Control method for an internal combustion engine

Abstract

Control method for an internal combustion engine with several cylinders, wherein cylinder-individual manipulated variables (z0, z1, z2, z3) are determined, wherein a manipulated variable limitation interval (d1, d2) is used to limit the manipulated variables (z0, z1, z2, z3), characterized in that an average value (M) of the manipulated variables (z0, z1, z2, z3) is determined, and that the manipulated variable limitation interval (d1, d2) is determined as a function of the average value (M).

Description

[0001] The invention relates to a control method for an internal combustion engine, in particular a method for cylinder pressure control, having the features of the preamble of patent claim 1.

[0002] One option for cylinder pressure control is to measure the pressure in the combustion chamber of a so-called lead cylinder using a cylinder pressure sensor and, based on the current pressure in the lead cylinder, determine the torque provided by the lead cylinder. This provided torque can also be referred to as cylinder torque. The cylinder torques of the other cylinders can be determined by evaluating the speed signal, including correction by the lead cylinder.

[0003] From the generic document DE 10 2010 051 370 A1, a control method for an internal combustion engine with multiple cylinders is known. The multiple cylinders have at least one lead cylinder and at least one further cylinder. The lead cylinder can also be referred to as a guide cylinder. Speed information, in particular angular velocities of a crankshaft of the internal combustion engine at various crankshaft angles, is recorded. A respective first torque is determined for each cylinder from the speed information. Furthermore, an internal cylinder pressure in the lead cylinder is recorded using a pressure sensor. A second torque for the lead cylinder is determined from the recorded internal cylinder pressure of the lead cylinder. An indicated torque is then determined for each cylinder based on the respective first and second torque.A torque difference is determined between the first torque of the lead cylinder and the first torque of the other cylinder. The indicated torques of the other cylinders are determined by correcting the respective second torque using the torque difference. A target fuel injection quantity is set for each cylinder depending on the respective indicated torque of the respective cylinder. For each cylinder, a manipulated variable is generated from the difference between the target fuel injection quantity and the actual fuel injection quantity. These manipulated variables are then limited to a manipulated variable limitation interval.

[0004] However, this generic method allows for relatively small control variables for the controllers to equalize the operating point, i.e., to equalize the torque to compensate for cylinder-specific deviations. Deviations can occur, for example, due to drift effects, e.g., injector drift over the operating time of the combustion engine.

[0005] DE 101 63 894 A1 discloses an internal combustion engine with direct injection. In order to compensate for tolerance differences during the production of injectors for a fuel injection system, the injectors of the individual cylinders are equalized by means of injector stroke and / or injection adjustment. One goal is to achieve the pilot control values of the application. The injection quantity of the individual injectors is varied in homogeneous and / or stratified operation until a target fuel quantity is set for all injectors. This achieves a nominal equalization of the fuel flow for all injectors. At an operating point, an injected fuel quantity is determined as an actual fuel quantity in each cylinder of the internal combustion engine.The actual fuel quantity is then compared with a target fuel quantity stored in the control device for the respective operating point, with an average being calculated from the actual fuel quantities of the individual cylinders. The injector is controlled by a control device in such a way that the injected fuel quantity corresponds to the average value determined from the actual fuel quantities. A difference value between the average value and a target fuel quantity stored in the control device for the respective operating point is also determined. The injected fuel quantity in each cylinder is then changed by the difference value in such a way that a second average value determined from the injected fuel quantities corresponds to the target fuel quantity. The injected fuel quantities are adjusted until an average value is set from the injected fuel quantities which corresponds to this second average value.The process compensates for tolerance differences that occur during the production of injectors for a power injection system.

[0006] DE 10 2008 042 104 A1 discloses a method for controlling an internal combustion engine, in which at least one quantity injected in at least one cylinder is corrected. The method uses a cylinder-specific correction quantity to compensate for irregularities in the speed curve due to different cylinder torques. Depending on whether the cylinder in question contributes a torque component that is above or below the average value, the correction quantity for this cylinder can assume positive or negative values. The correction value is not fully allocated to an individual injection in order to avoid an unintentional interruption of this injection in the case of negative correction values. An interruption threatens if the injector's shutdown threshold, i.e. the minimum quantity including an injection-type-dependent pressure wave correction, is not reached.The process aims to achieve the maximum possible negative correction quantity; the smallest injection quantity corresponds to the minimum quantity when the correction quantity is subtracted. The correction simultaneously reaches the minimum threshold during an injection, thus providing the maximum correction range.

