants. This means that it is important to maintain a stable amount of oxygen in the catalyst to ensure optimal depollution.

[0012] Document FR-A1 -3033364 sets out a method for calculating the amount of oxygen stored in the catalyst and its oxygen storage capacity. This method is used to deduce a set point value for the oxygen storage in the catalyst. Then, using a second regulator, it is possible to modify the richness set point (previously equal to 1 ) to regulate the amount of oxygen for said set point value.

[0013] Document FR-A1 -3101 110 discloses an improvement to this previous method. Given that the maximum storage capacity of a catalyst is not constant but permanently depends on the flow rate and temperature of the burnt gases passing through the catalyst, this document specifies how to determine the oxygen quantity set point. As set out in this document, the improvement consists of determining this flow rate and temperature to deduce the maximum and minimum thresholds for the amount of oxygen that can be stored in the catalyst to guarantee treatment of the aforementioned pollutants, and select a set point value between these thresholds (generally closer to the minimum threshold since treatment efficiency with respect to nitrogen oxides NOx by a catalyst is more sensitive to a decrease in richness than the treatment of carbon monoxide is to an increase of richness).

[0014] To be more specific, when the engine works with a rich mixture, the higher the gas flow rate, the more carbon monoxide CO the burnt gases contain. Then, the amount of oxygen stored in the catalyst drops and may fall below the minimum threshold, increasing the risk of carbon monoxide CO leaks downstream of the catalyst. These leaks are detected by toggling the voltage of a binary richness probe mounted at the output of the catalyst above a maximum voltage threshold.

[0015] On the contrary, when the engine works with a poor mixture, the higher the gas flow rate, the more nitrogen oxides NOx are present in the burnt gases. The quantity of oxygen stored in the catalyst then increases and may exceed the maximum threshold, generating a risk of NOx leaks downstream of the catalyst. These leaks are detected by toggling the voltage of the binary probe mounted at the output of said catalyst below a minimum voltage threshold.

[0016] The oxygen quantity set point chosen therefore makes it possible to avoid such leaks.

[0017] Using the method described in document FR-A1 -31011 10, it is reasonable to expect that no major leakage of carbon monoxide or nitrogen oxides will occur, at least when the engine is operating in a stabilised phase.

[0018] New regulations require the use of an additional catalyst downstream of the first catalyst on which a richness regulation method is applied, namely, a method according to publication FR-A1 -3101110, to treat leaks that might occur during the engine's transient operating phases. Indeed, if a sudden change in the engine's operating point will cause the minimum threshold to suddenly increase above the set point value that was chosen previously, there may be a leak of pollutants downstream of the first catalyst, which can then be treated by the second catalyst.

[0019] Therefore, taking into account the presence of two catalysts, it can reasonably be expected that no significant leakage of carbon monoxide or nitrogen oxides will be detected downstream of the catalysts during transient operating phases, and a fortiori during stabilised operating phases.

[0020] Furthermore, document FR-A1 -31011 10 proposes regularly recalibrating the calculations of the current oxygen quantity in order to prevent errors from accumulating and causing a regulation fault.

[0021] However, the applicant has noted that, surprisingly, carbon monoxide leaks remain in a very specific situation, namely during stabilised operating phases wherein the vehicle advances rapidly (typically on a motorway at 130 km / h). In particular, it was able to verify that these leaks occurred even after recalibration of these calculations.Description of the invention

[0022] In this context, the invention proposes an adjustment method as defined in the introduction, in which the richness set point is modified as a function of the values assumed by the flow of exhaust gases, the richness of the burnt gases measured by the intermediate probe, i.e., the richness probe located between the first and the second catalysts, and the richness of the burnt gases measured by the downstream probe.

[0023] In practice, carbon monoxide leaks into the atmosphere are caused by the absence of oxygen in both the first catalyst (upstream) and the second catalyst (downstream), meaning that oxidisation reactions can no longer take place as intended.

[0024] This absence of oxygen is the result of the chosen target set point for the amount of oxygen to be stored in the first catalyst, which, for the reasons mentioned above, is closer to the minimum limit than the maximum limit. During the aforementioned long stabilised phases, the slightly off-centre adjustment to favour the treatment of nitrogen oxides NOx is combined with a high flow rate of burnt gases (i.e., a high speed of passage of the molecules to be treated). As a result, the efficiency of carbon monoxide CO treatment decreases. Therefore, if this operating point is maintained for too long, carbon monoxide CO leaks will occur at the outlet of the upstream catalyst and then at the outlet of the downstream catalyst.

[0025] This is why the invention proposes the use of an intermediate probe, located between the first and the second catalysts, and a downstream probe located downstream of the second catalyst to detect leaks both at the outlet of the first and the second catalyst, and to modify the richness setting of the mixture injected into the combustion chamber as soon as such leaks are detected in the mentioned situation at a high exhaust flow rate.

