Search

KR-102962474-B1 - SYSTEM AND METHOD FOR DETECTING DAMAGE OF BEARING OF ENGINE USING KNOCK SENSOR

KR102962474B1KR 102962474 B1KR102962474 B1KR 102962474B1KR-102962474-B1

Abstract

A method for detecting bearing damage of an engine using a knocking sensor according to the present invention comprises: a data storage step (S110) in which a vibration signal output from a knocking sensor (11) installed in the engine is stored in a data storage unit (23); a frequency intensity calculation step (S120) in which a frequency intensity calculation unit (24) performs a Fast Fourier Transform (FFT) on the vibration signal input to the data storage unit (23) and calculates the intensity for each frequency; a detection frequency accumulation step (S130) in which a detection frequency selection unit (33) adds up the intensity of all selected detection frequencies to obtain an integrated detection frequency value; a noise judgment step (S140) in which a noise judgment unit (26) determines whether the exclusion frequencies selected by the exclusion frequency selection unit (34) correspond to a preset condition and determines whether the vibration signal of the knocking sensor (11) is a vibration signal unrelated to bearing damage; and the knocking sensor (11). If the vibration signal is a vibration signal caused by damage to the bearing, the method includes a counter increment step (S190) for increasing a damage counter (C_R)(C_I) if the accumulated value of the detection frequency is greater than a preset damage threshold (TH_R)(TH_I), and a damage confirmation step (S210) for confirming that the bearing is damaged if the damage counter (C_R)(C_I) is greater than a preset confirmation counter (TH_CR)(TH_CI).

Inventors

  • 한정석
  • 오창진

Assignees

  • 현대자동차주식회사
  • 기아 주식회사

Dates

Publication Date
20260508
Application Date
20210203

Claims (20)

