Determining a sensor error of a sensor in a motor vehicle exhaust system

Abstract

One aspect of the present invention relates to a method for ascertaining a sensor error of a sensor in a motor vehicle exhaust system. One step of the method is to determine at least one actual sensor signal of the sensor. Another step of the method is to determine at least one theoretical sensor signal of the sensor by means of a model. Another step of the method is to determine the sensor error of the sensor based on the deviation between the actual sensor signal of the sensor and the theoretical sensor signal of the sensor.

Description

Field of the Invention

[0001] The present invention relates to a method and a device for detecting sensor errors of sensors in a motor vehicle exhaust system. Background Art

[0002] It is known to detect sensor errors of sensors in a motor vehicle exhaust system, such as oxygen sensors, by controlling an internal combustion engine upstream of the exhaust system in a predetermined manner and method, so that a predetermined sensor signal is generated for an error-free sensor. If the actual sensor signal deviates from the predetermined sensor signal, a sensor error is inferred.

[0003] Known measures for detecting sensor errors of sensors in a motor vehicle exhaust system have the disadvantage that the internal combustion engine connected upstream of the exhaust system must be actively controlled here, which results in an adverse effect on driving comfort and / or results in the generation of undesired emissions. Summary of the Invention

[0004] The object of the present invention is to at least reduce the mentioned disadvantages.

[0005] It should be noted that the additional features of claims subordinate to an independent claim can form a separate invention independent of the entire feature combination of the independent claim without the features of the independent claim or only in combination with some features of the independent claim, which can be the subject of an independent claim, a divisional application or a later application. This applies in the same way to the technical teachings described in the specification, which can form an invention independent of the features of the independent claim.

[0006] A first aspect of the present invention relates to a method for detecting sensor errors of sensors for a motor vehicle exhaust system.

[0007] The sensor is especially installed in the exhaust system, for example, downstream of the exhaust flow before or after a catalytic converter.

[0008] One step of the method is to determine at least one actual sensor signal of the sensor. The actual sensor signal of the sensor is a signal generated by the sensor, by means of which the sensor provides the physical or chemical characteristics of its environment measured by it.

[0009] Another step of the method is to determine at least one theoretical sensor signal of the sensor by means of a model.

[0010] The theoretical sensor signal of the sensor is a virtual signal generated by the sensor. The theoretical sensor signal indicates what actual sensor signal an error-free sensor would generate according to the environmental conditions of the sensor.

[0011] Theoretical sensor signals can in particular be generated by an artificial neural network.

[0012] A neural network is an aggregation of a number of information processing units (neurons) arranged layer by layer in a network structure. When it comes to artificial intelligence, an artificial neural network is mentioned.

[0013] The neurons of an artificial neural network are arranged layer by layer in so-called layers and are usually interconnected in a fixed hierarchical structure. The neurons are mostly connected between two layers here, but in a few cases they are also connected within a layer.

[0014] Starting from the input layer, information flows through one or more intermediate layers (hidden layers) to the output layer. Here, the output of one neuron is the input of the next neuron.

[0015] The input layer is the starting point of the information flow in an artificial neural network.

[0016] Input signals are usually received by the neurons at the start of the layer and are weighted and transmitted to the neurons of the first intermediate layer at the end. Here, the neurons of the input layer transmit the corresponding information to the neurons of the first intermediate layer.

[0017] Between the input layer and the output layer, there is at least one intermediate layer (also called the active layer or hidden layer, in English: hidden layer) in each artificial neural network. The more intermediate layers there are, the "deeper" the neural network is, and in such a case, Deep Learning is also mentioned in English.

[0018] In theory, the number of possible hidden layers in an artificial neural network is not restricted. However, in practice, each additional hidden layer also causes an increase in the required computing power that is necessary for the operation of the network.

[0019] The output layer is behind the intermediate layer and forms the last layer in an artificial neural network. The neurons arranged in the output layer are respectively connected to the neurons of the last intermediate layer. The output layer forms the end point of the information flow in an artificial neural network and contains the result of the information processing through the network.

