Method and device for monitoring a high-pressure injection system, in particular a self-igniting internal combustion engine of a motor vehicle

Abstract

Method for monitoring a high-pressure injection system, in particular a self-igniting internal combustion engine of a motor vehicle, wherein a reaction chain (100 - 125) is provided for detecting and treating overpressure situations, wherein pressure values detected for detecting an overpressure situation are debounced (110) and wherein, in the event of an overpressure situation being detected, an error reaction of at least one actuator of the high-pressure injection system is generated, characterized in that a preliminary error reaction is generated (115) at least partially overlapping in time with the debouncing (110), that the preliminary error reaction (115) is maintained if the presence of an overpressure situation is confirmed after debouncing (110) has taken place, and that the preliminary error reaction (115) is canceled again if the presence of an overpressure situation is not confirmed after debouncing (110) has taken place.

Description

[0001] The invention relates to a method and a device for monitoring a high-pressure injection system, in particular a self-igniting internal combustion engine of a motor vehicle according to the preambles of the respective independent claims. Furthermore, the invention relates to a computer program that executes all steps of the method according to the invention when it runs on a computing device or a control device for controlling or monitoring such a high-pressure injection system. Finally, the present invention relates to a computer program product with program code stored on a machine-readable medium for carrying out the method according to the invention when the program is executed on a computing device or a control device for controlling or monitoring such a high-pressure injection system. State of the art

[0002] A method and device for monitoring a common rail injection system are disclosed in the previously published DE 196 26 689 C1. The injection system has a pressure relief valve (PRV) arranged in a high-pressure circuit, by means of which excessive fuel pressure in the high-pressure circuit or the rail is prevented by diverting fuel into a return line of the injection system when the PRV reaches its opening point. The PRV acts purely mechanically and passively and has neither an electrical control nor a sensor that detects, for example, the rail pressure. However, the pressure in the high-pressure area is regulated by means of a controllable pressure generator, typically a high-pressure pump with a metering unit (ZME). If the rail pressure exceeds a critical value, the PRV prevents system overload by opening mechanically. The system is subsequently transferred to an emergency mode.

[0003] A disadvantage of using systems with a pressure relief valve is that they have to discharge fuel through a permanent leak (typically an injector leak) in order to reduce the pressure again during normal operation without active injection, which is not very efficient. To increase efficiency, newer generation high pressure systems therefore use components without a pressure relief valve, with the pressure reduction functionality during normal operation being carried out by means of an (active) pressure control valve (PRV). On the other hand, the pressure limitation functionality of the PRV is also integrated into the PRV. In overpressure situations, the PRV is vented as quickly as possible, and therefore the venting of the DRV in safety-critical overpressure situations is particularly time-critical for system safety.

[0004] Despite the increased demands on response time, the quality and robustness of overpressure detection must also be ensured. For example, it is known to initiate a shutdown of the high-pressure system only after several critical rail pressure values have been detected by means of plausibility-based debouncing. However, such debouncing disadvantageously extends the response time for reducing the overpressure.

[0005] DE 103 29 073 B4 describes a high-pressure fuel supply device for internal combustion engines. A fuel pump with a pressure chamber draws in fuel and compresses it synchronously with the engine phase. An electromagnetic valve, controlled by a control unit, regulates the fuel flow to the injection system. The control unit adjusts the valve opening times to the engine rotation phase. If this is unknown, it controls the valve cyclically. A voltage sensor is new: At low voltage, the control unit extends the duty cycle control cycle and thus the current supply to the valve.

[0006] From DE 10 2005 014 161 B4 a method for determining the fuel pressure values of a high-pressure fuel system of an internal combustion engine is known.

[0007] The internal combustion engine has a fuel pump for pumping fuel into the fuel pressure accumulator, injection valves for introducing fuel from the fuel pressure accumulator into the cylinders, and a fuel pressure sensor associated with the fuel pressure accumulator. The measured values determined by the fuel pressure sensor are adapted to transient operating conditions of the internal combustion engine by calculating fuel pressure values by summing an average of the measured values from the fuel pressure sensor and a factor dependent on the temporal change in the measured values from the fuel pressure sensor.

