FR3086697A1 - ESTIMATING THE EXHAUST GAS TEMPERATURE - Google Patents

Abstract

The present invention relates to a method for controlling a heat engine (10) comprising the following steps: - estimation of an exhaust gas temperature generated by the engine (10) during a period comprising a transient phase and a stabilized phase during which the temperature of the exhaust gases passes respectively from a variable state to a stabilized state, and - control of an actuator (5) of the engine (10) as a function of the estimated temperature. According to the invention, the temperature of the exhaust gases is estimated by calculation from a first temperature from a sensor (20) measuring the temperature of said exhaust gases and from a second temperature from a model. preset. The invention also relates to a heat engine (10) controlled according to the present method as well as to a motor vehicle comprising such a heat engine (10).

Description

ESTIMATING THE EXHAUST GAS TEMPERATURE

TECHNICAL AREA

The present invention relates, in general, to the field of heat engines and relates more specifically to the determination of the temperature of the exhaust gases from such engines.

In terms of controls applied to combustion engines, in particular anti-pollution controls, it becomes necessary to know the state of the systems linked to the exhaust of gases in order to be able to meet requirements which are becoming more and more severe.

To do this, the present invention provides a method of controlling a heat engine based on an estimate of the temperature of the exhaust gases, as well as a heat engine controlled by this method and a motor vehicle comprising such an engine.

PRIOR ART

The engines concerned are preferably supercharged or atmospheric engines, of the petrol or diesel type. In use, the engine exhaust system essentially passes through two phases, namely a first phase called transient and a second called stabilized. During the transient phase the temperature of the gases varies and goes through different higher or lower levels. At the end of the transitional phase these variations fade and finally stabilize at a relatively constant level in the stabilized phase.

Nowadays, checks carried out on engines are increasingly based on theoretical or pseudo-theoretical models. These models allow modeling of measurement values using physical equations for example. In order for these checks to be successful, it is important in practice to be able to adjust the temperature of the exhaust gases as precisely as possible at all times, i.e. both during the transitional phase and during the stabilized phase.

Being able to determine the temperature of the exhaust gases, different solutions can generally be used.

The first known solution consists in carrying out measurements using a probe or a temperature sensor. Such a sensor is typically placed at the exhaust manifold, namely upstream of the pollution control line and, if necessary, upstream of the compressor drive turbine (turbocharger) and / or of the EG R circuit ( Exhaust Gas Recirculation) relating to exhaust gas recirculation.

The first drawback lies in the fact that, due to this location, the temperature measured by the sensor may differ from that measured further downstream, typically in the exhaust pipe during anti-pollution checks.

The second drawback comes from the construction of these sensors. In fact, to avoid any risk of breakage which could cause other subsequent damage to other components, these sensors are of a massive nature and generally have a large diameter which makes them rather bulky. However, any large sensor has the disadvantage of having high thermal inertia. Typically, the response time of such a sensor can be 10 seconds.

On the other hand, it has been found that these temperature sensors provide interesting precision values in stabilized phase, namely once the engine has reached characteristics or performances which one could qualify as nominal.

The second known solution for determining the temperature of the exhaust gases consists of modeling the measurement values using a model typically based on physical equations and / or empirical values. Generally, the modeling of the exhaust gas temperature is estimated on the basis of parameters such as the quantity of fuel injected, the injection advance or the ignition advance. Typically, the temperature model is obtained from a map of temperature differences between intake and exhaust. This temperature difference is a function of the speed and the fuel flow. Coefficients linked to different sensitivities (injection advance or ignition advance, engine water temperature, richness, boost pressure, etc.) are often still applied to these values.

This second solution has the drawback of being imperfect, in particular due to the fact that the temperature of the exhaust gases is very complicated to model. Indeed, due to the EGR gas recirculation circuit and / or complex phenomena in the combustion chamber, the results include errors whose importance depends on the quality of the model used, which makes this solution unusable for order a motor.

