Method for modeling a compressor intake temperature and/or a compressor discharge temperature of a compressor, and a control unit, and a motor vehicle

Abstract

The invention relates to a method for modeling a compressor intake temperature and/or a compressor discharge temperature of a compressor taking into account a compressor surge, wherein the method comprises: Identifying a pressure gradient across the compressor Identifying a mass flow gradient across the compressor Establishing that the compressor surge is present when the pressure gradient exceeds an upper pressure gradient limit and the mass flow gradient falls below a lower mass flow gradient limit; and Identifying the compressor intake temperature with a temperature correction factor that is dependent on the compressor surge and/or identifying the compressor discharge temperature on the basis of a corrected compressor discharge pressure that is dependent on the compressor surge.

Claims

1. A method for modeling a compressor intake temperature and/or a compressor discharge temperature of a compressor taking into account a compressor surge, the method comprising: identifying a pressure gradient across the compressor; identifying a mass flow gradient across the compressor; establishing that the compressor surge is present when the pressure gradient exceeds an upper pressure gradient limit and the mass flow gradient falls below a lower mass flow gradient limit; and identifying the compressor intake temperature with a temperature correction factor that is dependent on the compressor surge and/or identifying the compressor discharge temperature on the basis of a corrected compressor discharge pressure that is dependent on the compressor surge, the compressor intake temperature and/or the compressor discharge temperature being used to determine a turbocharger speed and a base boost pressure of an internal combustion engine.

2. The method according to claim 1, wherein the pressure gradient corresponds to a maximum pressure gradient over a predetermined time period and/or the mass flow gradient corresponds to a minimum mass flow gradient over the predetermined time period.

3. The method according to claim 1, wherein the identification of the compressor intake temperature and/or the compressor discharge temperature takes heat transfer effects into account.

4. The method according to claim 3, wherein the heat transfer effects include at least one of the following: wall-heat losses, stored heat, and thermal radiation emanating from components.

5. The method according to claim 3, wherein the heat transfer effects are determined with at least one of the following: characteristic curves and characteristic maps.

6. The method according to claim 1, wherein the temperature correction factor is identified based on a compressor surge temperature increase.

7. The method according to claim 6, wherein the compressor surge temperature increase is identified based on a pressure ratio between ambient pressure and a compressor intake pressure.

8. The method according to claim 7, wherein the compressor surge temperature increase is modeled with a differentiator with first order lag.

9. The method according to claim 1, wherein the identified compressor discharge temperature is based on an uncorrected compressor intake temperature, wherein the uncorrected compressor intake temperature has not been corrected with the temperature correction factor that is dependent upon the compressor surge.

10. The method according to claim 8, wherein the uncorrected compressor intake temperature is corrected based on a pressure ratio between ambient pressure and a compressor intake pressure.

11. The method according to claim 1, wherein the uncorrected compressor intake temperature is low-pass filtered.

12. The method according to claim 1, wherein for the purpose of error protection, the corrected compressor discharge pressure corresponds to a measured compressor discharge pressure when a pressure ratio across the compressor is less than a pressure ratio limit and a compressor discharge pressure is less than a last valid compressor discharge pressure.

13. The method according to claim 1, wherein the compressor discharge temperature is low-pass filtered.

14. A control unit that is equipped to carry out a method according to claim 1.

15. A motor vehicle having a control unit, wherein the motor vehicle is equipped and designed to carry out the method according to claim 1.

Description

BRIEF DESCRIPTION OF THE DRAWINGS

(1) The present invention will become more fully understood from the detailed description given hereinbelow and the accompanying drawings which are given by way of illustration only, and thus, are not limitive of the present invention, and wherein:

(2) FIG. 1 schematically shows a motor vehicle according to an embodiment;

(3) FIG. 2 shows an overall model for determining a compressor intake temperature and a compressor discharge temperature of a fluid;

(4) FIG. 3 shows a model for determining the compressor intake temperature;

(5) FIG. 4 shows a model for detecting compressor surge;

(6) FIG. 5 shows a model for pressure determination of a compressor discharge pressure;

(7) FIG. 6 shows a model for determining the compressor discharge temperature; and

(8) FIG. 7 schematically shows a control unit according to one embodiment.

