Method for operating an exhaust gas sensor, for example a broadband lambda probe

EP4705755A1Pending Publication Date: 2026-03-11ROBERT BOSCH GMBH
View PDF 0 Cites 0 Cited by

Patent Information

Authority / Receiving Office
EP · EP
Patent Type
Applications
Current Assignee / Owner
Filing Date
2024-03-27
Publication Date
2026-03-11

AI Technical Summary

Technical Problem

Existing methods for operating exhaust gas sensors, such as broadband lambda sensors, face suboptimal control behavior and measurement accuracy due to limitations in controller output, leading to issues like oscillation and toggling, especially during aging or inadequate heating conditions.

Method used

A method that adjusts the control gain of the pump current based on variables affecting the exhaust gas sensor, such as temperature and age, to improve the control loop's performance, allowing for adaptive control gain coefficients in the PID controller, and variable upper and lower limits for pump current protection.

Benefits of technology

This approach enhances the measurement accuracy and stability of exhaust gas sensors by adapting the control gain to changing conditions, preventing oscillations and ensuring precise oxygen management, thereby improving the sensor's operational performance.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure EP2024058228_07112024_PF_FP_ABST
    Figure EP2024058228_07112024_PF_FP_ABST
Patent Text Reader

Abstract

The invention relates to a method for operating an exhaust gas sensor which comprises a measurement chamber communicating with a measurement gas and which comprises an electrochemical pump cell, by which oxygen is transportable into the measurement chamber and out of the measurement chamber in accordance with the pump current Ip impressed in the pump cell, and which comprises an electrochemical Nernst cell, at which the Nernst voltage Un forms in accordance with the ratio of the oxygen content in the measurement chamber relative to the oxygen content in a reference chamber of the exhaust gas sensor (10). The method provides the operation of a control loop, the actual value of which is the Nernst voltage Un, the setpoint value of which is a predefined value UnSet and the manipulated variable of which is the pump current Ip calculated as the product of a control gain g and the difference e between the actual value and the setpoint value. The invention provides for the control gain to be variable as a function of at least one quantity relating to the exhaust gas sensor (10).
Need to check novelty before this filing date? Find Prior Art

Description

[0001] Description

[0002] title

[0003] Method for operating an exhaust gas sensor, for example a broadband lambda sensor

[0004] State of the art

[0005] From DE 10 2020 214 036 A1 of the applicant, a method for operating a broadband lambda probe is already known, which has a measuring chamber communicating with a measuring gas and which has an electrochemical pump cell with which oxygen can be transported into and out of the measuring chamber according to the pump voltage applied to the pump cell and according to a resulting pump current, and which has an electrochemical Nernst cell at which the Nernst voltage is formed according to the ratio of the oxygen content in the measuring chamber relative to the oxygen content in a reference chamber of the broadband lambda probe, wherein the method provides a control loop whose actual value is the Nernst voltage, whose setpoint is a predetermined value and whose manipulated variable is the pump current, wherein the manipulated variable pump current is actively limited by an upper limit and / or a lower limit to protect the pump cell,wherein the upper limit and / or the lower limit are each variably determined by determining a variable representing the temperature of the broadband lambda probe, and the upper limit and / or the lower limit are determined as a function of the variable representing the temperature of the broadband lambda probe.

[0006] Advantages of the invention

[0007] The present invention is based on the inventors' observation that, despite the limitation of the controller output by the upper or lower limit, the control behavior and thus the measuring accuracy of the exhaust gas sensor are not always optimal. There may indeed be individual operating points of the exhaust gas sensor for which the controller works optimally and, as a result, the measuring accuracy of the exhaust gas sensor is optimal. However, if the exhaust gas sensor is not in one of these operating points, e.g. due to aging or still insufficient or temporarily insufficient heating or for other reasons, then the control behavior is suboptimal. For example, oscillation or toggling (jumping back and forth) of the variable to be controlled and the manipulated variable may occur. The measuring function of the exhaust gas sensor is correspondingly suboptimal in these situations, in particular if the measured variable of the exhaust gas sensor is based on the controller output oris given by the controller output, as is usually the case with broadband lambda sensors.

