JP2009209865A - Device for diagnosing abnormality of knock detection system - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To accurately diagnose the abnormality of a knock sensor. <P>SOLUTION: A device A/D converts the output signals of the knock sensor 28 by an A/D conversion part 41 in a period in which an ignition timing is advanced to forcibly generate knock (or in a period in which knock is generated), takes therein the converted output signals, analyzes, with respect to time-frequency, the output signals of the knock sensor 28 by a time-frequency analysis part 42, and extracts vibration intensity varying patterns in a plurality of frequency ranges by simultaneously extracting the data on frequency, time, and vibration intensity from the output signals of the knock sensor 28. Then, the device calculates the quantity of the vibration intensity varying patterns in the frequency ranges started at the same timing. The presence of the abnormality of the knock sensor 28 is determined by confirming whether the quantity is less than a determination threshold or not. <P>COPYRIGHT: (C)2009,JPO&INPIT

Description

  The present invention relates to a knock detection system abnormality diagnosis device that determines whether or not there is an abnormality in a knock signal output means based on vibration intensities in a plurality of frequency ranges extracted from the output signal of the knock signal output means.

  When detecting an abnormality of the knock sensor, for example, as described in Patent Document 1 (Japanese Patent Laid-Open No. 2006-29158), a bias unit for pulling up and pulling down the output signal of the knock sensor is provided. There is one in which the presence or absence of an abnormality of a knock sensor is determined based on a direct current component.

Alternatively, as described in Patent Document 2 (Japanese Patent Application Laid-Open No. 2004-270626), the number of times that the output signal of the knock sensor falls below the determination value is counted by the abnormality counter for each cylinder, and the count of the abnormality counter is counted. In some cases, the knock sensor is determined to be abnormal when the value exceeds a predetermined value.
JP 2006-29158 A JP 2004-270626 A

  An object of the present invention is to provide a knock detection system abnormality diagnosis device capable of accurately determining whether or not there is an abnormality in a knock signal output means by a method different from the conventional abnormality detection technique of a knock sensor.

  In order to achieve the above object, the invention according to claim 1 is a knock signal output means for changing a waveform of an output signal according to knock vibration generated during operation of the internal combustion engine, and an output signal of the knock signal output means. Multiple frequency vibration intensity extraction means capable of extracting vibration intensity in a plurality of frequency ranges, and forcing knock by advancing the period or ignition timing during which the internal combustion engine is operating under the operating conditions in which knock occurs An abnormality diagnosing unit for determining whether or not the knock signal output unit is abnormal based on the number of vibration intensities in the frequency range that rises at the same timing extracted by the multiple frequency vibration intensity extracting unit during It is a configuration.

  When knocking occurs, not only the knocking vibration component of the fundamental frequency of knocking (the primary resonance frequency determined by the bore diameter of the cylinder, for example, around 7 kHz), but simultaneously, the knocking is also performed in a plurality of higher-order resonance frequency ranges. Vibration component appears.

  In view of such circumstances, in the present invention, a plurality of frequencies are generated during a period in which the internal combustion engine is operated under an operation condition in which knocking occurs or a period in which ignition is forcibly generated by advancing the ignition timing. If the number of vibration intensities in the frequency range that rises at the same timing extracted by the vibration intensity extraction means is large, it means that knock has been detected normally, so that the knock signal output means is determined to be normal, If the number of vibration intensities in the frequency range that rises at the same timing is small, it means that knocking cannot be detected. Therefore, it can be determined that the knock signal output means is abnormal.

  Further, as in claim 2, the number of vibration intensities in the frequency region rising at the same timing extracted by the multiple frequency vibration intensity extracting means during the period in which the vibration noise is generated or the period in which the vibration noise is forcibly generated. Based on this, the presence or absence of abnormality of the knock signal output means may be determined. Here, the vibration noise is, for example, valve seating noise or direct injection injector noise, and these vibration noises have a characteristic that vibrations rise in two frequency ranges at the same timing. Therefore, vibration noise can be detected normally if the number of vibration strengths in the frequency range that rises at the same timing during the period in which vibration noise is generated or the period in which vibration noise is forcibly generated is two. Therefore, it is determined that the knock signal output means is normal, and vibration noise cannot be detected normally unless the number of vibration intensity in the frequency range that rises at the same timing is two. Therefore, it can be determined that the knock signal output means is abnormal.