[0007] DE 10 2009 008 246 B3 discloses a method for predicting pre-ignition of a fuel-air mixture in a cylinder chamber of a gasoline-powered internal combustion engine. The combustion characteristics of each cylinder chamber are measured, and any pre-ignition is predetermined based on a comparison with the measured values of a previous engine cycle. These measures allow the occurrence of pre-ignition to be predicted before it occurs, allowing timely measures to modify the combustion process in the detected cylinder chamber to be initiated via an engine control unit.

[0008] DE 10 2007 023 553 B3 relates to a torque-based method for controlling the operation of an internal combustion engine. A target torque of the internal combustion engine is reduced in such a way that operational reliability of the internal combustion engine is ensured even if errors are diagnosed during monitoring of the internal combustion engine's operating parameters.

[0009] DE 102 29 238 A1 discloses a method and device for controlling the position of an actuating element. The actuating element is designed, for example, as a throttle valve, a charge movement flap, and / or an exhaust gas recirculation valve of a vehicle. The distance of the actuating element from a stop, in particular a mechanical stop, is monitored. The movement of the actuating element is limited in such a way that damage to the actuating element, in particular due to a collision with the stop, is avoided.

[0010] The invention is based on the object of improving the generic control method. In particular, a method should be provided in which the necessary manipulated variable limitations for correctly determining the cylinder torques are maintained while still taking global drift effects into account.

[0011] This problem underlying the invention is now solved by adding the mean values of the cylinder-specific manipulated variables of the torque control to the respective manipulated variable limits, both positively and negatively. This defines new manipulated variable limits that allow for greater correction. This has the advantage of significantly improving the performance of the connected torque control without affecting the stability of the control.

[0012] This problem underlying the invention is now solved by a control method having the features of patent claim 1.

[0013] An average value of the manipulated variables is determined, with the manipulated variable limitation interval being determined based on the average value. The manipulated variable limitation interval is determined periodically based on the average value. This defines larger manipulated variable limitations, which allow for greater correction; that is, the manipulated variable limitation interval can be selected to be larger than in the prior art. The cylinder-specific manipulated variables are limited to the manipulated variable limitation interval.

[0014] The upper limit of the manipulated variable limitation interval is equal to the sum of the mean value and half the width of a specified dispersion interval. The width of the dispersion interval is specified. The lower limit of the manipulated variable limitation interval is calculated as the difference between the mean value and half the width of the dispersion interval.

[0015] Preferably, a torque difference between the first torque of the lead cylinder and the first torque of the further cylinder is determined, wherein the indicated torque of the further cylinder is determined by correcting the second torque with the aid of the torque difference.

[0016] In order to ensure that the manipulated variables cannot assume too extreme values in the event of an excessive mean value shift, further safety mechanisms, i.e. additional limitations, are preferably incorporated into the process.

[0017] The mean value is limited by a mean value limit interval. In a preferred embodiment, the determined mean value of the manipulated variables is compared with a mean value limit interval. If the mean value exceeds a limit of the mean value limit interval, the manipulated variable limit interval is no longer determined based on the actual mean value; instead, the limit of the mean value limit interval is used to determine the manipulated variable limit interval. This prevents excessive mean value shifting.

[0018] The manipulated variable limitation interval is further limited by system boundaries. Preferably, the determined limits of the manipulated variable limitation interval are compared with system boundaries. The system boundaries consist, in particular, of a system upper limit and a system lower limit, which are not variable over time but are formed by absolute values. If the lower or upper limit of the manipulated variable limitation interval would actually exceed one of the system boundaries, the limit of the manipulated variable limitation interval is actually set to the system boundary. This allows the dispersion interval to be defined sufficiently large, while also avoiding excessively extreme manipulated variables.

[0019] These designs make it possible to allow a wide control variable limitation interval, i.e. a wide scatter interval, and at the same time to counteract possible emission and driving behavior abnormalities.

