The present invention relates to a method for operating an internal combustion engine.
Injection systems having injectors for injecting fuel are generally believed to be understood. It is also believed to be understood that, depending on the type of injector, the injection period changes for a certain control period according to a not exactly predeterminable service life drift.
It is believed that this change in the injection period results in disadvantages during the operation of the internal combustion engine. This means, in particular, that the pollutant emissions increase due to the imprecise injection period. To compensate for this change in the injection period, methods are known for the so-called zero quantity calibration. In the case of a zero quantity calibration, a control period is usually ascertained during which a defined rotary motion of the crankshaft occurs as a result of the fuel injected during the control period. The control period mentioned above is then used to compensate for the service life drift of every type of injector.
It is also believed to be understood that during a predefined control period of the injector, the drivetrain as well as the engaged gear, i.e., the transmission ratio of the drivetrain, has an influence on the rotary motion of the crankshaft. For each gear there usually exists a characteristic curve which links the control period to the rotary motion resulting from the control period. For this reason, it is necessary for each individual drivetrain to have appropriate data input into the control unit of the internal combustion engine in order to carry out a zero quantity calibration.
The object underlying the present invention is achieved by a method according to the description herein. Advantageous refinements are specified in the further descriptions herein. Features which are important for the present invention are furthermore specified in the following description and in the drawings; the features may be important for the present invention both alone and in different combinations without explicit reference being made thereto again.
By ascertaining an operating point-dependent control period of the injector as a function of a minimum control period during which the injector does not open just yet, the service life drift of the injector is advantageously compensated for. The minimum control period, which links a control period of the injector to a change in a rotary motion of the internal combustion engine, is advantageously ascertained from a function, the change in the rotary motion resulting from the control period.
A drivetrain-dependent input of data into the control unit is advantageously eliminated with the aid of the claimed method. Accordingly, it is not necessary to carry out tests with all drivetrain variants, design changes with regard to the components of a drivetrain may be implemented within a shorter period of time, and therefore all expenditures are eliminated which are necessary for a drivetrain-dependent data input for a zero quantity calibration.
In one advantageous specific embodiment of the method, the change in the rotary motion of the internal combustion engine is ascertained with the aid of an amplitude of a rotational speed change rate. A raw amplitude may be multiplicatively linked to the square of an averaged rotational speed of the internal combustion engine to form a compensated amplitude. In this way, it is advantageously achieved that a rotational-speed dependency of the raw amplitude is reduced. By multiplying by the square of the averaged rotational speed, the behavior of a centrifugal mass is emulated, the vehicle drivetrain being emulated as the centrifugal mass in a rough approximation. A precompensation of the raw amplitude according to the compensated amplitude thus results.
In one advantageous specific embodiment, the operating point-dependent control period is ascertained by additively linking the minimum control period to a multiplicative linkage of a setpoint injection quantity and a gradient. The operating point-dependent control period therefore advantageously represents a zero-quantity calibrated control period. Advantageously, the ascertainment of the operating point-dependent control period is thus easily ascertained in a zero-quantity-calibrating manner.
In one advantageous specific embodiment of the method, the function may be ascertained from value pairs with the aid of linear regression, each of the value pairs including the control period and the amplitude ascertained during the control period. A linear correlation between the control period and the amplitude is advantageously used to ascertain the minimum control period. Furthermore, the method advantageously minimizes the squared error after the linear regression.
In one advantageous specific embodiment of the method, the control period of one of the value pairs is greater than a starting control period. By using only those value pairs which have a control period which is greater than the starting control period, it is advantageously achieved that only those value pairs are incorporated into the computation of the function which are in an almost linear range of the function to be ascertained.
In one advantageous refinement of the method, multiple values are ascertained for the amplitude, the control period increases starting from a minimum control period, and the amplitude exceeds the starting amplitude during the starting control period. The starting control period thus ascertained shows the start of a range in which value pairs may be ascertained, each of which has an amplitude which has been reliably influenced by the metered fuel quantity.
In one advantageous refinement of the method, the starting amplitude is ascertained as a multiple, in particular a quadruple, of the standard deviation, the standard deviation being ascertained from multiple values for the amplitude. In this way, a limiting value, i.e., the starting amplitude, is advantageously established, this limiting value advantageously being related to the background noise generated by the drivetrain.
In one advantageous refinement of the method, the injector is controlled to ascertain the multiple values of the amplitude for such a short test control period that the injector reliably does not open. In this way, it is advantageously possible to ascertain a background noise or a corresponding performance figure with regard to the multiple values of the amplitude.
In one advantageous refinement of the method, the multiple values of the amplitude are ascertained in coasting mode. When ascertaining the multiple values in coasting mode, only the background noise of the drivetrain is advantageously ascertained, since injections reliably do not take place.
In one advantageous specific embodiment of the method, a defined number of value pairs is ascertained in each case for the ascertainment of the function in a state of the internal combustion engine in which a fixed transmission ratio of the power train is established. Since different functions would form for different gears or transmission ratios, it is prevented in this way that an incorrect minimum control period is ascertained.