  In this case, as described in claim 3, the ignition timing may be retarded during the abnormality diagnosis period to suppress the occurrence of knock. In this way, abnormality diagnosis can be performed under the condition that only vibration noise is generated in the state where knock does not occur, and the abnormality diagnosis accuracy of the knock signal output means can be improved.

  Further, as in claim 4, based on the number of vibration intensities in the frequency range rising at the same timing extracted by the plural frequency vibration intensity extracting means during the period in which the ignition noise is generated or the period in which the ignition noise is forcibly generated. Then, the presence or absence of abnormality of the knock signal output means may be determined. Here, since the ignition noise is a noise that occurs in one frequency range when the ignition timing is delayed too much, no knock is generated at the timing when the ignition noise occurs. Therefore, if the number of vibration intensities in the frequency range that rises at the same timing is only one during the period in which ignition noise is generated or the period in which ignition noise is forcibly generated, the ignition noise can be detected normally. In other words, it is determined that the knock signal output means is normal. On the other hand, if the number of vibration intensities in the frequency range that rises at the same timing is not one, ignition noise cannot be detected normally. Therefore, it can be determined that the knock signal output means is abnormal.

  In this case, as described in claim 5, the output signal of the knock signal output means may be time-frequency analyzed by the multi-frequency vibration intensity extracting means to extract the vibration intensities in a plurality of frequency ranges. In this case, time-frequency analysis (Time-Frequency Analysis) may use short-time Fourier transform (STFT), wavelet transform, Wigner distribution, and the like. Time and vibration intensity data can be extracted simultaneously.

  Alternatively, as described in claim 6, the output signal of the knock signal output means is processed by a plurality of bandpass filters that extract a plurality of specific frequency regions, thereby extracting vibration intensities in a plurality of frequency regions. good. Even in this case, vibration intensities in a plurality of frequency ranges can be easily extracted from the output signal of the knock signal output means.

  Hereinafter, four Examples 1 to 4 embodying the best mode for carrying out the present invention will be described.

A first embodiment of the present invention will be described with reference to FIGS.
First, a schematic configuration of the entire engine control system will be described with reference to FIG.
An air cleaner 13 is provided at the most upstream portion of the intake pipe 12 of the engine 11 that is an internal combustion engine, and an air flow meter 14 that detects the intake air amount is provided downstream of the air cleaner 13. On the downstream side of the air flow meter 14, a throttle valve 15 whose opening is adjusted by the motor 10 and a throttle opening sensor 16 for detecting the throttle opening are provided.

  Further, a surge tank 17 is provided on the downstream side of the throttle valve 15, and an intake pipe pressure sensor 18 for detecting the intake pipe pressure is provided in the surge tank 17. The surge tank 17 is provided with an intake manifold 19 for introducing air into each cylinder of the engine 11, and a fuel injection valve 20 for injecting fuel into the cylinder is attached to each cylinder of the engine 11. . The cylinder head of the engine 11 is provided with a spark plug 21 for each cylinder, and a high voltage is applied to the spark plug 21 by the ignition device 25 at each cylinder ignition timing to ignite the air-fuel mixture in the cylinder. The

  Further, the engine 11 is provided with a variable intake valve timing device 31 that varies the valve timing (opening / closing timing) of the intake valve 29 and a variable exhaust valve timing device 32 that varies the valve timing of the exhaust valve 30.

  On the other hand, the exhaust pipe 22 of the engine 11 is provided with a catalyst 23 such as a three-way catalyst that purifies CO, HC, NOx, etc. in the exhaust gas. / An exhaust gas sensor 24 for detecting lean is provided. A crank angle sensor 26 that outputs a pulse signal every time the crankshaft of the engine 11 rotates by a predetermined crank angle is attached to the cylinder block of the engine 11, and the crank angle and the crank angle are determined based on the output signal of the crank angle sensor 26. The engine speed is detected.

  Further, a knock sensor 28 (knock signal output means) for detecting knock vibration is attached to the cylinder block of the engine 11, and the output signal of the knock sensor 28 is digitally processed by a knock determination circuit 33 which will be described later to make a knock determination. In addition, an abnormality diagnosis of the knock sensor 28 is performed by a method described later. The knock determination result of the knock determination circuit 33 is input to an engine control circuit (hereinafter referred to as “ECU”) 34.

