BACKGROUND ART In connection with vehicles powered by a combustion engine, there is a gen- eral desire to reduce the vehicle fuel consumption as far as possible. This in turn is based upon environmental demands aiming at reducing the amount of detrimental discharges to the atmosphere, and upon demands regarding good fuel economy of the vehicle.

In today's motor vehicles, the supply of air and fuel to the engine is normally controlled by means of a computer-based engine control unit. This control unit is in a known manner arranged for detecting signals representing a number of different operating variables of the vehicle, e. g. engine speed, load, engine coolant temperature, vehicle speed, etc. From these signals, the amount of fuel to be supplied to the engine is continuously determined, and the supply is then made by means of an injection device.

With the intention of limiting the fuel consumption of a vehicle, the control unit may be arranged, in a known way, so as to ensure that, during operation, a stoicheiometric air/fuel mixture (i. e. a mixture where X=l) is fed to the engine.

This guideline value can however not be achieved for all points of operation, due to limitations regarding the maximum allowed thermal load on the components comprised in the engine and exhaust system. For example, the temperature of the engine cylinder head and exhaust system, and in any

existing turbocharger unit, must be held within certain predetermined maximum limits. Should these limits be exceeded, there would be a risk of damaging the components.

The risk for high thermal load on the engine system and its components is particularly marked at high loads and engine rotational speeds. For such operating cases the engine exhaust gas temperature must be limited, so as not to become so high that there will be a risk for damage to the engine and its associated components, as discussed above.

According to the known art, this cooling effect is obtained by supplying a certain excess amount of fuel to the engine during the above-mentioned operating conditions, like for example when the vehicle driver applies full throttle during overtaking. This will thus entail that the fuel mixture will be controlled so as to deviate from the stoicheiometric mixture. More precisely, this increase in fuel supply is controlled to reach a certain level, corresponding to the exhaust gas temperature remaining lower than a predetermined limit value. The magnitude of this limit value may be based on empirical criteria, which in turn would be determined by engine tests, and would be indicating a limit above which there is a risk of damage to certain sensitive components in the engine and exhaust system.

A major drawback with this known procedure relates to the fact that it is not always necessary to supply the excess fuel as quickly as the change in engine load, as the engine and exhaust system temperatures in any case do not increase as quickly as the load changes. This may in turn be attributed to thermal inertia in the various parts of the engine system. This often entails supplying an excess fuel amount to the engine at high loads and engine speeds, which is a drawback as it increases the vehicle fuel consumption.

Within the relevant technical area, a system for controlling the fuel supply to a combustion engine of a vehicle is previously known from the patent document US 5103791. This system comprises means for detection of the engine load and the engine coolant temperature. Based on these values of load and temperature, a value of the temperature in the engine exhaust system is estimated. This temperature value is the basis for a correction of

the amount of fuel fed to the engine. In this way, the exhaust system temperature can be limited, reducing the risk of damage.

Another system for controlling the fuel supply to a combustion engine is described in the patent document US 5158063. This system comprises means for estimating the temperature of at least one component in the engine system as a function of the current engine operating conditions. The air/fuel mixture supplied to the engine may then be controlled as a function of this estimated component temperature.

A common feature of the two previously known systems is that they include relatively simple models for the engine system temperature, in particular providing a control that does not account for the thermal inertia of the respective temperature-sensitive component, e. g. during a sudden increase of the load.

Consequently, there is a need for being able to provide temperature values that can be used in a better fashion when cooling the engine system.

DISCLOSURE OF INVENTION The object of the present invention is to provide an improved method for determination of temperature values that may be utilised for said control. This object is achieved by a method, the characteristics of which are stated in the appended claim 1. The object is also achieved by means of a device, the characteristics of which are stated in the appended claim 10.

The method according to the invention is intended for use with a control of a combustion engine in a vehicle, and includes detecting data regarding predetermined variables of the engine and vehicle operating conditions, deriving temperature values of the material in at least one component, arranged in connection with or inside the engine, as a function of said variables, whereby control of the thermal load of the engine can be performed dependent upon at least said temperature values. The invention is characterised in that said temperature values are derived dependent upon the thermal inertia inherent in said component when changing the rotational speed and/or the load of said engine.