[0020] Weights describe the strength of the information flow along the connections in a neural network. For this purpose, each neuron is assigned a weight for the information flowing through it and then transmits the information weighted and, if necessary, after adding a value for neuron-specific distortion to the neurons of the next layer. Usually, the weights and the distortion are initialized at the start of the training. The result of the weighted sum and the distortion is often guided through a so-called activation function and then the result is transmitted to the neurons of the next layer.

[0021] The weights and distortions are adapted during the training process such that the final result corresponds as accurately as possible to the requirements.

[0022] Another step of the method is to determine the sensor error of the sensor based on the deviation between the actual sensor signal of the sensor and the theoretical sensor signal of the sensor.

[0023] An advantage of the invention is that in order to determine the sensor error of the sensor, it is not necessary to actively control the internal combustion engine upstream of the exhaust system.

[0024] In an advantageous embodiment of the invention, determining the deviation between the actual sensor signal of the sensor and the theoretical sensor signal of the sensor includes: providing at least one error model for the actual sensor signal of the sensor, determining at least one error sensor signal of the sensor by correlating the theoretical sensor signal of the sensor with the corresponding error model, and determining the sensor error of the sensor based on at least one deviation between the actual sensor signal of the sensor and the corresponding error sensor signal of the sensor.

[0025] The error model is in particular a classification of the sensor error that results in a systematic deviation of the actual sensor signal.

[0026] In another advantageous embodiment, the at least one error model is a parameterizable error model that has at least one parameter. Determining the at least one deviation between the actual sensor signal of the sensor and the corresponding error sensor signal of the sensor here includes: estimating the at least one parameter of the error model such that the deviation between the actual sensor signal of the sensor and the corresponding error sensor signal of the sensor is minimized, and determining the sensor error based on the at least one parameter of the error model. For example, the estimated parameters of the error model can be determined as the sensor error, so that the sensor error is not only determined to be qualitatively present, but even quantified.

[0027] In another advantageous embodiment, the method includes providing at least two error models for the actual sensor signal of the sensor.

[0028] Furthermore, in this advantageous embodiment, the method includes: determining at least two error sensor signals of the sensor by correlating the theoretical sensor signal of the sensor with the corresponding error models, and determining the deviation between the actual sensor signal of the sensor and the corresponding error sensor signals of the sensor.

[0029] Additionally, in this advantageous embodiment, the method includes: selecting one of the error models based on the determined deviation between the actual sensor signal of the sensor and the corresponding error sensor signals of the sensor, and determining the sensor error based on the selected error model.

[0030] This advantageous embodiment is based on the recognition that it can be advantageous to compare the actual sensor signal of a sensor with a plurality of error sensor signals, since a sensor can be covered by different error models. Thus, for example, it is possible to determine during operation, without actively manipulating the internal combustion engine upstream of the exhaust system, the most likely error model present.

[0031] In another advantageous embodiment, the at least one error model includes at least one of the following six error models.

[0032] 1. A time delay of the actual sensor signal relative to the theoretical sensor signal when changing from a rich mixture to a lean mixture in the air-fuel ratio, and no time delay of the actual sensor signal relative to the theoretical sensor signal when changing from a lean mixture to a rich mixture in the air-fuel ratio.

[0033] The rich mixture is hereby characterized by an "air deficiency" with an air excess coefficient (Lambda value) of less than 1. In contrast, the lean mixture is characterized by an "air excess" with an air excess coefficient greater than 1.

[0034] 2. A time delay of the actual sensor signal relative to the theoretical sensor signal when changing from a lean mixture to a rich mixture in the air-fuel ratio, and no time delay of the actual sensor signal relative to the theoretical sensor signal when changing from a rich mixture to a lean mixture in the air-fuel ratio.