  The ECU 34 is mainly composed of a microcomputer, and executes various engine control programs stored in a built-in ROM (storage medium), thereby allowing the fuel injection amount of the fuel injection valve 20 and the ignition timing of the ignition plug 21 to be executed. The valve timings of the variable valve timing devices 31 and 32 on the intake side and the exhaust side are controlled. At this time, the ignition timing is controlled by repeating the process of advancing the ignition timing when no knock is detected by the knock determination circuit 33 and retarding the ignition timing when the knock is detected. Execute knock control to control near the limit.

  As shown in FIG. 3, when knocking occurs, not only the knocking vibration component of the fundamental frequency of knocking (the primary resonance frequency determined by the bore diameter of the cylinder) but also at the same time a plurality of higher-order resonance frequency regions. The knock vibration component also appears. Therefore, in the first embodiment, paying attention to the point that the vibration starts at the same timing in a plurality of (three or more) frequency regions when knocking occurs, the vibration intensities in the plurality of frequency regions are extracted from the output signal of the knock sensor 28. Thus, the knock determination is performed based on whether or not the number of vibration intensities in the frequency range that rises at the same timing is three or more.

  Further, in the first embodiment, the output signal of the knock sensor 28 is used during the period in which the engine 11 is operated under the operation condition in which knocking occurs or in the period in which knocking is forcibly generated by advancing the ignition timing. The presence or absence of abnormality of knock sensor 28 is determined based on whether or not the number of vibration strengths in the frequency range that rises at the same timing among the extracted vibration strengths in the plurality of frequency ranges is less than a determination threshold (for example, less than 3). I am doing so.

  Hereinafter, in the first embodiment, an example in which time-frequency analysis is used as a technique for extracting vibration intensity change patterns in a plurality of frequency ranges from the output signal of the knock sensor 28 will be described. In this case, the time-frequency analysis may use short-time Fourier transform (STFT), wavelet transform, Wigner distribution, or the like.

  This time-frequency analysis processing is performed by the time-frequency analysis unit 42 (multiple frequency vibration intensity extracting means) in the knock determination circuit 33. The output signal of knock sensor 28 is converted into a digital value by A / D converter 41 and processed by time-frequency analyzer 42. Thereby, when knocking occurs, as shown in FIG. 3, temporal change patterns of vibration intensity are extracted in a plurality of frequency ranges. The frequency range from which the vibration intensity is extracted includes the fundamental frequency (the primary resonance frequency determined by the bore diameter of the cylinder), which is the lowest of the knock vibration frequencies, and a plurality of higher-order resonance frequency regions. .

  Based on the analysis result of the time-frequency analysis unit 42, the knock determination / abnormality diagnosis unit 43 (abnormality diagnosis unit) calculates the number of vibration intensities in the frequency region that rises at the same timing among the vibration intensities in a plurality of frequency regions. The presence / absence of abnormality of knock sensor 28 is determined based on whether this number is less than a determination threshold value (eg, less than 3).

  The abnormality diagnosis process of the knock sensor 28 described above is executed by the knock determination / abnormality diagnosis unit 43 according to the abnormality diagnosis routine of FIG. The abnormality diagnosis routine of FIG. 4 is executed for each ignition of each cylinder, and plays a role as abnormality diagnosis means in the claims. When this routine is started, first, in step 101, it is determined whether or not an abnormality diagnosis execution condition is satisfied. The abnormality diagnosis execution conditions include, for example, (1) that an abnormality such as an ignition system is not detected by another abnormality diagnosis function, and (2) an operation state suitable for executing the abnormality diagnosis (for example, a steady operation state) ). If this abnormality diagnosis execution condition is not satisfied, this routine is terminated without performing the subsequent processing.

  On the other hand, if the abnormality diagnosis execution condition is satisfied, the routine proceeds to step 102 where the ignition timing is advanced and knocking is forcibly generated. Or you may make it wait until it becomes the driving | running condition which knocks.

  Thereafter, the process proceeds to step 103 where the output signal of the knock sensor 28 is A / D converted and captured by the A / D converter 41. In the next step 104, time-frequency analysis (STFT, wavelet transform, Wigner distribution, etc.) Is executed, and frequency, time, and vibration intensity data are simultaneously extracted from the output signal of the knock sensor 28 to extract vibration intensities in a plurality of frequency ranges.

