JP7803263B2 - Abnormal noise diagnosis device - Google Patents

Description

This disclosure relates to an abnormal noise diagnosis device that diagnoses abnormal noises generated in a vehicle.

It has been known that in internal combustion engines equipped with a supercharger, abnormal noise (blowback noise) occurs when air compressed by the supercharger's compressor flows backward and passes through the compressor when the accelerator pedal is released (see, for example, Patent Document 1). In such internal combustion engines, the generation of abnormal noise can be suppressed by defining a presurge line, which is the boundary between the region where the abnormal noise occurs and the region where it does not occur, and controlling the intake air volume so that the operating point of the supercharger (intake air volume and pressure ratio) does not fall within this region.

JP 2015-165124 A

However, abnormal noises caused by turbocharger operation can occur in a non-occurrence area based on the presurge line, making it difficult to determine based on the presurge line whether abnormal noises generated in a vehicle are caused by turbocharger operation.

The primary objective of this disclosure is to accurately identify abnormal noises that occur during the operation of a turbocharger.

The abnormal noise diagnosis device disclosed herein is an abnormal noise diagnosis device that diagnoses abnormal noise generated in a vehicle equipped with an internal combustion engine equipped with a turbocharger that includes an air bypass valve. If the frequency of the abnormal noise falls within a predetermined frequency range, the operating point of the turbocharger at the time the abnormal noise occurs falls within an abnormal noise generation region corresponding to the combination of the internal combustion engine and the turbocharger, and the vehicle is not decelerating when the abnormal noise occurs, it diagnoses that the abnormal noise has occurred due to pressure pulsation in the turbocharger. If the frequency of the abnormal noise falls within the frequency range, the operating point of the turbocharger at the time the abnormal noise occurs falls within the abnormal noise generation region corresponding to the combination of the internal combustion engine and the turbocharger, and the vehicle is decelerating when the abnormal noise occurs, it diagnoses that the abnormal noise has occurred due to operation of the air bypass valve.

This abnormal noise diagnosis device makes it possible to accurately identify abnormal noises caused by turbocharger operation. Furthermore, the abnormal noise diagnosis device disclosed herein may also include a diagnosis unit constructed using machine learning to diagnose the cause of the abnormal noise based on provided information when it is determined that no abnormal noises are occurring due to turbocharger operation.

1 is a schematic configuration diagram illustrating an abnormal sound diagnosis system including an abnormal sound diagnosis device according to the present disclosure. 3 is a flowchart showing a series of processes executed by the abnormal sound diagnosis device of the present disclosure. FIG. 2 is an explanatory diagram showing an abnormal noise generation region where abnormal noise occurs due to the operation of the supercharger.

Next, we will explain the form for implementing the invention of this disclosure with reference to the drawings.

FIG. 1 is a schematic diagram showing an abnormal sound diagnosis system 1 including a server 20 as an abnormal sound diagnosis device according to the present disclosure. The abnormal sound diagnosis system 1 shown in the figure is used to diagnose the cause of abnormal sounds generated in various vehicles, including vehicle X. Vehicle X is equipped with an internal combustion engine EG including a supercharger TC that compresses intake air using exhaust gas energy and an intercooler IC that cools the air compressed by the supercharger TC. In this embodiment, the supercharger TC is a turbocharger and includes a turbine wheel Wt rotatably disposed within a turbine housing formed in the exhaust pipe EP, a compressor wheel Wc rotatably disposed within a compressor housing formed in the intake pipe IP, a turbine shaft St that integrally connects the turbine wheel Wt and the compressor wheel Wc, a wastegate valve WGV, and an air bypass valve ABV.