[0035] 3. A low-pass filtering of the actual sensor signal relative to the theoretical sensor signal when changing from a rich mixture to a lean mixture in the air-fuel ratio, and no low-pass filtering of the actual sensor signal relative to the theoretical sensor signal (SSS) when changing from a lean mixture to a rich mixture in the air-fuel ratio.

[0036] 4. A low-pass filtering of the actual sensor signal relative to the theoretical sensor signal when changing from a lean mixture to a rich mixture in the air-fuel ratio, and no low-pass filtering of the actual sensor signal relative to the theoretical sensor signal when changing from a rich mixture to a lean mixture in the air-fuel ratio.

[0037] 5. A time delay of the actual sensor signal relative to the theoretical sensor signal when changing from a lean mixture to a rich mixture in the air-fuel ratio, and a time delay of the actual sensor signal relative to the theoretical sensor signal when changing from a rich mixture to a lean mixture in the air-fuel ratio.

[0038] 6. A low-pass filtering of the actual sensor signal relative to the theoretical sensor signal when changing from a lean mixture to a rich mixture in the air-fuel ratio, and a low-pass filtering of the actual sensor signal relative to the theoretical sensor signal when changing from a rich mixture to a lean mixture in the air-fuel ratio.

[0039] In a further advantageous embodiment of the invention, the sensor is an oxygen sensor which is arranged, in particular, upstream of the catalytic converter in the exhaust gas flow in the exhaust system, upstream of the catalytic converter.

[0040] In a further advantageous embodiment of the invention, the model for determining the at least one theoretical sensor signal of the sensor is a neural network.

[0041] In a further advantageous embodiment of the invention, the sensor is an oxygen sensor which is arranged downstream of the catalytic converter in the exhaust gas flow in the exhaust system, downstream of the catalytic converter.

[0042] Additionally, in this advantageous embodiment, the method includes: determining at least one actual sensor signal of the sensor and determining at least two theoretical sensor signals of the sensor by means of the respective model, wherein each model characterizes a specific aging state of the catalytic converter.

[0043] Furthermore, in this advantageous embodiment, the method includes: ascertaining the respective deviation between the actual sensor signal of the sensor and one of the theoretical sensor signals of the sensor, and selecting one of the theoretical sensor signals of the sensor according to the ascertained deviation, and ascertaining the sensor error according to the deviation between the actual sensor signal of the sensor and the selected theoretical sensor signal of the sensor.

[0044] A second aspect of the invention relates to a device for ascertaining the sensor error of a sensor in a motor vehicle exhaust system, wherein the device is configured to determine at least one actual sensor signal of the sensor, determine at least one theoretical sensor signal of the sensor by means of a model, and ascertain the sensor error of the sensor according to the deviation between the actual sensor signal of the sensor and the theoretical sensor signal of the sensor.

[0045] The above-described embodiments of the method according to the first aspect of the invention correspondingly also apply to the device according to the second aspect of the invention. The advantageous embodiments of the device according to the invention which are not explicitly described here and in the claims correspond to the above-described or claimed advantageous embodiments of the method according to the invention. Description of the Drawings

[0046] The invention will now be described with reference to the accompanying drawings by way of examples. In the drawings:

[0047] Figure 1 shows an embodiment of the method according to the invention;

[0048] Figure 2 shows some embodiments of the error model; and

[0049] Figure 3Shows other embodiments for the error model. Detailed Description

[0050] Figure 1 Shows a method for ascertaining a sensor error of a sensor for a motor vehicle exhaust system.

[0051] One step of the method is to determine at least one actual sensor signal ISS of the sensor S, in particular by the sensor S itself.

[0052] Another step of the method is to determine at least one theoretical sensor signal SSS of the sensor S by means of models NN1, NN2, in particular based on at least one parameter GMG of the basic engine. The at least one parameter GMG of the basic engine is in particular an adjustment or characteristic parameter of the basic engine, such as an adjustment or characteristic parameter of the crank or flywheel housing, the crank or camshaft, the drive wheels, the cylinder head or cylinder head cover, the connecting rod, the piston, the oil cooler, the oil separator, the oil filter and / or the injection system.