The temperature values derived in accordance with the invention may be utilised for control of the engine so as to cool it in an optimum way during e. g. sudden increases in load and speed. This in turn will secure that certain predetermined critical material temperature values are never exceeded. This cooling, i. e. limitation of the thermal load on the engine system, may for example be achieved by utilising the derived temperature values for control of the air/fuel mixture supplied to the engine, whereby an additional fuel amount is supplied as a function of the derived temperature values. In this manner particularly the enrichment of the air/fuel mixture can be delayed until its cooling effect is really needed. This leads to a lower fuel consumption of the engine compared to the known art.

The derivation according to the invention is active within a certain"critical area"of engine operation, which is characterised by high loads and high speeds. Within this"critical area"there is a risk that some engine component might experience a temperature exceeding a critical value, thereby risking damage to said component. This"critical area"is defined in this description as that area where the engine is normally controlled with an air/fuel mixture deviating from the stoicheiometric relationship.

The temperature values derived according to the invention allow the combus- tion engine to be controlled so as to limit the thermal load on the engine system. This can be achieved by using the derived temperature values for control of the air/fuel mixture supplied to the engine, whereby an additional fuel amount is supplied as a function of the derived temperature values. In this manner particularly the enrichment of the air/fuel mixture can be delayed until its cooling effect is really needed. As an alternative, the thermal load on the engine system may be limited by injecting water or a corresponding coolant directly into one or more of the engine cylinders. This will provide environmental and fuel economical advantages. Furthermore, the thermal load on the engine system may be limited by control of a thermostat belonging to the engine cooling system. According to a further alternative, particularly advantageous for engines provided with a turbocharger unit, the thermal load may be limited by controlling the charge pressure of the

turbocharger. This may in turn be accomplished by regulating a wastegate valve in the turbocharger unit.

The invention provides an improved engine control compared to the known systems, allowing the engine fuel consumption to be reduced, particularly for operating circumstances with high load and rotational speed. Notwithstanding this, the invention secures that no temperature-critical engine component will reach a temperature exceeding a critical limit value, at which damage might occur.

Preferably, the invention is implemented as a complementing software func- tion in an as such known engine control unit. Existing vehicle components are in this way to a high degree used in combination with auxiliary software functions, without having to introduce any additional hardware components.

Advantageous embodiments of the invention are described in the appended dependent claims.

BRIEF DESCRIPTION OF DRAWINGS The invention will be explained in greater detail below with reference to a preferred embodiment example and the enclose drawings, of which: Fig. 1 shows, in principle, an arrangement in connection with a combustion engine where the present invention may be applied, Fig. 2 is a flow chart showing the function of the control according to the inven- tion, Fig. 3 is a diagram that also illustrates the function of the invention and its effect on the fuel consumption of a combustion engine, Fig. 4 shows, in principle, an arrangement in connection with a combustion engine, according to a second embodiment of the invention, Fig. 5 is a flow chart showing the function of the control according to said second embodiment, Fig. 6 shows, in principle, an arrangement in connection with a combustion engine, according to a third embodiment of the invention, and Fig. 7 is a flow chart showing the function of the control according to said third embodiment.

MODES FOR CARRYING OUT THE INVENTION Fig. 1 shows, in principle, an arrangement in connection with a combustion engine where the present invention may be applied. According to a preferred embodiment, this arrangement is provided in a vehicle, in connection with the vehicle engine 1, which preferably consists of a conventional combustion engine. The engine 1 is fed in the normal manner with inflowing air through an air duct 2. The engine 1 is further provided with a cylinder head 3 and an engine block having a number of cylinders and a corresponding number of fuel injection devices 4, each connected to a central control unit 5. The control unit 5, which is preferably computer based, is functioning in a known manner to control each injection device 4, respectively, so as to supply, at each moment, an appropriate air/fuel mixture to the engine 1.