[0026] Of course, the signal emitted by the intermediate probe could also have other uses with respect to richness adjustment. Therefore, if this probe regularly detects carbon monoxide (or nitrogen oxides) leaks during transient operating phases, the fuel mixture during these phases can be adjusted accordingly.

[0027] Other advantageous and non-limiting features of the adjusting method in accordance with the invention, taken individually or in any technically possible combination, are as follows:- the method includes a step for calculating the amount of oxygen stored in the first catalyst, and a step for determining the oxygen storage set point value in the first catalyst, the richness set point being calculated by summing a basic set point corresponding to a stoichiometric mixture and a correction coefficient equal to a preliminary regulation coefficient by regulating said amount of oxygen stored according to said oxygen storage set point;- the method includes a step for acquiring the flow rate and temperature of the burnt gases passing through the first catalyst and then, a step for acquiring the minimum and maximum oxygen quantity thresholds according to said flow rate and temperature, and said oxygen storage set point value is chosen within a range between said minimum and maximum oxygen quantity thresholds, preferably closer to the minimum threshold than to the maximum threshold of oxygen quantity;- the method includes a step for acquiring the flow rate of the burnt gases passing through the first catalyst, wherein the richness set point will vary according to the richness of the burnt gases measured by the intermediate probe and the richness of the burnt gases measured by the downstream probe only if the flow rate exceeds a threshold;- the method includes a check to verify whether three conditions are simultaneously met. The first condition is that the richness of burnt gases is measured by the intermediate probe is rich. The second condition is that the richness of burnt gases is measured by the downstream probe is rich. The third condition is that the flow rate exceeds the said threshold. If all three conditions are simultaneously met, the correction coefficient will no longer be determined as equal to the preliminary regulation coefficient obtained by regulating said quantity of oxygen stored according to said oxygen storage set point, and the correction coefficient will be determined according to the last value of the correction coefficient before the three aforementioned conditions are simultaneously met;- if all three conditions are simultaneously met, the correction coefficient will be determined by subtracting a correction value from the aforementioned last value of the correction coefficient;- the correction value will vary according to the flow rate and the richness of the burnt gases measured by the intermediate probe;- as soon as the three conditions are no longer simultaneously met, the correction coefficient will once again be determined by regulating the quantity of stored oxygen according to the oxygen storage set point and the value of the quantity of stored oxygen will be corrected when regulation resumes;- the aforementioned control parameter is the duration of fuel injection into the combustion chamber at each operating cycle of the internal combustion engine.

[0028] The invention also relates to an internal combustion engine as defined in the introduction, comprising a computer programmed to implement the adjustment method mentioned above.

[0029] The features, variants and different embodiments of the invention may be associated with each other, in various combinations, as long as they are not incompatible or exclusive of each other.Detailed description of the invention

[0030] The following description, with reference to the attached drawings, given as nonlimiting examples, will clearly set out what the invention consists of and how it can be implemented.

[0031] Concerning the attached drawings:

[0032] [Fig- 1 ] is a schematic perspective view of an internal combustion engine according to the invention;

[0033] [Fig-2] is a diagram illustrating the steps to implement the richness adjustment method according to the invention;

[0034] [Fig.3] is a diagram illustrating the control loops used to implement the method illustrated in [Fig.2];

[0035] [Fig .4] illustrates the variations over time of four parameters measured on the internal combustion engine illustrated in [Fig.1 ].

[0036] In the description, the terms “upstream” and “downstream” will be used according to the normal direction of gas flow, from the point where fresh air is taken from the atmosphereto the outlet of the burnt gases into the atmosphere.

[0037] Fig. 1 shows schematically a motor vehicle internal combustion engine 1 provided with an accelerator pedal, comprising an engine block 10 enclosing a plurality of combustion chambers delimited by cylinders wherein pistons slide. Here, there are three cylinders 11 , but they could be fewer (e.g., two) or more (e.g., four, six or eight). The pistons are conventionally coupled to a crankshaft by connecting rods, allowing the crankshaft to be rotated.

[0038] The cylinders 1 1 are generally closed at the top by a cylinder head. This cylinder head has fuel inlets and exhaust gas outlets, which are controlled by valves. These valves are configured to open regularly to allow gases to pass through freely. They are actuated by a camshaft system.

[0039] Upstream of cylinders 11 , the internal combustion engine 1 comprises an intake line 20 which draws fresh air from the atmosphere, which then passes through an air distributor 25 designed to distribute the fresh air to each of the three cylinders 11 of engine block 10.

[0040] This intake line 20 comprises, in the direction of air flow, an air filter 21 which filters the fresh air taken from the atmosphere, a turbocharger compressor 22 which compresses the fresh air filtered by air filter 21 , a general intake valve 24 (also called a throttle body), which makes it possible to regulate the flow of fresh air flowing through air distributor 25, and a main air cooler 23 (typically an air / water exchanger) which cools the compressed air. Alternatively, the valve could be located elsewhere, e.g., downstream of the main air cooler.