[0008] To overcome these limitations, the invention proposes a method for operating an exhaust gas sensor.

[0009] The exhaust gas sensor can be, for example, a broadband lambda sensor or a NOx sensor. The invention merely requires that the exhaust gas sensor has a measuring chamber communicating with a measuring gas and an electrochemical pump cell with which oxygen can be transported into and out of the measuring chamber according to the pump current impressed into the pump cell, and an electrochemical Nernst cell at which the Nernst voltage develops according to the ratio of the oxygen content in the measuring chamber relative to the oxygen content in a reference chamber of the exhaust gas sensor.

[0010] The method further provides a control loop whose actual value is the Nernst voltage, whose setpoint is a predetermined value, and whose manipulated variable is the pump current. The pump current is determined or calculated as the product of a control gain and the difference between the actual value and the setpoint.

[0011] The control gain has the special feature that it is variable depending on at least one variable relating to the exhaust gas sensor.

[0012] The variable relating to the exhaust gas sensor can in particular be a variable that directly or indirectly influences the controlled system in the control loop, for example on the extent and dynamics of the generation of oxygen in the measuring gas space by the pumping current impressed into the pump cell and on the extent and dynamics with which this results in a change in the Nernst voltage. The variable relating to the exhaust gas sensor can, for example, be the temperature of the exhaust gas sensor. For example, the temperature of the exhaust gas sensor can be detected in the form of an electrical resistance measurement, for example a measurement of an internal resistance of the Nernst cell and / or the pump cell and / or the like. For example, an increase in the temperature of the exhaust gas sensor is accompanied by a decrease in the internal resistance of the Nernst cell and / or the pump cell and / or the like.Alternatively, the resistance measurement can be performed on a metallic conductor track within the exhaust gas sensor. This could, for example, be a conductor track whose electrical resistance increases with rising temperature.

[0013] On the other hand, the value relating to the exhaust gas sensor may be, for example, the age of the exhaust gas sensor, for example, the difference between the current point in time (e.g., date and, if applicable, time) and the time of manufacture or installation of the exhaust gas sensor; or the operating time of the exhaust gas sensor, for example, the total time the sensor was previously operated, i.e., for example, was actively heated and / or subjected to a pumping current.

[0014] The quantity relating to the exhaust gas sensor can in principle also be another relevant quantity.

[0015] The controller can, for example, be a so-called PID controller. The control gain can therefore be given according to the equation where: lp(t) is the pump current; e(t) is the difference between the setpoint and the actual value;

[0016] Kp is the coefficient of the proportional part of the control gain; Ki is the coefficient of the integral part of the control gain;

[0017] Kd is the coefficient of the derivative part of the control gain.

[0018] The fact that the control gain is variable depending on at least one variable relating to the exhaust gas sensor can then be expressed by the fact that the coefficient of the proportional component of the control gain is variable depending on the at least one variable relating to the exhaust gas sensor. Additionally or alternatively, the coefficient of the integral component of the control gain and / or the coefficient of the differential component of the control gain can also be variable depending on the at least one variable relating to the exhaust gas sensor.

[0019] For example, the control gain may have a coefficient of the proportional component that is variable depending on the temperature of the exhaust gas sensor, wherein the coefficient of the proportional component increases in a temperature interval with increasing temperature.

[0020] For example, the control gain may have a coefficient of the proportional component which is variable depending on the temperature of the exhaust gas sensor, wherein the coefficient of the proportional component has a first constant value below a first temperature and has a second constant value which is greater than the first constant value above a second temperature which is higher than the first temperature, and increases with increasing temperature in a temperature interval which is limited by the first and the second temperature.