During operation of the engine 1, the control unit 5 is functioning to control the air/fuel mixture to the engine 1 in such a manner that, at each moment, the fuel mixture will be adapted to the current operating conditions. The amount of air to be supplied to the engine 1 is controlled by a throttle 6, and the supply of fuel is made as a function of several parameters representing the current operating conditions of the engine 1 and the corresponding vehicle. For example, the engine control may be dependent upon the current throttle setting, the engine speed, the amount of air injected into the engine, and the oxygen concentration of the exhaust gases. The throttle 6 may be electrically controlled through a connection to the control unit 5, as indicated by a dashed line in the figure. In this case the throttle 6 is operated by an actuator motor (not shown), the position of which can be controlled by the control unit 5.

The engine 1 according to the embodiment is provided with the"multi-point" injection type, allowing the correct amount of fuel to the engine 1 to be supplied individually by means of the respective injection devices 4. The invention may, in principle, also be utilised for so called"single-point" injection, where a single fuel injection device is located in the engine inlet manifold.

The engine 1 illustrated in the figure has four cylinders. It should however be understood that the invention could be used for engines having different numbers of cylinders and cylinder configurations.

The exhaust gases from the engine 1 are discharged through an exhaust outlet in the form of a manifold 7. Moreover, the engine 1 illustrated is of the type equipped with a turbocharger unit 8. The invention is however not limited to this type of engine, but can also be used for engines without a turbocharger unit. According to the embodiment, the exhaust gases are transported through the exhaust manifold 7 and on through an exhaust pipe 9 connected to the manifold and a turbine 10 belonging to the turbocharger unit 8. From the turbine 10, the exhaust gases are transported on through an additional exhaust pipe 11 to an exhaust gas catalytic converter 12, and then on out to the atmosphere.

In a known manner, the turbocharger unit 8 comprises a compressor 13, rotatably arranged on a shaft 14, on which also is arranged the turbine. The compressor 13 functions to compress the air flowing in through an air inlet 15. In accordance with the above discussion, the incoming air is supplied to each cylinder through the air duct 2.

In a manner being as such previously known, there are a number of different sensors (not shown) provided in connection with the engine 1 and the adherent vehicle. These sensors are used for detection of different variables representing the operating conditions of the engine 1 and the vehicle.

Preferably, a lambda sensor 16 (located upstream of the catalytic converter 12) for detection of the oxygen concentration in the exhaust gases, a rotational speed sensor 17 for the engine 1, a load sensor in the form of an air flow meter 18 (for measuring the amount of air injected into the engine 1) arranged in the air inlet 15, a temperature sensor 19 for detecting the engine 1 coolant temperature, a temperature sensor 20 for the air flowing into the engine, and a sensor 21 for the vehicle speed, are used. All sensors are connected to the control unit 5 via electrical connections.

The turbocharger unit 8 further comprises, in a known manner, a so-called wastegate valve 22, which is electrically controllable and can be continuously controlled between two positions. Th first position is a closed position, in which a bypass duct 23 in the turbocharger unit is blocked so as to conduct the exhaust gases from the manifold 7 through the turbine 10. The other position is an open position, in which the passage through the bypass duct 23 is open. In the latter case, the exhaust gases will be bypassed directly to the exhaust pipe 11, without flowing through the turbine 10, which reduces the charge pressure from the turbocharger unit 8 during operation. For control of the wastegate valve, it is connected to the control unit 5. In this way, the turbocharger pressure can be influenced through controlling the function of the wastegate valve 22.

During operation of the engine 1, the control unit 5 functions to control the air/fuel mixture to the engine 1 so as to keep it, at all times, as close to the stoicheiometric mixture (i. e. X=1) as possible. According to the introductory discussion there is however, during certain operating conditions, particularly at high loads, a risk that the thermal load on the engine 1 and its associated components may cause damage to and a deteriorated strength in these components. As examples of particularly sensitive components, the exhaust manifold 7, the turbocharger unit 8, the cylinder head 3 and the catalytic converter 12 may be mentioned. Consequently, there is a need for limiting the temperature of those thermally sensitive components arranged in connection with the engine 1.