DETAILED DESCRIPTION

(9) FIG. 1 shows a motor vehicle 10 with an internal combustion engine 2 and a supercharging system 1. The supercharging system 1 includes a compressor 3 and a turbine 7. The compressor 3 is connected to the turbine 7 by a shaft 5, and consequently can be driven by the turbine 7.

(10) Air drawn in from the environment flows across the compressor 3 and an intake manifold 2a into cylinders of the internal combustion engine 2. The air/fuel mixture produced in the cylinders is burned to create a (vehicle) drive power. The exhaust gas arising from the combustion flows through an exhaust manifold 2b and the turbine 7. Consequently, the turbine 7 can be supplied with and driven by the exhaust gas from the internal combustion engine 2. In this way, the air that is drawn in can be compressed by the compressor 3 depending on the power of the turbine 7.

(11) FIGS. 2 to 6 show models that are used when carrying out an embodiment of the method according to the invention.

(12) FIG. 2 shows an overall model for determining a compressor intake temperature T.sub.V,E, α and a compressor discharge temperature T.sub.V,A while taking into account a compressor surge. The overall model comprises a compressor intake temperature model 100, a pressure determination model 300, and a compressor discharge temperature model 400.

(13) An engine temperature T.sub.M, an ambient temperature T.sub.0, an intake manifold temperature T.sub.S, a vehicle speed v, an ambient pressure p.sub.0, a compressor mass flow m.sub.ov across the compressor 3, a compressor intake pressure p.sub.1 (pressure before the compressor 3), and a compressor discharge pressure p.sub.2 (pressure after the compressor 3) enter into the compressor intake temperature model 100 as input quantities. These input quantities can be sensed through appropriate acquisition devices on the vehicle side, for example.

(14) A compressor intake temperature T.sub.1, the compressor intake temperature (corrected compressor intake temperature) T.sub.1,α that is dependent on compressor surge, a mass flow gradient ∇{dot over (m)} across the compressor 3, and a pressure gradient ∇p.sub.V across the compressor 3 result from the compressor intake temperature model 100 as output quantities. In the compressor intake temperature model 100, the compressor intake temperature T.sub.1,α is identified using a correction factor ΔT.sub.VP,α that is dependent on compressor surge.

(15) The mass flow gradient ∇{dot over (m)} and the pressure gradient ∇p.sub.V, the pressure p.sub.V,E before the compressor 3, and the pressure p.sub.V,A after the compressor 3 enter into the pressure correction model 300 as input quantities.

(16) A corrected compressor discharge pressure p.sub.2,α results from the pressure correction model 300 as the output quantity.

(17) The engine temperature T.sub.M, the compressor mass flow my, the compressor intake pressure p.sub.1, the corrected compressor discharge pressure p.sub.2,α, and the (uncorrected) compressor intake temperature T.sub.1 enter into the compressor discharge temperature model 400 as input quantities.

(18) The compressor discharge temperature T.sub.2 results from the compressor discharge temperature model 400 as the output quantity. In the compressor discharge temperature model 400, the compressor discharge temperature T.sub.2 is identified on the basis of a corrected compressor discharge pressure p.sub.2,α that is dependent on compressor surge.

(19) FIG. 3 shows the compressor intake temperature model 100 for determining the compressor intake temperature T.sub.1,α in detail. The ideal gas equation represents the physical basis for the compressor intake temperature model 100. The starting point is the ambient temperature T.sub.0, which in the novel approach is heated by thermal radiation from hot components such as the internal combustion engine 2 and the intake manifold 2a. Thus the compressor intake temperature model 100 comprises a stored heat model 120 for taking stored heat into account, a thermal radiation model 140 for taking thermal radiation into account, a temperature correction model 150 for taking a temperature increase due to the occurrence of compressor surge into account, and a pressure ratio correction model 160 for taking environmental conditions in the form of the ambient pressure into account.