[0053] The models NN1, NN2 for determining the at least one theoretical sensor signal SSS of the sensor S are in particular neural networks.

[0054] Another step of the method is to ascertain the sensor error of the sensor S based on the deviation between the actual sensor signal ISS of the sensor S and the theoretical sensor signal SSS of the sensor S.

[0055] In particular, it is possible to ascertain the deviation between the actual sensor signal ISS of the sensor and the theoretical sensor signal SSS of the sensor, wherein ascertaining the deviation between the actual sensor signal ISS of the sensor S and the theoretical sensor signal SSS of the sensor S includes providing at least one error model FM for the actual sensor signal ISS of the sensor S. The at least one error model FM is in particular a parameterizable error model and thus has at least one parameter.

[0056] Furthermore, ascertaining the deviation between the actual sensor signal ISS of the sensor S and the theoretical sensor signal SSS of the sensor S includes: determining at least one error sensor signal FSS of the sensor S by associating the theoretical sensor signal SSS of the sensor S with the corresponding error model FM, and ascertaining the sensor error of the sensor S based on at least one deviation between the actual sensor signal ISS of the sensor S and the corresponding error sensor signal FSS of the sensor S.

[0057] Determining the at least one deviation between the actual sensor signal ISS of the sensor S and the corresponding error sensor signal FSS of the sensor S in particular includes: estimating the at least one parameter of the error model FM such that the deviation between the actual sensor signal ISS of the sensor S and the corresponding error sensor signal FSS of the sensor S is minimized.

[0058] For this purpose, for example, conventional methods of compensation calculation, i.e., mathematical optimization methods (such as the least squares method), can be used, by means of which, for a series of measurement data, the unknown parameters of its geometric physical model or the parameters of a predetermined function should be determined or estimated.

[0059] Determining the at least one deviation between the actual sensor signal ISS of the sensor S and the corresponding error sensor signal FSS of the sensor S in particular further includes determining the sensor error 130 based on the at least one parameter of the error model FM.

[0060] Alternatively, the method according to the invention includes providing at least two error models FM for the actual sensor signal ISS of the sensor S.

[0061] In this case, the method according to the invention includes: determining 100 at least two error sensor signals FSS of the sensor S by correlating the theoretical sensor signal SSS of the sensor S with the corresponding error model FM, determining 110 the deviation between the actual sensor signal ISS of the sensor S and the corresponding error sensor signal FSS of the sensor S, selecting 125 one of the error models FM based on the determined deviation between the actual sensor signal ISS of the sensor S and the corresponding error sensor signal FSS of the sensor S, and determining the sensor error 130 based on the selected error model FM.

[0062] The sensor S is in particular an oxygen sensor in a motor vehicle exhaust system. For example, the sensor S is an oxygen sensor which is arranged upstream of the catalytic converter in the exhaust flow in the exhaust system.

[0063] Alternatively, the sensor S is an oxygen sensor which is arranged downstream of the catalytic converter in the exhaust flow in the exhaust system.

[0064] In this case, the method for example further includes determining at least two theoretical sensor signals SSS of the sensor S by means of the corresponding models NN1, NN2, wherein each model NN1, NN2 characterizes a specific aging state of the catalytic converter.

[0065] In this case, one step of the method is to determine 110 the corresponding deviation between the actual sensor signal ISS of the sensor S and one of the theoretical sensor signals SSS of the sensor S.

[0066] Another step of the method is to select one of the theoretical sensor signals SSS of the 127 sensors S based on the identified deviation. In particular, for this purpose, the smallest, i.e., the least deviation, can also be determined by means of a compensation calculation method, and then the theoretical sensor signal SSS of the sensor S that includes the smallest deviation is selected.