According to what will be described in greater detail below, according to the invention, a value of the temperature of at least one, from a temperature point of view critical, component is derived in the control unit 5. This temperature value is used in controlling the engine 1, e. g. for a calculation of the amount of surplus fuel to be supplied to the respective cylinder.

According to a preferred embodiment, the thermal load of the engine system may thus be controlled by the supply of surplus fuel in such a way that this temperature value will never exceed a predetermined limit value, corresponding to the presence of a risk of damage to the component in question.

In accordance with the preferred embodiment, preferably two temperature values are derived. The first value corresponds to the temperature of the material in the cylinder head 3. The second value represents the temperature in the turbocharger unit 8. The points in question are preferably selected as points in the respective components that from experience may be expected to be sensitive for high temperatures.

Fig. 2 is a flow chart that, in a somewhat simplifie way, shows the function of the invention according to the first embodiment. The engine control will follow a periodical course which is initiated by a number of data representing the vehicle operating conditions being detected by means of the sensors 16- 21 (compare Fig. 1) and registered in the control unit 5 (square 25). These data preferably comprise the engine speed, the engine load (e. g. the amount of air per combustion), the ignition angle, the engine coolant temperature, the temperature of the incoming air, and the vehicle speed.

From the detected data of engine speed and load, two values, here called base temperatures T, and T2 respectively, are modelled, which represent indications of the temperatures of the selected temperature-critical material points (preferably consisting of the cylinder head and the turbocharger unit, respectively) (square 26). For this purpose, a relationship between the base temperatures T1, T2 and the engine speed and load may be predetermined for the engine type in question. This is done through temperature measurements made in advance at a number of different speeds and loads, whereby the relationships are stored in the form of a table in the control unit 5. All other data concerning the vehicle operating conditions (i. e. the incoming air temperature, the injection time, the ignition angle, the coolant temperature and the vehicle speed) are at this stage assumed to be equal to their nominal values, i. e. values corresponding to an operating condition of the engine system at normal, continuous operation.

The next step of the procedure comprises a static correction being made of the base temperatures Ti, T2 (square 27). Hereby corrections ATa, AT2 are produced, dependent upon to what extent the recorded data for the engine injection time and ignition angle, coolant temperature, air temperature and vehicle speed are deviating from their respective normal values. For

example, the two different temperatures, in the cylinder head 3 and the turbocharger unit, are influence to a different extent by changes in the above parameters. These dependencies may also be produced by utilising tables stored in the control unit and defining a model for the temperatures of the cylinder head 3 and the turbocharger unit. In this way, statically corrected values can be determined as follows: Tis = Ti + ATi T2s = T2 + AT2 where Tis is the statically corrected value of the cylinder head temperature and T2s is the statically corrected value of the turbocharger unit temperature.

The statically corrected temperature values Tis, T2s are then subjected to a dynamic correction (square 28). This is preferably made by means of a low- pass filtration of said temperature values, producing dynamically corrected, modelled values T1M and T2M, respectively.

According to the embodiment, a low-pass filtration of the first order is used for the dynamic correction. Dynamic corrections of the statically corrected temperature values Tis, T2s are now obtained according to the relationships: TiM [t] = (1-h/x) TiM [t-1] + (h :) T1sft] T2M [t] = (1-h2/T2) T2M [t-1] + (h2/T2) T2s [t] where TAM is the output signal from the filter, corresponding to the final temperature estimation for the cylinder head 3, T2M is the output signal from the filter, corresponding to the final temperature estimation for the turbocharger unit, n and hi are the time constant and the sampling interval, respectively, for the cylinder head 3, and T2 and h2 are the time constant and the sampling interval, respectively, for the turbocharger unit. Preferably, the time constants are selected as suitable functions of the engine speed and load. Through this dynamic modelling according to the invention, the thermal inertia in association with the heating of the engine system can be utilised. In this context, the term"thermal inertia"is used to describe the inherent dynamic temperature filtration, i. e. the relatively slow adaptation to a

changed temperature existing between the exhaust gases and the material in the engine and the exhaust system. This thermal inertia is in turn due to the heat transfer between gas and wall material, the thermal capacity of the material, and the cooling effect of the surrounding media (e. g. air, water and material).