[0008] DE 10 2008 001 444 A1 discloses a method for determining an overpressure in a fuel reservoir of an injection system of an internal combustion engine, in particular in a common rail of a common rail system. The pressure in the fuel reservoir is detected. An overpressure in the fuel reservoir is determined when the derivative of the detected pressure with respect to time exceeds a predetermined gradient threshold and the detected pressure subsequently exceeds a predetermined pressure threshold. Disclosure of the invention

[0009] The invention is based on the idea of providing debouncing with a preventive shutdown of the high-pressure system in a reaction path (or reaction chain) underlying the monitoring of the injection system. The preventive shutdown occurs at the beginning of a debouncing phase and is only maintained if the presence of an overpressure situation is actually confirmed during or after the debouncing phase. The preventive shutdown corresponds to a preliminary error reaction, which serves to counteract a possible overpressure situation in the high-pressure system. However, if no overpressure situation exists after the debouncing phase has expired, the preventive shutdown is canceled again.

[0010] The aforementioned procedure therefore enables debouncing without, however, extending the overall reaction time of the aforementioned reaction path due to the debouncing. The preventive shutdown according to the invention therefore enables a significant improvement in the reaction speed of the safety reaction when monitoring an overpressure situation in an injection system affected here, and thus also the pressure-limiting capacity and operational reliability of the high-pressure circuit, without, however, compromising the required robustness of detecting a high-pressure situation.

[0011] In order to achieve the greatest possible time advantage in reaction speed, the preventive shutdown or the preliminary error reaction can, if possible, take place at the beginning of the debouncing or a corresponding debouncing phase.

[0012] According to a preferred embodiment, a debouncing threshold and a shutdown threshold are specified. The at least one actuator is transferred to a transition state upon reaching or exceeding the debouncing threshold. After debouncing, a check is performed to determine whether the shutdown threshold has also been exceeded at this time. In this case, the presence of a permanent overpressure situation is assumed, and preventive shutdown is therefore maintained. This approach ensures, in particular, robustness in detecting an overpressure situation.

[0013] In the case where the fault reaction is triggered by an actuator of the injection system, e.g., a DRV, the preventive shutdown corresponds to a transitional state of the actuator, with the transitional state preferably corresponding to a shutdown state. By implementing the transitional state as a shutdown state, the required operational reliability and monitoring reliability of the proposed method is ensured.

[0014] The invention can be used both in high-pressure injection systems of self-igniting internal combustion engines in motor vehicles and in corresponding internal combustion engines used in industrial plants, with the advantages described herein. It can also be used in any type of time-critical reaction chain in which debouncing processes are combined with dead-time elements, with the advantages described herein. Instead of an actuator subject to dead-time, such as a DRV described, the method can be applied accordingly to a quantity control of a high-pressure pump of an injection system concerned here, the so-called "metering unit" (ZME), which is also subject to delay times. However, such a ZME has an opposite direction of action, so that a fault reaction, in contrast to the DRV, requires an increase of the ZME control current to a maximum value instead of shutdown.In addition, the method can also be used for monitoring functions of other actuators such as turbochargers, throttle valves, engine brakes or the like.

[0015] Further advantages and embodiments of the invention will become apparent from the description and the accompanying drawings. It is understood that the features mentioned above and those to be explained below can be used not only in the respective combinations specified, but also in other combinations or on their own, without departing from the scope of the present invention. Short description of the drawings Fig. 1 shows a serial reaction chain for monitoring a high-pressure injection system according to the state of the art. Fig. 2 shows a reaction chain parallelized by means of the preventive shutdown method according to the invention for monitoring a high-pressure injection system affected here. Fig. 3a-d show measurement curves for the rail pressure and the control current of a pressure control valve (DRV) at four different switch-off intervals of the DRV. Fig. 4a, b show individual rail pressure curves in the event of a fault and the associated control currents of a DRV, in comparison with a control according to the state of the art ( Fig. 4a) and the invention ( Fig. 4b). Fig. Figure 5 shows filtered or averaged rail pressure curves with and without preventive shutdown over several tests. Fig. 6 shows similar rail pressure curves and control currents as in the Fig. 4a and Fig. 4b, but in the case of a relatively slow rail pressure increase. Description of implementation examples

[0016] The Fig. 1 schematically shows, using a block diagram, a reaction chain (or reaction path) known in the prior art for detecting and treating overpressure situations in a common rail high-pressure accumulator of a self-igniting internal combustion engine of a motor vehicle.