[0067] Furthermore, in this case, the method includes the step of identifying a 130 sensor error based on the deviation between the actual sensor signal ISS of the sensor S and the selected theoretical sensor signal SSS of the sensor S.

[0068] Figure 2 Some embodiments of an error model for an oxygen sensor are shown, in which the curves of the theoretical sensor signal SSS and the actual sensor signal ISS over time are depicted. The theoretical sensor signal SSS is represented by a solid line here. The deviation of the actual sensor signal ISS from the theoretical sensor signal SSS is shown by a dashed line. For the sake of simplicity of illustration, for the time period when the actual sensor signal ISS basically corresponds to the theoretical sensor signal SSS, a separate line is not shown for the actual sensor signal ISS.

[0069] Between time t0 and time t1, the actual sensor signal ISS shows an error model with respect to the theoretical sensor signal SSS in which there is a time delay of the actual sensor signal ISS with respect to the theoretical sensor signal SSS when the air-fuel ratio changes from a rich mixture to a lean mixture and there is no time delay of the actual sensor signal ISS with respect to the theoretical sensor signal SSS when the air-fuel ratio changes from a lean mixture to a rich mixture.

[0070] Between time t1 and time t2, the actual sensor signal ISS shows an error model with respect to the theoretical sensor signal SSS in which there is a time delay of the actual sensor signal ISS with respect to the theoretical sensor signal SSS when the air-fuel ratio changes from a lean mixture to a rich mixture and there is a time delay of the actual sensor signal ISS with respect to the theoretical sensor signal SSS when the air-fuel ratio changes from a rich mixture to a lean mixture.

[0071] Figure 3 In the same illustration as Figure 2 Other embodiments of an error model for an oxygen sensor are shown.

[0072] Between time t0 and time t1, the actual sensor signal ISS shows an error model with respect to the theoretical sensor signal SSS in which there is a low-pass filtering of the actual sensor signal ISS with respect to the theoretical sensor signal SSS when the air-fuel ratio changes from a rich mixture to a lean mixture and there is no low-pass filtering of the actual sensor signal ISS with respect to the theoretical sensor signal SSS when the air-fuel ratio changes from a lean mixture to a rich mixture.

[0073] Between time t1 and time t2, the actual sensor signal ISS shows an error model with low-pass filtering of the actual sensor signal ISS relative to the theoretical sensor signal SSS when the air-fuel ratio changes from a lean mixture to a rich mixture and with low-pass filtering of the actual sensor signal ISS relative to the theoretical sensor signal SSS when the air-fuel ratio changes from a rich mixture to a lean mixture.