  Thereafter, the process proceeds to step 105, and the number of vibration intensities in the frequency region rising at the same timing is calculated. In the next step 106, the number of vibration intensities in the frequency region rising at the same timing is calculated from a determination threshold (for example, 3). If the number is smaller than the determination threshold value, it means that knocking is not detected despite the occurrence of knocking. It is determined that the knock sensor 28 is abnormal. On the other hand, if it is determined in step 106 that the number of vibration intensities in the frequency range that rises at the same timing is greater than or equal to the determination threshold value (for example, 3 or more), this means that knock has been detected normally. Therefore, the process proceeds to step 108 and it is determined that the knock sensor 28 is normal.

  In the first embodiment described above, the time-frequency analysis unit 42 performs the period in which the ignition timing is advanced and the knock is forcibly generated (or the period in which the knock is generated under the operating condition). By determining whether or not the number of vibration intensities in the frequency range that rises at the same timing extracted is less than the determination threshold value, it is possible to determine whether knock sensor 28 is abnormal.

  In the first embodiment, during the period in which the ignition timing is advanced and the knock is forcibly generated (or the period in which the knock is generated under the operation condition), the vibration intensity in the frequency range that rises at the same timing is used. Whether the knock sensor 28 is abnormal or not is determined based on whether the number is less than the determination threshold value. However, in the second embodiment of the present invention, noise is detected by executing the abnormality diagnosis routine of FIG. Knock sensor based on whether or not the number of vibration intensity in the frequency range that rises at the same timing during the period when (vibration noise or ignition noise) occurs or during the period when noise is forcibly generated is the number at the time of noise generation 28 normal / abnormal are judged.

  Here, the vibration noise is, for example, valve seating noise or direct injection injector noise, and these vibration noises have a characteristic that vibrations rise in two frequency ranges at the same timing. Therefore, vibration noise can be detected normally if the number of vibration strengths in the frequency range that rises at the same timing during the period in which vibration noise is generated or the period in which vibration noise is forcibly generated is two. Therefore, if the knock sensor 28 is determined to be normal and the number of vibration intensities in the frequency range that rises at the same timing is not two, vibration noise cannot be normally detected. This means that the knock sensor 28 can be determined to be abnormal.

  Further, since the ignition noise is noise that occurs in one frequency range when the ignition timing is retarded too much, no knock is generated at the timing when the ignition noise occurs. Therefore, if the number of vibration intensities in the frequency range that rises at the same timing is only one during the period in which ignition noise is generated or the period in which ignition noise is forcibly generated, the ignition noise can be detected normally. Therefore, it is determined that the knock sensor 28 is normal, and the ignition noise cannot be normally detected unless the number of vibration intensities in the frequency range that rises at the same timing is one. Therefore, it can be determined that the knock sensor 28 is abnormal.

  The abnormality diagnosis routine of FIG. 5 executed in the second embodiment is merely the processing of steps 102 and 106 of the abnormality diagnosis routine of FIG. 4 changed to the processing of steps 102a and 106a, and the processing of other steps is the same. It is.

  In the abnormality diagnosis routine of FIG. 5, when the abnormality diagnosis execution condition is satisfied, the process proceeds to step 102a to forcibly generate noise (vibration noise or ignition noise). Or you may make it wait until it becomes the conditions which generate | occur | produce noise (vibration noise or ignition noise).

  Thereafter, in steps 103 to 105, the number of vibration strengths in the frequency region rising at the same timing is calculated, and in the next step 106a, the number of vibration strengths in the frequency region rising at the same timing is a predetermined value (two or one). ), And if the number of vibration intensities in the frequency range is a predetermined value, it means that noise (vibration noise or ignition noise) has been detected normally. It is determined that knock sensor 28 is normal.

  On the other hand, if the number of vibration intensities in the frequency range is not a predetermined value, it means that noise (vibration noise or ignition noise) cannot be normally detected. It is determined that there are 28 abnormalities.

In the second embodiment described above, the same effect as that of the first embodiment can be obtained.
It should be noted that during the abnormality diagnosis period of the knock sensor 28, the ignition timing may be retarded to suppress the occurrence of knock. In this way, abnormality diagnosis can be performed under the condition that only vibration noise is generated in a state where knock does not occur, and the abnormality diagnosis accuracy of the knock sensor 28 can be improved.