On the other hand, the model-based solution has the advantage of being able to provide a very dynamic response. Indeed, this response can be perceived as immediate or instantaneous given that it is not struck with any delay or offset either at the time level or at the level of the value of the quantity considered.

Document WO2010130539 discloses the use of a temperature sensor within the exhaust duct accompanied by the use of model values. This document only uses, in a separate way, model values to confirm the temperature values detected by the temperature sensor.

As explained above, each of these solutions has at least one drawback which prevents the best estimate of the temperature of the exhaust gases. Consequently, there is an advantage in finding a new, more suitable solution which at least partially makes it possible to resolve the abovementioned drawbacks.

SUMMARY OF THE INVENTION

To this end, a first aspect of the present invention relates to a method of controlling a heat engine based on an estimate of the temperature of the exhaust gases generated by this engine. The estimation of this temperature can preferably be made at any time, at least during a period comprising a so-called transient phase and a so-called stabilized phase during which the temperature of the exhaust gases respectively passes from a variable state to a stabilized state. . According to the present invention, the temperature of the exhaust gases is estimated by calculation from a first temperature originating from a sensor measuring the temperature of said exhaust gases and from a second temperature originating from a preset model.

Advantageously, the solution provided by the present invention makes it possible to obtain a better estimate of the temperature of the exhaust gases within the engine, in particular at the exhaust manifold, both in the transient phase and in the stabilized phase. This results in an improvement in the quality of the regulation systems, in particular the system for regulating the boost pressure as well as the air / EGR gas recirculation circuit in the engine block.

Advantageously still the present invention makes it possible to reduce the harmfulness rate of the pollutants emitted at the source, namely at the level of the engine, thanks in particular to an adequate control which depends on the temperature.

The solution provided by this invention also has the advantage of being able to be based on a combination of two complementary means embodied by the temperature sensor and the preset model. The complementarity of these means lies in the fact that at least one defect in one of these means can be compensated for by at least one advantage brought by the other means. More specifically, the information from the pre-established model is mainly used during the transient phase while that from the sensor is mainly used during the stabilized phase.

Preferably, the calculation allowing the estimation of the temperature of the exhaust gases is based on a corrected model which is obtained by applying at least one corrective function to the pre-established model. This corrective function can typically include at least one logical or mathematical function.

According to one embodiment, the corrective function generates a so-called calibration constant which is preferably representative of the enthalpy of the exhaust gases. Typically, this enthalpy can be determined by the product of two values where the first value is characteristic of the mass flow rate of the exhaust gases and the second value is characteristic of the temperature of the exhaust gases determined from the pre-established model.

More preferably, the estimate of the temperature of the exhaust gases is obtained by subtracting from the temperature obtained from the predetermined model an error from this model. This error is calculated by the difference between the temperature of the corrected model and the temperature of the sensor.

According to another aspect, the present invention also relates to a heat engine controlled according to any one of the embodiments of the above method. More preferably, the internal combustion or internal combustion engine is a supercharged or atmospheric type engine.

Finally, according to a last aspect, the present invention also relates to a motor vehicle comprising a heat engine such as that briefly described above.

BRIEF DESCRIPTION OF THE FIGURES

Other characteristics and advantages of the present invention will appear more clearly on reading the detailed description which follows and which presents different embodiments of the invention given by way of non-limiting examples and illustrated by the appended figures in which :

- Figure 1 shows schematically the flow of gases within an engine and through the main elements associated with it;

- Figure 2 shows, in the form of a first block diagram, the basic idea on which the present invention is based;

- Figure 3 is a graphic illustration of temperature curves obtained by different means, including that proposed by the solution proposed by the present invention, compared to the actual physical temperature curve; and

- Figure 4 shows, in the form of a second block diagram, the process for estimating the temperature of the exhaust gases according to the invention.