[0041] It should be noted that the intake line 20 comprises a bypass duct at the intake of the compressor fitted with a valve 26 (called a "pop off" valve), which is normally in the closed position and opens in the event that the throttle body closes abruptly when the accelerator pedal is released rapidly, so as to prevent air from passing through compressor 22 in the opposite direction to the normal direction of air flow.

[0042] At the outlet of cylinders 1 1 , the internal combustion engine 1 includes an exhaust line 80, which extends from an exhaust manifold 81 which is fed by the gases previously burnt in cylinders 11 , to an exhaust silencer (not shown) making it possible to expand the burnt gases before they are discharged into the atmosphere. It also includes, in the direction of flow of the burnt gases, a turbocharger compressor turbine 82, mounted on a common shaft with compressor 22, and allowing compressor 22 to be driven, and the means 83 for depolluting the burnt gases. In Fig.1 , the arrow on turbine 82 indicates that this is a variable geometry turbine, wherein the amount of energy acquired by turbine 82 from the burnt gases and delivered to the compressor is adjusted according to the position of the turbine's 82 variable inclination blades. As a variant of this system, the turbine 82 could also be a fixed geometry turbine associated to a bypass duct in the exhaust of the turbine 82 fitted with a valve (called a "wastegate" valve).

[0043] Alternatively, the engine could be supercharged without a turbocharger. Similarly, the compressor 22 could be driven by something other than turbine 82, such as an electric motor. As another variant, while not being preferential, the intake line could be a natural suction intake without comprising a compressor 22.

[0044] In the illustrated example, the engine does not comprise a partial exhaust gas recirculation line at the intake (called an EGR circuit). As a variant, the system could include a circuit for the partial recirculation of exhaust gases at high pressure and / or a circuit for partial recirculation of exhaust gases at low pressure.

[0045] The internal combustion engine 1 also includes fuel injection circuit 60, which comprises an injection pump 62, which takes fuel from a tank 61 in order to bring it under pressure via a common rail 63 into injectors 64 which open, for example, directly into the cylinders 11 in the case of a direct injection engine. As a variant, the injectors could open intoan air intake line 20, or more specifically, into the distributor 25. It should be noted that the opening time of the injectors 64 will be used to adjust the flow of fuel injected into cylinders 11.

[0046] Within the scope of the invention, this engine 10 is spark-ignited, four-stroke engine, which means that it includes spark plugs configured to generate sparks in cylinders 11 to start combustion of the fresh air-fuel mixture in the cylinders at the desired time (between compression and expansion strokes).

[0047] Given the type of engine, the means 83 for depolluting the burnt gases comprise two "three-way" catalysts, making it possible to oxidise at least part of the unburnt hydrocarbons (HC) and carbon monoxide (CO) and reduce at least part of the nitrogen oxides (NOx) emitted in the burnt gases.

[0048] Each catalyst is designed to operate optimally in a "catalytic window", based on a given temperature threshold, called the initiation temperature threshold (around 450°C).

[0049] A first of these catalysts, called "upstream catalyst 84", is to be placed as close as possible to engine block 10 in order to quickly rise in temperature after a cold start of the engine in order to treat the burnt gases quicker. In practice, it is located in the engine compartment under the bonnet of the vehicle.

[0050] This upstream catalyst 84 comprises a metal housing containing at least one catalytic converter. Here, there are two.

[0051] It may be equipped with electric heating means to heat at least one catalytic converter so that it reaches the initiation temperature threshold quicker.

[0052] The second catalyst, "downstream catalyst 86", is located further along the exhaust line, such as under the body of the car outside the engine compartment. It also comprises a metal housing containing a single catalytic converter.

[0053] These two catalysts, 84 and 86, are different. In other words, their metal housings are disjointed. In practice, downstream catalyst 86 is only present to compensate for the shortcomings of the upstream catalyst in certain operating ranges of the engine.

[0054] The depollution means preferably includes a particle filter 85. These particle filter 85 is ideally located between the two catalysts 84 and 86. It may be located inside the metal housing of upstream catalyst 84, downstream, or be a completely separate component.

[0055] To control the various components of an internal combustion engine 1 , particularly the intake valve 24 and the injectors 64, a computer 100 is provided, comprising a processor (CPU), a random access memory (RAM), a read-only memory (ROM), analogue-to-digital converters (A / D) and various input and output interfaces.

[0056] Through the input interfaces, the computer 100 is able to receive input signals from different sensors relating to engine's operating parameters.

[0057] It is particularly configured to receive a signal related to the angle of the accelerator pedal of the motor vehicle on which engine 10 is mounted, or to the pressure exerted on this pedal by the foot 30 of the driver of the vehicle. It can also receive the rpm of the engine and its load.

[0058] It is also designed to receive signals relating to the richness RA, of the fresh air-fuel mixture blown into the combustion chamber. It is important to remember that richness relates to the ratio of the fuel flow rate to the air flow rate, divided by the ratio of fuel flow to air flow in stoichiometric proportions.

[0059] In the context of the invention, the exhaust line 80 is equipped with three richness probes. These include oxygen probes which make it possible to measure the residual concentration of oxygen in the gases.