In one advantageous specific embodiment of the method, the value pairs are ascertained for the ascertainment of the function in a state of the internal combustion engine in which the internal combustion engine is operated in a certain range of the rotational speed. In this way, it may be prevented that value pairs, which may occur in a disadvantageous range of the rotational speed, are incorporated in the computation of the minimum control period.
Additional features, possible applications, and advantages of the present invention are derived from the following description of exemplary embodiments of the present invention, which are illustrated in the figures of the drawing. All features described or illustrated represent the object of the present invention alone or in any arbitrary combination, regardless of their recapitulation in the patent claims or their back-references, and regardless of their wording in the description or illustration in the drawing. The same reference numerals are used for functionally equivalent variables in all figures, even in different specific embodiments.
Exemplary specific embodiments of the present invention are explained below with reference to the drawings.
In
If inlet valve 20 is open, piston 18 takes in air from an intake manifold 28. Subsequently, the fuel is metered directly into combustion chamber 16 via an injector 30. The fuel auto-ignites when compressed. If outlet valve 22 is open, the combusted residual gases are expelled from combustion chamber 16 into exhaust gas system 14.
The control of internal combustion engine 12 takes place via a control unit 42 which, for example, processes signals of a rotational speed sensor 46, which cooperates with a sensor wheel 47, and of a driver input sensor 48. Rotational speed sensor 46 ascertains an angular position α, which is transmitted to control unit 42. Moreover, the signals of exhaust gas sensor 50 and the signals of other, not illustrated, sensors regarding pressures and/or temperatures in the area of internal combustion engine 12 or of exhaust gas system 14 may be supplied to control unit 42. With the aid of these and, if necessary, other input signals, control unit 42 forms control signals using which internal combustion engine 12 may be operated according to the driver input and/or according to preprogrammed requirements. In
The method described in the following is not limited to diesel internal combustion engines, but may also be used with gasoline internal combustion engines having an intake manifold injection or a direct injection, a spark plug being assigned to the combustion chamber in this case. Injector 30 may in this case be configured as a magnetic injector or as a piezoelectric injector, for example.
To control injector 30, control unit 42 applies a digital signal, which determines the time period of the control of the injector, to an output stage component (not shown). According to the digital signal, the output stage component generates a control variable, the control variable being a voltage U or a current I. With the aid of the control variable, an actuator of injector 30 is controlled by the output stage component for the generation of a fuel injection. The behavior of the injector is reflected in the control variable and the opening point in time and the closing point in time of injector 30 may be determined, for example. The control variable is measured by control unit 42. A control period AD (not shown in
A subarea of internal combustion engine 12 and a subarea of control unit 42 are shown in
Rotational speed signal n(t) which is ascertained by rotational speed sensor 46 is supplied to a block 64 of control unit 42. Alternatively to rotational speed signal n(t), a segment time signal, which indicates the time between the segments of sensor wheel 47, may also be supplied to block 64. The value of rotational speed signal n(t) is in this case reciprocal to the value of the segment time.
Block 64 generates a rotational speed change rate Δn(t)/n which normalizes rotational speed signal n(t) to an averaged rotational speed n. Averaged rotational speed n is ascertained from rotational speed signal n(t) with the aid of a mean value operation, for example. Alternatively to rotational speed change rate Δn(t)/n, another segment time signal may accordingly also be used. The effect of injection quantity q is reflected, among other things, in rotational speed change rate Δn(t)/n. A raw amplitude A, which is assigned to injection quantity q and test control period ADtest, may accordingly be ascertained from rotational speed change rate Δn(t)/n with the aid of a block 66. The ascertainment of raw amplitude A is explained in greater detail with reference to
In a linkage 68, raw amplitude A is multiplied by a factor k, the product of this multiplication being a compensated amplitude A*. Factor k corresponds to the square of averaged rotational speed n, for example. Amplitude A* is supplied to block 60. Furthermore, a setpoint injection quantity qsetpoint is supplied to block 60. Block 60 ascertains from setpoint injection quantity qsetpoint, among other things, an operating point-dependent control period AD*, which is ascertained as a function of an operating point of internal combustion engine 12, i.e., for example, as a function of the driver input, and essentially compensates for the service life drift of injection 30. Setpoint injection quantity qsetpoint may, for example, be ascertained from a setpoint value for compensated amplitude A* or from raw amplitude A. Alternatively to compensated amplitude A*, raw amplitude A may also be supplied to block 60.
Compensated amplitude A* is referred to in the following as amplitude A*.
A gradient ΔAD/Δq is stored as a constant, for example, and is derived from an engine characteristic map of control durations and injection quantities which is specific to the injector type. Gradient ΔAD/Δq is determined here from an almost linear range of the aforementioned engine characteristic map, the almost linear range being in the range of the control period during which the injector reliably opens. Gradient ΔAD/Δq is ascertained from a control period section dAD and an associated fuel quantity section dq. In linkage 76, gradient ΔAD/Δq is multiplied by setpoint injection quantity qsetpoint, the product of this multiplication being a differential control period ADΔ.