(20) In the stored heat model 120, a vehicle speed correction factor α.sub.v is identified from the vehicle speed v using a characteristic curve 122. The vehicle speed correction factor α.sub.v takes into account an effect of the vehicle speed v on the stored heat. In addition, a difference between the engine temperature T.sub.M and the ambient temperature T.sub.0 is calculated in the subtraction block 124. From this difference and the compressor mass flow m.sub.v, a stored heat temperature increase ΔT.sub.Stau due to the stored heat can be identified using a characteristic map 126. In the multiplication block, the vehicle speed correction factor α.sub.v and the stored heat temperature increase ΔT.sub.Stau are multiplied by one another in order to obtain a corrected stored heat temperature increase ΔT.sub.Stau,α that takes into account the vehicle speed v and is supplied to a summation block 102.

(21) In the thermal radiation model 140, a difference between intake manifold temperature T.sub.S and the ambient temperature T.sub.0 is calculated in a subtraction block 142. With the characteristic map 144, a thermal radiation temperature increase ΔT.sub.Strahlung due to thermal radiation can be identified from this difference and the compressor mass flow {dot over (m)}.sub.V. The thermal radiation temperature increase ΔT.sub.Strahlung is supplied to a summation block 104 and added to the result of the summation block 102. The result of the summation block 104 thus corresponds to the heating of the ambient temperature T.sub.0 due to stored heat and thermal radiation.

(22) With the stored heat model 120 and the thermal radiation model 140, the physical effects in the case of engine cold start and stored heat in the case of relatively long idle periods are taken into account.

(23) The result of the summation block 104 is a heat effect temperature increase ΔT.sub.W on account of heat effects, which is to say stored heat and/or thermal radiation in the present case. This temperature increase ΔT.sub.W is supplied to an (optional) low-pass filter 106.

(24) The pressure ratio between the ambient pressure p.sub.0 and the compressor intake pressure p.sub.1 is identified in the pressure ratio correction model 160 and taken into account by the compressor intake temperature model 100. With the pressure ratio correction model 160, the effect of the ambient pressure on the thermal capacity of the air can be taken into account. Thus, in a multiplication block 108, the heat effect temperature increase ΔT.sub.W is multiplied by the pressure ratio between the ambient pressure p.sub.0 and the compressor intake pressure p.sub.1 for a density correction. Consequently, the (uncorrected) compressor intake temperature T.sub.1 is obtained from the multiplication block 108. The compressor discharge temperature T.sub.1 thus obtained does not yet take any occurrence of compressor surge into account.

(25) In order to take the occurrence of compressor surge into account, the temperature correction model 150 is provided. The temperature correction model 150 comprises a compressor surge detection model 200, with which it is possible to detect an occurrence of compressor surge. On the input side, the compressor intake pressure p.sub.1, the compressor discharge pressure p.sub.2, and the compressor mass flow {dot over (m)}.sub.V enter into the compressor surge detection model 200. On the output side, a signal S.sub.PV is produced that indicates an occurrence of compressor surge.

(26) The compressor surge detection model 200 is shown in detail in FIG. 4. Characteristic of compressor surge is a sharply decreasing mass flow across the compressor 3 with a simultaneous (or occurring with a slight delay) comparatively sharp pressure increase after the compressor 3. This behavior is made use of in the compressor surge detection model 200 for detecting compressor surge.

(27) In the compressor surge detection model 200, the compressor intake pressure p.sub.1 and the compressor discharge pressure p.sub.2 are supplied to a block 202 for identifying the pressure gradient ∇p.sub.V across the compressor 3. Furthermore, the compressor mass flow {dot over (m)}.sub.V is supplied to a block 204 for identifying the mass flow gradient ∇{dot over (m)}.sub.V.

(28) In block 206, the results of the gradient calculations from the blocks 202, 204 are stored. Here, the results can come from multiple calculations. Thus five calculations can take place, for example. The gradient calculations in the blocks 202, 204 can be carried out every 5 to 15 milliseconds, for example. In some embodiments, the gradient calculations can be carried out every 10 milliseconds. By the means that the gradient calculations are performed over a predetermined time period, such as 50 milliseconds, it is possible for slightly delayed processes of the described characteristics to be taken into account for surge events. Consequently, the robustness of detecting compressor surge can be increased.

(29) In block 208, a maximum pressure gradient ∇p.sub.V,max is identified by means of the pressure gradient ∇p.sub.V across the compressor 3. Optionally, a minimum mass flow gradient V{dot over (m)}.sub.V,min can be identified in block 208 by means of the mass flow gradient ∇{dot over (m)}.sub.V.