DETAILED DESCRIPTION

With reference to FIG. 1, this illustrates the well-known circulation of gases, through a motor 10, more precisely of a motor system comprising an air intake at inlet 1 and an exhaust at outlet 2. As illustrated in this figure, the air admitted at the inlet successively passes through an air filter 11, a compressor 12 and a charge air cooler 13 before entering the engine block 15 through the intake manifold 14. The gases exit from the engine block 15 through the exhaust manifold 16. A portion of these gases can be reinjected into the intake manifold 14 by means of the exhaust gas recirculation system 17 followed by an exchanger 17a. This system reduces emissions of nitrogen oxides (NOx) and better meets emission standards without having to modify the engine's structure. The part of the exhaust gases which is not reinjected into this system is directed to a turbine 18 which makes it possible to drive the compressor by a mechanical connection. At the outlet of the turbine, the gases pass through a pollution control system 19 before leaving the exhaust circuit.

As shown in FIG. 1, the measurement of the temperature of the exhaust gases is generally carried out at the exhaust manifold 16 by a temperature sensor 20. This temperature measurement is typically intended to be used by an actuator 5, such as an injector or an ignition device, for controlling the engine 10.

Figure 2 shows, in the form of a block diagram, the basic idea on which the present invention is based. In order to be able to improve the estimation of a characteristic quantity of the exhaust gases, such as the temperature for example, the invention proposes to obtain this quantity on the basis of a combination of at least two pieces of information 11, I2. The first information 11 comes from a sensor, such as the temperature sensor 20 for example, and the second information I2 comes from a pre-established model 30. The combination of information I2, I2 coming from these two means 20, 30 is schematically illustrated by a dynamizer 40 from which a third piece of information I3 is obtained corresponding to the estimation of the quantity sought.

The pre-established model 30 makes it possible to obtain a modeling of the information I2 which it delivers on the basis of at least one input parameter P. These parameters may refer to different quantities and may relate, for example, to the flow in fuel, engine speed, different injection advance values, a temperature such as that of the plenum, post injection type flows / advances, pressure such as that of the plenum or the water temperature of engine cooling. This model is said to be pre-established because, on the basis of predefined algorithms, taking into account at least one parameter P as input data, it makes it possible to obtain information 12 as output. These algorithms can for example be obtained on a bench engine.

Preferably, at least one of the items of information 11, 12, 13 comprises, as a quantity, a temperature value. In the embodiments which will be presented in more detail below, all of this information 11, I2, I3 have the magnitudes of temperatures. However, some of the information, in particular information used as input values 11, I2, could reflect at least one other quantity such as for example a level of dioxygen (02) in the exhaust gases. Thus, in the description which follows, the sensor 20 will preferably be described as being a temperature sensor delivering a temperature denoted Te, the preset model 30 will preferably be configured to deliver a temperature denoted Tm and the dynamizer 40 will preferably be configured to deliver a temperature denoted Td, in particular the temperature of the exhaust gases estimated according to the invention.

Figure 3 gives a graphic illustration of different temperature curves obtained by different means. On the abscissa is time and on the ordinate the temperature. The time period presented in this graph includes a first portion or phase called transient PhT and a second portion or phase called stabilized Phs. Typically, one can consider being in a stabilized phase when the engine torque varies by less than 5%, in less than 1 second. Conversely, we can consider being in transient phase when the engine torque varies by more than 5%, in less than 1 seconds.

The first curve 21, illustrated in dashed lines at the end of this graph, is that given by the temperature sensor 20. The latter measures the temperature of the exhaust gases, preferably at the exhaust manifold 16, as illustrated in the figure 1.

The second curve 31, illustrated in phantom, is that given by the predetermined model 30. The latter models, by calculation, the temperature of the exhaust gases on the basis of at least one parameter P.

The third curve 41, illustrated in strong lines in FIG. 3, is that obtained by the dynamizer 40 according to the solution of the invention.

Finally, the last curve 51, illustrated in strong dashed lines, is representative of the actual physical temperature of the exhaust gases. It is therefore this temperature curve that we seek to obtain or to be able to imitate as best as possible thanks to the dynamizer 40.