In linkage 72, minimum control period AD** and differential control period ADΔ are additively linked to form operating point-dependent control period AD*. Operating point-dependent control period AD* is thus ascertained from an additive linkage of minimum control period AD** to a multiplicative linkage of setpoint injection quantity qsetpoint and gradient ΔAD/Δq.
Below minimum control period AD**, injector 30 does not open, fuel is not metered into combustion chamber 16, and the variation of control period AD accordingly does not have an influence on amplitude A*. Above minimum control period AD**, injector 30 is opened, injection quantity q is metered into combustion chamber 16, and increasing control period AD has an increasing influence on amplitude A*. Injection quantity q is approximately proportional to the opening time of injector 30, the opening time of injector 30 approximately being difference AD-AD**. Amplitude A* or raw amplitude A is approximately proportional to the generated torque oscillation on the crankshaft and thus almost proportional to associated injection quantity q. For a control period AD which is longer than minimum control period AD**, there is thus an almost linear correlation between amplitude A* and control period AD. Other correlations in
In the first step, starting control period ADstart is initially ascertained, among other things. For this purpose, a starting amplitude A*start is ascertained from multiple values for amplitude A*, control period AD being increased, e.g., starting form a minimum control period AD0. Amplitude A* exceeds starting amplitude A*start during starting control period ADstart. The exceedance of starting amplitude A*start means that the effects of test control periods ADtest are reliably measurable via starting control period ADstart. Starting control period ADstart is thus characterized in that injector 30 opens to the extent that a rotational speed oscillation is triggered which is differentiable from the background noise. According to amplitude A* of value pair M3, amplitude A* exceeds starting amplitude A*start and thereby establishes starting control period ADstart.
Starting amplitude A*start may be a multiple of the standard deviation of the ascertained multiple values for amplitude A*, for example. In particular, starting amplitude A*start is a quadruple of the standard deviation of the multiple values of amplitude A*. The multiple values for amplitude A* correspond, together with control period AD, to value pairs M2 of
In the first step, ending control period ADend is determined in such a way that the value range between starting control period ADstart and ending control period ADend is long enough to allow function f to be determined, and that ending control period ADend is short enough to prevent interfering noises of the internal combustion engine. Step width ΔAD is used to establish the distance between individual value pairs M1 or control periods AD of value pairs M1, so that value pairs M1 do not accumulate in a range of control period AD.
In the second step, block 84 now generates test control periods ADtest between starting control period ADstart and ending control period ADend and assigns an ascertained supplied amplitude A* to respective test control period ADtest, value pairs M1 being formed. A defined number of value pairs M1 is ascertained. Value pairs M1 are ascertained by block 84 in a first state of internal combustion engine 12, in the first state a transmission ratio of a power train of internal combustion engine 12 being established. A certain engaged gear corresponds to a fixed transmission ratio of the power train of the internal combustion engine. Value pairs M1 may also be ascertained in a second state of internal combustion engine 12, in the second state internal combustion engine 12 being operated in a certain range of the rotational speed. The first and the second states of internal combustion engine 12 may also take place together. Furthermore, other states and/or conditions are also possible which may determine the ascertainment of value pairs M1. The goal is to ascertain a series, i.e., a defined number of value pairs M1 in a gear or rotational speed range, whereby a straight line may be approximated. Multiple such series may be used to approximate a joint straight line.
Test control periods ADtest for ascertaining value pairs M1 may be changed incrementally by a step width ΔAD per step. Furthermore, test control periods ADtest for ascertaining value pairs M1 may be alternatingly increased and reduced by block 84 across the range between starting control period ADstart and ending control period ADend, it being additionally possible to use a different gear stage each time for such a sequence. The incremental change in or the alternating increase and reduction of test control period ADtest is suitable to achieve a uniform distribution of value pairs M1 with regard to the time and/or with regard to the rotational speed, and/or with regard to the range between staring control period ADstart and ending control period ADend.
Block 80 carries out the linear regression with the aid of at least two value pairs M1 to ascertain a straight line according to function f from the at least two value pairs M1. Function f thus forms the control behavior of one single type of injector 30. By updating or recalculating function f, a drift may be detected in the control behavior of injector 30 over its service life. The ascertainment of minimum control period AD** is ascertained in block 82 according to a point where function f and the AD coordinate axis intersect.
The above-described methods may be carried out as a computer program for a digital arithmetic unit. The digital arithmetic unit is suitable to carry out the above-described methods as a computer program. The internal combustion engine, in particular for a motor vehicle, includes a control unit which includes the digital arithmetic unit, in particular a microprocessor. The control unit includes a storage medium on which the computer program is stored.
Number | Date | Country | Kind |
---|---|---|---|
10 2010 063 097.7 | Dec 2010 | DE | national |
Filing Document | Filing Date | Country | Kind | 371c Date |
---|---|---|---|---|
PCT/EP11/71269 | 11/29/2011 | WO | 00 | 8/27/2013 |