[0021] Optionally, it can also be provided that the pump current is actively limited by an upper limit and / or a lower limit to protect the pump cell. The upper limit and / or the lower limit are each variably determined by determining the temperature of the exhaust gas sensor, and the upper limit and / or the lower limit are determined as a function of the temperature of the exhaust gas sensor. For example, the higher the temperature of the exhaust gas sensor, the higher the permitted pump currents and the lower the permitted pump currents.

[0022] The controller can basically be an analog controller or a digital controller.

[0023] In addition to the method described, the invention also relates to a corresponding computer program, a corresponding data carrier and a corresponding control device.

[0024] Drawing Figure 1 shows an example of a broadband lambda sensor and a

[0025] Control device as can be used in connection with the present invention.

[0026] Figure 2 shows an example of a detailed view of the Nerst voltage regulator of the control unit from Figure 1 .

[0027] Figure 3 shows an example of the dependence of the coefficient of the

[0028] Proportional part of the control gain depends on the temperature of the exhaust gas sensor.

[0029] Figure 4 shows an example of the method according to the invention as

[0030] Flow chart.

[0031] Figure 5 shows an example of the pump current and the Nernst voltage during a switch-on process of an exhaust gas sensor when carrying out the method according to the invention.

[0032] For comparison, Figure 6 shows the pump current and the Nernst voltage during a switch-on process of the exhaust gas sensor, analogous to Figure 5, when carrying out a process with fixed control parameters.

[0033] Description of the embodiments

[0034] Figure 1 shows an exemplary and schematic representation of a broadband lambda probe 10, such as can be used as an exhaust gas sensor in connection with the present invention. Of course, an exhaust gas sensor of a different design can also be used, e.g., a NOx sensor, provided it has the device features required in claim 1 with regard to the exhaust gas sensor.

[0035] The broadband lambda sensor 10 comprises an electrochemical pump cell 3 with a first electrode 116 and a second electrode 118. The second electrode 118 is arranged in a measuring chamber 130, which is connected to the measuring gas chamber 2 via a diffusion barrier 120 and a gas inlet hole 129. In the example, an exhaust gas to be sensed from an internal combustion engine (not shown) is located there. The first electrode 116 and the second electrode 118 are connected via a solid electrolyte 124. A pump current Ip can be impressed from the outside between the first electrode 116 and the second electrode 118.

[0036] The broadband lambda sensor 10 further comprises an electrochemical Nernst cell 4 with a first electrode 136 and a second electrode 138. The second electrode 138 of the Nernst cell 4 is arranged in the measuring chamber 130 and is identical to the second electrode 118 of the pump cell 3 in this example. Two different electrodes 118, 138, both arranged in the cavity 130 and optionally electrically connected to one another, would also be possible as an alternative. The first electrode 136 and the second electrode 138 of the Nernst cell 4 are connected to one another via the solid electrolyte 124. A Nernst voltage UN can be tapped and recorded with time resolution between the first electrode 136 and the second electrode 138 of the Nernst cell 4.

[0037] The first electrode 116 of the pump cell 3 can be referred to as an external pump electrode. In the example, it is arranged in the measuring gas chamber 2. The first electrode 136 of the Nernst cell 4 is arranged in the example in a reference gas chamber 126, i.e., in a gas chamber that is separated from the measuring gas chamber 2 in a gas-tight or at least essentially gas-tight manner. Furthermore, the broadband lambda probe 10 can comprise a heating element 128. Furthermore, the broadband lambda probe 10 can comprise a trimming resistor 180. The trimming resistor 180 can, for example, serve with a measuring resistor as a current divider for trimming the broadband lambda probe 10, in particular for calibration.

[0038] The broadband lambda probe 10 is connected to a control unit 112. The control unit 112 is configured to carry out a method according to the invention. The control unit 112 can be connected to the broadband lambda probe 10, for example, via an interface 134. The control unit 112 can comprise a current source (e.g., a constant current source) to inject the pump current Ip into the pump cell 3. For this purpose, the current source is connected to the second electrode 118 (also: inner pump electrode) of the pump cell 3 via the output IPE of the control unit 112 and via an inner pump electrode lead 142, and to the outer pump electrode 116 of the pump cell 3 via the output APE and via an outer pump electrode lead 144. For example, a voltage source can be provided to supply the heating element 128 via the + and - outputs and via two heater leads 188.The control unit 112 can be connected to the first electrode 116 of the pump cell 3 via the output MES via the line 146 and the balancing resistor 180.