[0020] A control method is used for an internal combustion engine with multiple cylinders. In a preferred embodiment, the multiple cylinders have at least one lead cylinder and at least one further cylinder. Speed information of a crankshaft of the internal combustion engine is detected at different crank angles, wherein a cylinder pressure in the lead cylinder is detected. A respective first torque is determined for each cylinder from the speed information. A second torque for the lead cylinder is determined from the internal cylinder pressure of the lead cylinder, wherein an indicated torque for each cylinder is determined on the basis of the respective first and second torques. A target fuel injection quantity is set for each cylinder as a function of the indicated torque of the respective cylinder, wherein a corresponding manipulated variable is formed for each cylinder.A difference between a target fuel injection quantity and an actual fuel injection quantity can be formed as a control variable.

[0021] The disadvantages mentioned above are therefore avoided and corresponding advantages are achieved.

[0022] There are now numerous possibilities for advantageously designing and developing the method according to the invention. For this purpose, reference is first made to the claims subordinate to claim 1. A preferred embodiment of the invention will be explained in more detail below with reference to the drawing and the accompanying description. The drawing shows: Fig. 1 shows a schematic flow diagram of part of a cylinder pressure control system, in which a manipulated variable limitation is carried out, and Fig. 2 in one diagram manipulated variables and various manipulated variable limits.

[0023] In Fig. 1 shows a controller 1 which supplies cylinder-specific manipulated variables z0, z1, z2, z3 for an internal combustion engine with four cylinders. The controller 1 can be designed as a PI controller. It is conceivable that the internal combustion engine has more or fewer than four cylinders. The manipulated variables z0, z1, z2, z3 can be formed in particular by comparing a target fuel injection quantity and an actual fuel injection quantity. A system limitation 2 then supplies system limits a1, a2, namely a corresponding upper system limit a2 and a lower system limit a1. These are used in a step 3 to limit the cylinder-specific manipulated variables z0, z1, z2, z3. In a method step 4, an average value M of the manipulated variables z0, z1, z2, z3 is formed. An offset limitation 5 provides limits b1, b2 of a mean value limitation interval, wherein in a method step 6 the calculated mean value M is limited to this mean value limitation interval b1, b2.This limited mean value M is then added in a method step 7 to a variation interval c1, c2 that is provided by a variation limiter 8. The result is a manipulated variable limiting interval d1, d2, which is used in method step 9 to further limit the manipulated variables z0, z1, z2, z3 that may already be limited by the system limit. In a method step 11, a setpoint structure 10 provides a feedforward control that is added to the manipulated variables z0, z1, z2, z3 limited by the manipulated variable limiting interval d1, d2. The cylinder-specific manipulated variables z0, z1, z2, z3 limited in this way are then used to further control the injectors of the internal combustion engine 12.

[0024] The effect of this control method on the cylinder-individual control variables z0, z1, z2, z3 can now be illustrated again using the example of Fig. 2 are described in more detail: In Fig.2. Four manipulated variables z0, z1, z2, and z3 are plotted. A system boundary interval a1, a2 is provided by system boundary 2. In the illustrated embodiment, a2 = -a1. The manipulated variables z0, z1, z2, and z3 describe the change in the injected fuel quantity in mg per stroke per main injection. The system boundaries here are centered around the value 0; however, it is also conceivable to use system boundaries not centered around the value 0 (a2 is not equal to -a1).

[0025] Furthermore, offset limits and mean limits d1 and d2 are shown. The mean limits d1 and d2 are also centered around the value 0, i.e., b2 = -b1. However, it is also conceivable to use mean limits not centered around the value 0 (b2 is not equal to -b1).

[0026] Furthermore, a scatter interval c1 and c2 is shown, where again c1 = - c1 in the case shown. However, it is also conceivable that a scatter interval not centered around the value 0 (c2 is not equal to -c1) is used.

[0027] The method according to the invention is implemented starting at time t0. Before time t0, the previously used control value limitation should be briefly explained: Until now, the manipulated variables z0, z1, z2, and z3 have been limited only by the dispersion limits c1 and c2. In the illustrated embodiment, the manipulated variable z0 is at the value c2 of the upper dispersion limit until time t1. Until time t0, the manipulated variable limitation interval is formed by the dispersion limits c1 and c2. All manipulated variables z0, z1, z2, and z3 lie within the manipulated variable limitation interval c1 and c2.