As shown in FIG. 1, the abnormal sound diagnosis system 1 includes a server 20 and a mobile terminal 10 capable of exchanging information with the server 20 via communication. The mobile terminal 10 is used by a worker (user of the abnormal sound diagnosis system 1) at a vehicle dealership, repair shop, or the like when responding to a user (owner) of a vehicle X or the like that is experiencing an abnormal sound, or when performing a reproduction test in which the vehicle X or the like is driven (operated) on a roadway or test bench to reproduce the abnormal sound. In this embodiment, the mobile terminal 10 is a smartphone that includes an SoC, ROM, RAM, an auxiliary storage device (flash memory) M, a touch-panel display 11, a communication module 12 capable of exchanging various information with the server 20 or an electronic control unit (ECU) (not shown) of the vehicle X or the like via wired or wireless communication, and a microphone (not shown). An abnormal sound diagnosis support application (program) is also installed on the mobile terminal 10. As shown in FIG. 1, the mobile terminal 10 includes a medical interview information acquisition unit 13, a sound acquisition unit 14, a vehicle state acquisition unit 15, a calculation processing unit 16, an extraction unit 17, and a display control unit 18, which are constructed in cooperation with an abnormal sound diagnosis support application (software) and hardware such as the SoC of the mobile terminal 10.

The medical interview information acquisition unit 13 acquires, via the display unit 11, medical interview information indicating the state of vehicle X, etc., at the time of the occurrence of an abnormal noise, provided by a user of vehicle X, etc. The medical interview information includes vehicle identification information including the vehicle model, vehicle identification number (chassis number), etc., the order, the date and time of occurrence, the frequency of occurrence, the location of the abnormal noise, the type of sound (onomatopoeia), physical quantities that change while vehicle X, etc., is traveling, the operating state of vehicle X, etc., the warm-up effect of an engine-equipped vehicle, options selected by the driver while driving vehicle X, etc., and driving environment information of vehicle X, etc., and is input by the operator or the user of vehicle X, etc. The sound acquisition unit 14 acquires time axis data of sound (sound pressure) when a reproduction test is performed by the operator. The vehicle state acquisition unit 15 acquires vehicle state data indicating the state of vehicle X, etc., in synchronization with the acquisition of the time axis data of sound by the sound acquisition unit 14 when a reproduction test is performed. The vehicle condition data includes multiple physical quantities corresponding to items of medical interview information, such as the intake air volume of the internal combustion engine EG, intake temperature, atmospheric pressure, and supercharging pressure of the supercharger TC. The calculation processing unit 16 performs analysis processing of the time axis data of the sound acquired by the sound acquisition unit 14. The extraction unit 17 narrows down the analysis results by the calculation processing unit 16 in accordance with the operator's selection, etc. The display control unit 18 controls the display unit 11.

The server 20, which serves as an abnormal sound diagnosis device, is a computer including a CPU, ROM, RAM, input/output devices, a communications module, etc., and is installed and managed, for example, by the automobile manufacturer that produces the vehicle X, etc. The server 20 is configured with an abnormal sound diagnosis unit 21 that diagnoses abnormal sounds generated in the vehicle X, etc., through cooperation between hardware such as the CPU and a pre-installed abnormal sound diagnosis application. The abnormal sound diagnosis unit 21 includes a neural network (convolutional neural network) as a diagnostic module constructed using supervised learning (machine learning) to diagnose the cause of abnormal sounds generated in the vehicle X, etc., and the part that is the source of the abnormal sound, based on the medical interview information and sound time axis data acquired by the mobile terminal 10.

When the server 20 detects the occurrence of a new abnormal noise in vehicle X or the like, it executes re-learning in the abnormal noise diagnosis unit 21, using the acquired sound time axis data for the new abnormal noise and the details of each item of the medical interview information as training data. The server 20 also includes a storage device 22 that stores an abnormal noise database containing information on multiple abnormal noises known to occur in each vehicle model. The abnormal noise database associates each of the multiple abnormal noises with information such as the sound time axis data, the cause of the abnormal noise, the part that is the source of the noise, the details of the medical interview information provided by the user or the like, and measures to resolve the abnormal noise. The server 20 also updates the abnormal noise database based on information acquired from vehicle X or the like, and information (reports) related to newly discovered abnormal noises sent from the automobile manufacturer (developer, etc.), vehicle dealership, repair shop, etc.