[0060] A first richness probe, called upstream probe 40, is located upstream of upstream catalyst 84. This upstream probe 40 is a proportional probe, meaning that the signal it emits (typically a voltage) is proportional to the measured richness.

[0061] A second probe, intermediate probe 41 , is located between the upstream catalyst 84 and the downstream catalyst 86 (typically in between the upstream catalyst 84 and the particle filter 85). This probe may be a binary probe, meaning that the signal it emits (typically a voltage) mainly indicates whether the mixture is rich or poor or stoichiometric. Such a probe is less expensive than a proportional probe and, for the purposes of the invention, it is sufficient to implement the method described below. The method relies on measuring the oxygen content of the burnt gases. In practice, it emits a constant high voltage when the richness is greater than a threshold just slightly greater than 1 (e.g., greater than 1.02). Similarly, it emits a constant low voltage when the richness is lower than a threshold just slightly below 1 (e.g., below 0.98). Between these two extreme voltage values, the output voltage varies almost proportionally with richness.

[0062] A third probe, downstream probe 42, is located downstream of the downstream catalyst 86. This may also be an oxygen probe. However, new standards may require a nitrogen oxide concentration sensor to be fitted downstream of the last catalyst, i.e., the downstream catalyst 86. Such a sensor would make it possible to determine the richness of the gases, so it would be preferable to use this sensor rather than a dedicated probe. To simplify matters, this probe will hereinafter be called downstream probe 42 and will be considered to emit a voltage that behaves like the binary intermediate probe 41 .

[0063] Other probes (or sensors) are also used here. For example, the method includes a sensor for measuring the temperature of upstream catalyst 84 and a means for determining the flow rate of the burnt gases passing through the catalyst 84. These may be "physical" probes located in or near the upstream catalyst 84 or so-called "software" probes, i.e., algorithms based, for example, in observers and allowing to calculate the temperature and / or the flow rate according to various other engine parameters. For example, the flow rate can be conventionally determined from a pressure value and a temperature value in the distributor 25 and a filling model.

[0064] Using mapping that is preset on a test bench and stored in the ROM, the computer 100 is able to generate output signals for each of the engine's operating stages.

[0065] Finally, through the output interfaces, the computer 100 is able to send output signals to the various engine components, in particular, to the intake valve 24 and the injectors 64.

[0066] Thanks to its memory, the computer stores a computer application, consisting of computer programs comprising instructions whose execution by the processor enables the computer to implement the method described below. It also stores maps

[0067] In the context of the invention, the richness of the mixture injected into the cylinders 11 is related to the characteristics of the upstream catalyst 84. The upstream catalyst 84 can therefore be further characterised here.

[0068] It has a finite Oxygen Storage Capacity OSC.

[0069] This Storage Capacity OSC is known when the catalyst is new and is annotated as OSCnew. However, it is not constant and decreases as upstream catalyst 84 ages.

[0070] Fortunately, the Storage Capacity OSC can be calculated and updated regularly aftereach start of the engine, to take into account the ageing factor V of the catalyst. Therefore, we can write:

[0071] [Math.1]

[0072] i / j i — I . i / J l ei

[0073] To determine the Storage Capacity OSC, the engine's computer can take advantage of the first occurrence of a fairly long foot-lift phase 30 of the accelerator pedal by the driver, after the engine has been started. Indeed, the operation of the engine with zero richness, corresponding to a cut-off of fuel injection, saturates the upstream catalyst 84 with oxygen, provided that the foot-lift phase is long enough. Following this saturation phase, the computer strictly applies a richness level greater than 1 during resumption after injection cut-off, e.g., a richness equal to 1 .05, to allow the upstream catalyst to gradually empty of oxygen, until the intermediate probe 41 (binary) switches above a pre-calibrated voltage threshold, indicating that the richness of the burnt gases at the outlet of the upstream catalyst is high.

[0074] The OSC value, taking into account the burnt gas flow rate and temperature, is then obtained using the following integral equation:

[0075] [Math.2]

[0076]

[0077] In this equation, the variables are defined as follows.

[0078] Qech refers to the flow rate of the burnt gases (equal, for example, to the sum of the flow rate of fresh gas circulating in the intake and fuel line).

[0079] RA refers to the richness measured by the upstream probe 40.

[0080] T02 refers to the mass level of oxygen in the air (about 0.23 or 23%).

[0081] t-tinit corresponds to the time before the value emitted by the intermediate probe 41 switches above the threshold.

[0082] The upstream catalyst 84 operates in the desired way when the richness of the burnt gases remains exactly equal to 1 . When this condition is not met, efficiency drops.

[0083] Indeed, when the richness increases, the amount of oxygen in the burnt gases and therefore in the upstream catalyst 84 decreases, meaning that the catalyst is no longer able to properly oxidise carbon monoxide CO.

[0084] On the contrary, when the richness decreases, the amount of oxygen in the burnt gases and therefore in the upstream catalyst 84 increases, meaning that the catalyst is no longer able to properly reduce nitrogen oxides NOx.