(30) In block 210, the maximum pressure gradient ∇p.sub.V,max and the minimum mass flow gradient ∇{dot over (m)}.sub.V,min are checked against appropriate thresholds or limits. If the maximum pressure gradient ∇p.sub.V,max exceeds an upper pressure gradient limit ∇p.sub.V,lim and the minimum mass flow gradient ∇{dot over (m)}.sub.V,min falls below a lower mass flow gradient limit V{dot over (m)}.sub.V,lim, then the compressor surge detection signal S.sub.VP is output in an AND block 212. Thus, it is identified in block 212 that the compressor surge is present if the pressure gradient ∇p.sub.V exceeds the upper pressure gradient limit ∇p.sub.V,lim and the mass flow gradient ν{dot over (m)}.sub.V,min falls below the lower mass flow gradient limit V{dot over (m)}.sub.V,lim. As mentioned above, the compressor surge detection signal S.sub.VP signals a presence/occurrence of compressor surge through the compressor 3.

(31) During compressor surge, a flow stall arises in which a backflow of hot compressed air (before the compressor 3) takes place. Consequently, an abrupt compressor surge temperature increase ΔT.sub.VP occurs at the compressor intake during the surge event, wherein the abrupt temperature increase diminishes over time. The compressor surge temperature increase ΔT.sub.VP is dependent on the ratio π.sub.0,V between the ambient pressure p.sub.0 and the compressor intake pressure p.sub.1, wherein the ratio is identified in block 152. The relationship between compressor surge temperature increase ΔT.sub.VP and the ratio π.sub.0,V is stored in a characteristic curve 154.

(32) The compressor surge temperature increase ΔT.sub.VP initially rises abruptly and then decreases over time. Therefore, the compressor surge temperature increase ΔT.sub.VP has a behavior that can be modelled with a differentiator with first order lag (DT1). This behavior can then be represented as a correction. To this end, the temperature correction model 150 has a block 156 into which the compressor surge temperature increase ΔT.sub.VP and the compressor surge detection signal S.sub.VP enter on the input side. Block 156 models a DT1 behavior. In other words, block 156 includes a DT1 element. A corrected compressor temperature increase ΔT.sub.VP,α is produced by block 156 on the output side. The corrected compressor temperature increase ΔT.sub.VP,α is thus the correction factor that is dependent on compressor surge.

(33) In the summation block 110, the compressor intake temperature T.sub.1 is summed with the corrected compressor temperature increase ΔT.sub.VP,α to identify a corrected compressor intake temperature T.sub.1,α. The corrected compressor intake temperature T.sub.V,E,α corresponds to a compressor intake temperature while taking compressor surge into account.

(34) In FIG. 5, the pressure determination model 300 is shown in detail. The pressure peaks that arise during compressor surge are brought about by the closing of the throttle valve and have no physical effect on the isentropic compression. For this reason, a corrected compressor discharge pressure p.sub.2,α can be identified with the aid of the pressure determination model 300.

(35) Thus, it is first checked in block 310 whether a pressure correction should be activated.

(36) To this end, the results of the gradient calculations from the blocks 202, 204 are stored in block 312. Here, the results may originate from multiple calculations. In addition, the maximum pressure gradient ∇p.sub.V,max and the minimum mass flow gradient ∇{dot over (m)}.sub.V,min are identified in block 312. In block 314, the maximum pressure gradient ∇p.sub.V,max and the minimum mass flow gradient ∇{dot over (m)}.sub.V,min are checked against appropriate thresholds or limits.

(37) In AND block 316, the following is checked: whether the maximum pressure gradient ∇p.sub.V,max exceeds the upper pressure gradient limit ∇p.sub.V,lim; whether the minimum mass flow gradient ∇{dot over (m)}.sub.V,min falls below the lower mass flow gradient limit ∇{dot over (m)}.sub.V,lim; and whether a pressure correction deactivation signal S.sub.korr,0 is present.