From this graph, it should be noted that the temperature curve 21 supplied by the sensor 20 delivers delayed and filtered information with respect to the actual physical information illustrated by the curve 51. This offset or deviation of the temperature supplied by the sensor 20 essentially results from the thermal inertia of this sensor. Thus, during the transient phase PhT, it can be seen on this graph that the response, namely the information, given by the sensor 20 does not comply with the expectations illustrated by curve 51 in this same phase. On the other hand, during the stabilized phase Phs, it was found that the response given by the sensor 20 was precise information, at least more precise than that given by the pre-established model 30. In fact, like the curve 31 , it is noted that the preset model provides a good response during the transient phase PhT, while this response is less good than that given by the sensor 20 during the stabilized phase Phs. During the transient phase PhT, it can be seen in FIG. 3 that, although sometimes deviating on the ordinate from the actual physical temperature values, the preset model 30 has the advantage of having a dynamic which is close to that of the curve 51.

The response provided by the curve 41 resulting, according to the invention, from the dynamizer 40 results from the fusion of the information I2 generated by the preset model 30 during the transient phase PhT, and the information 11 generated by the sensor 20 during the stabilized phase Phs. Thus, during most of the transient phase, the temperature from the dynamizer 40 advantageously tends to be close to the value supplied by the preset model 30, and at the end of this transient phase it converges towards the value supplied by the sensor. 20 to finally essentially correspond to the temperature given by the sensor during the stabilized phase.

On the basis of the above, the first object of the invention relates to a method of controlling a heat engine 10. This method comprises a first step in which an estimate is made of the temperature of the exhaust gases generated by the engine 10. This estimate during a period comprising a transient phase and a stabilized phase during which the temperature of the exhaust gases respectively passes from a variable state to a stabilized state, as illustrated in FIG. 3. It should be understood that the estimation of this temperature is not carried out only once during this period, but that it is repeated many times, as illustrated by all the points which form the curve 41 in FIG. no measurements, namely the time interval between two consecutive measurements, can be of the order of 0.1 seconds for example. Each point of the curve 41 corresponds to the estimate of the temperature from the dynamizer 40. For this reason, this temperature is said to be dynamized and is denoted Td.

As schematically illustrated in FIG. 2, the temperature of the estimated Td of the exhaust gases is, according to the invention, obtained by calculation from a first temperature Te (11) coming from the temperature sensor 20 and from a second temperature Tm (I2) from the pre-established model 30.

According to a preferred embodiment, the sensor 20 and the preset model 30 constitute two complementary means. In fact and as explained with reference to FIG. 3, these means complement each other in the sense that the information originating from one of these means is essentially used during the transient phase, while the information originating from the other means is essentially operated during the stabilized phase. More specifically, according to the invention, the pre-established model 30 is essentially used during the transient PhT phase while the sensor 20 is essentially used during the stabilized Phs phase.

From Figure 3, we could see that the value provided by the preset model 30 was not correct at all times. This is mainly due to the fact that the modeling of the temperature of the exhaust gases carried out by the algorithms which are at the base of this model is imperfect. Thus, there is an error generated by this model. This error can be viewed as the difference between the value (Tm) given by this model and the actual physical value (Tph).

For this reason, the calculation allowing the estimation of the temperature of the exhaust gases will preferably be based on a corrected model, obtained by applying at least one corrective function Fc to the pre-established model 30. According to a preferred embodiment, the function corrective includes at least one logic function. Preferably, this corrective function consists of a plurality of logical or mathematical functions. These can be materialized by programming steps for software or by flip-flops within a logic or electronic circuit. Such a circuit can be obtained from a plurality of electronic components interconnected using a printed circuit for example.

In a preferred embodiment, the corrective function Fc generates a calibration constant CST. More preferably, this calibration constant is at least representative or a function of the exhaust enthalpy. The exhaust enthalpy is a characteristic based on the product of two values, the first value of which is representative of the mass flow of the gases and the second value of which is representative of the temperature from the preset model.