[0085] In other words, for each pair of values for the burnt gas flow rate through the upstream catalyst 84 and the temperature of the catalyst, there exists a range of values for the quantity of oxygen OS between a minimum threshold OSmin and a maximum threshold OSmax, in which the conversion of the aforementioned pollutants is optimal. These thresholds can be determined during the design of the vehicle, for each pair of flow and temperature values, by observing the switching of the richness signal emitted by the intermediate probe 41.

[0086] The computer 100 is therefore programmed to regulate the quantity of oxygen OS stored in the upstream catalyst 84, in a closed loop, around a set point quantity of stored oxygen OSt, which is defined within said range by the following formula:

[0087] [Math.3]

[0088]

[0089] where K is a coefficient strictly between 0 and 1 and preferably between 0.25 and 0.75.

[0090] Preferably, coefficient K should be strictly less than 0.5 to favour nitrogen oxide reduction reactions instead of carbon monoxide oxidisation reactions. Indeed, the risk associated with carbon monoxide emissions is considered less than the risk associated with nitrogen oxide emissions, particularly because the amount of carbon monoxide emitted is less than that of nitrogen oxides when the richness deviates from 1 . For example, coefficient K could be equal to 0.3.

[0091] The computer memory stores a map whose inputs are pairs of flow rate and temperature values and whose outputs are pairs of minimum and maximum thresholds, OSmin_ _new, OSmax new of oxygen quantity for a new catalyst.

[0092] Now that the upstream catalyst 84 has been well defined, it is possible to describe the method for adjusting the richness of the mixture injected into the cylinders 1 1 .

[0093] However, it is first important to remember that, when the engine is started, the fresh air taken from the atmosphere by the intake line 20 is filtered by the air filter 21 , compressed by the compressor 22, cooled by the main air cooler 23 and then burnt in the cylinders 1 1 .

[0094] On leaving the cylinders 1 1 , the burnt gases are expanded in the turbine 82, treated by the depollution means 83 and then expanded again in the exhaust silencer before being released into the atmosphere.

[0095] The method implemented by the computer 100 then includes several main steps illustrated in Fig.2.

[0096] Just after starting the engine, the method includes a step E0, during which the computer 100 acquires the new oxygen storage capacity OSCnew of the new upstream catalyst 84, as well as the mapping of minimum and maximum oxygen quantity thresholds OSmin_new OSmax new- This data is stored in the computer’s memory.

[0097] It also determines the current value of the oxygen storage capacity OSC, in the manner described above, when the driver first lifts their foot on the accelerator pedal.

[0098] The current OSC value makes it possible to determine the ageing factor V.

[0099] This factor is therefore used to update the map, applying this ageing factor V to each of the values recorded on the map.

[0100] This step E0 is implemented only once after each engine start.

[0101] Then, the computer 100 is programmed to implement the following steps recursively, i.e., in a loop and at regular time intervals.

[0102] The first of these steps, E1 , consists of measuring the desired torque request and engine parameter values.

[0103] The desired torque set point is related to the torque that the driver wishes the engine to exert.

[0104] This request can, for example, be calculated by considering the engine speed and the angle p of the accelerator pedal 30 (received via an angular speed sensor and a position sensor connected to the computer 100).

[0105] It can also be calculated in a different manner, especially when the vehicle is driven (partially) autonomously.

[0106] The engine parameter values may include the engine speed, the burnt gas flow rate, the temperature of the upstream catalyst 84, etc.

[0107] During step E2, the computer calculates, according to the burnt gas flow rate value Qech and the temperature of upstream catalyst 84, the oxygen storage set point value Ost, using coefficient K=0.3, particularly if the vehicle is travelling at high speed, such as on a motorway, in other words, if the measured flow rate exceeds a predetermined threshold.

[0108] In the third step E3, the computer calculates the current value of the quantity of oxygen stored, OS, in upstream catalyst 84.

[0109] There are various possible calculation methods that could be used.

[0110] Typically, the one used will be the one described in document FR3033364, which is based on the following equation OS(RA).

[0111] [Math.1]

[0112]

[0113] In this equation, the variables are defined as follows.

[0114] OSinit refers to the amount of oxygen stored at time tOinit, of the beginning of the integration. This amount will be initially set to a predetermined value corresponding to, for example, the oxygen saturation of the catalyst. The beginning of the integration will then correspond to a moment wherein it is known that the catalyst is saturated with oxygen. This is typically the case when the fuel injection is cut off for a long enough time, especially when the driver completely lifts their foot off the accelerator pedal (“foot lift”). This value can be predetermined by prior testing.

[0115] OS refers to the quantity of oxygen stored at the current instant tO.

[0116] It is understood that such an integration calculation will be less accurate the longer the time between the instants tOinit and to, as the errors accumulate over time.

[0117] Preferably, means are provided for resetting the value of the quantity of stored oxygen, OS, based on measurements made by the intermediate probe 41 .