The modelled temperature values T1M and T2M thus represent the estimated temperatures of the cylinder head and the turbocharger unit, respectively, which have been compensated for the above-mentioned thermal inertia, and which will subsequently be used for controlling the surplus fuel supplied to the engine at full load. Hereby a comparison is made between the modelled temperature values T1M, T2M and the predetermined limit temperatures Tic, T2G representing critical temperatures at which the cylinder head 3 and the turbocharger unit, respectively, run the risk of being damaged (square 29), in accordance with what has been discussed above. The critical temperatures vary with the component in question, and also with the material used in that component.

From the above comparisons, corresponding values for a reduction of the amount of fuel injected into the engine are then determined (corresponding to the extent to which the injection time will be reduced in relation to the nominal case), which are to be used in controlling the engine injection device (square 30). This means that two different values for the reduction of the amount of injected fuel will be obtained, i. e. one value representing the calculation (T »- T1M) for the cylinder head 3 and one value representing the calculation (T2G- T2M) for the turbocharger unit. In order to safeguard that the critical temperature of the cylinder head 3, as well as that of the turbocharger unit, is never exceeded, the smaller of the two reductions is selected for the continued engine control (square 31). In this manner, a value of a corrected absolute amount of injected fuel is obtained (square 32), which is used in engine control for regulation of the respective injection device (square 33).

This will in turn create a limitation of the temperature within the system, as was explained above.

The corrected absolute amount of injected fuel will deviate to a certain degree from the nominal absolute amount. The respective injection device is

therefore controlled according to this corrected amount. The process then returns to square 25. When the process then restarts again, input signals from the various sensors will be detected anew. Hereby, the previously calculated value for the amount of injected fuel will be used as one variable in this detection (square 25). A dashed line in Fig. 2 indicates this.

Through the control described above, a reduction of the nominal amount of injected fuel is obtained, which in turn creates a fuel saving, but without exceeding the critical temperatures for the cylinder head 3 or the turbocharger unit. Furthermore, the corrected amount of fuel is preferably limited downwards by means of a maximum allowed X value (preferably X=1).

According to an alternative embodiment, the control of the added fuel amount in the"critical area"may be performed for the individual cylinder. For this purpose, the engine must then comprise separate injection devices and ignition angle control for each cylinder. This is frequently available in today's vehicles.

The function of the invention will now be explained further with reference to Fig. 3, which shows a diagram of the amount of surplus fuel supplied as a function of time. The diagram shows an operating sequence that at a certain point in time, tl, includes a situation with a large increase in load, i. e. into that "critical area"which is characterised by so high loads and rotational speeds that the air/fuel mixture would normally be made richer than the stoicheiometric mixture.

The amount of fuel supplied that is allowed according to the invention (i. e. the corrected absolute fuel amount) is shown in Fig. 3 by a continuous line, whereas the fuel amount according to the known art (i. e. the nominal absolute fuel amount) is shown by a dashed line. The level corresponding to zero of the y axis represents that case where the air/fuel mixture is stoicheiometric, i. e. where =1.

In the above situation, according to the known art, a sudden step occurs in the amount of fuel supplied, up to a level BN, causing a reduction of the exhaust gas temperature, according to what has been explained above. This

fuel amount BN corresponds to the exhaust gas temperature being limited to a critical limit value. Contrary to the known art, the invention is based on the understanding that such a large step BN in fuel supply is not necessary at once for the above load increase at tr, as the material temperatures (e. g. in the cylinder head 3 and the turbocharger unit) will not increase as rapidly anyway as the load change does. This will allow, for each time increment, a certain reduction of the surplus fuel amount that would normally be supplied to the engine. This reduction corresponds to a deviation AB from the nominal amount of fuel BN. In accordance with what is shown in Fig. 3, this deviation AB will successively be reduced to zero. In spite of a relatively small amount of surplus fuel being supplied during this process, the amount will still be large enough to prevent the material temperatures to exceed their critical values. Thanks to the invention, a lower fuel consumption than in the nominal case is achieved. The shaded area 34 of Fig. 3 will thus correspond to the fuel saving compared to the previously known art.