Next, the procedure for diagnosing an abnormal noise occurring in vehicle X using abnormal sound diagnosis system 1 will be described. When a user of vehicle X requests that an abnormal sound be resolved, a worker at a vehicle dealership or other such facility listens to the user's medical history and then performs a reproduction test to obtain the information necessary to diagnose the abnormal sound. To perform the reproduction test, the worker connects the mobile terminal 10 to the vehicle X's electronic control unit and places (fixes) the mobile terminal 10 or an external microphone connected to the mobile terminal 10 in an appropriate location in vehicle X (e.g., the passenger compartment or engine compartment). The worker also launches the abnormal sound diagnosis support application and turns on the vehicle X's start switch. Accordingly, the mobile terminal 10 obtains information such as vehicle identification number or chassis number of vehicle X from the electronic control unit. The worker then taps the recording start button displayed on the display unit 11 and drives (operates) vehicle X on a roadway or test stand to reproduce the driving conditions under which the abnormal sound occurred based on the medical history and information from the user of vehicle X.

While vehicle X is running (operating), the sound acquisition unit 14 of the mobile terminal 10 acquires time axis data of sound emitted from vehicle X at predetermined time intervals (micro time intervals) and stores the data in the auxiliary storage device M. Furthermore, the vehicle state acquisition unit 15 acquires vehicle state data specified by the operator in accordance with the medical interview information from the electronic control device of vehicle X at predetermined time intervals (micro time intervals) in synchronization with the acquisition of the sound time axis data by the sound acquisition unit 14, and stores the data in the auxiliary storage device M. When the operator taps the recording stop button displayed on the display unit 11, the acquisition of the sound time axis data and vehicle state data is completed.

After the reproduction test is completed, the calculation processing unit 16 performs a Short-Time Fourier Transform (STFT) on the sound time-axis data to obtain a spectrogram (acoustic spectrogram) showing the relationship between time, frequency, and sound pressure. The display control unit 18 then displays the spectrogram on the display unit 11. In this embodiment, the spectrogram is a color map with the horizontal axis representing time and the vertical axis representing frequency, and color-codes sound pressure levels to show the relationship between time and sound pressure level for each frequency. The operator also selects (specifies) on the display unit 11 the range of the spectrogram to be diagnosed (analyzed) by the abnormal sound diagnosis unit 21 (server 20) (hereinafter referred to as the "diagnosis range"). The operator then enters medical interview information into the input screen displayed on the display unit 11. When the operator taps the information transmission button displayed on the display unit 11, the information necessary for diagnosing the abnormal sound is transmitted from the communication module 12 of the mobile terminal 10 to the server 20. In this embodiment, the information sent from the mobile terminal 10 to the server 20 includes time axis data of the sound acquired by the sound acquisition unit 14, vehicle condition data acquired by the vehicle condition acquisition unit 15, medical interview information, and information defining the above-mentioned diagnostic range.

2 is a flowchart showing a series of processes executed by the server 20 in response to receiving information from the mobile device 10. As shown in the figure, the abnormal sound diagnosis unit 21 of the server 20 acquires sound time axis data, vehicle condition data, medical interview information, and information defining the diagnostic range from the mobile device 10 (S100). Then, it acquires the frequency ft of the abnormal sound recognized by the operator from the sound time axis data acquired in step S100 (S110). In step S110, the abnormal sound diagnosis unit 21 extracts characteristic frequencies within the diagnostic range selected by the operator according to a predetermined procedure, and sets the extracted frequency as the abnormal sound frequency ft. Furthermore, the abnormal sound diagnosis unit 21 determines whether the frequency ft of the abnormal sound acquired in step S110 is equal to or greater than a lower limit frequency f1 (e.g., 300 Hz) and equal to or less than an upper limit frequency f2 (e.g., 3 kHz) (S120). The lower limit frequency f1 and upper limit frequency f2 are determined in advance through experiments and analysis as the lower and upper limits of the frequency of abnormal noise caused by the operation of the turbocharger TC.