  In the first and second embodiments, the time-frequency analysis unit 42 of the knock determination circuit 33 performs time-frequency analysis on the output signal of the knock sensor 28 to extract vibration intensities in a plurality of frequency ranges. In the third embodiment of the present invention shown in FIG. 8, the knock determination circuit 50 extracts a plurality of vibration intensities in a plurality of frequency ranges from the output signal of the knock sensor 28 converted into a digital value by the A / D converter 41. Band pass filters (first BPF to fifth BPF) 51 to 55 are provided, and the knock determination / abnormality diagnosis unit 56 (abnormality diagnosis means) at the same timing among the vibration intensities in a plurality of frequency ranges extracted by the BPFs 51 to 55. Whether knock sensor 28 is abnormal or not is determined based on whether or not the number of vibration strengths in the rising frequency region is less than a determination threshold value.

  In this case, the frequency range extracted by each of the first to fifth BPFs 51 to 55 is a primary frequency range (a fundamental frequency range of knock vibration) and a plurality of higher order frequency ranges, and the primary frequency range ( The passband of the first BPF 51 is set to a frequency range including the fundamental frequency that is the lowest of the knock vibration frequencies (the primary resonance frequency determined by the bore diameter of the cylinder, for example, around 7 kHz). Yes. In addition, a plurality of higher-order frequency bands (passbands of the second to fifth BPFs 52 to 55) are set to, for example, around 12 kHz, around 15 kHz, around 17 kHz, and around 21 kHz as shown in FIG. Has been. The number of BPFs 51 to 55 is not limited to five, and may be three, four, or six or more.

  The abnormality diagnosis process of the third embodiment is executed by the knock determination / abnormality diagnosis unit 56 of the knock determination circuit 50 according to the abnormality diagnosis routine of FIG. In the abnormality diagnosis routine of FIG. 8, only the process of step 104 of the abnormality diagnosis routine of FIG. 4 is changed to the process of step 104a, and the processes of the other steps are the same.

In the abnormality diagnosis routine of FIG. 8, as in the first embodiment, the output signal of the knock sensor 28 is converted to an A / D converter in step 103 during a period in which knocking is forcibly generated (or a period in which knocking occurs). After A / D conversion at 41, the output signal of the knock sensor 28 is processed by the first to fifth BPFs 51 to 55 in step 104a to extract vibration intensities in a plurality of frequency ranges (see FIG. 7). Thereafter, the process proceeds to step 105, where the number of vibration intensities in the frequency range rising at the same timing is calculated. In the next step 106, whether the number of vibration intensities in the frequency range rising at the same timing is less than the determination threshold value. The presence or absence of abnormality of knock sensor 28 is determined (steps 107 and 108).
Also in the third embodiment described above, the same effect as in the first embodiment can be obtained.

  In the third embodiment, the output signal of the knock sensor 28 is time-frequency analyzed to extract vibration intensities in a plurality of frequency ranges. However, in the fourth embodiment of the present invention shown in FIG. Similarly to the first to fifth BPFs 51 to 51, the output signal of the knock sensor 28 subjected to A / D conversion during a period in which noise (vibration noise or ignition noise) is forcibly generated (or a period in which noise is generated) is generated. 55, the vibration intensities in a plurality of frequency regions are extracted from the output signal of the knock sensor 28 (step 104a), and whether the number of vibration intensities in the frequency region that rises at the same timing is a predetermined value or not. It is determined whether knock sensor 28 is normal or abnormal (steps 105 to 108).

In the fourth embodiment described above, the same effect as that of the first embodiment can be obtained.
In the first to fourth embodiments, the knock sensor 28 is used as the knock signal output means that changes the waveform of the output signal according to the knock that occurs during engine operation. However, the in-cylinder pressure sensor that detects the in-cylinder pressure, An ion current detecting means for detecting ions generated by combustion of the air-fuel mixture in the cylinder through the spark plug 21 or the like may be used as the knock signal output means.

  In addition, the present invention is not limited to the in-cylinder injection engine as shown in FIG. 1, but is also applied to an intake port injection engine or a dual injection engine having fuel injection valves mounted in both the intake port and the cylinder. Needless to say, the invention can be implemented with various modifications without departing from the gist of the invention, such as being applicable to an engine in which a variable valve timing device such as a variable valve timing device is not mounted.