  1. A data storage unit that stores a vibration signal output from a knock sensor installed in the engine, and A frequency intensity calculation unit that calculates the amplitude for each frequency by performing a Fast Fourier Transform (FFT) on the vibration signal input to the above data storage unit, and A detection frequency accumulation unit that accumulates the intensity of discriminative frequencies selected by the detection frequency selection unit, and A noise judgment unit that determines whether the exclusion frequency selected by the exclusion frequency selection unit corresponds to a preset condition, and determines whether the vibration signal of the knocking sensor is a vibration signal unrelated to bearing damage, and A bearing damage detection system for an engine using a knock sensor, comprising a damage determination unit that increases a damage counter and confirms that the bearing is damaged when the vibration signal of the knock sensor is a vibration signal caused by bearing damage and the accumulated value of the intensity of the detected frequencies is greater than a preset damage threshold.
  2. In paragraph 1, The above frequency intensity calculation unit is, An engine bearing damage detection system using a knocking sensor, characterized by calculating the amplitude of a vibration signal stored in the above-mentioned data storage unit at predetermined frequency intervals.
  3. In paragraph 1, The above noise judgment unit is, An engine bearing damage detection system using a knock sensor, characterized in that if the sum of the strengths of the above exclusion frequencies exceeds a preset standard, the vibration signal of the knock sensor is determined to be a vibration signal unrelated to bearing damage.
  4. In paragraph 1, The above noise judgment unit is, An engine bearing damage detection system using a knock sensor, characterized in that if the exclusion frequency with the greatest intensity among the above exclusion frequencies is within a predetermined rank among all frequencies, the vibration signal of the knock sensor is determined to be a vibration signal unrelated to bearing damage.
  5. In paragraph 1, The above noise judgment unit is, An engine bearing damage detection system using a knock sensor, characterized in that if the ratio of the exclusion frequency having the greatest intensity among the exclusion frequencies and the detection frequency exceeds a predetermined ratio, the vibration signal of the knock sensor is determined to be a vibration signal unrelated to bearing damage.
  6. In paragraph 1, The above damage assessment unit is, If the accumulated value of the above detection frequency is greater than the damage threshold, the damage counter is increased, and A bearing damage detection system for an engine using a knock sensor, characterized in that if the damage counter is greater than or equal to a confirmation counter that confirms the bearing is damaged, the bearing is confirmed to be damaged.
  7. In paragraph 6, The above damage assessment unit is, A bearing damage detection system for an engine using a knocking sensor, characterized by setting the damage threshold, the damage counter, and the confirmation counter according to the operating mode of the engine.
  8. In Paragraph 7, The above damage assessment unit is, An engine bearing damage detection system using a knock sensor, characterized by determining the damage to the bearing by dividing the case into when fuel is supplied and combustion proceeds steadily in the engine, and when fuel is cut off and combustion does not proceed steadily.
  9. In paragraph 8, The above damage assessment unit is, An engine bearing damage detection system using a knocking sensor, characterized by determining the bearing damage by dividing the case in which the engine is driven in a normal driving mode including full load, part load, and idle, and the case in which the engine is driven including entry into fuel cut, just before fuel cut, immediately after fuel cut, or tip out.
  10. In paragraph 1, A bearing damage detection system for an engine using a knock sensor, characterized by further including a limp groove control unit that limits the engine to operate at a speed below a preset safe rotational speed when the damage determination unit confirms the damage to the bearing.
  11. A data storage step in which a vibration signal output from a knock sensor installed in an engine is stored in a data storage unit, and A frequency intensity calculation unit, a frequency intensity calculation step in which a vibration signal input to the data storage unit is subjected to a Fast Fourier Transform (FFT) and intensity is calculated for each frequency, and A detection frequency accumulation step that calculates an integrated detection frequency value by adding up the intensities of all detection frequencies selected by the detection frequency selection department, and A noise judgment step in which a noise judgment unit determines whether exclusion frequencies selected by an exclusion frequency selection unit correspond to preset conditions, and determines whether the vibration signal of the knocking sensor is a vibration signal unrelated to bearing damage, and If the vibration signal of the above knocking sensor is a vibration signal caused by damage to the bearing, and the accumulated value of the detection frequency is greater than a preset damage threshold, a counter increment step for increasing the damage counter, and A method for detecting bearing damage in an engine using a knock sensor, comprising a damage confirmation step in which, if the damage counter is greater than a preset confirmation counter, the bearing is confirmed to be damaged.
  12. In Paragraph 11, The above data storage step is, A method for detecting bearing damage in an engine using a knock sensor, characterized by converting a vibration signal output from the knock sensor into a digital signal and storing it in a data storage unit while the measurement window is open.
  13. In Paragraph 11, The above frequency-specific intensity calculation step is, A method for detecting bearing damage in an engine using a knocking sensor, characterized by calculating the amplitude of a vibration signal stored in the above-mentioned data storage unit at predetermined frequency intervals.
  14. In Paragraph 11, The above noise judgment step is, A method for detecting bearing damage in an engine using a knock sensor, characterized in that if the sum of the strengths of the above exclusion frequencies exceeds a preset standard, the vibration signal of the knock sensor is determined to be a vibration signal unrelated to bearing damage.
  15. In Paragraph 11, The above noise judgment step is, A method for detecting bearing damage in an engine using a knock sensor, characterized in that if the exclusion frequency with the greatest intensity among the above exclusion frequencies is within a predetermined rank among all frequencies, the vibration signal of the knock sensor is determined to be a vibration signal unrelated to bearing damage.
  16. In Paragraph 11, The above noise judgment step is, A method for detecting bearing damage in an engine using a knock sensor, characterized in that if the ratio of the exclusion frequency having the greatest intensity among the exclusion frequencies and the detection frequency exceeds a predetermined ratio, the vibration signal of the knock sensor is determined to be a vibration signal unrelated to bearing damage.
  17. In Paragraph 11, If, in the noise judgment step above, it is determined that the vibration signal detected by the knocking sensor has the potential to damage the bearing, A method for detecting bearing damage in an engine using a knock sensor, characterized by performing a driving mode determination step to determine whether the current driving mode of the engine is a normal driving mode in which fuel is supplied and combustion occurs consistently in the engine.
  18. In Paragraph 17, Between the above driving mode determination step and the above counter increment step, An accumulated value application step for setting the accumulated value of the above detection frequency as a comparison target, and A method for detecting bearing damage in an engine using a knocking sensor, characterized by performing an accumulated value comparison step in which the accumulated value is compared with the damage threshold value.
  19. In Paragraph 18, If the above driving mode is determined to be the normal driving mode in the above driving mode determination step, In the above step of applying the accumulated value, the above detection frequency accumulated value is used as a reference value, and The above reference value is compared with a first damage threshold value set to detect damage to the bearing when driving in normal driving mode, and A method for detecting bearing damage in an engine using a knocking sensor, characterized by increasing a first damage counter when the above reference value exceeds the above first damage threshold.
  20. In Paragraph 19, The above driving mode determination step is, A method for detecting bearing damage in an engine using a knock sensor, characterized by including a state in which the engine is operated at any one of full load, part load, or idle.