If the abnormal sound frequency ft is within the frequency range from the lower limit frequency f1 to the upper limit frequency f2 (S120: YES), the abnormal sound diagnosis unit 21 acquires the intake air volume, intake temperature, atmospheric pressure, and boost pressure of the internal combustion engine EG within the above diagnosis range from the vehicle condition data acquired in step S100 (S130). Furthermore, based on the vehicle condition data within the above diagnosis range acquired in step S130, the abnormal sound diagnosis unit 21 calculates the intake air volume corrected for the intake temperature and calculates the pressure ratio (boost pressure/atmospheric pressure) obtained from the atmospheric pressure and boost pressure for each acquisition timing of the vehicle condition data within the diagnosis range (S140). Furthermore, the abnormal sound diagnosis unit 21 obtains the vehicle model of vehicle X, the model of the internal combustion engine EG and the model of the supercharger TC mounted on vehicle X from a vehicle information database stored in the storage device 22 or another server based on the vehicle identification number etc. included in the medical interview information, and reads from the storage device 22 the abnormal sound generation area corresponding to the combination of vehicle model, internal combustion engine EG (model) and supercharger TC (model) (S150).

The inventors conducted extensive research into abnormal noises generated by the operation of a turbocharger TC in a vehicle X equipped with an internal combustion engine EG including the turbocharger TC, and found that abnormal noises are generated by the operation of the turbocharger TC when the operating point of the turbocharger TC, defined by the intake air volume and pressure ratio, is included in the abnormal noise generation region shown in Figure 3, which corresponds to the combination of the vehicle model, internal combustion engine EG, and turbocharger TC. Based on this, in this embodiment, the abnormal noise generation region defined by the intake air volume and pressure ratio is determined in advance through experiments and analyses for each of multiple combinations of vehicle model, internal combustion engine EG, and turbocharger TC, and is stored in the storage device 22 (abnormal noise database) of the server 20. The abnormal noise generation region can be determined by actually measuring the pressure pulsation of the turbocharger TC alone and calculating the transfer function from the pressure measurement point to the compressor end to determine the volume velocity that is the driving force behind the abnormal noise.

The abnormal sound diagnosis unit 21 then determines whether at least one of the intake air volume and pressure ratio calculated for each acquisition timing of vehicle condition data within the diagnosis range in step S140, i.e., at least one of the operating points of the turbocharger TC when the abnormal sound occurs, is within the abnormal sound occurrence region read out in step S150 (S160). If at least one of the operating points of the turbocharger TC when the abnormal sound occurs is within the abnormal sound occurrence region (S160: YES), the abnormal sound diagnosis unit 21 calculates the amount of change in the intake air volume per unit time within the diagnosis range (S170). Here, according to research by the inventor, if the vehicle X is decelerating when the operating point of the turbocharger TC is within the abnormal sound occurrence region, an abnormal sound will occur due to operation of the air bypass valve ABV of the turbocharger TC. However, if the vehicle X is not decelerating when the operating point of the turbocharger TC is within the abnormal sound occurrence region, an abnormal sound (hereinafter referred to as "presurge sound") will occur due to pressure pulsation in the turbocharger TC.

Therefore, the abnormal sound diagnosis unit 21 determines whether or not the vehicle X is decelerating when the abnormal sound occurs based on the amount of change in the intake air volume calculated in step S170 (S180). If the vehicle X is decelerating when the abnormal sound occurs (S180: YES), the abnormal sound diagnosis unit 21 diagnoses (determines) that the abnormal sound generated by the vehicle X is caused by operation of the air bypass valve ABV of the turbocharger TC (S190). Furthermore, if the vehicle X is not decelerating when the abnormal sound occurs based on the amount of change in the intake air volume calculated in step S170 (S180: NO), the abnormal sound diagnosis unit 21 diagnoses (determines) that the abnormal sound generated by the vehicle X is a presurge noise caused by pressure pulsation in the turbocharger TC (S195). On the other hand, if the frequency ft of the abnormal noise is not within the frequency range from the lower limit frequency f1 to the upper limit frequency f2 (S120: NO), and if the operating point of the turbocharger TC at the time of abnormal noise occurrence is not within the abnormal noise occurrence region read out in step S150 (S160: NO), the abnormal noise diagnosis unit 21 diagnoses the cause of the abnormal noise occurring in vehicle X using the neural network based on the information acquired in step S100 (S125).