[0118] To be more specific, as long as the richness remains between a minimum threshold and a maximum threshold, the calculation of the amount of oxygen stored, OS, will be determined by the integral calculation method shown above.

[0119] However, if the voltage output of the intermediate probe 41 reaches its extreme low value, the stored oxygen amount OS is immediately reset to a value equal to the maximum oxygen threshold OSmax. Similarly, if the voltage output of intermediate probe 41 reaches its extreme high value, the stored oxygen amount OS is immediately reset to a value equal tothe minimum oxygen threshold OSmin

[0120] This reset makes it possible to compensate for rounding errors and differences in cumulative richness measurements that may distort the integral calculation.

[0121] It should be noted that this reset does not exclude the possibility of also resetting the amount of oxygen stored OS to the value of the oxygen stored capacity OSC after a long enough foot-lift phase, or to the zero value after a long enough pedal depression phase.

[0122] During a step E4, the computer determines a richness set point CA.

[0123] This richness set point CA, is equal to the sum of a basic set point C O and a correction coefficient aA

[0124] The basic set point typically corresponds to a richness equal to 1 (stoichiometric mixture).

[0125] For a standard approach, the correction coefficient aAis used to take into account the amount of oxygen stored OS in the upstream catalyst 84 to keep it strictly equal to the oxygen storage set point OSt.

[0126] As part of an exceptional approach that will be described later in this presentation, the correction coefficient cu, will be calculated differently to restore an oxygen defect in the two catalysts 84 and 86.

[0127] Firstly, let’s look at the standard approach.

[0128] With this approach, the computer 100 uses a servo loop, as illustrated in Fig.3, to determine the correction coefficient c

[0129] This loop includes a regulator R1 (e.g., a proportional-integral controller), which calculates a preliminary correction coefficient aAo according to the difference E2 between the quantity of oxygen stored, OS, in the upstream catalyst 84 and the oxygen storage set point OSt.

[0130] Alternatively, the regulator R1 may have a transfer function such as the one described in publication FR-A1 -3033364. More specifically, outside a deviation range E2 comprising the value 0, the richness set point correction aAis saturated at a constant value that is negative (when the deviation is below this range) or positive (when the deviation is above this range). Otherwise, within the deviation range, the preliminary correction coefficient Q O is a continuous, increasing function sharpened by parts of the deviation E2.

[0131] As part of the standard approach, the correction coefficient aA, is considered equal to the preliminary correction coefficient aAo.

[0132] Therefore, the computer calculates the richness set point C , by incrementing the basic set point C O by the value of the correction coefficient aA.

[0133] During a step E5, the computer 100, according to the data acquired and calculated, determines the control instructions for the components of engine 1 . These set points are calculated in order to adjust the fresh air flow rate and fuel flow rate to desired values.

[0134] In practice, the fresh air flow is regulated according to the position of the accelerator pedal.

[0135] The fuel flow rate is regulated such that the richness R of the mixture that enters the cylinders 11 (oxygen and fuel) remains generally equal to the richness set point C

[0136] To do this, the computer 100 uses a second feedback loop, as illustrated in Fig.3.

[0137] This second feedback loop receives an input related to the richness RA, calculated by the upstream probe 40 and the richness set point CA. It then outputs a set point value for the flow rate of fuel to be injected, in the form of a set point T, for the duration of opening of the injectors 64 during each cycle.

[0138] Using a regulator R0 (.e.g, a proportional-integral controller), the difference £ between the measured richness RA and the richness set point CA makes it possible to determine a correction value Tcof the duration of opening of injectors 64.

[0139] The opening time set point T sent to injectors 64 is then equal to the sum of correction value Tcand a predetermined value t;(which corresponds to the stoichiometric richness, taking into account the operating point of the engine, i.e., its speed and load).

[0140] In summary, the richness of the mixture injected into cylinders 1 1 is adjusted by regulating the amount of oxygen in upstream catalyst 84, which relies on controlling the richness measured upstream of this catalyst.

[0141] Therefore, as a general rule, the amount of oxygen stored, OS, stored in the upstream catalyst 84 will be stable and, in practice, small enough to prevent nitrogen oxide leaks (which occur faster than carbon monoxide when the amount of oxygen stored is outside the calculated OSmin-OSmax range). As a general rule, it can be observed that the amount of oxygen stored in the downstream catalyst is quite high (since only a small level of nitrogen oxide reduction reactions occur at this point).

[0142] This standard approach for regulating the intake mixture richness is, in practice, implemented in a loop as long as three cumulative conditions are not simultaneously met. Otherwise, the exceptional approach, different from the standard one, is implemented instead.

[0143] The three conditions correspond to a situation wherein the engine is running at high flow for a significant amount of time (typically when the vehicle is on the motorway). They are as follows.

[0144] The first condition is that the richness of the burnt gases measured by the intermediate probe 41 is high, which means that these gases contain an abnormal amount of carbon monoxide CO downstream of the upstream catalyst 84. This first condition is satisfied as soon as the richness exceeds an activation threshold and remains above a deactivation threshold which is lower than the activation threshold.