Practical tests have shown that the invention achieves a substantial reduction of the fuel consumption at high loads and engine speeds. The invention works particularly well during highway driving with frequent overtaking with generally fully open throttle.

Instead of comparing to fixed, nominal values (compare square 27, Fig. 2), the modelling process according to the invention can be made adaptive. This might be necessary because one of the sensors 16-21 (see Fig. 1) is providing measurement values that drift over time and will provide differing measurement results, or because different engines will differ even if they are of the same model, making individual adaptation necessary. Furthermore, ageing of the engine and its associated components may require adaptive control. A detection of changes can be made by means of separate sensors or through empirical relations stored in tables in the control unit. Such possible changes may for example be detected by a temperature sensor (not shown) for measuring the exhaust gas temperature. As the measured tem- perature changes, the static calculation model will then be updated by being corrected. This adaptive calculation model (square 35) may then be included into the flow chart according to Fig. 2 by correcting on one hand the

modelling of the base temperatures (square 26) and on the other the calculation model used for the static correction (square 27).

Consequently, the values obtained for the injected amount of fuel (see square 32, Fig. 2) can be utilised for controlling the engine 1 at high loads and speeds. According to what was discussed above, this control may be performed by regulating the amount of surplus fuel to the engine. Alternatively, the control may also be performed by a regulation of the total amount of fuel and air supplied to the engine, in which case a lower engine power output entails a temperature decrease. This may in turn be controlled by means of the throttle 6, if the latter is an electrically controlled throttle.

According to a further embodiment, cooling the respective engine combustion chambers by means of a suitable coolant, for example water, may also perform the control of the engine thermal load. Fig. 4 shows, in principle, how such a cooling system may be arranged. The arrangement according to Fig. 4 corresponds to that shown in Fig. 1, with the exception of a particular injector 36 for water, located at the respective cylinder of the engine 1. The injector 36 is furthermore connected to a water pump 37 that is functioning to deliver water under high pressure during operating conditions characterised by high loads and speeds.

Fig. 5 shows a flow chart for the cooling system according to Fig. 4. The ref- erence figures 25-29 correspond to what was stated above in connection with Fig. 2. When the comparison is made between the modelled temperature values T1M. T2M and the respective limit values T1G. T2G. it is determined to what extent water injection by each injector 36 is deemed necessary in order to limit the material temperatures in the cylinder head and the turbocharger unit, respectively. Consequently, the required amount of water to be supplied to each cylinder in order to achieve the required cooling is determined (square 38). Hereby, two different values will be obtained, corresponding to the temperature of the cylinder head 3 and the turbocharger unit, respectively. These values are based on a modelling, performed in advance, of the effect of the water amount on the respective temperatures, as a func- tion of the operating point. In order to safeguard that the critical temperature of the cylinder head 3, as well as that of the turbocharger unit, is never

exceeded, the larger of the two water amounts is selected for the continued control (square 39).

Thereupon, the respective injector 36 is activated for the cylinder or cylinders where cooling is required (40). Then, when the process restarts, a feedback is obtained in that the selected value of water amount supplied is used as an input signal for the temperature model (square 25).

According to yet another variant embodiment, the cooling of the engine is controlled by means of controlling the engine coolant temperature. Fig. 6 shows an arrangement by which such a control may be utilised. The arrangement according to Fig. 6 corresponds to that shown in Fig. 1, with the exception of using the coolant system of the engine 1 for controlling the motor as a function of load and speed variations. The engine 1 is in a known manner provided with a radiator 41 for a water-based coolant, which is made to circulate inside the engine by means of a coolant pump 42. In the figure, arrows indicate the coolant flow direction. A thermostat 43 governs the coolant flow. The thermostat 43 (and preferably also the pump 42) is electrically controlled and connected to the control unit 5. The system, in a known manner, further comprises a cooling fan 44.