Claims

1. A method for ascertaining a sensor error of a sensor (S) for a motor vehicle exhaust system, the method comprising the following steps: · Determining at least one actual sensor signal (ISS) of the sensor (S), · Determining at least one theoretical sensor signal (SSS) of the sensor (S) by means of a model (NN1, NN2), and · Ascertaining the sensor error of the sensor (S) based on the deviation between the actual sensor signal (ISS) of the sensor (S) and the theoretical sensor signal (SSS) of the sensor (S), wherein, Determining the deviation between the actual sensor signal (ISS) of the sensor (S) and the theoretical sensor signal (SSS) of the sensor (S) includes the following steps: · Providing at least one error model (FM) for the actual sensor signal (ISS) of the sensor (S), wherein the error model is a classification of the sensor errors that cause systematic deviations of the actual sensor signal, · Determining (100) at least one error sensor signal (FSS) of the sensor (S) by correlating the theoretical sensor signal (SSS) of the sensor (S) with the respective error model (FM), and · Ascertaining the sensor error of the sensor (S) based on at least one deviation between the actual sensor signal (ISS) of the sensor (S) and the respective error sensor signal (FSS) of the sensor (S), wherein the at least one error model (FM) includes at least one of the following error models: · A time delay of the actual sensor signal (ISS) relative to the theoretical sensor signal (SSS) when the air-fuel ratio changes from a rich mixture to a lean mixture, and no time delay of the actual sensor signal (ISS) relative to the theoretical sensor signal (SSS) when the air-fuel ratio changes from a lean mixture to a rich mixture, · A time delay of the actual sensor signal (ISS) relative to the theoretical sensor signal (SSS) when the air-fuel ratio changes from a lean mixture to a rich mixture, and no time delay of the actual sensor signal (ISS) relative to the theoretical sensor signal (SSS) when the air-fuel ratio changes from a rich mixture to a lean mixture, · A low-pass filtering of the actual sensor signal (ISS) relative to the theoretical sensor signal (SSS) when the air-fuel ratio changes from a rich mixture to a lean mixture, and no low-pass filtering of the actual sensor signal (ISS) relative to the theoretical sensor signal (SSS) when the air-fuel ratio changes from a lean mixture to a rich mixture, · A low-pass filtering of the actual sensor signal (ISS) relative to the theoretical sensor signal (SSS) when the air-fuel ratio changes from a lean mixture to a rich mixture, and no low-pass filtering of the actual sensor signal (ISS) relative to the theoretical sensor signal (SSS) when the air-fuel ratio changes from a rich mixture to a lean mixture, · A time delay of the actual sensor signal (ISS) relative to the theoretical sensor signal (SSS) when the air-fuel ratio changes from a lean mixture to a rich mixture, and a time delay of the actual sensor signal (ISS) relative to the theoretical sensor signal (SSS) when the air-fuel ratio changes from a rich mixture to a lean mixture, and / or · A low-pass filtering of the actual sensor signal (ISS) relative to the theoretical sensor signal (SSS) when the air-fuel ratio changes from a lean mixture to a rich mixture, and a low-pass filtering of the actual sensor signal (ISS) relative to the theoretical sensor signal (SSS) when the air-fuel ratio changes from a rich mixture to a lean mixture.

2. The method according to claim 1, wherein, The at least one error model (FM) is a parameterizable error model and has at least one parameter, and ascertaining the at least one deviation between the actual sensor signal (ISS) of the sensor (S) and the corresponding error sensor signal (FSS) of the sensor (S) comprises the following steps: · Estimating (120) the at least one parameter of the error model (FM) such that the deviation between the actual sensor signal (ISS) of the sensor (S) and the corresponding error sensor signal (FSS) of the sensor (S) is minimized, and · Ascertaining the sensor error based on the at least one parameter of the error model (FM).

3. The method according to claim 1 or 2, wherein, The method comprises the following steps: · Providing at least two error models (FM) for the actual sensor signal (ISS) of the sensor (S), · Determining (100) at least two error sensor signals (FSS) of the sensor (S) by associating the theoretical sensor signal (SSS) of the sensor (S) with the corresponding error model (FM), · Ascertaining the deviation between the actual sensor signal (ISS) of the sensor (S) and the corresponding error sensor signal (FSS) of the sensor (S); · Selecting (125) one of the error models (FM) based on the ascertained deviation between the actual sensor signal (ISS) of the sensor (S) and the corresponding error sensor signal (FSS) of the sensor (S), and · Ascertaining the sensor error based on the selected error model (FM).

4. The method according to claim 1 or 2, wherein, The sensor (S) is an oxygen sensor.

5. The method according to claim 1 or 2, wherein, The models (NN1, NN2) for determining the at least one theoretical sensor signal (SSS) of the sensor (S) are neural networks.

6. The method according to claim 1 or 2, wherein, The sensor (S) is an oxygen sensor which is arranged downstream of the catalytic converter in the exhaust gas flow in the exhaust system, wherein the method comprises the following steps: · Determining at least one actual sensor signal (ISS) of the sensor (S), · Determining at least two theoretical sensor signals (SSS) of the sensor (S) by means of the corresponding models (NN1, NN2), wherein each model (NN1, NN2) characterizes a specific aging state of the catalytic converter, · Ascertaining the corresponding deviation between the actual sensor signal (ISS) of the sensor (S) and one of the theoretical sensor signals (SSS) of the sensor (S), · Selecting (127) one of the theoretical sensor signals (SSS) of the sensor (S) based on the ascertained deviation, and · Ascertaining the sensor error based on the deviation between the actual sensor signal (ISS) of the sensor (S) and the selected theoretical sensor signal (SSS) of the sensor (S).

