DE102006042874A1 - Method for estimation of temperature in intake manifold of internal combustion engine, involves determining estimated value for temperature in intake manifold of internal combustion engine by kalman filter - Google Patents

Abstract

The method involves determining an estimated value for the temperature in the intake manifold of an internal combustion engine (2) by a kalman filter. A measured value is provided for the temperature in the intake manifold by a temperature sensor and during a stationary condition the estimated value for the temperature in the intake manifold is corrected on basis of this measured value. An independent claim is also included for a control device for a motor vehicle.

Description

The The present invention relates to a method for estimating the Temperature in the intake manifold an internal combustion engine, and a control device for a motor vehicle, which one for carrying out the Contained method estimator.

A Regulation of the temperature in the intake manifold of an internal combustion engine, in particular a compensation of unwanted temperature fluctuations, is of great importance for ignition delay, for example. Accordingly, there is a need for a most reliable determination this temperature size.

Out US 6,748,313 B2 For example, a method of estimating air charge in a cylinder of an internal combustion engine using a mass air flow (MAF) sensor and an intake manifold absolute pressure (MAP) sensor is known, where primarily the signal is used to estimate the air charge during a transient condition of the MAP sensor, in a transition period between a transient state and a stationary state, primarily a combination of the signals from the MAP sensor and the MAF sensor using a smoothing algorithm and during a steady state only the signal from the MAF sensor is used.

On the other hand is a sufficiently accurate sensor-based determination of the temperature in the intake manifold while transient phases of the engine problematic, and in particular because of the pronounced dynamics the temperature fluctuations associated with the dynamics of variation of Gas composition is comparable.

It is therefore an object of the present invention, a method for determining the temperature in the intake manifold of an internal combustion engine and a control device for to provide a motor vehicle, so that in particular during transient Phases an increased reliability the estimate can be achieved.

These The object is achieved by a method according to the independent claim 1 or a device according to the independent claim 4 solved.

According to the invention is a estimated value for the Temperature in the intake manifold determined by means of a Kalman filter. This is one way increased reliability even while transient phases of the internal combustion engine reaches than the determined estimated value the pronounced Gas dynamics in the intake manifold follows in transient phases. The invention according to a temperature sensor determined measured value for the temperature in the intake manifold is further during a stationary one Condition for correcting the estimated value for the Temperature in the intake manifold used. The inventive method thus contributes to one of the while transient phases of the internal combustion engine existing, pronounced gas dynamics in the intake manifold Invoice and minimizes the other estimation error in the steady state.

Further Embodiments of the invention are the description and the dependent claims to remove.

The Invention will be described below with reference to a preferred embodiment with reference to the attached Illustrations explained. Show it:

1 a schematic representation of an exhaust gas recirculation circuit for an internal combustion engine; and

2 a discrete Kalman filter algorithm used in the intake manifold temperature estimation method of the present invention.

The model underlying the method for estimating the temperature in the intake manifold according to the invention will be described below with reference to the in 1 schematically illustrated structure of an exhaust gas recirculation circuit 1 for an internal combustion engine with turbocharger explained.

Shown is an internal combustion engine 2 , in the exhaust stream, an exhaust gas turbine 3 is arranged, which is a compressor 4 drives. The from the compressor 4 exiting, compressed air is via an intercooler 5 before entering the internal combustion engine 2 cooled. In addition, a part of the internal combustion engine 2 exiting exhaust gases via an exhaust gas recirculation cooler 6 the from the intercooler 5 supplied incoming fresh air.

Also in 1 are designated at certain positions within the exhaust gas recirculation loop 1 present temperatures, wherein the exhaust gas temperature at the exit from the internal combustion engine 2 with T _exh , the temperature of the compressor 4 entering fresh air with T _amb , the air temperature downstream of the compressor 4 with T _comp , the temperature downstream of the intercooler 5 with T _ic , the air temperature downstream of the exhaust gas recirculation cooler 6 with T _egr and the temperature in the intake manifold of the internal combustion engine 2 with T _man is designated.