[0039] Furthermore, the control unit 112 can include a voltage measuring device for measuring the Nernst voltage UN between the first and second electrodes 116, 118 of the Nernst cell 4. For this purpose, the voltage measuring device is connected to the first electrode 136 of the Nernst cell 4 (also known as the reference electrode) via the output RE and the line 140, and to the second electrode 138 of the Nernst cell 4 via the output IPE and the line 142.

[0040] The control unit 112 also has a Nernst voltage regulator 114 (synonym: pumping current regulator), which is shown in detail in Figure 2. The Nernst voltage regulator 114 can, as in the example, be a digital regulator. Since the broadband lambda sensor 10 naturally has analog inputs and outputs, AD conversions are performed upstream of the Nernst voltage regulator 114 and DA conversions are performed downstream of the Nernst voltage regulator 114. However, for the sake of clarity, this is not shown in Figure 2 and will not be discussed further below. Alternatively, the Nernst voltage regulator 114 can also be an analog one.

[0041] The Nernst voltage regulator 114 receives the Nernst voltage Un as the actual value and a predetermined value Unset as the setpoint, e.g., 450 mV. The Nernst voltage regulator 114 calculates the difference e between the actual value and the setpoint using a difference former 115 and further processes this difference in an amplifier 119 according to a control gain g to obtain a preliminary value of the pump current lp'.

[0042] To protect pump cell 3, the pump current Ip is actively limited in a limiter 117 of the pump current controller 114. For this purpose, the limiter 117 receives the preliminary value of the pump current Ip' and outputs a value of the pump current Ip. The value of the pump current Ip is determined by the limiter 117 as follows:

[0043] - Identical to the provisional value of the pump current I p' if it lies between an upper limit IpMax and a lower limit IpMin.

[0044] - Identical to an upper limit IpMax if the preliminary value of the pump current IP' is greater than the upper limit IpMax. - Identical to a lower limit IpMin if the preliminary value of the pump current IP' is less than the lower limit IpMin.

[0045] The upper limit IpMax and the lower limit IpMin can be fixed values. In the example, however, they depend on the temperature θ of the broadband lambda sensor 10, which is available in the control unit 112 based on the internal resistance of the pump cell 3 and / or the Nernst cell 4. The upper limit IpMax and the lower limit IpMin differ in their sign, and larger pump currents Ip are permitted at higher temperatures T of the broadband lambda sensor 10.

[0046] In the example, it is provided that the control gain g is variable depending on at least one variable relating to the broadband lambda sensor 10. In the example, this variable is the temperature T of the broadband lambda sensor 10, which is available in the control unit 112 based on the internal resistance of the pump cell 3 and / or the Nernst cell 4 or in another way.

[0047] In the example, the control gain g has a PID characteristic, i.e., it is given by the equation where:

[0048] I p'(t) is the preliminary value of the pump current; e(t) is the difference between the setpoint and the actual value;

[0049] Kp is the coefficient of the proportional part of the control gain;

[0050] Ki is the coefficient of the integral part of the control gain and

[0051] Kd is the coefficient of the derivative part of the control gain.