The coolant circulating in the engine 1 absorbs heat. By means of the thermostat 43, the coolant flow inside the engine 1 can be controlled. When the engine 1 is cold, no coolant circulates through the radiator 41, as the thermostat 43 is set to a certain limit temperature and will block coolant flow to the radiator 41 when the engine temperature is lower than the limit temperature. In accordance with what can be seen in Fig. 6, the coolant will however circulate inside the engine 1 also when the thermostat 43 is blocking the flow to the radiator 41. When the engine has been warmed up to the limit temperature of the thermostat 43, the latter will open and allow coolant flow to the radiator. In this way the engine can be cooled, so that the temperature- critical components are not damaged.

According to the embodiment, the limit temperature of the thermostat 43 can be adjusted according the cooling needs, e. g. if a sudden increase occurs in load and speed. This is then performed according to the flow chart shown in

Fig. 7. The reference figures 25-29 correspond to what was stated above in connection with Figs. 2 and 5. When the comparison is made between the modelled temperature values T1M, T2M and the respective limit values Tic, T2G, it is determined what coolant flow through the radiator 41 is necessary for the required degree of cooling (square 45). In order to safeguard that the critical temperature of the cylinder head 3, as well as that of the turbocharger unit, is never exceeded, the larger of the two calculated water flow rates is selected for the continued control (square 46). Consequently, the cooling of the engine will be performed depending upon the selected limit value of the thermostat. This value will also be used in the continued detection of variables regarding the engine operating conditions (square 25).

According to still another variant embodiment, the engine cooling can be achieved by regulation of the above-mentioned wastegate valve 22 (see Fig.

1), which for this purpose will be electrically controlled by means of the control unit 5. Unlike the methods described above, the wastegate valve 22 can, according to this embodiment, be regulated, more specifically by changing it to a variable mode, in order to lower the charge pressure from the turbocharger unit. This entails that the temperature in the turbocharger unit 8 is reduced. By utilising previously known relationships between the charge pressure of the turbocharger unit 8 and the modelled values of the temperatures T1M. T2M of the cylinder head 3 and the turbocharger unit, respectively, the wastegate valve may be controlled so as to obtain the required charge pressure.

The invention is not limited by the embodiments described above and shown in the drawings, but may be varied within the scope of the appended patent claims. For example, a multitude of different material points may be used, i. e. not only the cylinder head 3 and the turbocharger unit as stated above.

These material points are selected in those components associated with the engine that are judged to be temperature-critical. Examples of other material points that may be utilised are the catalytic converter and the lambda probe.

When selecting material points, preferably one point associated with the engine combustion chamber and one point downstream of the engine are selected.

Aside from the above embodiments, where various types of engine cooling are used, other forms of cooling can be utilised. As an example, the vehicle cooling fan may be controlled for this purpose.

The embodiment variant described in connection with Figs. 6 and 7 could suitably be arranged so as to activate the control when the coolant has reached a certain predetermined limit temperature.

The temperatures of one or more of the thermally critical components could alternatively be determined with the aid of a hardware type temperature sensor that may be fitted in connection with the component in question. Consequently, also directly measured values could be used, instead of modelled values, in the control used for cooling the engine.

Also other variables concerning the vehicle and the engine operating condi- tions than the ones stated above, may be utilised and considered in determining the current temperature values. For example, the k value obtained in the exhaust gases, during full load modelling according to the invention, could be fed back and used as an input variable to the control unit.

Furthermore, existing systems for detection of erroneous ignition (so called "misfire") of the engine could be utilised for the modelling, as an incomplete ignition will also influence the exhaust gas temperature.

The invention can also be used for engines without turbocharger units. Pref- erably, the exhaust manifold would then be used as a temperature-critical component, the temperature of which you would wish to model.

The cooling by means of thermostatic control according to Figs. 6 and 7 is preferably used as a complement to one of the other types of cooling described above, as its influence is slower, and may primarily be used for control of the temperature in the cylinder head 3.

Finally, it should be understood that the cooling of the engine might be real- ised through various combinations of the embodiments described above.