From the law of mass conservation at the position designated P downstream of the intercooler 5 and the exhaust gas recirculation cooler 6 follows the relationship

die zeitliche Änderung des Gas-Masse im Saugrohr des VerbrennungsmotorsIn this case, MAF denote the fresh air mass flow, EGR the exhaust gas recirculation mass flow, ASP the mass flow of the intake from the engine gas flow and

the temporal change of the gas mass in the intake manifold of the internal combustion engine

From the law of conservation of energy from position P to the intake valves, the relationship follows

Where the variable Cν represents the specific heat capacity of the gas in the intake manifold, the variable M _MAN is the mass of the gas in the intake manifold. H _AIR is the incoming enthalpy of the fresh air, H _EGR is the inflow enthalpy of the EGR system, and H _ASP is the enthalpy sucked from the intake manifold.

mit den Komponenten

• – Messwert der Ansaugkrümmertemperatur T_{MAN_MEAS}(k)
• – Temperatur des in den Ansaugkrümmer eintretenden, rückgeführten Abgases ("EGR-Gas") unter Berücksichtigung der Massentransport-Zeitverzögerung T_{EGR_DLY}(k) und
• – Korrigierte Ansaugkrümmertemperatur T_{MAN_COR}(k) gewählt.

On the basis of these laws, an estimated value for the temperature in the intake manifold is determined by means of a Kalman filter. The overall concept is named as a temperature observer. According to the invention, this is a three-component state vector

with the components

• - _Measurement of intake manifold _temperature T _{MAN_MEAS} (k)
• The temperature of the recirculated exhaust gas entering the intake manifold ("EGR gas") taking into account the mass transport time delay T _{EGR_DLY} (k) and
• - Corrected intake manifold _temperature T _{MAN_COR} (k) selected.

The transition of the system from time (k-1) to time k is modeled by x k = Φ k-1 .x k-1 + Λ k-1 u k-1 + w k-1 z k = H k x k + V k (3)

The variable Λ _{k-1 describes} the effect of the input quantities u _k-1 on the system state x _k . The input quantities here are indirectly the enthalpies H _AIR and H _EGR , but taking into account that the point P (see 1 ) has a small volume, can sucked in Ansaugkrümmen gas temperature be determined follows:

If the temperature behind the intercooler is not measured, the compressor outlet temperature T _COMP can be determined. T IC = T COMP - ε IC (T COMP - T AMB )

Wherein the variable ε _IC dastellt the cooling efficiency of the intercooler.

The compressor exit temperature T _COMP can be determined by measurement and / or modeling.

The P _COMP and P _{AMB are} the absolute air pressure behind and in front of the compressor. The variable η _c represents the compressor efficiency.

The exhaust gas recirculation temperature T _EGR is estimated to be determined as follows T EGR = T EXH - ε EGR (T EXH - T COOL )

Where the variable ε _{EGR represents} the heat exchange efficiency of the EGR cooling circuit (see 1 , Position 6 ), T _EXH is the exhaust temperature before turbine (see 1 , Position 3 ), and T _COOL is the temperature of the coolant used for the EGR gas (normally, engine cooling water is used as the coolant, and T _COOL is estimated to be equal to engine coolant temperature). If the exhaust gas temperature T _{EXH is} not measured, or the measuring sensor is too slow to provide the dynamic exhaust gas temperature with sufficient accuracy, the following estimate can be used:

Where ζ _{EXH provides} the fraction of the fuel energy released by combustion for heating the exhaust gas mass flow. FUEL is the injected fuel mass flow, E _HV is the hot value of the fuel, c man / p is the average heat capacity of the gas in the intake manifold, and finally c exh / p is the average heat capacity of the gas in the exhaust manifold

The state transfer function, Φ, specifies the following set of state equations for the physical description of the system:

By adjusting the calibration parameters c ₁ and c ₂ , the estimated value for the temperature in the intake manifold can be corrected on the basis of the sensor-supported, measured value for the temperature in the intake manifold during a stationary state.