[0145] The second condition is that the richness of the burnt gases measured by the downstream probe 42 is high, which means that these gases contain an abnormal amount of carbon monoxide CO downstream of the downstream catalyst 86. This second condition is satisfied as soon as the richness exceeds an activation threshold and remains above a deactivation threshold which is lower than the activation threshold.

[0146] The third condition is that the flow rate QeCh of the burnt gases is high. This third condition is satisfied as soon as the flow rate exceeds an activation threshold SaeCh and then remains above a deactivation threshold SdeCh which is lower than the activation threshold.

[0147] The exceptional approach is only implemented if all three of these conditions are met simultaneously. This approach consists of interrupting the regulation of the amount of oxygen stored, OS, according to said oxygen storage set point Ost. In other words, the correction coefficient aA, is no longer considered equal to the preliminary correction coefficient aAo (Fig .3) .

[0148] Instead, with the exceptional approach, the correction coefficient aA, is determined according to the last value of the preliminary correction coefficient taken before the aforementioned three conditions are simultaneously met. The latter value is annotated as c o.

[0149] More precisely, the correction coefficient aA, is determined by subtracting a correctionCo from this latter value aA0.

[0150] The correction Co is preferably a variable value according to the burnt gas flow rate Qech and the richness of the burnt gases measured by the intermediate probe 41 . The idea is that it is greater if the flow is greater and it is greater if the richness is greater. On the other hand, it will be smaller as the richness decreases, so as to avoid causing the opposite situation, i.e., nitrogen oxide leaks.

[0151] The value of this correction may vary continuously or incrementally.

[0152] By not maintaining the quantity of oxygen stored, OS, at OSt target, it is then possible to deplete the intake mixture so that oxygen once again begins to be stored in the upstream catalyst 84, so that the treatment of burnt gases, especially carbon monoxide CO, will take place as normal once more.

[0153] It should be noted that, when the three conditions are no longer simultaneously met, the correction coefficient aA, will be determined once again in the standard way, by regulating the quantity of oxygen stored, OS, in accordance with said oxygen storage set point Ost. However, once the standard approach has resumed, a special transition may be applied to prevent the computer from poorly regulating the amount of oxygen.

[0154] Indeed, as long as the three conditions are met, the engine is controlled so that the upstream catalyst is filled with oxygen again. The model for calculating the amount of oxygen stored, OS, which uses an integration calculation, therefore risks the consideration that the amount of oxygen stored, OS, is very high and adjusting the intake mixture with too high a richness, which would be counter-productive.

[0155] Therefore, the transition when resuming the standard approach consists of recalibrating the value of the quantity of oxygen stored OS by assigning it the value of the oxygen quantity set point OSt.

[0156] In FIG. 4, the variations over time t of four parameters are represented by four curves to clearly illustrate the invention.

[0157] From top to bottom, these parameters are the flow rate Qech of the burnt gases passing through upstream catalyst 84, the output voltage U41 of the intermediate probe 41 , the output voltage U42 of the downstream probe 42 and the richness set point CA

[0158] Before time ti, the flow rate Qech remains moderate. The two voltages also remain below activation thresholds Sa4i, Sa42 (meaning that the richnesses measured by the probes remain below the corresponding activation thresholds). Accordingly, the three conditions are not met and the method is carried out according to the standard approach, i.e., the correction coefficient cu, is equal to the preliminary correction coefficient aA0.

[0159] At time ti, the flow rate Qech increases but the voltages do not exceed activation thresholds Sa4i, Sa42. It should be noted, however, that this increase in flow rate first causes voltage U41 and then voltage U42 to increase.

[0160] At time t2, all three conditions are met. Therefore, the standard approach is interrupted and the correction coefficient cu, is determined by subtracting the correction Co from the last value OAO-

[0161] It should be noted that the richness set point CA, then decreases abruptly, allowing the oxygen content of the burnt gases to increase in order to raise the amount of oxygen stored in upstream catalyst 84.

[0162] At time ts, at least one of the three conditions is no longer met, in this example, because voltage U41 falls below deactivation threshold Sd4i. Alternatively, it could be thevoltage U42 that first passes below the deactivation threshold Sd42 (although this situation will be less common). Another alternative is that the flow rate Qech falls below deactivation threshold SdeCh.

[0163] In any case, at this moment, the standard regulation resumes by recalibrating the value of the quantity of oxygen stored OS.

[0164] At this stage, it should be noted that the deactivation thresholds must each be below the corresponding activation threshold to avoid oscillation between the standard approach and the exceptional approach.

[0165] The present invention is in no way limited to the embodiment described and shown and a suitably skilled person will know how to make any variant thereof in accordance with the invention.

[0166] Typically, in the embodiment described and shown, the exceptional approach starts when all three conditions are met. In practice, the exceptional approach could be triggered by just two conditions, a high burnt gas flow rate and a high output voltage from the downstream probe 42. Indeed, it will be very unusual for the output voltage of the downstream probe 42 to exceed the activation threshold without the output voltage of the intermediate probe 41 also exceeding it.