(38) If the above three conditions are answered in the affirmative, then a pressure correction activation signal S.sub.korr,1 and a last valid pressure value p.sub.2,val for the compressor discharge pressure are produced from the AND block 316 on the output side. The last valid compressor discharge pressure p.sub.2,val is that which is present before the occurrence of the compressor surge. In other words, the last valid compressor discharge pressure p.sub.2,val is the value for the compressor discharge pressure p.sub.2 before an increase in the compressor discharge pressure p.sub.2 takes place on account of compressor surge.

(39) Since the blocks 312, 314, and 316 are, in principle, directed toward the detection of compressor surge, a check can be made in block 316 for a presence of the compressor surge detection signal S.sub.VP from the compressor surge detection model 200 as an alternative to the output quantities from blocks 312, 314.

(40) The pressure correction model 300 comprises error protection 320 that uses two boundary conditions. Firstly, the corrected compressor discharge pressure p.sub.2,α must be less than the measured compressor discharge pressure p.sub.2. Secondly, compressor surge only occurs at a high pressure ratio π.sub.V across the compressor 3. If this falls below a minimum pressure ratio limit π.sub.V,lim, then the pressure correction is terminated.

(41) For this purpose, it is identified in block 322 of the error protection 320 whether the pressure ratio π.sub.V across the compressor 3, which is to say the compressor discharge pressure p.sub.2 divided by the compressor intake pressure p.sub.1, is less than the pressure ratio limit π.sub.V,lim. Here, the pressure ratio limit π.sub.V,lim can be between 1.3 and 2.8, for example.

(42) In block 324, it is identified whether the measured compressor discharge pressure p.sub.2 is less than the last valid compressor discharge pressure p.sub.V,A,val. If the pressure ratio π.sub.V is less than the pressure ratio limit π.sub.V,lim and the compressor discharge pressure p.sub.2 is less than the last valid compressor discharge pressure p.sub.2,val, then a pressure correction deactivation signal S.sub.korr,0 is produced by an AND block 326.

(43) In block 340, it is identified whether the pressure correction is carried out. In other words, the value that the corrected compressor discharge pressure p.sub.2,α takes on is identified in block 340. For this purpose, the pressure correction activation signal S.sub.korr,1 from the block 310 and, if applicable, the pressure correction deactivation signal S.sub.korr,0 from the error protection 320 go into the block 342 on the input side. When the pressure correction deactivation signal S.sub.korr,0 is present at the input side on block 342, then the pressure correction deactivation signal S.sub.korr,0 also results from the block 342 on the output side. If no pressure correction deactivation signal S.sub.korr,0 results from the error protection 320 and only the pressure correction activation signal S.sub.korr,1 is present at the input side on block 342, the pressure correction activation signal S.sub.korr,1 is produced by the block 342 on the output side. The output quantity of the block 342 is supplied to a switch 344.

(44) The corrected compressor discharge pressure p.sub.2,α is produced by the switch 344 on the output side. When the pressure correction activation signal S.sub.korr,1 is present in the switch 344 on the input side, then the corrected compressor discharge pressure p.sub.2,α corresponds to the last valid compressor discharge pressure p.sub.2,val. When the pressure correction deactivation signal S.sub.korr,0 is present in the switch 344 on the input side, then the compressor discharge pressure p.sub.2,α corresponds to the measured compressor discharge pressure p.sub.2.

(45) In FIG. 6, the compressor discharge temperature model 400 is shown in detail. The temperature modeling at the discharge of the compressor is based on the above-described compressor intake temperature T.sub.1 in which the temperature increase on account of compressor surge is excluded, on the isentropic compression, and on corrections, such as for taking engine cold start into account. The exclusion of the temperature increase on account of compressor surge is necessary because a reversal of the flow direction of the mass flow through the compressor 3 occurs during compressor surge. Accordingly, a causal relationship no longer exists between the compressor discharge temperature T.sub.2 and the compressor intake temperature T.sub.1, and a correction of the compressor intake temperature T.sub.1 becomes necessary. The reason for this is that chaotic conditions are present across the compressor 3. Thus the compressor intake temperature T.sub.1 is increased by a backflow. However, since almost no compressor output is being provided any more, the compressor discharge temperature T.sub.2 is no longer increasing significantly. As a result, the effects of compressor surge are taken into account in determining the compressor intake temperature

(46) In the compressor discharge temperature model 400, the following output quantity is first identified in a block 410:

(47) $\frac{{\left(\frac{p_{2,\alpha }}{p_1}\right)}^{\frac{\kappa -1}{\kappa }}}{{\eta }_V}$

(48) For this purpose, in block 412 a corrected pressure ratio π.sub.V,α across the compressor 3 is identified that corresponds to a ratio of the corrected compressor discharge pressure p.sub.2,α divided by the compressor intake pressure p.sub.1. The relationship between the isentropic exponent κ and the corrected pressure ratio π.sub.V,α is stored in a

(49) ${\left(\frac{p_{2,\alpha }}{p_1}\right)}^{\frac{\kappa -1}{\kappa }}$

(50) compression model 414. Consequently, the isentropic compression term is identified as the output quantity from the compression model 414 using the corrected pressure ratio π.sub.V,α and the isentropic exponent κ.

(51) With knowledge of the compressor mass flow {dot over (m)}.sub.V and of the corrected pressure ratio π.sub.V,α, the compressor efficiency η.sub.V is identified using a compressor efficiency characteristic map 418.

(52) In block 416, the isentropic compression term

(53) ${\left(\frac{p_{2,\alpha }}{p_1}\right)}^{\frac{\kappa -1}{\kappa }}$
is divided by the compressor efficiency η.sub.V, so the above-mentioned output quantity of block 410 is obtained.

(54) Using a compressor-discharge-side wall heat loss characteristic map 420, with knowledge of the compressor mass flow {dot over (m)}.sub.V, a wall heat loss correction factor α.sub.konv is identified that is then multiplied by the result of the block 416 in the multiplication block 422 so that a factor is produced by the multiplication block 422 on the output side that takes into account an effect of the compression and of wall heat losses on the compressor discharge temperature T.sub.2.

(55) In the optional block 426, the temperature unit of the compressor intake temperature T.sub.1 (without taking compressor surge into account) is converted from degrees Celsius into Kelvin.

(56) In the summation block 424, the value “1” is added to the output quantity of the block 422.

(57) In the multiplication block 428, the compressor intake temperature T.sub.1 that has been converted into Kelvin is multiplied by the output quantity from block 424.

(58) In the subtraction block 430, the output quantity from block 428, whose temperature unit is given in Kelvin, is converted into degrees Celsius.

(59) With knowledge of the engine temperature T.sub.M, a thermal radiation temperature increase ΔT′.sub.Strahlung on the compressor discharge side on account of thermal radiation is identified using a compressor-discharge-side thermal radiation characteristic map 432.

(60) The temperature increase is added to the output quantity from the block 430 in the summation block 434.

(61) Optionally, with knowledge of the corrected pressure ratio π.sub.V,α, a filter time t.sub.filter for dynamic forming is identified using a filter time characteristic map 436.

(62) The filter time t.sub.filter is used in the (optional) low-pass filtering 438. The compressor discharge temperature model 400 can be made dynamic using the low-pass filtering 438 as a function of the filter time t.sub.filter.

(63) On the output side, the (modeled) compressor discharge temperature T.sub.2 results from the low-pass filtering 438. Here, the compressor discharge temperature model 400 takes into account any compressor surge that may occur.

(64) FIG. 7 schematically shows an exemplary control unit 20 that is equipped to carry out the above-described methods/models. The control unit 20 comprises a processor 22, a memory 24, and an interface 26. The memory 24 serves to store data such as, e.g., the above-mentioned characteristic maps, characteristic curves, or models 122, 126, 144, 154, 414, 418, 420, 432, 436. These characteristic maps and characteristic curves may have been identified beforehand on test stands. Furthermore, software that is designed to carry out the above-described methods is also stored in the memory 24. The processor 22 is designed to carry out program instructions of the software. The interface 26 is also designed to receive and transmit data. It can be, for example, an interface to a CAN bus of the motor vehicle 10, through which the control unit receives measured values from sensors, such as the input quantities for the compressor intake temperature model 100 and the like, and sends control commands.

(65) The invention being thus described, it will be obvious that the same may be varied in many ways. Such variations are not to be regarded as a departure from the spirit and scope of the invention, and all such modifications as would be obvious to one skilled in the art are to be included within the scope of the following claims.