[0052] In the example, only the coefficient of the proportional component of the control gain Kp depends on the temperature T of the broadband lambda sensor 10, while the coefficients of the integral component of the control gain and the differential component of the control gain Ki, Kd are constant values. In the example, Kp = Kp (T) and Kp = Kp_adaptive, respectively. The dependence of the coefficient of the proportional component Kp of the control gain g on the temperature T of the broadband lambda sensor 10 is shown as an example in Figure 3: Below a first temperature T1, Kp assumes a first constant value Kp1; above a second temperature T2 that is higher than the first temperature T1, Kp assumes a second constant value Kp2 that is greater than the first constant value Kp1; and in a temperature interval that is bounded by the first and second temperatures T1, T2, Kp increases with increasing temperature T.For example, T1 can be in the temperature range between 150°C and 300°C and T2 can be in the temperature range between 550°C and 700°C or for example in the temperature range between 550°C and 800°C.

[0053] In alternative embodiments, additionally or alternatively, the coefficient of the integral component of the control gain Ki and / or the coefficient of the differential component of the control gain Kd may also be dependent on the temperature T of the broadband lambda probe 10.

[0054] The method according to the invention was carried out by way of example using the device described with reference to Figures 1 to 3, which is shown as a flow chart in Figure 4. Figure 5 shows the corresponding time profiles of the upper limit IpMax, the lower limit IpMin, the pump current Ip, and the Nernst voltage lln.

[0055] In the example, at time tO (Figure 5) in method step V1 (Figure 4), a cold broadband lambda probe 10 is exposed to a stoichiometric exhaust gas from an internal combustion engine, and the self-heating of the broadband lambda probe 10 is activated in method step V2, so that the temperature T of the broadband lambda probe 10 subsequently rises. A Nernst voltage Un applied to the Nernst cell 4 is measured in method step V3; at time tO, it initially has a value that is significantly lower than the specified value UnSet. The Nernst voltage regulator 114 calculates the difference e between the Nernst voltage Un and the specified value in method step V4. In method step V5, the temperature T of the broadband lambda probe 10 is determined, for example on the basis of the internal resistance of the Nernst cell 4 and / or the pump cell 3. The value of the temperature T of the broadband lambda probe 10 is transmitted to the amplifier 119 and the limiter 117 in method step V6.In method step V7, the amplifier 119 determines the control gain from the temperature T, i.e., in the example, it determines the value of the coefficient Kp of the proportional component of the PID-like control gain (see above) according to the characteristic curve shown in Figure 3. In the example, the coefficient of the proportional component Kp of the control gain g has a relatively small value at time tO because the broadband lambda probe 10 is cold. The amplifier 119 hereby determines a low preliminary value of the pump current I p', method step 8. Since this value lies between the upper limit IpMax and the lower limit IpMin, it is adopted by the limiter 117 identically as the value of the pump current Ip, method step 9, and accordingly, in the example, oxygen is removed from the measuring chamber 130 and released in the form of O. 2'- Ions are transported from the inner pumping electrode 118 through the solid electrolyte 124 to the outer pumping electrode 116. Process steps 1 to 9 continue cyclically as long as the broadband lambda sensor 10 is in operation.

[0056] By a subsequent time t1 (Figure 5), the temperature T of the broadband lambda probe has increased to an average value, so that the value of the coefficient Kp of the proportional component of the PID-like control gain g now has a higher value than at time t0. The upper limit IpMax and the lower limit IpMin now also have values ​​that are greater in magnitude than at time t0. As a result of the oxygen transport out of the measuring chamber 130, which occurs between times t0 and t1, and to a certain extent also as a result of the increased temperature T of the broadband lambda probe 10, the value of the Nernst voltage Un has increased, but is still smaller than the setpoint Unset.

[0057] By a subsequent time t2 (Figure 5), the temperature T of the broadband lambda probe has increased to a high value that largely corresponds to the operating temperature of the broadband lambda probe 10, e.g., 785°C. The value of the coefficient of the proportional component Kp of the PID-like control gain g and the upper limit IpMax and the lower limit assume corresponding values.

[0058] The Nernst voltage Un has reached its setpoint without overshoot by time t2 and the pump current Ip decreases accordingly to a value that is proportional to the oxygen partial pressure in the measuring gas chamber 2.