[0017] In block 100, the current rail pressure is first recorded and then subjected to error classification in the subsequent block 105. The result of the error classification may be that the current rail pressure is above a predetermined threshold value, possibly indicating a permanent and therefore critical overpressure situation. In the subsequent block 110, this classification result is checked for plausibility; in this example, this is done by debouncing or plausibility checking by recording additional rail pressure values, in order to then assess whether the recorded overpressure situation was only brief, i.e., temporary, or permanent. Only a permanent overpressure situation can be assessed as system-critical.

[0018] If the debouncing (or plausibility check) 110 confirms that a permanent overpressure situation actually exists, an electrical error response is generated according to block 115 to reduce the overpressure. Block 120 indicates that there is usually a dead time between the generation of the electrical error response 115 and the actual hydraulic error response 125 executed by an actuator (e.g., a pressure control valve (DRV) mentioned above). This dead time is caused by a calculation time in the control unit or by electrically, hydraulically, or mechanically induced actuator delay.

[0019] As from the Fig. 2, the invention takes advantage of said dead time in that the debouncing 110 is no longer carried out sequentially with the generation of the electrical error reaction 115 and the said dead time 120, but is carried out in parallel during the entire time interval from the generation of the electrical error reaction 115 to the end of the dead time 120. At the beginning of the debouncing phase 110, the said preventive shutdown of the actuator is carried out, ie it is assumed, at least temporarily, that the assumed overpressure situation is also confirmed by the debouncing 110.

[0020] By executing debouncing 110 and the electrical control of the dead-time-dependent DRV in parallel, the reaction chain for pressure limitation in the event of a fault is shortened by a maximum of the duration of debouncing phase 110, thus shortening the overall reaction time despite the debouncing. The reaction time is thus shortened without sacrificing the robustness-enhancing plausibility check of the measured rail pressure values in debouncing phase 110.

[0021] It should be noted that the described method can also be applied in cases where the debounce time is greater than the dead time. However, the reaction chain 100 - 125 must be modified so that the aforementioned preventive shutdown only begins at a predefined debounce level. It is also worth mentioning that with variable debounce / plausibility durations, it must be ensured that the preventive shutdown of the actuator can still be reversed before an unjustified, premature error response occurs. Depending on the actuator properties, it is also possible to switch to a cyclical ON / OFF mode in this situation, in which the hydraulic error response 125 does not yet occur, but an advantage in dead time bridging can still be achieved.

[0022] The following describes exemplary embodiments of the control behavior of such an actuator, using a DRV as an example. A suitable control behavior enables or even improves the aforementioned effect of preventive shutdown.

[0023] The Fig. The rail pressures measured at four different shutdown durations of a DRV, shown in Figures 3a to 3d, illustrate a reaction or dead time behavior of the DRV suitable for preventive shutdown. In the diagrams, the left ordinate represents the DRV's drive current in amperes, and the right ordinate represents the rail pressure in 10 6 hPa plotted over time in seconds. The control current of the DRV was achieved in a conventional manner using pulse width modulation (PWM).

[0024] During these measurements, the high-pressure circuit of the common rail injection system was set to a constant fuel pressure. As a result, the pump delivery and thus also the fuel flow were reduced, and any injector injection was prevented, thus preventing any fuel discharge. Since the DRV was subjected to a sufficiently large electrical holding current, the pressure-tight system in question was Fig. 3a, a constant rail pressure corresponding to the relatively smooth (i.e., low-oscillation) curve 300 occurs. The oscillation curve 305, also shown, represents the corresponding course of the control current of the DRV.

[0025] Subsequently, the drive current of the DRV was completely switched off in a series of short switch-off times (in this case 4, 6 and 8 ms). The resulting rail pressure curves are shown in the Fig. 3b, Fig. 3c and Fig. 3d with the reference symbols 300', 300'' and 300'''. Afterwards, the DRV was energized again with the initial value. As can be seen from the Fig. 3b, Fig. 3c and Fig. As can be seen in Figure 3d, the following two effects were found which are significant for the reaction and dead time behavior of the DRV: a) After the control current is switched off, the electrical current in the DRV decreases gradually due to the electrical energy stored in the magnetic circuit. This means that the holding force of the DRV also decreases only with a time delay after switching off. Only after longer shutdowns over a duration specific to the respective DRV is the reaction of the DRV noticeable not only in the current but also in the rail pressure, in this case starting from 6 ms ( Fig. 3c). As can be seen from the Fig. As can be seen in Figure 3d, the rail pressure gradually decreases with longer shutdown times of the respective shutdown duration. b) Even with a relatively brief power cut to the DRV, which has no effect on the rail pressure, the current build-up after the DRV is re-energized is also delayed and requires several PWM cycles to regain the initial level. If the DRV is permanently shut down during this current build-up time or the corresponding dead time after the DRV is re-energized, this shutdown occurs at a lower output level than the previous shutdown, but still has the reaction time advantage described above.