1 is a schematic configuration diagram showing an entire engine control system in Embodiment 1 of the present invention. It is a block diagram which shows the circuit which processes the output signal of the knock sensor of Example 1, and determines a knock. It is a figure which shows typically the time change pattern of the vibration intensity of the some frequency range extracted from the output signal of the knock sensor of Example 1. FIG. 3 is a flowchart illustrating a flow of processing of an abnormality diagnosis routine according to the first embodiment. 6 is a flowchart showing a flow of processing of an abnormality diagnosis routine of Example 2. It is a block diagram which shows the circuit which processes the output signal of the knock sensor of Example 3, and determines a knock. 10 is a time chart illustrating an example of a vibration waveform after bandpass filter (BPF) processing according to the third embodiment. 10 is a flowchart showing a flow of processing of an abnormality diagnosis routine of Example 3. 10 is a flowchart showing a flow of processing of an abnormality diagnosis routine of Example 4.

Explanation of symbols

  DESCRIPTION OF SYMBOLS 11 ... Engine (internal combustion engine), 12 ... Intake pipe, 16 ... Throttle valve, 20 ... Fuel injection valve, 21 ... Spark plug, 22 ... Exhaust pipe, 25 ... Ignition device, 28 ... Knock sensor (knock signal output means), DESCRIPTION OF SYMBOLS 29 ... Intake valve, 30 ... Exhaust valve, 31 ... Variable intake valve timing apparatus, 32 ... Variable exhaust valve timing apparatus, 33 ... Knock determination circuit, 34 ... ECU, 41 ... A / D conversion part, 42 ... Time-frequency analysis Part (multiple frequency vibration intensity extracting means), 43 ... knock determination / abnormality diagnosis part (abnormality diagnosis means), 50 ... knock determination circuit, 51-55 ... band pass filter (first BPF to fifth BPF), 56 ... knock determination / Abnormality diagnosis unit (abnormality diagnosis means)

Claims (6)

Knock signal output means for changing the waveform of the output signal in response to knock vibration generated during operation of the internal combustion engine;
Multiple frequency vibration intensity extracting means capable of extracting vibration intensity in a plurality of frequency ranges from the output signal of the knock signal output means;
A frequency range that rises at the same timing extracted by the multiple frequency vibration intensity extracting means during a period in which the internal combustion engine is operated under an operation condition in which knocking occurs or a period in which ignition timing is advanced to forcibly generate knocking A knock detection system abnormality diagnosing device, comprising: an abnormality diagnosing unit that determines whether or not the knock signal output unit has an abnormality based on the number of vibration intensities of the knock detecting system. Knock signal output means for changing the waveform of the output signal in response to knock vibration generated during operation of the internal combustion engine;
Multiple frequency vibration intensity extracting means capable of extracting vibration intensity in a plurality of frequency ranges from the output signal of the knock signal output means;
Abnormality of the knock signal output means based on the number of vibration intensities in the frequency range that rises at the same timing extracted by the multiple frequency vibration intensity extraction means during a period in which vibration noise is generated or a period in which vibration noise is forcibly generated A knock detection system abnormality diagnosis device, comprising: an abnormality diagnosis means for determining the presence or absence of an abnormality.   3. The knock detection system abnormality diagnosis apparatus according to claim 2, wherein the abnormality diagnosis means suppresses the occurrence of knock by retarding the ignition timing during the abnormality diagnosis period. Knock signal output means for changing the waveform of the output signal in response to knock vibration generated during operation of the internal combustion engine;
Multiple frequency vibration intensity extracting means capable of extracting vibration intensity in a plurality of frequency ranges from the output signal of the knock signal output means;
Abnormality of the knock signal output means based on the number of vibration intensities in the frequency range that rises at the same timing extracted by the multiple frequency vibration intensity extraction means during the period in which ignition noise is generated or the period in which ignition noise is forcibly generated A knock detection system abnormality diagnosis device, comprising: an abnormality diagnosis means for determining the presence or absence of an abnormality.   5. The knock detection according to claim 1, wherein the multiple frequency vibration intensity extracting unit extracts a vibration intensity of the plurality of frequency ranges by performing time-frequency analysis on an output signal of the knock signal output unit. System abnormality diagnosis place.   The multi-frequency vibration intensity extracting unit extracts vibration intensities in the plurality of frequency regions by processing the output signal of the knock signal output unit with a plurality of band-pass filters that extract a plurality of specific frequency regions. The knock detection system abnormality diagnosis device according to any one of claims 1 to 4.

Images