[0059] Due to the adaptivity of the coefficient of the proportional component Kp of the control gain g during the heating phase of the broadband lambda probe 10 shown in Figures 4 and 5, the Nernst voltage Un was able to reach its setpoint Unset very quickly and yet without overshoot, and the oxygen partial pressure in the measuring gas chamber 2 could be determined early and precisely from the pump current. The measurement accuracy of the broadband lambda probe 10 was thus improved during this phase compared to the prior art, which operates with constant control parameters. For comparison, Figure 6 shows the pump current Ip and the Nernst voltage Un during a switch-on process of the broadband lambda probe 10, analogous to Figure 5, when implementing a process with fixed control parameters Kp, Ki, Kd. Due to the initially inappropriately high control gain, a so-called toggling (jumping back and forth between extreme values) of the Nernst voltage Un and the pump current Ip occurs during the heating phase.

Claims

Claims 1 . A method for operating an exhaust gas sensor, in particular a broadband lambda probe (10), which has a measuring chamber (130) communicating with a measuring gas and which has an electrochemical pump cell (3) with which oxygen can be transported into and out of the measuring chamber (130) according to the pump current (Ip) impressed into the pump cell (3), and which has an electrochemical Nernst cell (4) at which the Nernst voltage (Un) is formed according to the ratio of the oxygen content in the measuring chamber (130) relative to the oxygen content in a reference chamber of the exhaust gas sensor (10), wherein the method provides a control loop whose actual value is the Nernst voltage (Un), whose setpoint is a predetermined value (UnSet), and whose manipulated variable is the pump current (Ip), which is calculated as the product of a control gain (g) and the difference (e) between the actual value and the setpoint, characterized in thatthat the control gain (g) is variable depending on at least one variable relating to the exhaust gas sensor (10).

2. Method according to claim 1, wherein the control gain (g) has a proportional component.

3. Method according to claim 2, wherein the coefficient of the proportional component (Kp) is variable depending on at least one variable relating to the exhaust gas sensor (10).

4. Method according to claim 2 or 3, wherein the control gain (g) further comprises an integral component and / or a differential component.

5. The method according to claim 4, wherein the coefficient of the integral component (Ki) and / or the coefficient of the differential component (Kd) is variable as a function of at least one or of the at least one variable relating to the exhaust gas sensor (10) 6. Method according to one of the preceding claims, wherein the at least one variable relating to the exhaust gas sensor (10) is the temperature (T) of the exhaust gas sensor (10).

7. Method according to one of the preceding claims, wherein the at least one variable relating to the exhaust gas sensor (10) is the age of the exhaust gas sensor (10).

8. The method according to claim 1, wherein the control gain (g) has a proportional component whose coefficient (Kp) is variable as a function of the temperature (T) of the exhaust gas sensor (10), wherein it increases in a temperature interval with increasing temperature (T).

9. The method according to claim 1, wherein the control gain (g) has a proportional component whose coefficient (Kp) has a first constant value (Kp1) below a first temperature (T) of the exhaust gas sensor (10); and above a second temperature (T2) which is higher than the first temperature (T1) has a second constant value (Kp2) which is greater than the first constant value (Kp1) and increases with increasing temperature (T) in a temperature interval which is limited by the first and second temperatures (T1, T2).

10. Method according to one of the preceding claims, wherein the manipulated variable pump current (Ip) is actively limited by an upper limit (IpMax) and / or by a lower limit (IpMin) to protect the pump cell (3), wherein the upper limit (IpMax) and / or the lower limit (IpMin) are each variably set by determining the temperature (T) of the exhaust gas probe (10) and determining the upper limit (IpMax) and / or the lower limit (IpMin) as a function of the temperature (T) of the exhaust gas probe (10).

11. A computer program comprising instructions which, when executed by a computer, cause the computer to carry out the method according to any one of the preceding claims.

12. A computer-readable storage medium comprising instructions which, when executed by a computer, cause the computer to carry out the method according to any one of claims 1 to 10.

13. Analog and / or digital control device configured to carry out the method according to one of claims 1 to 10.