[0026] It should be noted that the Fig. The test results shown in Figures 3a to 3d were conducted on a PCVN2-25 DRV from Bosch GmbH. However, other actuator types exhibit similar response behavior and can therefore also be used for the preventive shutdown method described herein, with the advantages described herein.

[0027] The Fig. 4a and Fig. 4b show in the upper area rail pressure curves to illustrate a pressure limitation function or overpressure shutdown of the high-pressure circuit of an injection system concerned here, namely Fig. 4a with an overpressure cut-off known in the prior art and Fig. 4b when using a preventive shutdown as described above. The lower section shows the corresponding control currents of the DRV, both as "ON" and "OFF" values.

[0028] As can be seen from the Fig. 4a and Fig. As can be seen in Figure 4b, two pressure thresholds are defined for the rail pressure limitation function, namely a debounce threshold 400 and a shutdown threshold 405 located above the debounce threshold. The exact value of the debounce threshold 400 is selected such that the pressure values occurring during fault-free operation of the injection system do not exceed the debounce threshold 400. Any exceedance of the debounce threshold 400 is considered a malfunction. The aforementioned plausibility check (debouncing) of the error is performed by multiple recording or measuring of the rail pressure value, whereby an aforementioned error classification is performed if the debounce threshold 400 is exceeded continuously or again within a specified debounce time 415.The switch-off threshold 405 is determined on the basis of the available measuring range of the pressure sensor used and / or on the basis of system-related strength limits of the injection system, in particular of the high-pressure circuit or the high-pressure accumulator.

[0029] The Fig. The scenario shown in Figure 4a is based on the Fig. 1, i.e., it does not include the preventive shutdown method according to the invention. In the fault scenario assumed here, the rail pressure 410 rises relatively quickly, with both the debounce threshold 400 and the shutdown threshold 405 being exceeded well before the debounce time 415 expires. This means that it is necessary to wait until the debounce time 415 expires before a pressure-reducing reaction can be triggered. Since, in the present example, the shutdown threshold 405 is still exceeded at the end of the debounce time 415, the control current 420 is transitioned from the "ON" state 425 to the "OFF" state 430.

[0030] From the Fig. Figure 4a shows that, in a serial reaction chain, the robustness gained through debouncing directly results in a reaction delay. This disadvantage is remedied by the preventive shutdown method, as described below.

[0031] The Fig. The pressure cut-off scenario shown in Figure 4b is based on a Fig. 2 parallelized reaction chain according to the invention. As in Fig. 4a, a relatively rapid rail pressure increase 440 occurs, whereby both the debounce threshold 400 and the shutdown threshold 405 are exceeded in a very short time sequence. In contrast to Fig. 4a, however, the first time the debounce threshold 400 is reached, the drive current 445 of the DRV is transferred from the "ON" state 425 to a transition state 450 and marked as "P_OFF" 435. At the end of the debounce time 415, the shutdown threshold 405 is still exceeded in the present scenario, whereby the existing pressure curve is classified or verified as a permanent overpressure situation, and thus the temporary shutdown "P_OFF" 435 of the drive current 445, 450 transitions without interruption to the "OFF" state 430.

[0032] In contrast to the Fig. In the known pressure cutoff approach shown in Figure 4a, the control current of the DRV is cut off earlier by the amount of the debouncing time, without, however, having to forego debouncing itself. The earlier current cutoff shortens the response time in the event of a fault, so that the maximum rail pressure in the high-pressure system can advantageously be kept at a lower level.

[0033] This pressure relief effect is due to the Fig. 5, in which maximum filtered rail pressure values for the two scenarios are shown over several tests according to the Fig. 4a and Fig. 4b. The upper pressure curve 510 corresponds to the first scenario according to Fig. 4a and the lower pressure curve 515 according to the second scenario Fig. 4b. The mentioned reduction of the maximum pressure results in the pressure difference 520. In the lower part of the diagram in Fig. 5 shows the timely progression of the control current 500 of the DRV for the first scenario as well as the progression of the control current 505 according to the second scenario, including the early preventive shutdown.