wobei die Kovarianzmatrix Q_k im Falle des Temperaturbeobachters die folgende Struktur hat:

The variable w _k indicates the noise within the system (eg due to model errors), which is assumed to be white and normally distributed with expected value zero with the covariance matrix Q _k , ie w _k ∈ N (0, Q _k ), with

wherein the covariance matrix Q _k in the case of the temperature observer has the following structure:

wobei die Kovarianzmatrix R_k im Falle des Temperaturbeobachters die folgende Struktur hat: Rk = r2SENSOR The variable ν _k indicates the noise due to measurement noise, which is assumed to be white and normally distributed with expected value zero with the covariance matrix R _k , ie, ν _k ∈ N (0, R _k ), with

wherein the covariance matrix R _k in the case of the temperature observer has the following structure: R k = r 2 SENSOR

The Variable r 2 / SENSOR is the temperature measurement error covariance during the Intake manifold gas temperature measurement by means of the temperature sensor prolapse.

The algorithm of the discrete Kalman filter will be described below with reference to FIG 2 explains:
The system is refreshed at fixed time intervals or at fixed events, for example at fixed crank angle intervals of eg 120 °. This is called the main step.

N_sub_samples is a constant representing the desired number of between indicates intermediate steps performed on successive main steps.

According to 2 At the beginning of the algorithm, the value of the current sub-sampling step, I, is first zeroed in step S10. In step S20, a propagation of the state vector according to FIG x ^ k-1 (l + 1) = Φ k-1 (l) · x ^ k-1 (l) + Λ k-1 (L) u k-1 for lε [0; N_sub_samples - 1] (9) and a propagation of the error covariance matrix according to P k-1 (l + 1) = Φ k-1 (L) · P k-1 (L) · Φ T k-1 (l) + Q k-1 for lε [0; N_sub_samples - 1]. (10)

Of the Value of the current sub-sampling step, l, is incremented by one in S30, and it is in step S40 queried whether this value already the Has reached the value of N_sub_samples. If the answer is "No", it goes Algorithm returns to step S20, d. H. the value of the current sub-sample step, l, is incremented by one, and it is again in step S40 queried whether this value already has reached the value of N_sub_samples etc.

As soon as the answer in step S40 is "yes", the estimate is updated with the results of the measurement z _k , in this case the intake manifold gas temperature measurement, in accordance with FIG x ^ k (0) = x ^ k-1 (N_sub_samples) + K k · [Z k - H k · X ^ k-1 (N_sub_samples)] (11)

The update of the error covariance matrix with the results of a measurement is carried out according to P k (0) = [I - K k H k ] · P k-1 (N_sub_samples) (12) where Pk-1 (N_sub_samples) denotes the Prädiktionsfehlerkovarianzmatrix, where H _k describes how the current measurement is received in the system state, and where K _k is the Kalman gain matrix is determined, the weighting of the measurement to the prediction and given by K k = P k-1 (N_sub_samples) · H T k ·[H k · P k-1 (N_sub_samples) · H T k + R k ] -1 (13)

These Both updates form the basis for the re - run to estimate the next System condition, d. H. the algorithm returns to step S10 and a new one Main step can begin.

Claims (4)

Method for estimating the temperature in the intake manifold of an internal combustion engine, characterized in that an estimated value for the temperature in the intake manifold is determined by means of a Kalman filter. Method according to claim 1, characterized in that that a reading for the temperature in the intake manifold is determined by a temperature sensor and during a stationary State of the estimate for the Temperature in the intake manifold is corrected on the basis of this measured value. A method according to claim 1 or 2, characterized in that a state vector estimated by means of the Kalman filter includes at least one of the following values: intake manifold temperature reading (T _{MAN_MEAS} (k)), temperature of recirculated exhaust gas entering the intake manifold taking into account the mass transport Time delay (T _{EGR_DLY} (k)) and intake manifold _temperature (T _{MAN_COR} (k)). Control device for a motor vehicle, characterized characterized in that it comprises at least one estimator, which is suitable for carrying out a method according to one of the preceding claims.