7. According to the method of claim 4, wherein, The oxygen sensor is arranged upstream of the catalytic converter in the exhaust gas flow in the exhaust system.

8. A device for ascertaining a sensor error of a sensor (S) in a motor vehicle exhaust system, wherein, The device is configured to · Determine at least one actual sensor signal (ISS) of the sensor (S), · Determine at least one theoretical sensor signal (SSS) of the sensor (S) by means of the models (NN1, NN2), and ·Determine the sensor error of the sensor (S) based on the deviation between the actual sensor signal (ISS) of the sensor (S) and the theoretical sensor signal (SSS) of the sensor (S). Among them, determining the deviation between the actual sensor signal (ISS) of the sensor (S) and the theoretical sensor signal (SSS) of the sensor (S) includes the following steps: ·Provide at least one error model (FM) for the actual sensor signal (ISS) of the sensor (S), where the error model is a classification of the sensor error that causes the systematic deviation of the actual sensor signal. ·Determine (100) at least one error sensor signal (FSS) of the sensor (S) by correlating the theoretical sensor signal (SSS) of the sensor (S) with the corresponding error model (FM), and ·Determine the sensor error of the sensor (S) based on at least one deviation between the actual sensor signal (ISS) of the sensor (S) and the corresponding error sensor signal (FSS) of the sensor (S). Among them, the at least one error model (FM) includes at least one of the following error models: ·There is a time delay of the actual sensor signal (ISS) relative to the theoretical sensor signal (SSS) when the air-fuel ratio changes from a rich mixture to a lean mixture, and there is no time delay of the actual sensor signal (ISS) relative to the theoretical sensor signal (SSS) when the air-fuel ratio changes from a lean mixture to a rich mixture. ·There is a time delay of the actual sensor signal (ISS) relative to the theoretical sensor signal (SSS) when the air-fuel ratio changes from a lean mixture to a rich mixture, and there is no time delay of the actual sensor signal (ISS) relative to the theoretical sensor signal (SSS) when the air-fuel ratio changes from a rich mixture to a lean mixture. ·There is a low-pass filtering of the actual sensor signal (ISS) relative to the theoretical sensor signal (SSS) when the air-fuel ratio changes from a rich mixture to a lean mixture, and there is no low-pass filtering of the actual sensor signal (ISS) relative to the theoretical sensor signal (SSS) when the air-fuel ratio changes from a lean mixture to a rich mixture. ·There is a low-pass filtering of the actual sensor signal (ISS) relative to the theoretical sensor signal (SSS) when the air-fuel ratio changes from a lean mixture to a rich mixture, and there is no low-pass filtering of the actual sensor signal (ISS) relative to the theoretical sensor signal (SSS) when the air-fuel ratio changes from a rich mixture to a lean mixture. ·There is a time delay of the actual sensor signal (ISS) relative to the theoretical sensor signal (SSS) when the air-fuel ratio changes from a lean mixture to a rich mixture, and there is a time delay of the actual sensor signal (ISS) relative to the theoretical sensor signal (SSS) when the air-fuel ratio changes from a rich mixture to a lean mixture, and / or ·There is a low-pass filtering of the actual sensor signal (ISS) relative to the theoretical sensor signal (SSS) when the air-fuel ratio changes from a lean mixture to a rich mixture, and there is a low-pass filtering of the actual sensor signal (ISS) relative to the theoretical sensor signal (SSS) when the air-fuel ratio changes from a rich mixture to a lean mixture.

Images