[0034] As in the Fig. 4a, Fig. 4b and Fig. As shown in Figure 5, a critical overpressure situation is effectively prevented by the preventive shutdown and the transition to the permanent shutdown state of the DRV during relatively rapid rail pressure increases. However, there are situations in which the rail pressure may rise more slowly but can still reach a critical overpressure.

[0035] In the Fig. Figure 6 illustrates such a scenario in which the preventive shutdown is activated ("ON" state 425; "P_OFF" state 455), but at the end of the debounce time 440, not all conditions are met to transition to the permanently switched-off state 450 ("OFF" state). This is because, by the end of the debounce time 440, the shutdown threshold 405 has not yet been reached or exceeded, so that the preventive shutdown must be canceled again, i.e., the control current of the DRV is switched on again 430 ("ON" state 445). Otherwise, an unjustified drop in rail pressure occurs, since a permanent overpressure situation does not yet exist.

[0036] In the present scenario, the rail pressure rises steeply after the debounce time 440 has elapsed, thereby exceeding the shutdown threshold 405. In this case, a rail pressure above the shutdown threshold 405 is determined in a calculation point carried out again after the debounce time 440 has elapsed, whereby the control current is classified as permanently switched off (state "OFF" 450) and is therefore switched off immediately. The short restart phase 430 is not ideal. However, even in the present scenario, the preventive shutdown offers the time advantage described above compared to a serial reaction path according to Fig. 1.

[0037] The described method can be implemented either in the form of a control program in an existing control unit for controlling an injection system concerned here or in the form of a separate control unit for pressure monitoring of such an injection system.

Claims

[1] Method for monitoring a high-pressure injection system, in particular a self-igniting internal combustion engine of a motor vehicle, wherein a reaction chain (100 - 125) is provided for detecting and treating overpressure situations, wherein pressure values detected for detecting an overpressure situation are debounced (110) and wherein, in the event of detecting an overpressure situation, an error reaction of at least one actuator of the high-pressure injection system is generated, characterized by that a preliminary error reaction is generated (115) at least partially overlapping in time with the debouncing (110), that the preliminary error reaction (115) is maintained if the existence of an overpressure situation is confirmed after debouncing (110), and that the preliminary error reaction (115) is canceled again if the existence of an overpressure situation is not confirmed after debouncing (110). [2] Method according to claim 1, characterized by that the preliminary error reaction (115) occurs at the beginning of a debouncing phase (110). [3] Method according to claim 1 or 2, characterized by that a debouncing threshold (400) and a switch-off threshold (405) are specified, wherein the at least one actuator is transferred to a transition state (450) when the debouncing threshold (400) is reached or exceeded and, after debouncing (110) has taken place, it is checked whether the switch-off threshold (405) has been exceeded, wherein, in the event of the switch-off threshold (405) being exceeded, a permanent overpressure situation is detected and the switch-off from the transition state (450) of the actuator is maintained. [4] Method according to claim 3, characterized by that the debouncing threshold (400) and the switch-off threshold (405) are combined into a common threshold. [5] Method according to claim 3 or 4, characterized bythat the transition state (450) of the actuator corresponds to an at least temporary switch-off state. [6] Computer program which carries out all the steps of a method according to one of claims 1 to 5 when it runs on a computing device or a control device for controlling or monitoring the high-pressure injection system. [7] Computer program product with program code stored on a machine-readable carrier for carrying out the method according to one of claims 1 to 5, when the program is executed on a computer or a control device for controlling or monitoring the high-pressure injection system. [8] Control device for controlling or monitoring a high-pressure injection system, in particular a self-igniting internal combustion engine of a motor vehicle, which has a reaction chain (100 - 125) for detecting and treating overpressure situations, wherein pressure values detected for detecting an overpressure situation are debounced (110) and wherein, in the event of detecting an overpressure situation, an error reaction of at least one actuator of the high-pressure injection system is generated, characterized by Means by means of which a preliminary error reaction is generated (115) at least partially overlapping in time with the debouncing (110), wherein the preliminary error reaction (115) is maintained if the presence of an overpressure situation is confirmed after debouncing (110) has taken place, and wherein the preliminary error reaction (115) is canceled again if the presence of an overpressure situation is not confirmed after debouncing (110) has taken place.

Images