Claims

CLAIMS1 Method for adjusting the richness of a mixture of fresh air and fuel injected into the combustion chamber of an internal combustion engine (1 ) of the spark ignition type which comprises an exhaust line (80) of burnt gases equipped with a first catalyst (84), a second catalyst (86) situated downstream of the first catalyst (84), an upstream richness probe (40) situated upstream of the first catalyst (84), an intermediate richness probe (41 ) located between the first and second catalysts (84, 86), and a downstream richness probe (42) located downstream of the second catalyst (86), the adjustment method being implemented by a computer (100) and comprising :- a step for acquiring the richness (RA) of the burnt gases measured by the upstream probe (40),- a step for calculating a richness set point (CA), and- a step for determining a control parameter (Ti) for the internal combustion engine (1 ) to regulate the richness of the burnt gases acquired according to said richness set point (CA), characterised in that the richness set point (CA) varies according to the the burnt gas flow rate (Qech), to the richness of the burnt gases measured by the intermediate probe (41 ) and to the richness of the burnt gases measured by the downstream probe (42).2.- Method for adjusting according to claim 1 , in which there is provided:- a step of calculating a quantity of oxygen stored (OS) in the first catalyst (84),- a step for determining an oxygen storage set point (OSt) in the first catalyst (84), and wherein the richness set point (CA) is calculated by summing a base set point (CAO) corresponding to a stoichiometric mixture and a correction coefficient (aA) determined as equal to a preliminary regulation coefficient (aAO) by regulating said stored oxygen quantity (OS) according to said oxygen storage set point (OSt).3.- Method for adjusting according to claim 2, in which there is provided:- a step for acquiring a flow rate (Qech) of the burnt gases passing through the first catalyst (84),- a step for acquiring a temperature of the first catalyst (84),- a step for determining a minimum oxygen quantity threshold (OSmin) and a maximum oxygen quantity threshold (OSmax) as a function of said flow rate (Qech) and said temperature, and wherein said oxygen stock set point (OSt) is chosen within a range between said minimum oxygen quantity threshold (OSmin) and said maximum oxygen quantity threshold (OSmax), preferably closer to the minimum oxygen quantity threshold (OSmin) than to the maximum oxygen quantity threshold (OSmax).4.- Method for adjusting according to one of claims 1 to 3, in which a step is provided for acquiring a flow rate (Qech) of the burnt gases passing through the first catalyst (84), and in which the richness set point (CA) varies as a function of the richness of the burnt gases measured by the intermediate probe (41 ) and of the richness of the burnt gases measured by the downstream probe (42) only if the flow rate (Qech) exceeds a threshold (Saech).5.- Method for adjusting according to claims 3 and 4, in which provision is made for:- checking whether three conditions are simultaneously met, a first condition being that the richness of the burnt gases measured by the intermediate probe (41 ) is rich, a second condition being that the richness of the burnt gases measured by the downstream probe (42) is rich, and a third condition being that the flow rate (Qech) exceeds the said threshold (Saech), and- if the three conditions are simultaneously fulfilled, interrupting the determination of the correctioncoefficient (aA) as equal to the preliminary regulation coefficient (aAO) obtained by regulating said stored oxygen quantity (OS) according to said oxygen storage set point (OSt), and determining the correction coefficient (aA) as a function of the last value of the correction coefficient (aA) present just before the three aforementioned conditions are simultaneously fulfilled.6.- Method for adjusting according to claim 5, wherein if the three conditions are simultaneously met, the correction coefficient (aA) is determined by subtracting a correction factor (cO) from said last value of the correction coefficient (aA).7.- Method for adjusting according to claim 6, wherein the correction factor (cO) varies as a function of the said flow rate (Qech) and of the richness of the burnt gases measured by the intermediate probe (41 ).8.- Method for adjusting according to one of claims 5 to 7, wherein, as soon as the three conditions are no longer simultaneously fulfilled, the determination of the correction coefficient (aA) is resumed by regulating the said quantity of oxygen stored (OS) according to the said oxygen storage set point (OSt), the value of the quantity of oxygen stored (OS) being corrected when the regulation is resumed.9.- Method for adjusting according to one of claims 1 to 8, wherein said control parameter (Ti) is a duration of injection of fuel into the combustion chamber at each cycle of operation of the internal combustion engine (1 ).10.- Internal combustion engine (1 ) of the spark ignition type comprising :- a combustion chamber,- an exhaust line (80) for burnt gases from the combustion chamber which is equipped with a first catalyst (84), a second catalyst (86) located downstream of the first catalyst (84), as well as an upstream richness probe (40) located upstream of the first catalyst (84), an intermediate richness probe (41 ) located between the first and second catalysts (84, 86), and a downstream richness probe (42) located downstream of the second catalyst (84, 86), and- a computer programmed to implement a method for adjusting according to one of claims 1 to 9.

Images