After processing step S190, S195, or step S125, the abnormal sound diagnosis unit 21 generates a diagnosis result including the name of the abnormal sound, the part that is the source of the abnormal sound, the cause of the abnormal sound, and measures to resolve the abnormal sound read from the storage device 22 (S200). Furthermore, the abnormal sound diagnosis unit 21 transmits the diagnosis result generated in step S200 to the mobile device 10 (S210), thereby ending the series of processes in FIG. 2. When the diagnosis result from the server 20 is received by the mobile device 10, the diagnosis result is displayed on the display unit 11. This allows the operator to accurately explain the diagnosis result from the server 20 to the user of vehicle X, etc., and promptly proceed with measures to address the abnormal sound.

As described above, the server 20 as an abnormal sound diagnosis device in the abnormal sound diagnosis system 1 diagnoses that a pre-surge sound caused by pressure pulsation in the turbocharger TC has occurred (S195) if the frequency ft of the abnormal sound is within the frequency range from the lower limit frequency f1 to the upper limit frequency f2 (S120: YES), the operating point (intake air volume and pressure ratio) of the turbocharger TC at the time of the abnormal sound occurrence is within the abnormal sound occurrence region corresponding to the combination of the vehicle model, internal combustion engine EG and turbocharger TC (S160: YES), and the vehicle X is not decelerating when the abnormal sound occurs (S180: NO). Furthermore, if the frequency ft of the abnormal noise falls within the above frequency range (S120: YES), the operating point of the turbocharger TC when the abnormal noise occurs falls within the above abnormal noise occurrence region (S160: YES), and the vehicle X is decelerating when the abnormal noise occurs (S180: YES), the server 20 diagnoses that the abnormal noise is caused by operation of the air bypass valve ABV (S190). This makes it possible to accurately identify abnormal noises caused by operation of the turbocharger TC.

The server 20 also includes an abnormal sound diagnosis unit 21 that is constructed using machine learning to diagnose the cause of the abnormal sound based on the provided information when it is determined that no abnormal sound is occurring due to the operation of the turbocharger TC. This further improves the versatility of the server 20 and, in turn, the abnormal sound diagnosis system 1. The abnormal sound diagnosis support application installed on the mobile terminal 10 may also be installed on a tablet terminal or personal computer, etc., and such a tablet terminal may be used in place of the mobile terminal 10. Furthermore, some of the functions of the abnormal sound diagnosis unit 21 may be provided on the mobile terminal 10, and the abnormal sound diagnosis system 1 may be configured as a single information processing device such as a personal computer.

The disclosed invention is extremely useful for diagnosing abnormal noises that occur in vehicles equipped with internal combustion engines equipped with superchargers that include air bypass valves.

1 Abnormal sound diagnosis system, 10 Mobile terminal, 20 Server (abnormal sound diagnosis device), 21 Abnormal sound diagnosis unit, ABV Air bypass valve, EG Internal combustion engine, TC Turbocharger, X Vehicle.

Claims (1)

An abnormal noise diagnosis device for diagnosing abnormal noise generated in a vehicle equipped with an internal combustion engine equipped with a turbocharger including an air bypass valve,
an abnormal noise diagnosis device that diagnoses that the occurrence of an abnormal noise is due to pressure pulsation in the supercharger, if the frequency of the abnormal noise is within a predetermined frequency range, the operating point of the supercharger at the time the abnormal noise is generated is within an abnormal noise generation region that corresponds to the combination of the internal combustion engine and the supercharger, and the vehicle is not decelerating when the abnormal noise is generated; and that the occurrence of an abnormal noise is due to operation of the air bypass valve, if the frequency of the abnormal noise is within the frequency range, the operating point of the supercharger at the time the abnormal noise is generated is within the abnormal noise generation region that corresponds to the combination of the internal combustion engine and the supercharger, and the vehicle is decelerating when the abnormal noise is generated.