More preferably, the estimate of the temperature of the exhaust gases, namely the dynamic temperature Td, is obtained by subtracting from the temperature Tm from the pre-established model an error Err from this model calculated by the difference between the temperature Tmc provided by the corrected model and the temperature Te of the sensor. This embodiment is illustrated in Figure 4 and follows from the reasoning explained below.

As mentioned previously, the predefined model is marred by an error Err (or correction to be made) which can be perceived as being the difference between the temperature resulting from this model Tm and the real physical value Tph. From this observation, we obtain the following first expression:

(1) Err = Tm-Tph.

However, in practice the physical temperature Tph will be given by the sensor 20 because it is this means which physically takes the temperature of the exhaust gases. In a theoretical situation where the sensor 20 would be ideal, the temperature value Te supplied by this sensor 20 should at all times be equal to the physical (real) value Tph coming from the corresponding curve 51 (FIG. 3). However, to obtain the physical value Tph, it has been observed that it is also necessary to make a correction to the value Te supplied by the sensor, because of the delayed information that the latter delivers. This correction can be compared to the CST calibration constant.

Since the value Te, set for the value Tph in expression (1), must be corrected by the calibration constant CST, it follows that an equivalent correction must also be applied to the value Tm (from pre-established model) so that expression (1) remains correct. Applied to the value Tm, this correction (given by the calibration constant CST) makes it possible to obtain the corrected value of the model, noted Tmc:

(2) Err = Tmc-Tc.

The Err error of the model is nothing other than the correction that we will have to make to this model. Thus, the dynamic temperature Td sought is obtained by correcting the value Tm with a value equal to the error Err:

(3) Td = Tm - Err.

In other words, we obtain using the expression (2) in (3):

(4) Td = Tm - (Tmc - Te).

Thus according to this last expression (4), it appears that the dynamic temperature Td, supplied by the dynamizer 40, can be obtained by subtracting from the temperature Tm coming from the pre-established model, the error of this model which is calculated by the difference between the temperature Tmc from the corrected model and the temperature Te from the sensor.

Advantageously, it can be seen that this embodiment is based on a single expression (4) which is applicable at all times, namely both during the transient phase and during the stabilized phase. In other words, the solution proposed by the present invention, in this embodiment, does not require any switching of steps or algorithms in order to be able to pass from one phase to another. This embodiment therefore has the advantage of being applicable over the entire period that the measurements last, in particular the measurements of the temperature of the exhaust gases carried out by means of the sensor 20. In fact, in reverse resonance, we have the following expression on which the dynamic temperature Td is based:

(1 ’) Td = Tm - (Tmc - Te).

From this expression, the two phases are successively considered:

A) Transitional phase:

In the transient phase PhT and in the perfect theoretical case where the delay and filtering correction, provided by the calibration constant CST, is perfectly calibrated, namely determined or adjusted, we have:

(2 ’) Tmc = Te.

So, by placing the expression (2 ’) in (T) we get:

(3 ’) Td = Tm.

This last expression (3 ’) corresponds well to the goal initially sought, namely that in the transient phase PhT the energized temperature Td coming from the energizer 40 corresponds to that given by the pre-established model 30.

B) Stabilized phase:

In stabilized phase Phs we have:

(4 ’) Tmc = Tm.

So, by placing the expression (4 ’) in (T) we get:

(5 ’) Td = Tc.

This last expression (5 ’) also corresponds well to the goal initially sought, namely that in the stabilized phase Phs the energized temperature Td coming from the energizer 40 corresponds to that given by the sensor 20.

With reference to FIG. 4, this gives an illustration of the materialization of this embodiment, namely of the determination of the dynamic temperature Td on the basis of the expression (4) or (T) mentioned above. As mentioned previously, this materialization can be, for example, obtained by means of a logic circuit whose components such as flip-flops, memories or registers are schematically illustrated by the elements located inside the dynamiser 40. The latter is connected to on the one hand to the temperature sensor 20 and on the other hand to the preset model 30. The preset model provides at its output information comprising at least one temperature value Tm which it has modeled on the basis of a parameter Pi or d ' a plurality of parameters denoted Pi to Px supplied at the input of the model. The sensor 20 provides the temperature value Te that it has measured.