[0028] From time t0 onwards, a mean value shift is permitted. For this purpose, an arithmetic mean is calculated from the manipulated variables z0, z1, z2 and z3, which is described here by the curve M. This mean value M can also be referred to as the global offset. From time t0 onwards, the manipulated variable limitation interval is described by the limits d1 and d2. The limits d1, d2 of the manipulated variable limitation interval are now calculated by shifting the scatter limits c1, c2 by the mean value M. The mean value M is added to the scatter limits c1 and c2. In this case, the upper limit d2 of the manipulated variable limitation interval can be represented in general form by the formula M + (c2 - c1) / 2. The lower limit can be represented by the formula d2 = M - (c2 - c1) / 2.From time t1, however, the system limitation takes effect, so that the lower limit d1 of the manipulated variable limitation interval is set to the lower system limit a1, since otherwise the value calculated according to the formula would fall below this limit a1. From time t2, the offset limitation 5, namely the corresponding mean value limitation 5, takes effect. The value M is set to the lower mean value limit b1 until time t3, since otherwise the calculated mean value would fall below this limit b1. From time t3, the mean value M rises again above the limit b1, whereby the lower system limit a1 still forms the lower limit of the manipulated variable limitation interval d1 until time t4.

[0029] This limitation makes it possible to take global drift effects into account while simultaneously increasing the maximum permissible control variables for the operating point equalization controllers. This reduces the risk of abnormal driving behavior and emissions impacts. LIST OF REFERENCE SYMBOLS 1 controller 2 System limitations 3 Process step: Limiting the manipulated variables by system boundaries 4 Process step: Calculating the mean value of the manipulated variables 5 Mean value limitation or offset limitation 6 Process step: Limiting the mean value 7 Step: Adding the limited mean and a scatter interval 8 Scattering limitation 9 Process step: Limiting the manipulated variables already limited by the system boundaries 10 Setpoint structure 11 Process step: Adding a feedforward control 12 internal combustion engine a1, a2 system boundary b1, b2 mean limit c1, c2 scatter limit M mean d1, d2 Limits of the manipulated variable limitation interval t0, t1, t2, t3, t4 time points z0, z1, z2, z3 cylinder-specific control variables

Claims

[1] Control method for an internal combustion engine with several cylinders, wherein cylinder-individual manipulated variables (z0, z1, z2, z3) are determined, wherein a manipulated variable limitation interval (d1, d2) is used to limit the manipulated variables (z0, z1, z2, z3), characterized by that a mean value (M) of the manipulated variables (z0, z1, z2, z3) is determined, and that the manipulated variable limitation interval (d1, d2) is determined as a function of the mean value (M). [2] Control method according to claim 1, characterized by that an upper limit of the manipulated variable limitation interval (d2) is equal to the sum of the mean value (M) and half the width of a scatter interval (c1, c2), wherein the lower limit value (d1) of the manipulated variable limitation interval (d1, d2) is formed by the difference between the mean value (M) and half the width of the scatter interval (c1, c2). [3] Control method according to one of the preceding claims, characterized bythat the mean value (M) is limited by a mean value limiting interval (b1, b2). [4] Control method according to one of the preceding claims, characterized by that the manipulated variable limitation interval (d1, d2) is limited by system boundaries (a1, a2). [5] Control method for an internal combustion engine according to one of the preceding claims with several cylinders, characterized byin that the plurality of cylinders have at least one lead cylinder and at least one further cylinder, wherein rotational speed information of a crankshaft of the internal combustion engine is detected at different crank angles, wherein a cylinder pressure in the lead cylinder is detected, wherein a respective first torque for each cylinder is determined from the rotational speed information, wherein a second torque for the lead cylinder is determined from the internal cylinder pressure of the lead cylinder, wherein an indicated torque for each cylinder is determined on the basis of the respective first and second torques, wherein a desired fuel injection quantity for each cylinder is set as a function of the indicated torque of the respective cylinder, wherein a manipulated variable (z0, z1, z2, z3) is formed for each cylinder. [6] Control method according to the preceding claim, characterized bythat a torque difference is determined between the first torque of the lead cylinder and the first torque of the further cylinder, wherein the indicated torque of the further cylinder is determined by correcting the second torque with the aid of the torque difference. [7] Control method according to one of the preceding claims, characterized by that a difference between a target fuel injection quantity and an actual fuel injection quantity is formed as the manipulated variable (z0, z1, z2, z3).

Images