Description

System and Method for Detecting Damage to Bearing of Engine Using Knock Sensor The present invention relates to a system and method for detecting bearing damage using a vibration signal input by a knocking sensor attached to an engine. More specifically, the invention relates to a system and method for detecting bearing damage in an engine using a knocking sensor, which can improve accuracy by preventing false detection through comparison of vibration signals in frequency bands other than the frequency band used to detect bearing damage. Bearings are installed at the points in the engine where the crankshaft connects to the cylinder block and connecting rod to reduce friction. The engine operates under the control of the ECU and generates the power necessary for driving the vehicle. Under normal conditions, the above engine generates noise and vibration through combustion, but if a malfunction occurs in a component constituting the engine, it generates noise and vibration different from that caused by combustion. For example, if the above bearing is damaged, it generates noise and vibration different from a normal engine. In addition, if the above bearing is damaged, it can cause a serious problem of seizure occurring in the engine. In order to detect damage to the bearing, a bandpass filter was used to monitor the vibration magnitude in a specific frequency band when the bearing is damaged, thereby detecting the damage to the bearing. When the bearing is damaged, the vibration component in the bearing damage frequency band of approximately 2 kHz to 5 kHz becomes stronger during engine operation, so the damage to the bearing was detected by monitoring the vibration signal in the said band. However, according to the prior art described above, since only vibration signals in the bearing damage frequency band are monitored, there was a problem in that the bearing was detected as damaged even when vibration signals in the entire frequency band increased together due to strong knocking, etc. That is, as shown in FIG. 3a, if the ignition angle is advanced for a certain period of time so that knocking occurs only in cylinder 2 of an engine with an undamaged bearing, it can be seen that not only the bearing damage frequency band (R1) but also the signal of the knocking frequency band (R2) monitoring the knocking of the engine increases together. In the prior art, only the signal of the bearing damage frequency band (R1) was converted using the bandpass filter (22) and monitored. The damage to the bearing was detected by comparing it with a threshold value of the vibration magnitude of the bearing damage frequency band. When an engine with a damaged bearing is operated under the same conditions as in FIG. 3a, only the signal in the bearing damage frequency band (R1) should increase, as in FIG. 3b. However, when strong knocking occurred, the signal in the knocking frequency band (R2) increased along with the bearing damage frequency band (R1), and there was a problem in that the bearing was misidentified as damaged based only on the increased vibration in the bearing damage frequency band (R1). Furthermore, when using the aforementioned bandpass filter, since peripheral frequency components as well as the center frequency are used for bearing damage detection, the bearing damage detection performance was inevitably degraded. In addition, since the natural frequency components generated upon bearing damage differ for each engine, there was a problem in that a significant amount of effort was required to select a suitable bandpass filter for each engine. FIG. 1 is a block diagram illustrating a bearing damage detection system for an engine using a knock sensor according to the present invention. FIGS. 2A and 2B are flowcharts illustrating a method for detecting bearing damage of an engine using a knock sensor according to the present invention. Figure 3a is a graph showing the intensity by frequency obtained by FFT transforming the vibration signal when strong knocking occurs. Figure 3b is a graph showing the intensity by frequency obtained by FFT transforming the vibration signal when the bearing is damaged. Figure 4 is a graph showing the intensity by frequency by FFT transforming the vibration signal of a specific cylinder. Figure 5a is a graph showing the signal strength when bearing damage is detected using a band filter method. Figure 5b is a graph showing the signal intensity when bearing damage is detected using the FFT method. The engine bearing damage detection system and method using a knock sensor according to the present invention will be described in detail below with reference to the attached drawings. A bearing damage detection system for an engine using a knocking sensor according to the present invention comprises, as illustrated in FIG. 1, a data storage unit (23) in which a vibration signal output from a knocking sensor (11) installed in an engine is stored; a frequency intensity calculation unit (2