The calibration constant CST makes it possible to determine the correction to be made to the temperature value Tm, in order to obtain a corrected model temperature Tmc. This calibration constant is preferably determined on the basis of a so-called time constant Csît and a so-called filter constant Csîf. Typically, the calibration constant is obtained from the exhaust enthalpy which can be determined by an entity 42. As illustrated in this FIG. 4, this entity is part of the dynamizer 40. However, as a variant, this entity 42 could be external to the dynamizer 40. As illustrated in this figure, the exhaust enthalpy can be determined from a parameter coming from the model and / or from at least one additional parameter P + coming from another external entity.

The error Err of the temperature value Tm from the preset model 30, or in other words the correction to be made to the value from this model, is determined by the entity 43 on the basis of the values Tmc and Te. Finally, this error Err is applied to the value Tm by a last entity 44 so as to obtain at the output of the dynamizer 40 the value corresponding to the energized temperature Td.

Preferably, the entities 42, 43 and 44 are logical units materialized by electronic components, such as flip-flops, or implemented by means of programming steps of software configured to be implemented in a motor control 5.

The second object of the present invention relates to a heat engine 10 configured to be controlled by any of the embodiments of the method described above. To do this and as illustrated in FIG. 1, the heat engine 10 can be controlled by an actuator 5 configured according to this method. Such a configuration can be obtained, for example, by means of the dynamizer 40.

Finally, the last object of the present invention relates to a motor vehicle comprising a heat engine 10 according to the second object mentioned above.

Although the objects of the present invention have been described with reference to specific examples, various obvious modifications and / or improvements could be made to the embodiments described without departing from the spirit and scope of the invention. invention.

Claims (10)

1. Method for controlling a heat engine (10) comprising the following steps: - estimation of an exhaust gas temperature generated by said engine (10) during a period comprising a transient phase (Ρήτ) and a stabilized phase (Phs) during which the temperature of the exhaust gases passes respectively from a state variable to a stabilized state, - control of an actuator (5) of the engine (10) as a function of said estimated temperature (Td), characterized in that the temperature of the exhaust gases is estimated by calculation from a first temperature (Te) after a sensor (20) measuring the temperature of said exhaust gases and a second temperature (Tm) from a predetermined model (30). 2. Method according to claim 1, characterized in that the sensor (20) and the preset model (30) constitute two complementary means in which the preset model (30) is essentially used during the transient phase (Ρήτ) and the sensor ( 20) is mainly used during the stabilized phase (Phs). 3. Method according to one of the preceding claims, characterized in that said calculation is based on a corrected model which is obtained by applying at least one corrective function to the pre-established model. 4. Method according to claim 3, characterized in that said corrective function comprises at least one logical function. 5. Method according to claim 3 or 4, characterized in that the corrective function generates a calibration constant (CST) representative of the exhaust enthalpy. 6. Method according to claim 5, characterized in that the exhaust enthalpy is determined by the product of a first value, characteristic of the mass flow rate of the exhaust gases, by a second value characteristic of the temperature of the exhaust gas determined from the pre-established model (30). 7. Method according to one of claims 3 to 6, characterized in that the estimation of the temperature of the exhaust gases is obtained by subtracting from said second temperature (Tm), from the pre-established model, an error (Err) of the preset model (30) calculated by the difference between the temperature (Tmc) of the corrected model and the temperature (Te) of the sensor (20). 8. Method according to one of the preceding claims, characterized in that said heat engine (10) is a supercharged or atmospheric type engine. 9. Heat engine (10) controlled according to the method of any one of claims 1 to 8. 10. Motor vehicle comprising a heat engine (10) according to claim 9.